ISSN 0031-1480
VOL. 48, NO 1-2, MARCH-JUNE 2005
Medical Society of Papua New Guinea
Executive 2005
President: Mathias Sapuri Vice-President: Nicholas Mann Secretary: Sylvester Lahe Treasurer: Harry Aigeeleng Executive Member: Uma Ambihaipahar Published quarterly by the Medical Society of Papua New Guinea
Papua New Guinea Medical Journal
ISSN 0031-1480
March-June 2005, Volume 48, Number 1-2
EDITORS: PETER M. SIBA, JOHN REEDER, NAKAPI TEFUARANI
Editorial Committee
I. Kevau A. Saweri G. Mola
Editorial Assistant: Cynthea Leahy Emeritus Editor: Michael Alpers
Email: [email protected] Web page: http://www.pngimr.org.pg
Registered at GPO, Port Moresby for transmission by Post as a Qualified Publication.
Printed by PNG Printing for the Medical Society of Papua New Guinea.
Authors preparing manuscripts for publication in the Journal should consult ‘Information for Authors’ inside back cover. Papua New Guinea Medical Journal Volume 48, Number 1-2, March-June 2005
CONTENTS 50TH YEAR CELEBRATION
FOREWORD 50th year celebration P.M. Siba 1
ORIGINAL ARTICLE The first 50 years of the Papua New Guinea Medical Journal: a celebration of its Golden Jubilee M.P. Alpers 2
CLASSIC PAPERS Some epidemiological features of pig-bel T.G.C. Murrell 27
The pathogenesis of pig-bel in Papua New Guinea G. Lawrence 39
Stronglyloides species infestation in young infants of Papua New Guinea: association with generalized oedema J.D. Vince, R.W. Ashford, M.J. Gratten and J. Bana-Koiri 50
Strongyloides infection in a mid-mountain Papua New Guinea community: results of an epidemiological survey R.W. Ashford, J.D. Vince, M.J. Gratten and J. Bana-Koiri 58
Weanling diarrhoea J. Biddulph and P. Pangkatana 66
The bacteriology of acute pneumonia and meningitis in children in Papua New Guinea: assumptions, facts and technical strategies M. Gratten and J. Montgomery 73
Nutrition and morbidity: acute lower respiratory tract infections, diarrhoea and malaria D. Lehmann, P. Howard and P. Heywood 87
Small-area variations in the epidemiology of malaria in Madang Province J.A. Cattani, J.S. Moir, F.D. Gibson, M. Ginny, J. Paino, W. Davidson and M.P. Alpers 95
Hereditary ovalocytosis in Melanesians D. Amato and P.B. Booth 102
Trends in sexually transmitted disease incidence in Papua New Guinea C.K. Lombange 109
A New Guinea population in which blood pressure appears to fall as age advances I. Maddocks and L. Rovin 122
MEDLARS BIBLIOGRAPHY 127
PNG Med J 2005 Mar-Jun;48(1-2):1
FOREWORD
50th year celebration
This issue of the Papua New Guinea with people being willing to contribute papers Medical Journal is a special one as it for publication and the Editors have always celebrates the 50th year of the Journal’s encouraged authors and especially new existence. The issue has been put together authors to submit their papers. for the purpose of honouring those who have contributed and worked tirelessly to ensure The editorial staff are now preparing a that this quality Journal continues to be print-ready copy of the Journal and this will published. reduce the time it takes to produce an issue of the Journal. There are many issues in the Michael Alpers has meticulously recorded pipeline and it is hoped that we will shortly the history of the first 50 years of the Journal get back on track. (1) and together with the Editors has selected 11 articles from past issues of the Journal. Peter M. Siba Some of these articles will bring back memories of past research and Editor, Papua New Guinea Medical Journal achievements ofPapua New Guinean Papua New Guinea Institute of Medical scientists and doctors and for others, who Research might be reading them for the first time, they will be examples of why the Journal has been PO Box 60 described as a national asset (2). Goroka EHP 441 Over the years there have been problems Papua New Guinea with the printing of the Journal and the publication schedule has fallen badly behind, REFERENCES but despite this the Journal has continued to be published. Finance has also been a 1 Alpers MP. The first 50 years of the Papua New Guinea Medical Journal: a celebration of its Golden problem and the Medical Society has at times Jubilee. PNG Med J 2005;48:2-26. struggled to find the funds to pay for its 2 Naraqi S. Papua New Guinea Medical Journal – a publication. There has never been a problem national asset. PNG Med J 1977;20:156-157.
1 PNG Med J 2005 Mar-Jun;48(1-2):2-26
The first 50 years of the Papua New Guinea Medical Journal: a celebration of its Golden Jubilee
MICHAEL P. A LPERS1
Centre for International Health, Curtin University of Technology, Perth, Australia
The Papua New Guinea Medical Journal It began that way, with 4 issues in the first was founded in May 1955. In a Foreword to year and the publication of Volume 2 No 1 in the first issue the Founding Editor, John T. May 1956. However, after that its publication Gunther, who was Director of Public Health, record became spasmodic: in 1957, 1961, outlined the purpose of the Journal and its 1962 and 1964 there was only one issue and philosophy. It was initially an instrument of there was none in 1958 and 1963. the Department of Public Health and, after Fortunately, once Ian Maddocks took charge the first 2 issues, the editorial responsibility as Editor in March 1965 the quarterly for it devolved on Eric J. Wright, who was publication schedule very quickly became Assistant Director (Medical Training) in the firmly established and persisted, with minor Department. The Medical Society of Papua exceptions, until 1999; since then the Journal New Guinea was established on 28 July 1964 has been reduced, as a survival measure, and very quickly, in March 1965, took over to two double issues a year (Table 1). responsibility for the Journal, though publication was continued by the Department The first issue included an editorial, of Public Health. This final link as an original articles, clinical notes, reviews of ‘instrument of the Department’ came to an patrol reports and other sections such as end in June 1974, when the Government literature reviews and an obituary. The first printed its last issue of the Journal. The article was a paper, which has often been September 1974 issue came out under the quoted, by J. T. Gunther on the epidemiology imprint of Kristen Pres and was paid for by of malaria in PNG (1). the Society. The Journal had achieved independence (a year before the nation), and In the second issue of the Journal one with it came the beginning of its financial editorial was about the Journal itself and uncertainties. expanded on the point made in the first issue that the Journal should be for all members However, other parts of John Gunther’s of the health professions in PNG, at all levels. purpose and philosophy have persisted to the This intention was carried out for a while but, present day. The Journal was to be “devoted as the Journal became more scientific and to the disease pattern and other problems more successful, the interest for nurses, peculiar” to Papua New Guinea (PNG) – then health extension officers and junior doctors called the Territory of Papua and New diminished. For this reason, a series of Guinea. The Journal was intended “to keep Clinical Practice articles was initiated in all our varied grades of medical personnel December 1985 and over the next 20 years abreast of world progress in the prevention 47 such articles were published (Table 2). and treatment” of the diseases found in PNG, “to promote medical research locally into The next commentary on the Journal these diseases” and “to facilitate an itself was in Ian Maddocks’ editorial (2) in exchange of ideas between…geographically the first issue that was ‘edited and prepared’ isolated” members of the health professions. by the Medical Society of Papua New Guinea. Firstly, he announced the formation The Journal was to be published quarterly. of the Society and stated its objectives:
1 Curtin University of Technology, Centre for International Health, Shenton Park Campus, GPO Box U1987, Perth, Western Australia 6845, Australia Formerly Director of the Papua New Guinea Institute of Medical Research, PO Box 60, Goroka, EHP 441, Papua New Guinea
2 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
(a) To promote the advance of medical he reviewed the role of the Journal (3). He knowledge, and to assist in the planned to extend the scope of the Journal, dissemination of such knowledge in the same search for a Papua New amongst its members; Guinean character as had engaged Ian Maddocks – but with a different outcome. His idea was to present medical reviews of (b) To promote such ethical rules, conduct both research projects and medical problems and practice as will best maintain the related to PNG. In addition, he felt strongly honour of the profession; and that visiting research workers should be obliged to contribute in some way to the local (c) To represent the corporate interest of all journal in order to try and bridge the medical practitioners in the country communication gap between research [Territory]. workers and health care practitioners. In a letter published a few issues later John Rooney (4) suggested that the Journal In accord with these objectives the should cater more for the needs of Papua Medical Society undertook to prepare and New Guinean doctors by adopting a less edit the PNG Medical Journal. The Editor formal and forbidding style. How to cater expressed the hope of increasing for different needs is a dilemma faced by all involvement of Papua New Guinean doctors editors and articulated by many of them. in the research and administrative problems Sirus Naraqi, when editing his first issue, of their country and looked for evidence of described the Journal as a national asset (5). this in the pages of the Journal. In this issue He pointed out that the Journal was indexed Papua New Guineans appear for the first by Index Medicus and Current Contents and time as authors, and there was a photograph he intended to maintain the high standard of of the first three graduates of the Papuan the Journal set by previous editors. He Medical College, Dr Jeffrey Tuvi, Dr Amelia encouraged all health workers and Homba and Dr Ilomo Batton. researchers to publish the ‘significant results of their clinical or research work’ in the Journal. The Society has continued to flourish. In 1965 it held its first symposium in Goroka, on the theme of pigbel. The Medical The editor of any journal is critical to its Symposium has been held every year since success. In celebrating the success of the then: the Society can boast that it has never Papua New Guinea Medical Journal we must missed holding its Annual Symposium, a celebrate all of its editors: they are set out in significant achievement indeed. The themes Table 3. Their close colleagues and from the past are listed in the Program and assistants have also been important: in Table Abstracts of each symposium and are not 4 are listed the Assistant Editors, in Table 5 forgotten. These annual meetings have the Business Managers and in Table 6 always been successful and have perhaps the most important of all, the maintained a very high standard, Editorial Assistants, who keep the gears of scientifically, medically and socially. the Journal in constant motion. Although the Journal comes out as a quarterly event – and when there are printing problems we all The Journal has also flourished since the wonder whether it will come out at all – the Society took it over. For most of this period toil of reaching these occasional outputs is there were, as has been noted, four issues daily and unremitting. Manuscripts have to a year though recently, because of financial be accessioned and sent for review, and logistic problems with its printing, the reviewers and authors have to be chased Journal has been reduced to two double up, manuscripts have to be edited into final issues annually. We hope that the newly shape to comply with the House Rules of acquired desktop publishing capability in the the Journal, proofs have to be checked and editorial office of the Journal will make future authors’ final – often impossible – changes printing both efficient and affordable. assessed and, since the Journal is a family affair, accommodated wherever possible. How much work goes on in the back room In the first issue edited by D. Graeme can only really be appreciated by those who Woodfield, when he was still Acting Editor, toil there, but with all the display of the
3 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
1ELBAT
CATALOGUE OF THE PAPUA NEW GUINEA MEDICAL JOURNAL
Vremulo Neebmu taD
1416,3,2, 591yraurbeF;5591rebmevoN,tsuguA,yaM
2217, 591yraurbeF,6591yaM
3319,2, ,yraurbeF 591rebmeceD,tsuguA
4210, 691yluJ,lirpA
51 691hcraM 1
61 691yluJ 2
71 691rebmeceD 4
8315,2, 691rebmeceD,tsuguA,hcraM
9416,3,2, ,rebotcO,yluJ,lirpA 691rebmeceD
140 17,3,2, 691rebmeceD,rebmetpeS,enuJ,hcraM
141 18,3,2, 691rebmeceD,rebotcO,yluJ,hcraM
142 19,2, ,3 691rebmeceD,rebmetpeS,enuJ,hcraM
143 10,3,2, 791rebmeceD,rebmetpeS,yluJ,hcraM
144 11,3,2, tpeS,enuJ,hcraM 791rebmeceD,rebme
145 12,3,2, 791rebmeceD,rebmetpeS,enuJ,hcraM
1-6 4,3,2,1 )1oN,71loVhtiwdenibmoc4oN()4791hcraM(;3791rebmetpeS,enuJ,hcraM
147 14,3,2, 791rebmeceD,rebmetpeS,enuJ,hcraM
148 15,3,2, craM 791rebmeceD,rebmetpeS,enuJ,h
149 16,3,2, 791rebmeceD,rebmetpeS,enuJ,hcraM
240 17,3,2, 791rebmeceD,rebmetpeS,enuJ,hcraM
241 18,3,2, eD,rebmetpeS,enuJ,hcraM 791rebmec
242 19,3,2, 791rebmeceD,rebmetpeS,enuJ,hcraM
243 10,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
244 11,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
245 12,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
246 13-3,2, -rebmetpeS,enuJ,hcraM 891rebmeceD
247 14-3,2, 891rebmeceD-rebmetpeS,enuJ,hcraM
248 15,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
4 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
249 16,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
340 17,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
341 18,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
342 19,3,2, 891rebmeceD,rebmetpeS,enuJ,hcraM
343 10,3,2, 991rebmeceD,rebmetpeS.enuJ,hcraM
344 11,3,2, 991rebmeceD,rebmetpeS,enuJ,hcraM
345 12,3,2, 991rebmeceD,rebmetpeS,enuJ,hcraM
346 13,3,2, 991rebmeceD,rebmetpeS,enuJ,hcraM
347 14,3,2, 991rebmeceD,rebmetpeS,enuJ,hcraM
348 15,3,2, 991rebmeceD,rebmetpeS,enuJ,hcraM
349 16,2, ,3 991rebmeceD,rebmetpeS,enuJ,hcraM
440 17-3,2, 991rebmeceD-rebmetpeS,enuJ,hcraM
441 18-3,2, 991rebmeceD-rebmetpeS,enuJ,hcraM
442 19-3,2- 991rebmeceD-rebmetpeS,enuJ-hcraM
443 10-3,2- 002rebmeceD-rebmetpeS,enuJ-hcraM
444 11- -3,2 002rebmeceD-rebmetpeS,enuJ-hcraM
445 12-3,2- 002rebmeceD-rebmetpeS,enuJ-hcraM
446 13-3,2- eS,enuJ-hcraM 002rebmeceD-rebmetp
447 14-3,2- 002rebmeceD-rebmetpeS,enuJ-hcraM
Journal in this celebratory issue we must of having focus issues. The concept was take a moment to reflect on the work that introduced by D. Graeme Woodfield, who makes it possible. edited many focus issues, but Ian Maddocks had produced one on pigbel six years before and, subsequently, another on health in the Talking about display, we think naturally village, true focus issues before the name about the cover of the Journal, the dress was created. The first guest editor for a focus which it shows to the world. In all, in its first issue was Richard W. Hornabrook, who 50 years, the Journal has had 7 different edited one on neurology, the last issue put kinds of cover. A cover change was usually out during D. G. Woodfield’s term as editor. associated with a new editor, clearly hoping After that time, with 3 exceptions, all focus to stamp his character on the Journal from issues have had a guest editor. The titles, the beginning. The current cover, with its numbers and editors of the 40 focus issues annual colour variation in a three-year cycle, are set out in Table 8. has survived 8 changes of editor since it was first introduced by Euan Scrimgeour. It has established itself as a familiar face, having Finding papers on particular topics or by been used now for 76 issues, almost half particular authors in the Journal in recent the total number in the Journal’s history issues is easy for those with access to (Table 7). Representative covers are Medlars and PubMed. Furthermore, for illustrated in Figure 1, unfortunately not in those searching their hard copies the Journal colour. staff have since 1982 produced a series of three-year indexes (coinciding with the same colour change on the Journal’s cover) that One of the special features of the Papua have been published in the Journal itself, in New Guinea Medical Journal is the practice the following December issue (always in a
5 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
2ELBAT
CLINICAL PRACTICE ARTICLES IN THE JOURNAL,2002-5891
Aerohtu Telti cnerefeRlanruoJ
ByJhpluddi parehtnoitardyherlarO 903-303:)4(82;ceD5891
EeJVtterev -tsoP gahrromeahmutrap 011-901:)1(92;raM6891
CnAKJyzel eelpsderutpurehtfotnemeganamtnerruC 191-981:)2(92;nuJ6891
,JGnamedniL lacitcarp:stludanisitigninemsuolucrebuT 372-962:)3(92;peS6891 SiqaraN sisongaidehtnostnemmoc
,JnosugreF lacitcarp:stludanisitigninemsuolucrebuT 943-543:)4(92;ceD6891 qaraN Si sisongorpehtnostnemmoc
,AlesaelB lacitcarp:stludanisitigninemsuolucrebuT 07-36:)1(03;raM7891 SiqaraN ctnemmo nemtaertehtnost
EsKsdrawd oohdlihcfosisongaidehT isolucrebutd 871-961:)2(03;nuJ7891
AyWInekti sorpelnisnoitcaerfotnemeganamehT 642-932:)3(03;peS7891
CaAKJyzel nluderutcarfehtfotnemtaertrelpmiS 613-513:)4(03;ceD7891
RdBtseW,Jsnehci T3robaldnalacinilceh iohpytfosisongaidyrota 7-96:)1(13;raM8891
WnPtta erdlihcniaimeanA 841-741:)2(13;nuJ8891
,APeralG rephtiwstneitapothcaorppaedisdebA larehpi 012-702:)3(13;peS8891 SiqaraN amedeo
MsGalo i'noitroba'nanehw:gnideelbeniretulanoitcnufsyD 792-592:)4(13;ceD8891 noitrobanaton
LDehcseW sitamohcartaidymalhC cituepareht:snoitcefni 07-56:)1(23;raM9891 snoitacilpmi
EdKsdrawd oohdlihcfotnemeganamdnasisongaidehT 051-341:)2(23;nuJ9891 noitirtunrednu
AnYFnosnhoJhatt isredrosidcirtaihcysp-oruendecudni-gurD 422-912:)3(23;peS9891 weivrevona–ecitcarplacidem
CeAKJyzel russerplainarcartnidesiaR 092-782:)4(23;ceD9891
,JBnosduH M5aeniuGweNaupaPniswayfotnemegana 6-95:)1(33;raM0991 ,LJesuohkcaB GEekrahcsT
BaJMnair imeacitpeslatanoeN 661-161:)2(33;nuJ0991
OsSGioreb sdnaaisehtseanA ispesniyregru 652-352:)3(33;peS0991
,KIyeliaB,IxocliW esaesidyranorocfotnemssessalacinilC 513-103:)4(33;ceD0991 HIuaveK
SsNSahni M8nrublamrehtfotnemegana 7-57:)1(43;raM1991
RsJsnehci gaidehT mrofnommocfotnemeganamdnasison 451-941:)2(43;nuJ1991 aeniuGweNaupaPnistludanisitirhtrafo
RsBApagno noitcefniyrotaripseretucafotnemeganamlacinilC 422-022:)3(43;peS1991 aeniuGweNaupaPniselsaemdna
,ACoifulK yrevileddnanoitatneserphceerB 592-982:)4(43;ceD1991 BAaomA
6 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PeAGsnosra T0yedereh 7-76:)1(53;raM2991
VnDJecni erdlihcnisnoisluvnoC 151-441:)2(53;nuJ2991
,JAtteverT ocinehpmarolhctakoolA?rennisrotniaS l 612-012:)3(53;peS2991 SiqaraN
,AanohiniroK lacitcarpa:stludaniesodrevoeniuqorolhC 813-113:)4(53;ceD2991 ,FInosneruaL tnemeganamothcaorppa SiqaraN
,AatpuGneS A0noisnetrepyhothcaorppacituepareht 7-36:)1(63;raM3991 HIuaveK
WkKADsretta cohscimealovopyhfotnemeganamylraeehT 552-942:)3(63;peS3991
ACoifulK , ycnangerpniniaplanomodbA 253-243:)4(63;ceD3991 ,BAaomA OuaegaR
JeretsiShpeso tucafonoitatneserpdeyaledfotnemeganamehT 35-05:)1(73;raM4991 siticidneppa
SfAnotae otnemeganamehtnisdioretsocitrocfoelorehT 031-521:)2(73;nuJ4991 aeniuGweNaupaPnisnoitcefni
CaAKJyzel isehtseanalacolrednuriaperainrehlaniugnI 191-981:)3(73;peS4991
OeSGioreb htfoeloreht:tnemeganamniapcinorhC 032-622:)4(73;ceD4991 tsitehtseana
,KADsrettaW stneitapcitebaidfotnemeganamevitarepoirepehT 732-132:)4(73;ceD4991 SGiorebO yregrusgniogrednu
GyVkaplo ltneuqerfa–msyruenaeslafcitamuart-tsoP 16-75:)1(83;raM5991 seirujnignitartenepninoitidnocdessim
,ACoifulK ,sesuac:egahrromeahmutraptsopyramirP 941-331:)2(83;nuJ5991 ,BAaomA dnanoitneverp,srotcafksirlacigoloitea GagiwiraK tnemeganam
M,LDGalo sllaftipdnasmelborp,euqinhcet:ymotoisyhpmyS 832-132:)3(83;peS5991 mehtdiovaotwohdna
WdKADsretta eenaeohrraidhtiwtneitapehtseodnehW 833-233:)4(83;ceD5991 ?yregrus
SnARnotae isitigninemlaccocotpyrcfotnemeganamehT 37-76:)1(93;raM6991 aeniuGweNaupaP
RsNHAstrebo eirujnidnahfotnemtaertehT 241-531:)2(93;nuJ6991
,IuaveK tfahslaromeffotnemeganamevitavresnoC 151-341:)2(93;nuJ6991 KADsrettaW serutcarf
VaDJecni rdlihcniamhtsafotnemeganamehT upaPnine 733-923:)4(93;ceD6991 aeniuGweN
WwKADsretta eNaupaPniseirujnidaeherevesgniganaM 56-36:)2-1(44;nuJ-raM1002 aeniuG
JeSAnoskca tomernitneitapamuartehtfoeracycnegremE -222:)4-3(54;ceD-peS2002 aeniuGweNaupaPfosnoiger 232
7 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
3ELBAT
EDITORS OF THE JOURNAL IN ITS FIRST 05 YEARS, WITH THEIR FIRST ISSUE, THE NUMBER OF ISSUES THEY EDITED AND THEIR AFFILIATION
Enrotid Aeoitailiff Frussitsri ebmuN seussifo detide
,rehtnuG.TnhoJ cilbuPfotnemtrapeD,rotceriD V25591yaM,1oN1lo rotidEgnidnuoF htlaeH
ElthgirW.Jcir acideM(rotceriDtnatsissA 5591voN,3oN1loV 21 cilbuPfotnemtrapeD,)gniniarT oN4loVnidemantsrif( htlaeH )0691luJ,2
IrskcoddaMna etal(enicideMnirerutceL V5*5691raM,1oN8lo 2 naupaP,)naeDdnarosseforP ytlucaFretal(egelloClacideM )enicideMfo
emearG.D doolBssorCdeR,rotceriD **1791peS,3oN41loV 71 dleifdooW ecivreSnoisufsnarT )rotidEgnitcA( 2791raM,1oN51loV † )rotidE(
DyotamAkcinimo bseroMtroP,tsigolotameaH V76791raM,1oN91lo latipsoHlareneG
SeiqaraNsuri nicideMnirerutceLroineS 7791ceD,4oN02loV † 8 foytisrevinU,)rosseforPretal( aeniuGweNaupaP
J,yzelC)neK(.A.K. naeDdnayregruSforosseforP V29791ceD,4oN22lo GNPU,enicideMfoytlucaF
.MnauE ,enicideMnirerutuceLroineS 0891nuJ,2oN32loV ¶ 11 ruoegmircS GNPU,enicideMfoytlucaF
,eiruoLnhoJ ,ygoloiBnamuHforosseforP V23891raM,1oN62lo dnauaveKisI enicideMnirerutceLroineS letaPdemahoM nirerutceLdna)rosseforPretal( ,enicideMfoytlucaF,enicideM GNPU
J,eiruoLnho ygoloiBnamuHforosseforP ceD-peS,4-3oN62loV 2 GNPU,enicideMfoytlucaF 3891
MasreplAleahci eniuGweNaupaP,rotceriD V84891nuJ,2oN72lo 5 hcraeseRlacideMfoetutitsnI
enogM.SselrahC tnatsissAdnarotceriDytupeD nuJ-raM,2-1oN34loV 3 abiS.MretePdna RMIGNP,rotceriD 0002
CRenogM.Sselrah DcMIGNP,rotceriDytupe eD-peS,4-3oN44loV 1 1002
enogM.SselrahC tnatsissAdnarotceriDytupeD nuJ-raM,2-1oN54loV 2 abiS.MretePdna RMIGNP,rotceriD 2002
8 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
dnaabiS.MreteP dnaRMIGNP,rotceriDytupeD nuJ-raM,2-1oN64loV 2 inaraufeTipakaN ,scirtaideaPnirerutuceLroineS 3002 htlaeHdnaenicideMfoloohcS GNPU,secneicS
,abiS.MreteP ,rotceriDdnarotceriDytupeD nuJ-raM,2-1oN74loV 2 dnaredeeRnhoJ etaicossAdnaRMIGNP 4002 inaraufeTipakaN ,scirtaideaPforosseforP htlaeHdnaenicideMfoloohcS GNPU,secneicS
* 5691guA,2oN8loV,eussitneuqesbusehthtiw,niaga,dnaegnahcrevocahtiwdetaicossA ** lacideMaeniuGweNaupaPotlanruoJlacideMaeniuGweNdnaaupaPmorfegnahcemanehthtiwdetaicossA lanruoJ † egnahcrevocahtiwdetaicossA ¶ 2891raM,1oN52loVeussiehthtiwnagebegnahcrevoC 651sisraey05tsrifehtniseussiforebmunlatotehT year with a blue cover). Earlier, indexes were screen identified over 40 papers that merited produced for the Journal and published in inclusion but the choice had to be made and, separate booklets. Of these the Journal in doing so, a balance was maintained office is missing a copy of the one for 1972, between subject matter, geographical and there are 7 years, 1975-1981, for which location and publication date. no index has been prepared (Table 9). Here is a task for a retired librarian with time on their hands! Further, if this were done, we The special diseases of Papua New could compile all the indexes into one and Guinea are kuru, pigbel and swollen belly put the key to the Journal’s 50 years on our syndrome. Of cosmopolitan diseases Papua website. New Guinea has contributed significantly to the control and understanding of neonatal tetanus, infant malnutrition and diarrhoeal Selecting a First Eleven of the Journal’s disease, pneumonia, malaria, endemic classic papers cretinism, filariasis and leprosy. There have been many fine research investigators who have lived and worked in Papua New There have been many excellent papers Guinea: it would be good if we could published in the Journal since its inception acknowledge them all. Furthermore, it would 50 years ago. The Medical Society of Papua be a shame if some of the good papers New Guinea has decided to reproduce a written by Papua New Guineans were not Classic Eleven in this celebratory Jubilee included. There are papers too that are issue. Choosing this short list has not been striking, unusual and memorable for their an easy matter. A few simple rules helped style and panache and we would all be to narrow the choice – and by necessity reluctant to see our particular favourites left eliminated some well-deserving papers. To out. In one issue of the Journal we cannot be ‘classic’ we decided that a paper must be satisfy all these desirables: compromises at least 10 years old – so the field was necessarily have had to be made. If anyone restricted to papers published before 1996. is not happy with the selection, I encourage To be eligible for reproduction a paper should them to go back and read or re-read all the not be too long – this automatically excluded papers in the Journal since May 1955 – a few great papers (6-9). Also if we are to which is an entirely pleasurable undertaking include papers that are substantial, they – and make their own selection. should not be too short. Ideally the papers should reflect the theme of the Annual Symposium of the Medical Society being One problem arising from the important held in Goroka this year: Medical Research contributions made by Papua New Guinea – a tool for health care delivery in the new to the world of tropical medicine is that most millennium. Some of those selected do not of the key papers reporting these discoveries fit the theme but have particular merit and have been published in the international special interest to warrant a place in the literature. In some fields, such as neonatal Classic Eleven in their own right. The initial tetanus, this leaves nothing for the Journal;
9 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 in others, such as malaria, there is still great (Table 8). With good vaccination coverage wealth here – in fact, we could easily make in the highlands pigbel disappeared. Then a good Classic Eleven out of malaria alone. the vaccine ran out. By the time new stocks You are encouraged to tap into the wider of the vaccine had been secured it looked literature on medical research in Papua New as if pigbel, to our great relief, had in most Guinea in a manuscript, as yet unpublished, places stayed low and had not sprung back that is available electronically through the (32). Its incidence needs now to be carefully Papua New Guinea Institute of Medical monitored. Research (PNGIMR) (M.P. Alpers, Medical research in Papua New Guinea: a review). Swollen belly syndrome was discovered in Papua New Guinea. Its cause proved to Kuru secures 11 explicit references in the be an unusual intestinal nematode first Journal though it is also mentioned or described elsewhere in PNG 6 years earlier discussed elsewhere. One reference has (33). This enabled the disease to be treated been cited already (7). The others are listed so it has now effectively disappeared – here (10-19) together with 3 on the before we could fully investigate the life cycle geographically related neurological condition of the organism or the pathogenesis of the of kogaisantamba (20-22). There has not disease. The causative organism was been a focus issue on kuru, though the one named Strongyloides fuelleborni kellyi after on neurology (Table 8) includes 2 papers on its discoverer, Alan Kelly. The disease kuru (17,18) and 1 on Creutzfeldt-Jakob investigation was conducted by John Vince, disease in PNG (23). Alpers (24) and Richard Ashford, Michael Gratten and Joe Lindenbaum (25) provide more recent, yet Bana Koiri (34,35). Later, Jennifer Shield still classical, updates on various aspects of (36) and Guy Barnish (37) and their the kuru story. Kuru has now all but colleagues studied infection with this parasite disappeared (M.P. Alpers, The epidemiology in various communities in the fringe of kuru in the period 1987 to 1995, submitted highlands. manuscript), which is a blessing for us all, but globally it has become more significant than ever with the advent of bovine In the September 1977 issue of the spongiform encephalopathy and its human Journal (Vol 20 No 3, p 150) is the following form, variant Creutzfeldt-Jakob disease. news item: “Baby feeding bottle bill passed in Parliament early this year, will be gazetted this month.” The passing of this bill, which There have been many good papers made baby feeding bottles available only on published on pigbel in the Journal, including prescription, was a triumph for Professor two focus issues devoted to the subject John Biddulph and is another of the medical (Table 8). The early workers had first to achievements of PNG. The justification for discover the disease: by identifying the legislation came from the work of John necrotizing enteritis out of all the other Biddulph and his colleagues on weanling conditions subjected to abdominal surgery. diarrhoea (38) and the advantages of They then had to define it, and name it, and breastfeeding (39) in the face of increasing establish its cause. This was all done most use of artificial feeding in urban areas (40). efficiently by Timothy Murrell, Lajos Roth, Six years later John Biddulph reviewed the John Egerton and Peter Walker between outcome of the Baby Feed Supplies (Control) 1961 and 1964 (26-28). These ideas were Act of 1977; he concluded that it had been a not accepted by the medical establishment, success but warned that maintenance of that as can be seen in the first focus issue: this success would require constant vigilance story is recounted by Peter Walker in his (41). John Biddulph’s contributions to child obituary of Tim Murrell (29). Tim’s work was health, in Papua New Guinea and extended and justified by Gregor Lawrence, internationally, were so outstanding that, who clarified the pathogenesis of pigbel (30) after his death, a memorial issue of the and conducted a successful trial of a toxoid Journal was devoted to his achievements vaccine against the beta toxin of Clostridium (Table 8). perfringens type C to prevent the disease in highland children (31). The second focus issue on pigbel was edited by Greg Research on pneumonia in PNG began Lawrence, Tim Murrell and Peter Walker with Robert Douglas and Ian Riley (42). The
10 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
4ELBAT
ASSISTANT EDITORS OF THE JOURNAL IN ITS FIRST 05 YEARS
M)ecirPleahci latipsoHlareneGybseroMtroP,naicisyhP( 1791ceD,4oN41loVdna1791peS,3oN41loV )seussi2(
htimSdivaD aeniuGweNaupaP,snoitcefnIyrotaripseRetucA,wolleFhcraeseR( )hcraeseRlacideMfoetutitsnI 4891ceD-peS,4-3oN72loVdna4891enuJ,2oN72loV )seussi2(
K)aitahBpeedlu RMIGNP,tsiciteneGnamuH( ;5891nuJ,2oN82loVdna5891raM,1oN82loV ;9891peS,3oN23loVot7891raM,1oN03loV 0991peS,3oN33loVot0991raM,1oN33loV )seussi61(
C)snikneJlora RMIGNP,tsigoloporhtnAlacideM( ;6891ceD,4oN92loVot5891peS,3oN82loV 9891ceD,4oN23loV )seussi7(
J)snehciRnho latipsoHesaBakoroG,naicisyhP( 0991peS,3oN33loVot8891peS,3oN13loV )seussi9(
M)sreplAleahci ytisrevinUnitruC,htlaeHlanoitanretnIforosseforP( )rotidEsutiremE( 4002ceD-peS,4-3oN74loVot0002nuJ-raM,2-1oN34loV )seussi01(
5ELBAT
BUSINESS MANAGERS OF THE JOURNAL IN ITS FIRST 05 YEARS
yelgiWnatS )seussi4(2791ceD,4oN51loVot2791raM,1oN51loV
dwoDdivaD )seussi5(7791raM,1oN02loVot6791raM,1oN91loV
nedyaH.P )seussi51(0891ceD,4oN32loVot7791nuJ,2oN02loV
gnooF.I.J )seussi6(2891nuJ,2oN52loVot1891raM,1oN42loV
retxaBnasuS )seussi8(4891ceD-peS,4-3oN72loVot2891peS,3oN52loV
setiahWboR )seussi21(7891ceD,4oN03loVot5891raM,1oN82loV
litaP.GittirviN )seussi21(0991ceD,4oN33loVot8891raM,1oN13loV
apuaKdranoeL )seussi71(5991raM,1oN83loVot1991raM,1oN43loV
11 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
6ELBAT
EDITORIAL ASSISTANTS OF THE JOURNAL IN ITS FIRST 05 YEARS
reilliHecnereT )seussi4(8791nuJ,2oN12loVot7791peS,3oN02loV
J;nosmalliWtena 8791ceD,4oN12loV )seussi4(9791ceD,4oN22loVot9791nuJ,2oN22loV
mahneBynneJ )seussi8(6891ceD,4oN92loVot5891raM,1oN82loV
J;sggirBeilu ,2oN23loVot7891raM,1oN03loV 9891nuJ )seussi62(4991ceD,4oN73loVot1991raM,1oN43loV
tnayrBennaoJ )seussi2(7891nuJ,2oN03loVdna7891raM,1oN03loV
nosraePtaP )seussi2(7891nuJ,2oN03loVdna7891raM,1oN03loV
rogergcaMhtepslE )seussi2(7891ceD,4oN03loVdna7891peS,3oN03loV
yeraCneelhtaK )seussi2(8891ceD,4oN13loVdna8891peS,3oN13loV
nairBesiuoL )seussi7(0991ceD,4oN33loVot9891nuJ,2oN23loV
thginKneraK )seussi8(1991raM,1oN43loVot9891nuJ,2oN23loV
inaWbocaJ )seussi5(3991ceD,4oN63loVot2991ceD,4oN53loV
yhaeLaehtnyC )seussi03(4002ceD-peS,4-3oN74loVot4991raM,1oN73loV
tamoPseirroN )seussi81(1002nuJ-raM,2-1oN44loVot5991nuJ,2oN83loV
E)sewoLannal J-raM,2-1oN54loV eussi1(2002nu pneumococcus (Streptococcus bacteriology of pneumonia and meningitis pneumoniae) was the predominant in children in PNG for a focus issue of the causative organism. To the surprise (and, Journal (49). Epidemiological studies have characteristically, disbelief) of those working been conducted in Tari since 1970 and have in the first world, isolates of pneumococcus formed the basis for many research in PNG frequently showed intermediate investigations and interventions, in particular resistance to penicillin – 12% in David on acute lower respiratory tract infections Hansman’s series (43). The other surprising (ALRI) (50). We should note that ALRI, in result, which has occasioned similar levels the PNG context, is equivalent to of disbelief, is the protection attained against pneumonia. In 2002 a focus issue of the death from pneumonia in young children by Journal was devoted to the wide range of immunization with the pneumococcal research conducted in Tari (Table 8). How polysaccharide vaccine (44-46). Aetiological nutritional status interacts with ALRI – and studies have been undertaken over many with diarrhoea and malaria – has been years, confirming the importance of the studied by Deborah Lehmann and pneumococcus and Haemophilus influenzae colleagues (51). Children with low in both pneumonia and meningitis (47-49). birthweight have 4 times the risk of dying The achievement of years of work in PNG from ALRI in the first year of life; children and the solution of many technical problems under two years of age who are led to Michael Gratten being chosen to write malnourished are 4 times more likely to be the manual for respiratory bacteriology for admitted with pneumonia and, if admitted, the World Health Organization. He and are 4 times more likely to die (51). Obtaining Janet Montgomery comprehensively the evidence for these interactions has been reviewed all aspects of their work on the a major achievement.
12 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
7ELBAT
COVERS OF THE JOURNAL IN ITS FIRST 05 YEARS
)seussi7(9591beF,1oN3emuloVot5591yaM,1oN1emuloV ,)lanruoJlacideMaeniuGweNdnaaupaP(eltiteht,revoctnorfehtno,htiwsrevocdeR ybseroMtroP,htlaeHcilbuPfotnemtrapeD,etaddnarebmuneussidnaemuloveht .edisnistnetnoC.smrafotaocnailartsuAehtdna
)seussi7(4691ceD,1oN7emuloVot9591guA,2oN3emuloV .dexobsieltiteht;revoctnorfehtnonoitamrofniemasehthtiwngisedweN .edisnistnetnoC.ffubseussi5,yergelapseussi2:ruoloC
5691raM,1oN8emuloV dnaemulovehtdnaeltitehtrednurevoctnorfnostnetnoc,ruolocffub,ngisedweN tnorfehtfomottobehttA.dettimosmrafotaocnailartsuA.etaddnarebmuneussi yteicoSlacideMehtybderaperpdnadetidE".tnemetatsgniwollofehtsierehtrevoc ".ybseroMtroP,htlaeHcilbuPfotnemtrapeDehtybdehsilbuP.aeniuGweNaupaPfo
)seussi62(1791ceD,4oN41emuloVot5691guA,2oN8emuloV eltitdexobyllaitraparednurevoctnorfnostnetnoC.ruolocneergelap,ngisedweN sitnemetatsgnihsilbupdnagnitideehT.etaddnarebmuneussidnaemulovehtdna sawlanruojehtfoemaneht1791peS,3oN41loVnI.revoctnorfedisniehtno .lanruoJlacideMaeniuGweNaupaPotdegnahc
)seussi22(7791peS,3oN02emuloVot2791raM,1oN51emuloV ruoloC.revoctnorfehtno,ngisedlanoitidartni,elopdnadriB.ngisedwenyletelpmoC lacideMaeniuGweNaupaP,eltitehtedisdnah-thgirehtnO.eussiyb,elbairav fopamllamsadnaetaddnarebmuneussidnaemuloveht,srettelkcolbni,lanruoJ aeniuGweNaupaP,sdnalsinaeniuGweNaupaPehtsulpaeniuGweNfodnalsieht rosthgilhgihehtrenrocdnah-tfelpotehttA.ngisedrevocehtforuolocehtnignieb .eussiehtfosucoF .lanruoJlacideMaeniuGweNaupaPehtfoogolahtiwrevockcabehtnostnetnoC ,eniltuoni,ylnoaeniuGweNaupaPotdegnahcsawpameht5791raM,1oN81loVnI .desopmirepusraepsagniniwtneekansahtiw
)seussi71(1891ceD,4oN42emuloVot7791ceD,4oN02emuloV rMyb,erugiflartsecnanadnayllacitrevgninnursksamhtiw,ngisedtralanoitidartweN .anobaSiuhaL edisdnah-thgirehtnoetaddnarebmuneussidnaemulovehtdnaeltitlanruojehT naeniuGweNaupaPehthtiwpamllamsemaseht,erugifehtevoba,tfelehtno,dna ehtfoogolahtiwrevockcabehtnostnetnoC.eussiyb,elbairavruoloC.'suecudac' regnolonogoleht9791raM,1oN,22loVmorf;lanruoJlacideMaeniuGweNaupaP ceD,4oN02loVmorF.edisniegaps'rotideehtnodeniatertubrevockcabehtno ceD,4oN42loVot0891raM,1oN32loVmorfdna8791nuJ,2oN12loVot7791 .edisnidetnirposlaerewstnetnoCeht)seussi11(1891
)seussi67(4002ceD-peS,4-3oN74emuloVot2891raM,1oN52emuloV tnorfehtfopotehttasrettelkcolbegralnieltitlanruojehthtiw,ngisedwenyletelpmoC ehtdna,revocehtfoezisehtflahtsomlaaeniuGweNaupaPfopamarevo,revoc ybgniyrav,desusruoloceerhT.mottobehttaetaddnarebmuneussidnaemulov ruolocemasehtyltcaxe,sedahssuoiravni,neergdnader,eulb:yllaunna,emulov .emulovenofoseussiehtrofdesugnieb .edisnidna,revockcabehtnostnetnoC nuJ,2oN52loVmorf,dnaetadeht,rebmuneussidnaemulov,emanlanruojehT NSSIeht5891nuJ,2oN82loVmorF.enipsehtnodetnirpsegapevisulcnieht,2891 morfsemulovllA.renrocdnah-thgirpotehtnirevoctnorfehtotdeddasaw0841-1300 -3neerg-der-eulbehtswollof7891retfagnixedniehtdnadexednierano52emuloV .eussirebmeceDeulbtxenehtnigniraeppa,elcycraey
13 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
14 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Figure 1. Representative examples of the 7 different types of cover that the Journal has had in its first 50 years. See also Table 7.
15 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Chronic lung disease and asthma have not heeded. Terry Spencer’s opinion was been extensively studied in PNG and some based on detailed ecological studies that he of this work has been reported in the Journal conducted with Margaret Spencer and (52-56). The epidemic of asthma in adults colleagues in several areas of PNG (63). It in Okapa and neighbouring areas, without is hard to recapture the religious fervour of affecting children, is a puzzle that has been the arguments about malaria eradication well defined but not yet completely solved. through residual house spraying with DDT This is another remarkable and unexpected but the conflicts and the seeds of doubt even fact of medical life in Papua New Guinea, among the faithful can be recognized though few outside of the Eastern Highlands between the lines of many, otherwise seem to be aware of it. Unlike kuru, pigbel austere, papers on malaria in the Journal. or swollen belly syndrome, rural adult asthma Wallace Peters, the Malariologist in the is still prevalent, still a major worry for people Department of Public Health, described the living in this part of the highlands and is eradication campaign in detail in a paper crying out for further investigation. The entitled ‘Malaria control in Papua and New combined topics of chronic lung disease and Guinea’ (64). In his earlier, meticulous asthma were reviewed by Ross Anderson survey of Western Province (65) Peters and Ann Woolcock in 1992 (57). raised doubts about whether “total control of transmission is feasible” in areas such as Western Province but he remained optimistic There have been 3 focus issues on that “success may be achieved in the next malaria (Table 8), indicating not only its few years in this type of situation”. How importance to the health of Papua New wrong this proved to be. Guineans but also the amount of research that has been done on it. Unfortunately, it is still a major problem. However, we are One aspect of malaria for which PNG is beginning to understand malaria better – and famous is hyperreactive malarious much of that new understanding has come splenomegaly (HMS) or, as it used to be from work carried out in PNG. We have the called, tropical splenomegaly syndrome. knowledge and tools to control malaria, even Gregory Crane spent much of his life working if they are not being fully utilized. One of on this disease. He has 8 substantive papers these new insights was the frequent on HMS in the PNG Medical Journal, occurrence of small-area variations in including the effect of regular malaria epidemiology and the importance of chemoprophylaxis in reducing the local ecologies in explaining differences in prevalence and severity of the disease. Greg the burden of malaria. In the second focus Crane’s obituary includes his bibliography, issue on malaria Jacqueline Cattani and where his achievements on HMS and his colleagues reported on their findings in contributions to the Journal, among other Madang (58) which demonstrated variation important work, are recorded (66). in malaria at the village level. The ecology was studied in relation to humans by James Moir and Paul Garner (59) and in relation to Malaria has been present among the mosquito vectors by Derek Charlwood and people of Papua New Guinea for long colleagues (60). This built on years of work enough for them to have evolved together. by malariologists in the past, much of it There are several polymorphisms that reported in the Journal, especially in the first protect against severe malaria that are focus issue on malaria (Table 8). However, peculiar to Papua New Guinea or the region. the philosophy had changed from One of the most interesting is ovalocytosis. eradication to control between the two focus The first paper specifically about issues. There was “a move away from ovalocytosis in Papua New Guineans was centrally organized campaigns applying a published in the Journal by Dominick Amato single method towards more flexible control and Peter Booth (67). The protection programs. Such programs should be based afforded by ovalocytosis against severe on a knowledge of the local epidemiology of malaria that was predicted by Amato and malaria with the aim of using a variety of Booth proved to be highly significant, and different control methods in an integrated uniquely so: absolute protection against approach” (59). Indeed, others, such as cerebral malaria (68). This extraordinary John Gunther (61) and Terence Spencer result should enable us to find clues to the (62), had been saying this before but were pathogenesis of cerebral malaria, and
16 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 investigation of this continues to be problem of endemic cretinism and its solution vigorously pursued by the Papua New (76). Mass campaigns of iodized oil Guinea Institute of Medical Research. injections were carried out in PNG and all salt for sale in bulk has, since 1973, been required by law to be iodized. Follow-up Another important advance in our studies were conducted by Peter Heywood understanding of malaria has been achieved and colleagues in 1986 (77) and by Paul through the use of the polymerase chain Pharoah and Peter Heywood in 1994 (78): reaction (PCR) to detect parasites. This was goitre rates were found to be low but there pioneered by Ingrid Felger and her was some persistent endemic cretinism. colleagues in PNG (69). Through the use of Since then efforts have been made to ensure this more sensitive method of detection that all salt sold in remote areas is correctly parasite rates in adults increased from 23%, iodized. Continued vigilance and monitoring as detected by microscopy, to 47%. The use are still required to ensure that this serious of PCR also enables strains of malaria disease does not recur. parasites to be described. Adult immunity in an area of high endemicity is maintained by low-level, asymptomatic chronic infection Of the great work done in PNG on with multiple strains of malaria parasites. We maternal immunization to prevent neonatal have to be cautious in any control program tetanus there is much in the international about disturbing this ecological balance. literature but nothing in the Journal. This was true also for the innovative and successful studies on filariasis until the focus Some of the earliest work on the use of issue on filariasis in 2000 (Table 8). In that impregnated bednets to control malaria was issue is a report on the study in Dreikikir, carried out in PNG, there have been many where there are communities with rates of studies on drug resistance and on malaria filariasis among the highest recorded in the in pregnancy, recent achievements have world – another distinction for PNG. The included malaria mapping in the highlands results showed that community-wide, annual … The list goes on. Unfortunately, we cannot single-dose treatment with cover them all. However, in discussing diethylcarbamazine (DEC) and ivermectin aspects of malaria that have been could lead to complete interruption of prominently reported in the Journal we transmission and ultimately elimination of cannot omit the drive towards a malaria lymphatic filariasis (79), an outcome of great vaccine led by the Papua New Guinea significance, not only for PNG. Now the Institute of Medical Research. The carefully global elimination of filariasis is within our planned project with complex baseline grasp, and we must seize the opportunity – studies (70) resulted in a very encouraging especially in PNG, where these studies were outcome (71), though we still have a long first carried out. way to go before an appropriate malaria vaccine will be available. We hope that Papua New Guineans will be among the first For leprosy, mercifully a disease now in to benefit from a malaria vaccine when it steep decline in PNG, the focus issue came comes, but this will not just happen because early, in 1973. Douglas Russell described of good work done in the past: it will have to leprosy in PNG in 1960 (80) and also in 1973 be constantly fought for. (81), when he made brief reference to studies being carried out in Karimui, in particular to a field trial, which began in 1962, In the past Papua New Guinea made an of BCG vaccine as a prophylaxis against important contribution to the global control leprosy. The successful outcome of the trial of iodine deficiency diseases through the was reported later (82). Ken Clezy is a work of Basil Hetzel and his colleagues. The surgeon who worked in Papua New Guinea problem of goitre was identified by for many years. He became Professor of McCullagh and a control project with iodized Surgery and Dean of the Medical School. oil was successfully initiated by him (72,73). He made a great contribution to teaching and His work on goitre was followed up by training. He was briefly Editor of the Journal Buttfield and Hetzel (74) and by Peter in 1979-1980 and contributed 4 Clinical Pharoah on endemic cretinism (75). A Practice articles to our series. His major monograph of the PNGIMR is devoted to the interest, however, was in the surgery of
17 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
8ELBAT
FOCUS ISSUES OF THE JOURNAL IN ITS FIRST 05 YEARS
FreltiTsuco FrebmuNeussIsuco otidEtseuG
P6lebgi V)691luJ,2oN9lo rotide(skcoddaM.I
H0egalliVehtnihtlae V)791raM,1oN31lo rotide(skcoddaM.I
dleifdooWemearG.DybdecudortnisaweussIsucoFehtfotpecnocehT
H2ygolotamea V)791raM,1oN51lo rotide(dleifdooW.G.D
L2esaesiDrevi V)791nuJ,2oN51lo rotide(dleifdooW.G.D
R3esaesiDyrotaripse V)791raM,1oN61lo rotide(dleifdooW.G.D
L3ysorpe V)791nuJ,2oN61lo rotide(dleifdooW.G.D
M,airala 1oN71loV-4oN61loV )rotide(dleifdooW.G.D 4791raM-3791ceD
snareclumuiretcabocyM esaesiD V)4791nuJ,2oN71lo rotide(dleifdooW.G.D
C4esaesiDralucsavoidra V)791peS,3oN71lo rotide(dleifdooW.G.D
O4scirtetsb V)791ceD,4oN71lo rotide(dleifdooW.G.D
O5ygolomlahthp V)791nuJ,2oN81lo rotide(dleifdooW.G.D
N5ygolorue Vk791ceD,4oN81lo oorbanroH.WdrahciR
P6yrtaihcys Vy791raM,1oN91lo eldarB-notruB.GnotruB
O6htlaeHlanoitapucc Vg791nuJ,2oN91lo nikcoHecurB
G6ygoloretneortsa V)791peS,3oN91lo rotide(otamA.D
H7ygolotamea V)791raM,1oN02lo rotide(otamA.D
C7enicideMytinummo Vt791peS,3oN02lo tayWegroeG
T8ygolonummIlacipor Vs791raM,1oN12lo replA.PleahciM
P8yrtaihcys Vy791peS,3oN12lo eldarB-notruB.GnotruB
P9lebgi Ve791raM,1oN22lo cnerwaL.WrogerG llerruM.C.GyhtomiT reklaW.DreteP
D9eraChtlaeHfoyrevile V)791peS,3oN22lo rotide(iqaraN.S
T0scitamosohcysPlacipor Vy891raM,1oN32lo eldarB-notruB.GnotruB
S5enicideMdnaecneicSlaico Vs891peS,3oN82lo nikneJloraC
M6airala Vi891raM,1oN92lo nattaCenileuqcaJ
T7scirtaideaPlacipor Vh891nuJ,2oN03lo pluddiBnhoJ
N8noitirtu VduJ,2oN13lo 891n oowyeHreteP
18 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
H9sciteneGnamu Va891ceD,4oN23lo itahBpeedluK
C0enicideMralucsavoidra Vu991ceD,4oN33lo aveK.HisI
A1snoitcefnIyrotaripseRetuc Vn991peS,3oN43lo namheLharobeD
M2airala Vi991ceD,4oN53lo qaraNsuriS
Q3ecnarussAytilau Vn991nuJ,2oN63lo osamohT.AenaJ leraK.GnehpetS
A4feileRniaPdnaaisehtsean Vi991ceD,4oN73lo orebO.SrednijaG
E5snoitcefnIciretn Vy991ceD,4oN83lo essaPnageM
T6amuar Vs991nuJ,2oN93lo rettaWdivaD
H6SDIAdnaVI Va991peS,3oN93lo mreVnitiN
M0hpluddiBnhoJrofeussIlairome Ve002nuJ-raM,2-1oN34lo kuDroverT
F0sisairali Ve002ceD-peS,4-3oN34lo irakcoB.JsesoM
D1setebai Ve002ceD-peS,4-3oN44lo lgO.DmaharG
ehtnitnemnorivnEehtdnahtlaeH Vn2002nuJ-raM,2-1oN54lo ellA.JtnayrB aerAiraT liaVnhoJ
S4htlaeHlauxe Vr002nuJ-raM,2-1oN74lo ammaHecnerwaL leprosy, and his contributions to this field STDs exploded. By the time of the second were warmly appreciated throughout PNG survey they had been joined by HIV (human and recognized internationally. The immunodeficiency virus), and AIDS achievements of Ken Clezy and Doug (acquired immune deficiency syndrome) Russell are recounted and praised in an added a frightening new dimension to the excellent history of leprosy management and problem of STDs. In 1994 another wake-up control in PNG by Hugo Ree (83). call was published: the PNGIMR Monograph on Sexual and Reproductive Knowledge and Behaviour (88). Unfortunately, its small Sexually transmitted diseases (STDs) do voice was insufficient to calm the epidemic. not get much coverage in the early pages of Since then many papers have been the Journal until syphilis spread up the published on the extraordinarily high rates Highlands Highway to affect a newly of sexually transmitted infections (often emerging population without the traditional unrecognized) in our communities and on immunity acquired from yaws (84,85). In the increasing prevalence of HIV infection. 1984 Candy Lombange reported on The Journal’s first focus issue on sexually gonorrhoea and syphilis in the previous 10 transmitted infections was on HIV and AIDS years and found a significant increase in both in 1996; in 2004 we had a second one, on despite the introduction of a National STD sexual health (Table 8). Control Program. He analyzed the data by region and in the major urban centres of Port Moresby and Lae. He discussed operational Papers on nutrition and growth have been problems and made recommendations for published in the Journal since the beginning improving the Control Program (86). This (6). The one focus issue on nutrition contains paper was a wake-up call for Papua New a paper on the 1982/83 National Nutrition Guinea, later re-emphasized by a multicentre Survey by Peter Heywood and colleagues laboratory-based study conducted in 1989- (89) and the account of nutrition and 1990 (87). Unfortunately our progress was morbidity (51) discussed previously in slow while the many organisms causing relation to ALRI/pneumonia. How virus
19 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
9ELBAT
INDEXES TO THE JOURNAL IN ITS FIRST 05 YEARS
ruomyeS.HannAybdelipmoC9691-5591,21-1emuloV 2791nuJniretnirPtnemnrevoGehtybtelkoobetarapesasadetnirP
ruomyeS.HannAybdelipmoC0791,31emuloV 2791nuJniretnirPtnemnrevoGehtybtelkoobetarapesasadetnirP
enicideMfoytlucaF,lluHnyboRsrMybderaperP1791,41emuloV 2791peSniretnirPtnemnrevoGehtybtelkoobetarapesasadetnirP
morfeussihcaeforevockcabedisniehtnodesitrevdadnaderaperpxednI2791,51emuloV *)seussi31(7791nuJ,2oN02loVot4791nuJ,2oN71loV
enicideMfoytlucaF,yrarbiLehtybderaperP4791-3791,71-61emuloV serPnetsirKybtelkoobetarapesasadetnirP
]gnissimsraey7[
162-352:3891ceD-peS,4-3oN62loVnI3891-2891,62-52emuloV
oVnI7891-4891,03-72emuloV 222-512:8891peS,3oN13l
513-013:1991ceD,4oN43loVnI0991-8891,33-13emuloV
942-342:4991ceD,4oN73loVnI3991-1991,63-43emuloV
981-481:7991ceD-peS,4-3oN04loVnI6991-4991,93-73emuloV
532-232:0002ceD-peS,4-3oN34loVnI9991-7991,24-04emuloV
102-891:3002ceD-peS,4-3oN64loVnI2002-0002,54-34emuloV
* roftiworrobro,mehtmorftieriuqcaotyppahyrevebdluowlanruoJehtxednIsihtfoypocasahenoynafI gniypocotohp infections and protein-energy malnutrition Bundi (96), Oro Bay (97) and Pari village interact had been reviewed 10 years earlier (98,99), to restrict ourselves to some early by John Mackenzie (90). Peter Heywood examples. Roy Scragg studied four and Alison Heywood, in studying the populations continuously for 19 years – functional significance of protein-energy Lemankua and Solas on Buka and Tabar and malnutrition, showed that growth retardation Tigak (near Kavieng) in New Ireland. He is a risk factor not only for certain infectious showed a decline in mortality rates, overall diseases but also is associated with delayed and for specific diseases, over this period. motor development, delayed eruption of This meticulously conducted study was teeth and reduced visual attention (91). To reported in the Journal in 1969 (100). obtain such results careful studies must be Another remarkable study of that time was carried out in the community, with full participation by members of the hamlet, the epidemiological sample survey of the village or settlement. The extensive and whole country by Peter Vines; he described comprehensive work done in Tari has his methodology in the Journal in 1965 (101). already been mentioned. During the last 50 The results of the survey were published by years other communities have been studied the Department of Public Health in 1970 demographically and epidemiologically, with (102). In this context we should not forget many variables measured, and the results Peter Sinnett’s monumental work on the published in the Journal: Anguganak Engan people of Tukisenta, which was (92,93), Kiriwina (94), Baiyer River (95), published as a PNGIMR monograph (103).
20 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Other population studies concentrated on Burkitt lymphoma is one of the special specific variables or diseases. Blood diseases of PNG, though the focus of the pressure has been widely studied and most disease here ranks only number two in the emphasis has been on the changes that world. Among various studies carried out have occurred with the modernization which over many years some promising new work now affects every community, to a variable shed light on the relationship of the tumour degree, in the country (104). Some studies, with Epstein-Barr virus and malaria (108) in contrast, have provided a link with the and these findings need to be explored traditional past, which is now lost to us further. Here is yet another local research forever. Ian Maddocks and Luke Rovin problem waiting to be tackled by a keen measured blood pressures in communities young scientist. However, it is by going back in Simbu Province, where, as they reported in time that we reach the most interesting in the Journal in 1965, blood pressure fell connection between Burkitt lymphoma and with age in both males and females (105). I PNG: it was first described here, and should recorded the same findings in Waisa village not be called by that name at all. Jan Saave in the Okapa District in 1962 (M.P. Alpers, reported on ‘lymphadenosis’ in PNG (109) 3 unpublished data). Moreover, as they did, I years before Denis Burkitt described ‘a investigated this further by studying the sarcoma involving the jaw in African electrolytes in 24-hour urine samples. The children’, which eventually became ‘his’ samples were meticulously collected and tumour. sent to the United States for analysis. My colleagues refused to believe or publish the results: the urinary volumes were considered Betelnut chewing is another special too small to be true 24-hour collections and feature of life in Papua New Guinea. It has the sodium and potassium levels were an important social function but, bizarre! This of course was exactly what the unfortunately, does help to explain the high study was intended to show – and document, incidence of oral cancer. However, the exact in the last opportunity we would ever have roles of smoking, the lime used, the person’s of doing so. Traditionally, salt was made diet and other factors in the pathogenesis of from potash and contained potassium, not their cancer have not been worked out. This sodium; the diet was high in potassium, low is important to investigate since it may be in sodium; and fluid intakes in the highlands possible to establish rules for ‘safe chewing’ were remarkably low, partly because of the if we had better evidence about the critical burden of carrying water in bamboo cylinders combination of factors in the pathogenesis. to hamlets built on high ridges for security. Other aspects of betelnut chewing have The paper by Maddocks and Rovin is an been addressed in the Journal: its possible important link to this not so remote but now nutrient value (110); its effect of reducing totally obliterated past in the highlands of dental caries (111); its possibly beneficial PNG. It is a tribute too to a remarkable effects in pregnancy (112); its short-term Papua New Guinean, who was tragically elevation of heart rate without a central effect killed in 1972: Luke Rovin’s obituary on visual processing or alertness (113); its appeared in the Journal (106) and in A aggravation of asthma (114,115); and its History of Medicine in Papua New Guinea association with diabetes (116). (107). In all our work, communicating with The selection of the First Eleven of the people, as individuals, as families, as Journal’s classic papers has now been made communities, is vital. This is true whether (Table 10) and it is time to draw this we are in clinical medicine, public health or discursive review to a close. The number of medical research. I have particular affection topics not covered is large and to include for three papers from the Journal that relate them all would mean extending the review to these issues, and I would not like to leave to more than twice its current length. A this review without citing them and quoting glance at the list of focus issues in Table 8 briefly from them. Vincent Zigas and Jettie will reveal immediately the most important van Delden discuss communication and topics not considered; and there are many community response in health programs more. However, there are just a few small, (117). They make the point forcefully that but interesting, items that I would like to draw health workers are often unpopular and that attention to before I close. “much of this unpopularity has been
21 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
01ELBAT
AFIRST ELEVEN OF THE JOURNAL'S CLASSIC PAPERS, LISTED IN THE ORDER OF CITATION IN THIS REVIEW
.CGTllerruM .leb-gipfoserutaeflacigoloimedipeemoS JdeMGNP .05-93:)2(9;luJ6691
.GecnerwaL .aeniuGweNaupaPnileb-gipfosisenegohtapehT JdeMGNP 9791 .94-93:)1(22;raM
.JirioK-anaB,JMnettarG,WRdrofhsA,DJecniV sediolygnortS ninoitatsefniseiceps .amedeodezilareneghtiwnoitaicossa:aeniuGweNaupaPfostnafnignuoy deMGNP J .721-021:)2(22;nuJ9791
.JirioK-anaB,JMnettarG,DJecniV,WRdrofhsA sediolygnortS -dimaninoitcefni .yevruslacigoloimedipenafostluser:ytinummocaeniuGweNaupaPniatnuom GNP JdeM .531-821:)2(22;nuJ9791
.PanatakgnaP,JhpluddiB .aeohrraidgnilnaeW JdeMGNP .31-7:)1(41;raM1791
.JyremogtnoM,MnettarG ninerdlihcnisitigninemdnaainomuenpetucafoygoloiretcabehT .seigetartslacinhcetdnastcaf,snoitpmussa:aeniuGweNaupaP JdeMGNP 1991 .891-581:)3(43;peS
.PdoowyeH,PdrawoH,DnnamheL tcartyrotaripserrewoletuca:ytidibromdnanoitirtuN .airalamdnaaeohrraid,snoitcefni JdeMGNP .611-901:)2(13;nuJ8891
.PMsreplA,WnosdivaD,JoniaP,MynniG,DFnosbiG,SJrioM,AJinattaC aera-llamS .ecnivorPgnadaMniairalamfoygoloimedipeehtnisnoitairav JdeMGNP 6891 .71-11:)1(92;raM
.BPhtooB,DotamA .snaisenaleMnisisotycolavoyratidereH JdeMGNP ;raM7791 .23-62:)1(02
.KCegnabmoL .aeniuGweNaupaPniecnedicniesaesiddettimsnartyllauxesnisdnerT JdeMGNP .751-541:)4-3(72;ceD-peS4891
.LnivoR,IskcoddaM sallafotsraeppaerusserpdoolbhcihwninoitalupopaeniuGweNA .secnavdaega JdeMGNP .12-71:)1(8;raM5691
1991,8891,6891,4891,)3(9791,7791,1791,6691,5691-sraeynoitacilbuP generated by an aloof, paternalistic attitude factors in determining immunization status”. by some health officials towards the people Better knowledge about immunization did they are trying to help. This attitude is mainly not lead to improved practice. They due to a lack of understanding and rapport recommended that “maximal use must be with the community.” “In order to be truly made of local community organizations to successful, the health worker must be able disseminate the key immunization to make the educational mechanisms information”. Carol Jenkins enlarges on the function as a two-way process in which he difficulties of changing hygiene behaviour learns from the people even as they learn (119); indeed, essentially the same problems from him.” Paul Freeman, Jane Thomason have to be faced in trying to achieve any and Gilbert Bukenya studied the factors that desirable behaviour change. In the context of her paper she argues that “improved made urban settlement dwellers take their hygiene and sanitation could be achieved if children for immunization (118). They found sufficient resources were focused on the that “the provision of information to mothers people who use water and sanitation on when to start immunization and how often systems and not simply on the systems the child should be immunized were key themselves”.
22 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
I have a few other favourites scattered REFERENCES among the papers of the Journal. Wilfred Moi describes his cultural background, his 1 Gunther JT. Epidemiology of malaria in Papua and growing up in Ambasi in Milne Bay Province New Guinea. PNG Med J 1955;1:1-9. 2 Maddocks I. Editorial. The search for a national and the relevance of this to his career as a character. PNG Med J 1965;8:1-2. psychiatrist (120). This paper should be an 3 Woodfield DG. Editorial. The role of the “Medical inspiration to other Papua New Guineans to Journal”. PNG Med J 1971;14:71-73. realize that their own background and life 4 Rooney J. The role of the Medical Journal. Letter. PNG Med J 1973;16:226. story will be equally fascinating and relevant 5 Naraqi S. Papua New Guinea Medical Journal – to others, and that they should convince a national asset. PNG Med J 1977;20:156-157. themselves to write it up. I cite it also to 6 Oomen HAPC. Assessment and prevention of honour the memory of Wilfred Moi, a gentle child malnutrition. PNG Med J 1956;2:1-20. 7 Zigas V, Gajdusek DC. Kuru: clinical, pathological man of great achievement, who also died in and epidemiological study of a recently discovered tragic circumstances. Peter Sharp acute progressive degenerative disease of the contributed several excellent, well-written central nervous system reaching ‘epidemic’ papers to the Journal; three of them are proportions among natives of the Eastern Highlands of New Guinea. PNG Med J 1959;3:1- memorable not only for their content but also 24. for their title (121-123). I am pleased to cite 8 Spencer M. The history of malaria control in the them for that reason alone, but also as a way Southwest Pacific Region, with particular reference of encouraging everybody to read – or re- to Papua New Guinea and the Solomon Islands. read – them. PNG Med J 1992;35:33-66. 9 Allen JS, Johnson FYA. Eye movements and schizophrenia in Papua New Guinea: qualitative analyses with case histories. PNG Med J Finally, though the Journal articles are 1995;38:106-126. usually about Papua New Guinea, not all are, 10 Papua and New Guinea Medical Journal. and some even come from outside our local Editorial. Kuru. PNG Med J 1959;3:25-27. 11 Glasse S. The social effects of kuru. PNG Med J region. Takeo Doi was Professor of 1964;7:36-47. Psychiatry at the University of Tokyo and he 12 Gajdusek DC. Kuru in New Guinea: a definitive introduced the Japanese concept of amae bibliography. PNG Med J 1964;7:48-51. to readers of our Journal (124): it means ‘to 13 Burnet FM. Kuru – the present position. PNG Med depend and presume upon another’s love J 1965;8:3-7. 14 Hornabrook RW. Kuru – some misconceptions or bask in another’s indulgence’. As I close and their explanation. PNG Med J 1966;9:11-15. this review I hope, as Takeo Doi 15 Mathews JD. The epidemiology of kuru. PNG Med recommends, that amae is something that J 1967;10:76-82. we all can feel dear to our hearts. This will 16 Chandor SB, Hornabrook RW. Serum immunoglobulin patterns in kuru. PNG Med J make my neglect here of so much good work 1969;12:40-41. published in our Journal easier to bear. 17 Hornabrook RW, King HOM. Kuru. PNG Med J 1975;18:203-206. 18 Gajdusek DC, Alpers MP. Recent data on the ACKNOWLEDGEMENTS properties of the viruses of kuru and transmissible virus dementias. PNG Med J 1975;18:207-213. 19 Benfante RJ, Gajdusek DC. Antibody studies in the kuru region. Immunological epidemiology of I acknowledge the commission from the population groups in and adjacent to the kuru President, Mathias Sapuri, and the Executive region in Papua New Guinea. I. Enteroviruses. of the Medical Society of Papua New Guinea PNG Med J 1978;21:124-133. 20 Hornabrook RW. The Auyana head nodders. PNG to prepare this review of the first 50 years of Med J 1970;13:90-92. our Journal and I thank them warmly for 21 Zigas V. Auyana head nodders. Letter. PNG Med according me this privilege. I acknowledge J 1971;14:39. the Editors of the Journal, Peter Siba, Nakapi 22 Hornabrook RW, Nagurney JT. Auyana head Tefuarani and John Reeder, and our nodders. Essential heredo-familial tremor in the Eastern Highlands. PNG Med J 1975;18:214-219. dedicated and skilful Editorial Assistant, 23 Hornabrook RW, Wagner F. Creutzfeldt-Jakob Cynthea Leahy, and thank them very much disease. PNG Med J 1975;18:226-228. for all the support they have given me. I 24 Alpers MP. Kuru. In: Attenborough RD, Alpers MP, also thank Jason Kovacs for his eds. Human Biology in Papua New Guinea: The Small Cosmos. Oxford: Clarendon Press, indispensable help in establishing the 1992:313-334. Journal’s newly acquired publishing 25 Lindenbaum S. Kuru Sorcery: Disease and capability and, especially for this issue, in Danger in the New Guinea Highlands. Palo Alto: enabling us to reproduce the classic papers Mayfield, 1979. 26 Murrell TGC. Some epidemiological features of from the past. pig-bel. PNG Med J 1966;9:39-50.
23 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
27 Roth L. Diagnosis and management of pig-bel. by standard clinical criteria. J Infect Dis PNG Med J 1966;9:51-54. 1989;159:348-352. 28 Egerton JR. Bacteriology of enteritis necroticans 49 Gratten M, Montgomery J. The bacteriology of in New Guinea highlanders. PNG Med J 1966;9:55- acute pneumonia and meningitis in children in 59. Papua New Guinea: assumptions, facts and 29 Walker PD. Timothy George Calvert Murrell. technical strategies. PNG Med J 1991;34:185-198. Obituary. PNG Med J 2003;46:87-89. 50 Lehmann D. Mortality and morbidity from acute 30 Lawrence G. The pathogenesis of pig-bel in Papua lower respiratory tract infections in Tari, Southern New Guinea. PNG Med J 1979;22:39-49. Highlands Province 1977-1983. PNG Med J 31 Lawrence G, Walker PD, Freestone DS, Shann 1991;34:174-184. F. The prevention of pig-bel in Papua New Guinea. 51 Lehmann D, Howard P, Heywood P. Nutrition and PNG Med J 1979;22:30-34. morbidity: acute lower respiratory tract infections, 32 Poka H, Duke T. In search of pigbel: gone or just diarrhoea and malaria. PNG Med J 1988;31:109- forgotten in the highlands of Papua New Guinea? 116. PNG Med J 2003;46:135-142. 52 Woolcock AJ, Colman MH, Blackburn CRB. 33 Kelly A, Voge M. Report of a nematode found in Chronic lung disease in Papua New Guinea and humans at Kiunga, Western District. PNG Med J Australian populations. PNG Med J 1973;16:29- 1973;16:59. 35. 34 Vince JD, Ashford RW, Gratten MJ, Bana-Koiri 53 Anderson HR, Cunnington AM. House dust J. Strongyloides species infestation in young mites in the highlands of Papua New Guinea. PNG infants of Papua New Guinea: association with Med J 1974;17:304-308. generalized oedema. PNG Med J 1979;22:120- 54 Turner KJ. The conflicting role of parasitic 127. infections in modulating the prevalence of asthma. 35 Ashford RW, Vince JD, Gratten M, Bana-Koiri PNG Med J 1978;21:86-104. J. Strongyloides infection in a mid-mountain Papua 55 Green W, Woolcock AJ, Dowse G. House dust New Guinea community: results of an mites in blankets and houses in the highlands of epidemiological survey. PNG Med J 1979;22:128- Papua New Guinea. PNG Med J 1982;25:219-222. 135. 56 Dowse GK, Turner KJ, Woolcock AJ, Alpers MP. 36 Shield JM, Hide RL, Harvey PWJ, Vrbova H, Emerging asthma in the Okapa District of the Tulloch J. Hookworm (Necator americanus) and Eastern Highlands Province of Papua New Guinea: Strongyloides fuelleborni-like prevalence and egg the problem and its implications. PNG Med J count with age in highlands fringe people of Papua 1983;26:33-41. New Guinea. PNG Med J 1987;30:21-26. 57 Anderson HR, Woolcock AJ. Chronic lung 37 Barnish G, Harari M. Possible effects of disease and asthma in Papua New Guinea. In: Strongyloides fuelleborni-like infections on children Attenborough RD, Alpers MP, eds. Human Biology in the Karimui area of Simbu Province. PNG Med in Papua New Guinea: The Small Cosmos. Oxford: J 1989;32:51-54. Clarendon Press, 1992:289-301. 38 Biddulph J, Pangkatana P. Weanling diarrhoea. 58 Cattani JA, Moir JS, Gibson FD, Ginny M, Paino PNG Med J 1971;14:7-13. J, Davidson W, Alpers MP. Small-area variations 39 Murtagh K. The advantages of breast feeding. in the epidemiology of malaria in Madang Province. PNG Med J 1977;20:157-159. PNG Med J 1986;29:11-17. 40 Lambert J, Basford J. Port Moresby infant feeding 59 Moir J, Garner P. Malaria control through health survey. PNG Med J 1977;20:175-179. services in Papua New Guinea. PNG Med J 41 Biddulph J. Legislation to protect breast feeding. 1986;29:27-33. PNG Med J 1983;26:9-12. 60 Charlwood JD, Graves PM, Alpers MP. The 42 Douglas RM, Riley ID. Adult pneumonia in Lae – ecology of the Anopheles punctulatus group of 99 consecutive cases. PNG Med J 1970;13:105- mosquitoes from Papua New Guinea: a review of 109. recent work. PNG Med J 1986;29:19-26. 43 Hansman D. The development of antibiotic 61 Gunther JT. The early history of malaria control resistance in pneumococcus and its relevance to in Papua New Guinea. PNG Med J 1974;17:4-7. the treatment of pneumonia in tropical countries. 62 Spencer M. Obituary of Terence Edward Thornton PNG Med J 1973;16:15-19. Spencer. PNG Med J 2003;46:90-97. 44 Riley ID. Pneumococcal vaccine and mortality 63 Spencer T, Spencer M, Venters D. Malaria from pneumonia. PNG Med J 1985; 28:231-232. vectors in Papua New Guinea. PNG Med J 45 Lehmann D. Pneumococcal polysaccharide 1974;17:22-30. vaccine prevents death from pneumonia in Papua 64 Peters W. Malaria control in Papua and New New Guinean children. PNG Med J 1986;29:281- Guinea. PNG Med J 1959;3:66-75. 283. 65 Peters W. A malaria survey in the Western District 46 Riley ID, Lehmann D, Alpers MP, Marshall of Papua. PNG Med J 1957;2:25-38. TFdeC, Gratten H, Smith D. Pneumococcal 66 Bashir H. Obituary of Gordon Gregory Gibson vaccine prevents death from acute lower- Crane, with Bibliography of G.G. Crane. PNG Med respiratory-tract infections in Papua New Guinean J 2002;45:238-240;243-245. children. Lancet 1986;2:877-881. 67 Amato D, Booth PB. Hereditary ovalocytosis in 47 Shann F, Gratten M, Germer S, Linnemann V, Melanesians. PNG Med J 1977;20:26-32. Hazlett D, Payne R. Aetiology of pneumonia in 68 Genton B, Al-Yaman F, Mgone CS, Alexander children in Goroka Hospital, Papua New Guinea. N, Paniu MM, Alpers MP, Mokela D. Ovalocytosis Lancet 1984;2:537-541. and cerebral malaria. Nature 1995;378:564-565. 48 Barker J, Gratten M, Riley I, Lehmann D, 69 Felger I, Tavul L, Narara A, Genton B, Alpers Montgomery J, Kajoi M, Gratten H, Smith D, M, Beck HP. The use of the polymerase chain Marshall TFdeC, Alpers MP. Pneumonia in reaction for more sensitive detection of children in the Eastern Highlands of Papua New Plasmodium falciparum. PNG Med J 1995;38:52- Guinea: a bacteriologic study of patients selected 56.
24 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
70 Alpers MP, Al-Yaman F, Beck HP, Bhatia KK, of sexually transmitted diseases in five STD clinics Hii J, Lewis DJ, Paru R, Smith TA. The Malaria in Papua New Guinea. PNG Med J 1994;37:152- Vaccine Epidemiology and Evaluation Project of 160. Papua New Guinea: rationale and baseline studies. 88 The National Sex and Reproduction Research PNG Med J 1992;35:285-297. Team, Jenkins C. National Study of Sexual and 71 Genton B, Betuela I, Felger I, Al-Yaman F, Reproductive Knowledge and Behaviour in Papua Anders RF, Saul A, Rare L, Baisor M, Lorry K, New Guinea. Papua New Guinea Institute of Brown GV, Pye D, Irving DO, Smith TA, Beck Medical Research Monograph No 10. Goroka: HP, Alpers MP. A recombinant blood-stage malaria Papua New Guinea Institute of Medical Research, vaccine reduces Plasmodium falciparum density 1994. and exerts selective pressure on parasite 89 Heywood P, Singleton N, Ross J. Nutritional populations in a phase 1-2b trial in Papua New status of young children – the 1982/83 National Guinea. J Infect Dis 2002;185:820-827. Nutrition Survey. PNG Med J 1988;31:91-101. 72 McCullagh SF. Goitre control project. PNG Med 90 Mackenzie JS. Virus infection, vaccination and J 1959;3:43-47. protein-energy malnutrition. PNG Med J 73 McCullagh SF. The Huon Peninsula endemic: I. 1978;21:134-151. The effectiveness of an intramuscular depot of 91 Heywood P, Heywood A. The functional iodized oil in the control of endemic goitre. Med J significance of protein-energy malnutrition. PNG Aust 1963;1:769-777. Med J 1988;31:103-108. 74 Buttfield IH, Hetzel BS. The aetiology and control 92 Sturt RJ, Stanhope JM. Mortality and population of endemic goitre in Eastern New Guinea. PNG patterns of Anguganak. PNG Med J 1968;11:111- Med J 1966;9:119-126. 117. 75 Pharoah POD. Endemic cretinism in New Guinea. 93 Sturt RJ. Infant and toddler mortality in the Sepik. PNG Med J 1971;14:115-119. PNG Med J 1972;15:215-220. 76 Hetzel BS, Pharoah POD, eds. Endemic 94 Stanhope JM. Mortality and population growth – Cretinism. Proceedings of a Symposium held at Losuia area, Kiriwina, Trobriand Islands. PNG Med the Institute of Human Biology, Goroka, 27-29 Jan J 1969;12:42-48. 1971. Papua New Guinea Institute of Medical 95 Becroft TC, Stanhope JM, Burchett PM. Research Monograph No 2. Goroka: Papua New Mortality and population trends among the Kyaka Guinea Institute of Medical Research, Dec 1971. Enga, Baiyer Valley. PNG Med J 1969;12:48-55. 77 Heywood PF, Buttfield IH, Buttfield BL, Anian 96 Malcolm LA. Child mortality and disease pattern: G. Endemic cretinism and endemic goitre in two recent changes in the Bundi area. PNG Med J areas of Madang Province, Papua New Guinea. 1969;12:13-17. PNG Med J 1986;29:149-152. 97 Dowell M, Stanhope JM. Mortality and population 78 Pharoah PDP, Heywood PF. Endemic goitre and trends, Oro Bay, Northern Papua. PNG Med J cretinism in the Simbai and Tep-Tep areas of 1970;13:132-136. Madang Province, Papua New Guinea. PNG Med 98 Maddocks DL, Maddocks I. Pari village study: J 1994;37:110-115. results and prospects, 1971. PNG Med J 79 Bockarie MJ, Ibam E, Alexander NDE, Hyun P, 1972;15:225-233. Dimber Z, Bockarie F, Alpers MP, Kazura JW. 99 Maddocks DL, Maddocks I. The health of young Towards eliminating lymphatic filariasis in Papua adults in Pari village. PNG Med J 1977;20:110- New Guinea: impact of annual single-dose mass 116. treatment on transmission of Wuchereria bancrofti 100 Scragg RFR. Mortality changes in rural New in East Sepik Province. PNG Med J 2000;43:172- Guinea. PNG Med J 1969;12:73-83. 182. 101 Vines AP. Methodology of the Papua and New 80 Russell DA. Leprosy in Papua and New Guinea. Guinea epidemiological sample survey. PNG Med PNG Med J 1960;4:49-54. J 1965;8:35-38. 81 Russell DA. Leprosy in Papua New Guinea. PNG 102 Vines AP. An Epidemiological Sample Survey of Med J 1973;16:83-85. the Highlands, Mainland and Islands Regions of 82 Bagshawe A, Scott GC, Russell DA, Wigley SC, the Territory of Papua and New Guinea. Port Merianos A, Berry G. BCG vaccination in leprosy: Moresby: Department of Public Health, 1970:621p. final results of the trial in Karimui, Papua New 103 Sinnett PF. The People of Murapin. Papua New Guinea, 1963-79. Bull World Health Organ Guinea Institute of Medical Research Monograph 1989;67:389-399. No 4. Faringdon: EW Classey, 1977:208p. 83 Ree GH. Leprosy – the development of an 104 King H, Collins V, King LF, Finch C, Alpers MP. integrated service. In: Burton-Bradley BG, ed. A Blood pressure, hypertension and other History of Medicine in Papua New Guinea: cardiovascular risk factors in six communities in Vignettes of an Earlier Period. Sydney: Papua New Guinea, 1985-1986. PNG Med J Australasian Medical Publishing Company, 1994;37:100-109. 1990:205-229. 105 Maddocks I, Rovin L. A New Guinea population 84 Rhodes FA, Anderson SEJ. An outbreak of in which blood pressure appears to fall as age treponematosis in the Eastern Highlands. PNG advances. PNG Med J 1965;8:17-21. Med J 1970;13:49-52. 106 Papua New Guinea Medical Journal. Luke 85 Garner MF, Hornabrook RW, Backhouse JL. Robert Rovin. Obituary. PNG Med J 1973;16:72. Treponematosis along the Highlands Highway. 107 Sawa TL. Luke Robert Rovin. In: Burton-Bradley PNG Med J 1972;15:139-141. BG, ed. A History of Medicine in Papua New 86 Lombange CK. Trends in sexually transmitted Guinea: Vignettes of an Earlier Period. Sydney: disease incidence in Papua New Guinea. PNG Australasian Medical Publishing Company, Med J 1984;27:145-157. 1990:333-337. 87 Hudson BJ, van der Meijden WI, Lupiwa T, 108 Moss D. Epstein-Barr virus and malaria in relation Howard P, Tabua T, Tapsall JW, Phillips EA, to Burkitt’s lymphoma in Papua New Guinea. PNG Lennox VA, Backhouse JL, Pyakalyia T. A survey Med J 1986;29:41-43.
25 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
109 Saave JJ. Lymphadenosis amongst New Guinea special reference to Papua New Guinea. PNG Med natives. Med J Aust 1955;1:358-360. J 1974;17:80-83. 110 Weegels P, Heywood P, Jenkins C. 118 Freeman PA, Thomason JA, Bukenya GB. Consumption of betel nut and its possible Factors affecting the use of immunization among contribution to protein and energy intakes. PNG urban settlement dwellers in Papua New Guinea. Med J 1984;27:37-39. PNG Med J 1992;35:179-185. 111 Howden GF. The cariostatic effect of betel nut 119 Jenkins C. Changing hygiene behaviour in Papua chewing. PNG Med J 1984;27:123-131. New Guinea. PNG Med J 1995;38:320-324. 112 Taufa T. Betel-nut chewing and pregnancy. PNG 120 Moi W. Growing up in Ambasi. PNG Med J Med J 1988;31:229-233. 1976;19:14-18. 113 Frewer LJ. The effect of betel nut on human 121 Sharp PT. “Pierced by the arrows of this ghostly performance. PNG Med J 1990;33:143-145. world” – a review of arrow wounds in Enga 114 Kiyingi KS. Betel-nut chewing may aggravate Province. PNG Med J 1981;24:150-163. asthma. PNG Med J 1991;34:117-121. 122 Sharp PT. Ghosts, witches, sickness and death: 115 Kiyingi KS, Saweri A. Betelnut chewing causes the traditional interpretation of injury and disease bronchoconstriction in some asthma patients. PNG in a rural area of Papua New Guinea. PNG Med J Med J 1994;37:90-99. 1982;25:108-115. 116 Benjamin AL. Community screening for diabetes 123 Sharp PT. The searching sun: the Lyeime in the National Capital District, Papua New Guinea: Movement – crisis, tragic events and folie à deux is betelnut chewing a risk factor for diabetes? PNG in the Papua New Guinea highlands. PNG Med J Med J 2001;44:101-107. 1990;33:111-120. 117 Zigas V, van Delden J. Communication and 124 Doi T. Introducing the concept of amae. PNG community response in health programmes with Med J 1990;33:147-150.
26 PNG Med J 2005 Mar-Jun;48(1-2):27-38
SOME EPIDEMIOLOGICAL
BY T. G. C. MURRELL. * FEATURES OF PIG-BEL.
SPONTANEOUS gangrene of the small and Rassfeld-Sternberg, 1950 ; Hermann, 1948 ; intestine, without obvious vascular or mech- Siegmund 1948; Kloos and Brummond, anical cause, is an extremely rare disease. It is 1951). As a result of the recovery rarely associated with the known enteric and of Type C strains from man in New Guinea, dysenteric infections of the bowel. Within the the strains of Cl. perfringens Type F, believed framework of the present pathological services to cause ‘Darmbrand’, have been reclassified available in the Highland hospitals of New on toxicological grounds as a Type C variety Guinea, the story of the ecology and epi- (Sterne and Warrack, 1964). demiology of a widespread gangrenous enteritis was rather slow in unfolding. The condition Heat resistant Type A strains are the only was first noticed in subjects undergoing laparo- other recognized we1chii enteric pathogens of tomy for an “acute abdomen” in Goroka early man (Hobbs et al., 1953), and cause food in 1961. It was then termed a “necrotizing poisoning usually via a medium of re-heated jejunitis” because initial bacteriological investi- meat dishes. These strains produce alpha toxin gations failed to establish a cause (Murrell and as the major antigen and are one of the organ- Roth, 1963). Beta toxin producing strains of isms responsible for the ordinary gas gangrene Clostridium perfringens (Cl. welchii) were sub- of wounds. Strains from other groups, produc- sequently recovered from the bowel contents and ing different toxins, cause enterotoxaemic faeces of subjects with the condition (Egerton diseases in animals, each type having a limited and Walker, 1964; Murrell et al. 1966). host range. Thus Type B is associated chiefly with lamb dysentery (Dalling, 1928), Type C This condition, synonymous with “enteritis with enterotoxaemias of sheep, calves, and pig- necroticans”, can be defined as an acute necro- lets (Griner and Bracken, 1953; Field and tizing inflammatory disease of the small bowel, Gibson, 1955), Type D with pulpy kidney of patchy distribution and commencing in the disease of sheep (Bennett, 1932), and Type E is duodenum or jejunum. In New Guinea Clos- occasionally found as a saprophyte in the intes- tridium welchii (Cl. perfringens) Type C is tines of calves (Bosworth, 1943). Overfeeding thought to be one of the aetiological components and dietary change play a significant part in of its cause. Because the term includes a group the aetiology of these diseases which may be of conditions known by a multiplicity of names, analogous to Pig-bel of man in New Guinea. the New Guinea syndrome warrants definition by a specific name. ‘Pig-bel’ is suggested as a Clinically the disease in New Guinea is suitable name for the disease because of its characterized by the symptom triad of severe and aetiological association with the pig-feasting prac- spasmodic upper abdominal pain, bloody diar- tices of the New Guinea Highlander. rhoea with melaena and nausea with occasional vomiting. In advanced cases there is constipa- A similar disease, ‘Darmbrand’, appeared in tion, abdominal distension and constant general- epidemic form in North-West Germany in the ized pain. The signs may be those of dehydra- latter and post-war years, and was believed to tion from a severe enteritis, shock and toxaemia, be due to Clostridium perfringens Type F (Han- or those of an acute or sub-acute small bowel sen et al., 1949; Oakley, 1949; Marcuse and obstruction with localized or diffuse abdominal Konig, 1950). Zeissler and Rassfeld-Sternberg tenderness. The disease progresses, if untreated, (1949) and Hain (1949) established a food- to complete segmental gangrene of parts of the borne origin for the disease but other small intestine with the development of ileus, workers believed that additional dietary, both mechanical and paralytic, oligaemic shock nutritional and possibly viral factors were and severe toxaemia. Gas bubbles can some- associated with the pathogenesis – Pietzonka times be seen in the distal mesenteric venous arches and the mesenterium is oedematous, with * District Medical Officer, Department of Public Health. swollen ‘spongy’ regional lymph glands. Tissue
VOLUME 9, NO. 2, JULY, 1966 39
27 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 emphysema along the mesenteric attachment and Mason 1931; Glenny, Bar, Llewellyn-Jones, in the subserosa is a frequent finding at opera- Dalling and Ross, 1933). tion or autopsy. Epidemiological information included a sur- A small proportion of persons with the dis- vey of beta antitoxin levels in the normal popu- ease survive the initial stages without treatment, lation in high and low prevalence areas; a bac- or with conservative management, but perfora- teriological survey of faeces from the normal tion and peritonitis or the development of an human and porcine populations to determine, if acute malabsorption syndrome results in death. any, a ‘carrier rate’ of Cl. perfringens strains; The malabsorption is due to chronic small bowel a bacteriological search for these organisms in obstruction by adhesive bands, stenosis by cica- cooked and uncooked pig meat; and general trization or the development of short circuits, observations in the preparation of food at pig- blind loops and rigid scarred bowel segments killing ceremonies in Western and Eastern Highland clans. A disease prevalence rate was denuded of normal mucosa. Follow-up studies established at Goromougo in the Upper Chimbu indicate that certain features of the malabsorption during 1964 and trends in the prevalence of syndrome develop in a group with residual diarrhoeal disease were obtained from hospital chronic jejunitis. records of the Goroka, Kundiawa, Mount Hagen, Baiyer River, Wabag and Tari hospitals follow- This paper describes the cumulative investiga- ing known pig-killing activities. tions into the condition during the four years, 1961 to 1964, and attempts to review some of Specific Cl. perfringens Type C antiserum was the epidemiological and aetiological features. given to consecutive patients with pig-bel, and the mortality compared to an unmatched series MATERIALS AND METHODS. of patients treated consecutively prior to the arrival of the anti-serum. Case records were kept of patients with known and suspected enteritis necroticans, during the RESULTS. period January, 1961, to November, 1964. A total of 210 cases are reviewed in this study. Distribution. The diagnosis was made on clinical grounds alone in 73 instances and on the pathological Age and sex distribution of the 210 cases are features at autopsy or operation in 137 patients. listed in Table 1., and illustrated in Figures They were classified clinically into acute toxic, 1 and 2. The condition was most common in pre-adolescent children. 52.3 per cent. of patients acute surgical, sub-acute surgical and mild were two to ten years old. Children in the groups. The diagnosis was established on the six to ten year group made up the largest seg- history of pork consumption, the clinical ment of cases (29 per cent.). Males were features, X-ray findings, bowel appearances at affected more than females in the ratio of operation or autopsy, the isolation of Cl. per- 2.2 : 1. This ratio was generally maintained fringens Type C from a significant proportion in all age groups except in the 16 to 20 year of resected intestinal segments, the detection of group when more females were recorded with rising Cl. perfringens beta antitoxin levels in the disease. The numbers recorded in this the sera of recovered patients, and marked clini- group, however, were too small to be significant. cal improvement following the intravenous administration of specific Type C antiserum (Murrell et al., 1966).
Bacteriological and radiological investigations have been presented in other papers. Antitoxin levels to the beta toxin of Cl. perfringens were estimated in 58 cases, serial samples being taken from 21 persons with the disease. The serum was separated, kept under refrigeration and one drop of 50 per cent. o-cresol-ether added to each ml. of serum as a preservative. These were despatched in batches to the Wellcome Research Laboratories for estimation of Cl. perfringens beta antitoxin (Glenny, Llewellyn-Jones and
40 PAPUA AND NEW GUINEA MEDICAL JOURNAL
28 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Of the 210 persons reviewed in this series, 1st June, 1962, to 30th November, 1964. These only one was European. All other persons, with rates were significantly higher in the 1 to 12 one exception, were inhabitants from areas where year age group. At Baiyer River 16.1 per they contracted the disease. The exception was cent. of deaths occurring in this age group were a Mount Hagen native working as a plantation due to Pig-bel. The enteritis necroticans group labourer on Kar Kar Island, in the Madang Dis- forms a considerable proportion of deaths from trict. The European, a 28-year old Polish lines- all diarrhoeal diseases in all age groups. This man, contracted his disease at Tari following a was 15.3 per cent. at Goroka, 17.3 per cent. at meal of native pork. Kundiawa and 26.7 per cent. at Baiyer River.
Incidence of pig-bel in general population.
Assessment of the incidence of the disease was made from census figures taken of five clans at the Government Rest House at Goromaugo (Upper Chimbu) in October of 1964. The total population, including absentees, was 1,448 (Table 2). Following pork feasting in May and August of the same year, seven persons were known to have contracted Pig-bel, and two died.
Figure I.—Age Distribution of 210 cases of Pig-bel
Figure 2.—Sex Distribution and Survival Figures. Mortality.
The overall death rate of cases was 36 per cent. In the severe forms of the disease the case mortality was 49.8 per cent. This was highest in the acute toxic group (84.6 per cent.). The rate was 57.7 per cent. in children under five years of age and 46.2 per cent. in persons over 40 years of age. The lowest death rate occurred in the 11 to 15 year group. The over- all rate for females was slightly higher (38.5 pcr cent.) than that for males (35.2 per cent.). (Figure 3.)
For all ages, enteritis necroticans accounted for 2.1 per cent. of deaths at the Goroka and Kundiawa hospitals and 3.5 per cent. at the Baiyer River hospital during the period 1st April, 1962, to 31st August, 1964. The Kundiawa period reviewed was slightly longer, Figure 3.—Case Mortality by Age Groups
VOLUME 9, NO. 2, JULY, 1966 41
29 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
The diagnosis was established bacteriologically and sex distribution showed no marked difference in three cases, serologically in two and by autopsy from hospital cases. in a further two. The incidence in this area was therefore assessed at 48.3 per 10,000. A Outbreaks and geographical distribution. high prevalence of Cl. perfringens beta antitoxin The geographical distribution of cases over the in population groups sampled in the Upper four years 1st January, 1961, to 31st Decem- Chimbu supported the impression that Goro- ber, 1964, is shown in Figure 4. Epidemics maugo represented a high incidence area. Age occurred in the Upper Asaro during the months
Figure 4.-Geographical Distribution of Cases.
42 PAPUA AND NEW GUINEA MEDICAL JOURNAL
30 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
June to September of 1961 ; in the Tari Basin of jejunum, denuded of normal mucosa, was just prior to this; most of Baiyer River in the removed at laparotomy on the 24th day. The Western Highlands in March to June, 1962, and patient also had a rise and fall in beta antitoxin. in Bundi and the Chimbu throughout 1964. There was further eyidence of local spread in When pig feasting activities commenced in August when pork exchanges between clans at May in the Upper Chimbu, the prevalence of Mai and Goromaugo resulted in two further enteritis necroticans rose. There was a further acute cases (Nos. 150 and 155). The disease spread following the pig cycles in the Central was firmly diagnosed three times in members of Chimbu (Dom, Kup and Koronigl Census Div- one family (Nos 89, 90, 92, 93, 132 and isions) . Following the ‘Te’ festival of the Enga people of the Western Highlands late in 134) . A dietary history of persons consuming 1963, only six cases of the disease were reported. the same pork meal was recorded on 42 occasions Serological sampling earlier in the year confirmed and in only 17 instances (38 per cent.) were mild the impression that the Wabag-Wapenamanda symptoms of food poisoning recorded in persons area was a low incidence area. at risk.
Cl. perfringens Type C was isolated from cases Hospital admissions. at the Upper Asaro, Chimbu, Baiyer River, Wabag, Wapenamanda and Mendi. Bacterio- At the Kundiawa hospital for a 12-month logical examinations from cases elsewhere at period 1st December, 1963, to 30th November, Bundi, Kainantu, Henganofi, Tari and Minj 1964, admission for diarrhoea conditions under were not undertaken. categories 045-049, 571 and 785.6 of the W.H.O. International Disease classification, were A specific focus or geographical location of the disease cannot be defined and it seems likely reviewed. There were also 71 (1.4 per cent.) that the affection has always been endemic in admissions for mild and severe forms of pig-bel. New Guinea and that recognition and detection The two sets of figures are shown together in of the cause was not made until 1961-1962. Figure 5. Peaks in admissions for the disease Whether Cl. perfringens Type C has been intro- corresponded to the maximal rates for the diar- duced with the advent of civilization remains rhoeal disease admissions except in December, open to speculation. 1963. These periods were preceded by periods Evidence of local spread.
During 1964, seven cases of pig-bel were detected from clans near Goromaugo in the Upper Chimbu (Nos. 91 to 93, 97, 99, 101 and 155). Cl. perfringens Type C was recovered from three of these, significant beta antitoxin was found in two others and the diagnosis established at autopsy in two more.
Following the pig killing at Goromaugo there was a bridal exchange of pork between mem- bers of a Pagakaune line and one man, Ambane Umba (Case 96), of a Kurumogl line. Ambane subsequently developed a more protracted form of the disease which was confirmed by a rising beta antitoxin and recovery of Cl. perfringens Type C from his faeces. Sera taken from eight relatives exposed to the same meal and from the man who had prepared the meat all con- tained significant amounts of beta antitoxin.
A further case (No. 100) occurred in a man at Gena (Upper Chimbu) who had eaten pork originating from a line near Goromaugo. Con- tacts again had immunological evidence of exposure to beta toxin. A “lead pipe” piece
VOLUME 6, NO. 2, JULY, 1966 43
31 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
known to occur from October to December of 1962 in the Western Highlands. Gastro- enteritis was prevalent prior to the pig-killing near Baiyer River in April, 1962.
Most cases were seen during the months of June and August, 1961, in the Upper Asaro, in April and May of 1962 (Baiyer River), the Upper Chimbu in May and June, 1964, and the Central Chimbu later that same year.
Pig-feasting investigations.
Figure 6.—Analysis of 301 operations of laparotomy 1. Description of a pig-kill. Preparations performed at Goroka, 1961-1964. months in advance take place when long houses are built to accommodate residents and visitors of heavy pork consumption in areas within two round central courtyard clearings. Tables are hours walk of the hospital. The relative rise also built in readiness to receive the pork slabs in diarrhoeal diseases treated as compared with for distribution. In the Eastern Highlands two other diseases in May and September-October to four weeks before a large pig-kill, a smaller was statistically significant. celebration takes place, the smaller pigs being sacrificed on such an occasion. Operative records analysed for almost a four- year period (1st February, 1961, to 30th Novem- ber, 1964) at the Goroka hospital emphasised All large pig-killing ceremonies take place at the importance of enteritis necroticans as a sur- the time of a full moon. Whenever possible they gical disease in the Highlands (Figure 6). are held during favourable weather which is usually during the dry season. Dancing and Laparotomy was undertaken on 301 occasions singing festivities are held before the killing during this period. Sixty-nine patients (22.8 commences. In the Chimbu the pork prepara- per cent.) had acute and complicated forms tion and distribution takes place over a period of Pig-bel and this was the largest group of three or four days, whereas only a little over requiring surgical exploration of the abdomen. a day is needed in the Huri and Enga areas. For acute and sub-acute surgical conditions The description which follows is basically applic- excluding trauma, the frequency rose to 31.2 able to all groups except for the important dif- per cent. As a surgical disease, therefore, Pig- ference just mentioned. Pig-kills were witnessed bel was the most common cause of an “acute at Korfena (July, 1961), Tambul and Wapena- abdomen”. It is possible also that some of manda (September, 1963) and the Upper the causes of strangulation by post-inflammatory Chimbu (May, 1964). bands, listed in the second group of Figure 6, were also sequelae of the disease.
Seasonal distribution.
In Figure 7 admission to hospitals in the Highlands and Madang from enteritis necroticans are grouped at three-monthly intervals. Peaks occurred mostly in the ‘dry’ season between the months of April and September, and coincided with the larger pig-killing ceremonies held at this time of the year because of the prevailing climate and good harvest.
The relationship of Pig-bel and epidemics due to other diseases was not investigated in detail. A widespread influenza epidemic preceded the commencement of pig-killing in the Upper Figure 7.—Admisions for Pig-bel grouped at Chimbu in 1964 and a measles epidemic was three-monthly intervals.
44 PAPUA AND NEW GUINEA MEDICAL JOURNAL
32 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Plate I.—Tethered pigs are clubbed to death.
The tethered pigs are clubbed to death by a behind the anus and the inner two meet anterior near relative of the owner, four to five sharp to it. Here spillage may occur and the butcher’s blows to the head with a two to three inch hands can become contaminated directly by pig diameter stick usually being sufficient to kill the faeces (Plates II and III). The abdominal skin animal (Plate I). There is no bleeding of the flaps are dissected up towards the head and the carcase although some blood from the head thoratic cage opened laterally by axe cuts. A wounds spills out over the ground. Cassowaries, careful butcher then removes the diaphragm, chickens and dogs are sometimes killed along peritoneal sac and contents intact. However, with the pigs but are cooked separately. The sometimes the peritoneal cavity is penetrated by Chimbu people place the dead pigs in radiating the less adept and very occasionally bowel may lines from a central spirit or ‘Bolim’ house, be perforated at this stage. This evisceration rather like the spokes of a cartwheel, each spoke is preferably performed with bamboo knives, representing pigs to be distributed from one which being new, are more hygienic than man’s house. The clan leaders then welcome unwashed steel knives. Women take no part visitors and incantate upon the significance of in this process and sit aside watching the pro- the occasion. After about two hours, the butcher- ceedings, stray dogs, piglets, fowls and children ing starts. This was well under way by 10 wander at will about the dissectors, all contribut- a.m. in the Enga situation. The haste of Enga ing to a cumulative spoiling of the fresh meat. in preparing and cooking the pork is probably The boys also help mop up pooled blood due to the longer distances over which exchanges with a parsley type of plant Oenanthe jarvanica must take place and the greater number of pigs squeezing it into a bowl under the head. Thor- slaughtered. acic viscera, tongue and oesophagus are then removed and the solid organs separated. Not The hair is singed over an open fire and the infrequently, butchers had infected and dressed carcases then lifted onto leaf mats of banana, sores on their hands from which meat became breadfruit and tree fern leaves. An intricate contaminated. dissection in the dorsal position is commenced with double lateral incisions in the anterior The women take the bowels in their ‘bilums’ axillary line. The outermost continues down (string carrying-bags slung over the head) to
VOLUME 9, NO. 2, JULY. 1966 45
33 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 to the nearest stream for washing. The Chimbu the viscera is consumed soon after cooking. and Gahuku (Asaro) women evert the small Organs such as liver, spleen and kidney are intestine by intussuscepting a stick into the regarded as having special magic-religious signifi- bowel lumen, then siphon and blow water cance. Cooked bowel is eaten mainly by the through the large bowel, and anal skin and women and children. anus acting as a funnel. Enga fashion is a little less crude as the whole bowel is slit open After the butchering is completed, which is and cleaned more thoroughly. The bowels are in about two to four hours, the night is given plaited and wrapped in leaves ready for cook- to more singing and dancing. Earth pits are ing. Stomach is packed with chopped fat, greens dug, long and shallow for the half sides of and herbs and cooked in the form of a haggis. meat and deep and wide for the offal and prime It is eaten after ‘maturing’ for two or three cuts. The ovens are lined with banana leaves, tree weeks. While the bowel washing is in progress, fern fronds (Cyathea contaminans) and leaves the men continue the final filleting of the car- of a variety of breadfruit (Ficus dammaropni); case: Skull, rib cage and backbone being removed the meat, corn-cobs, taro, sweet potato and in one section. Here, contamination by feet was chopped greens are mixed with the pre-heated most apparent as assistants were required to pull stones. Men handling them use long wooden the head and spine forward and ventrally. tongs with great dexterity (Plate IV). Layer by layer the oven fills up until the large leaves Children eat raw morsels such as ear tips and roof over a final insulating layer formed by the snouts. Parents wrap entrails and sex organs pig’s quarters and flanks. Water is poured in about their wrists to promote future fertility. and the oven is again sealed by another layer Bladders on inflation are sometimes used as of leaves and sometimes mounded up with dirt, balloons or footballs. The extraneous blood is so that the food is cooked under steam pressure used as a garnish but the bulk is mixed in its own moisture. The juices of all the con- with chopped fat and leaves, rolled tents are thus retained in the cooking process. in a banana leaf and cooked over The Chimbu women also cook bowels and other an open fire as a blood sausage. Most of morsels in wooden barrels or ‘stone ovens’.
Plate II.—Technique of dissection.
46 PAPUA AND NEW GUINEA MEDICAL JOURNAL
34 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
to its destination, a half side of pork may change hands two to five times and reach consumption point one to four days later. A day or more may then elapse before the pork is further cut up and a second distribution held. The meat is re-heated in a smaller earth oven so that any prior bacterial spoiling becomes a potent culture with a high infective dose. It may be assumed that the cooking temperatures will be the same. The redistribution and piece-meal consumption of this meat increases further the possibilities of food poisoning occurring and it is probably this meat which causes the prolonged upper abdomi- nal pain of which so many people complain after pork eating (Nilles, 1950).
From the foregoing, it may be concluded that the likelihood of disease spread is much greater in the large pig ceremonies than in the much smaller marriage and death distributions, where there is a limit to the number of pigs killed and a smaller number of recipients. Bridal payments do take place within the framework
Plate III.—Showing opportunity for spillage during dissection.
2. Temperature observations. Temperatures failed to rise above 110 degrees Cent. in the centre of the ovens, this temperature being reached one-half hour after completion of the oven. Cooking time was between 1 ½ to 2 hours. The centre of a hind quarter immediately after removal from an oven after two hours’ cooking had a mean temperature of 78 degrees Cent. for five pieces sampled. The stones after cooking were a little hotter than warm and could be handled with bare hands.
The results indicated that large chunks of meat were not thoroughly cooked and the lower mean temperatures at Tambul were due to the shallow ovens which also contained relatively fewer hot stones. Following cooking, the meat rapidly cools, further handling occurs, and the same sources of contamination are present—feet, flies, dogs and so on. It is after cooking that contamination is most significant and the argu- ments and discussions concerning the distribu- tion may continue for a whole day. In transit Plate IV.—Showing use of long wooden tongs.
VOLUME 9, NO. 2, JULY, 1966 47
35 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 of the larger ceremonies and stacks of pork such detectable levels. In this normal popula- quarters and halves were seen at Goglme for tion there were significantly different levels this purpose. between individuals from high and low prevalence areas, and between this group and Sources of Infection. patients with Pig-bel and their contacts (Murrell et al., 1966). Males had a slightly greater 1. Pigs and pork. Cl. perfringens was immune status than females but the difference recovered from the intestinal contents of 53 and was not significant. These investigations 322 pigs chosen at random from pigs killed at together with a rise in beta antitoxin levels Korfena, Tambul and Goglme. All strains in the clinical disease lent further support isolated proved to be Type A and no Type C strains of this organism were recovered. From to the role Cl. perfringens plays in the aetiology 92 samples of cooked and uncooked pork, col- of Pig-bel. Specific Cl. perfringens Type C anti- lected mostly from the medial aspect of the hind- serum also significantly reduced case mortality. quarters where contamination was considered to be most likely, 15 isolations of Cl. perfringens DISCUSSION. Type A were obtained. Again no Type C strains were recovered. These samples included 18 The aetiology suggests that enteritis necro- cooked bowel specimens. On only five occasions, ticans or Pig-bel can be included in with the remnants of a pig (hair, bone and fat) sus- well-documented group of enterotoxaemic dis- pected as a source of origin for a case of enter- eases in animals. The mystery remains however itis necroticans, and Cl. perfringens Type A was as to whether Cl. perfringens Type C produces obtained each time but no Type C strains. The disease by exogenous ingestion or endogenous failure to enlarge the samples from this latter proliferation secondary to other local factors group was due primarily to the late arrival of in the upper intestinal tract. In animals cases to hospital and the difficulty in obtaining exogenous contaminated food types seem to meal remnants. All traces of meat had gone occur in young animals whereas stimulation of by the time attempts to get such samples were resident gut population by dietary changes pro- made. The general reluctance of the people duces the endogenous situation in older animals. to blame pork as a cause also added to this difficulty. The German outbreaks of Darmbrand had well 2. Human population. A one in ten sys- defined seasonal distributions over the years 1947- tematic sample of faeces of a normal human 1948, reaching pandemic levels during the mid population failed to yield any Cl. perfringens and late summer months of those years (Kloos Type C strains. This survey was carried out and Brummond, 1951). The disease was more independently of any pig feasting activity, prevalent in cities than in the rural areas. A although in the Upper Chimbu sampling at noteworthy difference in Pig-bel from its German Goglme, Goromaugo and Kurumogl was under- counterpart was the age distribution. In Darm- taken one month following the pig killing there. brand people in the fourth to sixth decades were The results were as follows :— most affected. The occurrence of the disease in childhood in New Guinea is readily explained by the dietary practice associated with pork feasting. With the larger pig cycles occurring at three to ten yearly intervals, it is possible that the younger age groups have an initial exposure to massive pork meals. Infants without teeth are not offered solid foods, and toddlers, who are breast fed up to the age of two and one half to three years are given only token amounts of pork. Only three patients in the series reviewed Serological Studies. were under two years. The belief that pork imparts strength into the individual is prac- In a normal population sample of 216 per- tically demonstrated by encouraging children, sons, 154 (71.3 per cent.) had 0.5 units or especially males, to consume as much as pos- more of Cl. perfringens beta antitoxin whereas sible. Particular organs, such as the genitalia, only 4 in a control group of 42 Europeans had liver and kidneys, are also reserved for children
48 PAPUA AND NEW GUINEA MEDICAL JOURNAL
36 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 to eat. Burchett (1964) reported to the author occurred at these times. A lot more anthropolo- that in the Baiyer River outbreaks in 1962 child- gical work will be necessary in order to trace ren actively refused further meals offered to the origins and destinies of pork transactions in them by their parents because of their fear of order to reveal the true epidemiological story. becoming ill. A significantly high level of Cl. perfringens The reluctance of older people to seek medical beta antitoxin was found in contact persons at care, the general regard that illness in the older risk. Perhaps some toxin neutralizing sub- generation is due to ‘old age’ and the lower stance in the bowel limits the distribution of life expectancy, account for the fewer case the disease in persons at risk. This has been reports in people over 40 years of age. The postulated in the case of food poisoning due to life expectancy of the Highlander is probably Cl. perfringens Type A (Goudie and Duncan, lower than that estimated in life tables by Scragg 1956) . (1954) for New Ireland natives of New Guinea. It is postulated that the pathogenesis com- The sex distribution in New Guinea corres- mences with the action of powerful toxins in ponded with that in Germany (2.2 males: 1 the upper intestine and this genesis is then female) . This distribution may be influenced stimulated by additional factors, intrinsic and by the greater frequency with which males seek extrinsic, favouring an overwhelming toxaemic medical care as indicated by a higher bed occu- infection of the host. The role that other faecal pancy rate both in Papua (Campbell and Arthur, and oral flora play in this toxaemia requires 1964), and the Territory as a whole (Depart- further investigation. Differential bacteriological ment of Public Health, 1964). analysis of organisms of the bowel flora such as B. coli, Paracolon sp., Bacteroides sp., Strep. More females contracted the disease than faecalis and Proteus mirabilis normally present males in the 16 to 20 year age group. At in bowel obstructed areas needs to be worked marriage feasts the bride and prospective brides out in the New Guinea disease. are encouraged to consume unusually large amounts of pork to encourage fertility. This The large pork meal, which may or may not possibly explains the distribution of the disease contain preformed toxin and a high helminthic in this group. infestation, possibly predisposes to motility changes, spasm on one side and stony on the The evidence that large-scale pig-killing other. This favours further bacterial growth activities influence the prevalence and spread of and toxin production setting in motion a chain enteritis necroticans in the Highlands is cir- of events which eventually leads to total bowel cumstantial. It would appear from the trends necrosis. in admission figures that this cultural practice influences the prevalence of diarrhoeal disease The beta toxin of the New Guinea strains in the localities of the hospitals reviewed. The possibility that coincident gastro-enteric infections when injected subcutaneously into guinea pigs due to a virus or other causes occurred has still gave a bluish gelatinous necrotic lesion. This to be excluded. In the presence of such a coin- was the basis of the toxin-antitoxin neutraliza- cidence, the spread of these pathogens would be tion method of determining antitoxin levels greatly assisted by consuming contaminated pork (Glenny et al., 1931). Zeissler maintained that segments involved in exchanges. The relative the variable pathogenicity in his Type F strains increase in hospital admissions for diarrhoea at was probably due to slight differences in toxin Kundiawa following this activity in two separate production by the organism and host suscepti- areas during two different times of the same bility. He believed that individual variations in year suggest an aetiological relationship. this isolate pathogenicity explained the wide range of clinical types of Darmbrand. Most A more accurate index of the morbidity and German writers on the subject concluded that Cl. mortality following pig-feasting could be obtained perfringens Type F was probably the initiating by measuring the diarrhoeal attack rate in a infective agent introduced in the diet. The population sample. Difficulties apparent in infection failed to establish itself unless other this approach were found to be a reluctance of dietary indiscretions occurred (Pietzonka and the people to implicate pork as a cause of Rassfeld-Sternberg, 1950). This situation existed diarrhoea, and the problem of accounting for in the post-war occupation of Germany, and is people before, during and after the feasting analogous to the dietary changes invoked by pork due to the transient migrations which naturally feasting in New Guinea.
VOLUME 9, NO. 2, JULY, 1966 49
37 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
McLennan (1962) In his excellent monograph EGERTON, J. R. AND WALKER, P. D. (1964). The on the histotoxic clostridial infections of man isolation of Clostridium perfringens Type C. from states “ Clostridia are at most facultative patho- necrotic enteritis of man in Papua-New Guinea. gens, able to gain a footing and produce those J. Path. Bact. 88 : 275. effects which we speak of as disease, only when FIELD, H. I. AND GIBSON, E. A. 1955). Studies on piglet mortality: Clostridium welchii infection. Vet. the tissues of the primary lesion have been Rec. 67 : 31. altered to the most important pre-requisite for clostridial infection (as distinct from clostridial GLENNY, A. T., BAR, M., LLEWELLYN-JONES, M., DALLING, T. AND Ross, H. E . (1933). Multiple contamination), that there must be present an toxins produced by some organisms of the Cl. area of lower oxidation-reduction potential.” welchii group. J. Path. Bact. 37 : 53.
GLENNY, A. T., LLEWELLYN-JONES, M. AND MASON, Because a direct food-borne epidemiological J. H. (1931). The intracutaneous method of test- sequence was not demonstrated bacteriologically, ing the toxins and antitoxins of the gas gangrene further research directed to answer such questions organisms. J. Path. Bact. 34 : 201. as to why some patients failed to develop ileus, GOUDIE, J. G. AND DUNCAN, I. B. R. (1956). whether other exciting stimuli were operative Clostridium welchii and neutralizing substances for before or after the pig meal, and why all par- Cl. welchii alpha toxin in faeces. J. Path. Bact. ticipants of a contaminated meal were not equally 72 : 381. affected seems indicated. GRINER, L. A. AND BRACKEN, F. K. (1953). Cl. per- fringens type C in acute haemorrhagic enteritis of SUMMARY. calves. J. Amer. Vet. Med. Ass. 122 : 99. HAIN, E. (1949a). Origin of Cl. welchii type F The cause of epidemic intestinal gangrene infection. Brit. med. J. 1 : 271. known as Pig-bel or enteritis necroticans is due HANSEN, K., JECKELN, E., JOCHIMS, J., LEZIUS, A., to invasion by Cl. welchii, in particular, to Type MEYER-BURGDORFF, H. AND SCHUTZ, F. (1949). C strains which produce beta toxin. This toxin Darmbrand-Enteritis Necroticans. Georg. Thiem, is lethal, necrotizing and probably induces Verlag, Stuttgart. arteriolar thrombosis and necrosis. The sequence KLOOS, K. AND BRUMMOND, W. (1951). Epidemi- of events leading to the establishment of this ologische Studien uber der Enteritis gravis in Nor- infection may involve a number of mechanisms, dosthalstein, assuer Lubeck. Zeitschr. f. Hyg. 132 : 64 because a simple food poisoning epidemiology McLENNAN, J. D. (1962). The histotoxic clostridial was not firmly established. The radical dietary infection of man. Bact. Rev. 26 : 177. changes invoked by pork feasting, subsequent MARCUSE, K. AND KONIG, J. (1950). Bakteriolische pork engorgement and pre-existing immunity Befunde be Jejunitis Necroticans. Zbl. Bakt. 156 : 107 probably influence the pathogenesis. This aspect MURRELL, T. G. C., EGERTON, J. R., ROTH, L., of the disease seems analogous to the entero- SAMELS, JANET, WALKER, P. D. (1966). Pig-bel— toxaemias of animals caused by Types B, C and Enteritis necroticans. A study in Diagnosis and D organisms of the Cl. welchii (Cl. perfringens) Management. Lancet 1 : 217. group. MURRELL, T. G. C. AND ROTH, L (1963). Necrotizing jejunitis - a newly discovered disease in the High- lands of New Guinea. Med. J. Aust. 1: 61. ACKNOWLEDGMENTS. I am indebted to the Director and to members of NILLES (1950). The human of the Chimbu region, the Papua and New Guinea Research Advisory Council Central Highlands, New Guinea. Oceania 21 : 25. for their guidance and help. Drs. R. Roderigue, OAKLEY, C. L. (1949). The toxins of Cl. welchii L Roth, C. Conner, C. Mathews, L Malcolm and Type F. Brit. med. J. 1 : 269. F. Smith graciously provided clinical data. PIETZONKA, H. AND RASSFELD-STERNBERG, L. (1950). I owe deep gratitude to Mr. J. Egerton and Dr. P. Weitere bakteriologisch gesicherte falle von Enteritis Walker who isolated and typed the strains of Cl. Necroticans in Hamburg und Dithmarsch hen perfringens and to Mrs. J. Samels of the Wellcome (Marz 1948 bis Marz 1949). Klin . Wschr. 28: 425. Laboratories also provided the specific Type C anti- SCRAGG, F. R. (1954). Depopulation in New Ireland. serum. Brit. med. J. 1 : 267. REFERENCES. SIEGMUND, H. (1948). Zus Pathologie un Patho- BENNETT, H. W. (1932). Infectious enterotoxaemia of sheep—West Australia. Aust. Counc . Sc. lnd . Res. genese der als Darmbrand bezeichneten Enteritis Bull., 57. necroticans. Klin. Wschn. 26: 33. CAMPBELL, C . H. AND ARTHUR , A. (1964). A study of 2,000 STERNE, M. AND WARRACK, G. H. (1964). The admissions to the medical ward of the Port Moresby types of Clostridium perfringens . J. Path. Bact. General Hosp. Med. J. Aust. 1 : 989. 88 : 279. DALLING, T. (1928). Lamb dysentery. Vet. Rec . ZEISSLER, J. AND RASSFELD-STERNBERG, L. ( 1949) 8 : 841. Enteritis necroticans due to Clostridium Welchii DEPARTMENT OF PUBLIC HEALTH (1964). Hospital Statistics, 1962-1963. Type F. Brit. med. J. 1 : 267.
50 PAPUA AND NEW GUINEA MEDICAL JOURNAL
38 PNG Med J 2005 Mar-Jun;48(1-2): 39-49
THE PATHOGENESIS OF PIG-BEL IN PAPUA NEW GUINEA
Gregor Lawrence FRACP*
Introduction:
In Western countries necrotising enter- -ary diseases with similar pathology, patchy itis is a relatively unusual disease, except necrosis of the upper small gut, also caused in the premature infant and apart from a by CWC. Clostridium welchii Types pro- dramatic temporally limited epidemic duce a number of different toxins. In involving people in Northern Germany1, the case of CWC â toxin is the most im- and Denmark at the end of World War II. portant, it is lethal when given to animals The recognition of a similar disease2 end- intravenously and causes local necrosis if emic in Papua New Guinea, was a very injected into the tissues.8 important finding both in regard to the pathogenesis of the disease and for health In the animal and human diseases the workers involved in treatment and pre- typical necrotic lesions are the result of vention. absorbed toxins, principally â, on the gut wall. â toxin is a protein that is ex- Pig-bel usually follows a high protein tremely sensitive to proteolysis. The meal, there is a latent period of some basic chain of events in pig-bel is consum- hours to some days before the victim ption of high protein food which leads develops abdominal pain, vomiting and to rapid growth of CWC, either contami- diarrhoea which often contains blood. nating the meat meal, or already present Diarrhoea is usually short lived and fol- in the gut lumen, with production of â lowed in moderate and severe cases by toxin that damages the intestinal wall. constipation as intestinal obstruction due to the damaged gut develops. The clinical This paper will consider the factors features of pig-bel are described in other involved in allowing this damage to take papers in this issue. place.
The incidence of pig-bel is very varia- The Organism ble. Using hospital admission data the incidence in areas served by the hospital Clostridium welchii is an anaerobic, changes greatly from year to year. In gram positive, spore forming rod that is about 7500 children aged 1 to 15 in Sina divided into types on the basis of the pat- Sina followed in the vaccine trial, there tern of exotoxins formed during growth. has been an annual incidence of 3.3/ CWC produces mainly á and â toxin,9 1,000/year over a two year period.3 Sina Sina is considered to be a high incidence it causes enterotoxaemias with similar area; in some other areas the incidence gut necrosis in many other animals as has been higher.4 Mortality in these pat- well as in man. The types of Cl. welchii ients was 12%, lower than most hospital cannot be distinguished by morphology 10 figures,2,5,6 however two of the six or on colonial appearance, Type A children involved died before reaching organisms are much more common than hospital. any of the other types. Cl. welchii is commonly a faecal organism, it multiplies The organism causing pig-bel has been and sporulates in the intestine of man and shown to be Clostridium welchii Type C other animals. Identification of types (CWC).7 There are number of veterin- can be done in mixed cultures by using fluorescent antibodies produced against *Institute of Medical Research, P.O. Box 60, the bacterial cells themselves.11 The Goroka, EHP. types have different surface antigens and
PAPUA NEW GUINEA MEDICAL JOURNAL 39
39 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 this method allows the easy demonstr- Cl. welchii Type A food poisoning is ation of CWC in the presence of a great characteristically caused by strains of Cl. numerical excess of Type A cells. welchii that produce an enterotoxin at the time of sporulation.15 CWC strains CWC has been shown to be ubiquitous have been shown to produce this entero- in the Highland environment, present in toxin too.16 The diarrhoea and vomiting the stools of over half the people and seen at the beginning of pig-bel may be their pigs and easily demonstrated in in some cases related to enterotoxin. The village soil samples.12 enterotoxin probably has no importance in the production of the characteristic Another aspect of the biology of CWC pathology seen in pig-bel. that is important in relation to pig-bel is its sporulation. CWC readily forms spores The Toxins that are relatively heat stable, withstand- ing heating for some minutes to over CWC produces the major toxins á and 95°C. In the case of the German disease â, other toxins may be produced too. á CWC was initially named Type F because toxin is a lethal necrotising toxin that is of its extreme heat resistance surviving a lecithinase, and it is produced by all boiling at 100°C for four to six hours.13 strains of Cl. welchii. Although enteritis The New Guinea strains are not as heat in fowls can be caused by Type A strains stable as this but stand the temperatures and Type A strains may contribute to reached in parts of the ground oven cook- the disease after gut damage by â toxin, ing used in the Highlands. The lower á toxin is probably not of major import- temperature of boiling water here (95°C) ance in the aetiology of pig-bel. also assists its survival. CWC has been found in the soil of mumu ovens in which â toxin is essential for the production of pig meat involved in a case of pig-bel was the veterinary enterotoxaemias17 and cooked. probably for pig-bel in man. It is a protein exotoxin made by CWC during active Spores of CWC exhibit the pheno- growth. It is lethal, necrotising and ex- menon called heat shocking. When a piece tremely rapidly destroyed by proteases.18 of contaminated food is cooked the The susceptibility of â toxin to proteolysis vegetative organisms are killed. As the is central to the present theory on the food cools the spores, not killed by the pathogenesis of pig-bel. temperature reached, all begin to grow at once. After a short time they begin to Purified â toxin has some interesting make spores again. This ability of Cl. wel- properties. When injected intradermally chii to form relatively heat resistant in to the skin of depilated guinea pigs a spores, the phenomenon of heat shocking necrotic patch develops. The appearance and the very rapid growth of Cl. welchii of these patches is very reminiscent of the necrotic patches seen in the gut in make it a frequent contaminant of food pig-bel - a yellowish centre surrounded and a common cause of food poisoning. by a haemorrhagic ring at its periphery. 14 Repeated heating and cooling of meat, as is often practiced in the village, is Given intravenously in mice, â toxin particularly dangerous as a multiplier is lethal. In doses of 1 µg death occurs in effect occurs which can lead to extremely forty minutes to two hours or so. If high levels of Cl. welchii on the food. So antibody is given intravenously the Highland meat may often be heavily animal can be protected, but only if the populated with vegetative and spore delay between giving toxin and antibody forms of CWC some of which will survive is short, ten minutes or less. This suggests passing through the stomach and begin that â toxin has its effect very quickly, growing rapidly in the small intestine. or that it is bound in the body and then
40 VOLUME 22, NO. 1, MARCH, 1979
40 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 becomes inaccessible to antibody. toxic filtrate containing high levels of penicillin and lincomycin produced small Many patients who develop pig-bel bowel pathology, showing that the toxin’s show a rise in â anti-toxin after the illness. effect was direct and the changes were The presence of detectable â antitoxin not due to secondary overgrowth of path- in the blood is thought to indicate pre- ogenic organisms already present in the vious pig-bel, in many cases of mild lumen. severity. But in Rooney and Shepherd’s series of patients who had a laparotomy The Role of Low Protease Levels and for pig-bel19 and in whom antibody Trypsin Inhibitors levels were measured, seven out of seven- teen did not develop raised â antitoxin When considering the pathogenesis of levels. This may be due to the small pig-bel the epidemiology is important. amount of â toxin involved in causing Papua New Guineans get pig-bel; Europ- tissue damage and to its rapid destruction eans resident in Papua New Guinea, with in the tissues by lysosomal and other one exception, do not. Judging from the proteases released with tissue death. â history of Darmbrand, as the disease in toxin has been said to reduce the motility Germany was called, and from sporadic of intestinal villi,20 perhaps initially cases in the literature, Europeans are not leading to local stasis in the small gut intrinsically immune. We have demonstr- and allowing more favourable conditions ated CWC in the stools of Europeans here, for toxin, being rapidly formed by org- so, presumably their food is sometimes anisms overgrowing in the bowel con- contaminated with the organism. Why tents, to cause more damage to the did people in Europe suffer with this mucosa and for attachment of organism disease over a few years only, while it to the villi. The motility changes des- continues in Papua New Guinea endemic- cribed20 were non-specific and were ally? One of the striking things about the accompanied by severe haemorrhagic German population at the time was a high effects in the mucosa. This work has not incidence of malnutrition; in the area separated any effects of â toxin on mot- where Darmbrand was very common ility from its general necrotising prop- about 70% 21 of the population was erties. clinically malnourished. Eating of unus- ually rich food was also thought to be a factor. In animal experiments the pathological lesions of necrotising enteritis can be Low protein diet and protein malnutri- caused by giving a toxic filtrate of CWC 17 tion lead to decreased levels of proteases containing toxins but no organisms. in the pancreatic secretions. The malnutri- In these experiments in the neonatal pig tion in Germany may have led to very low and in the guinea pig the conditions are levels of tryptic activity, so that â toxin not those found in the natural disease here. With guinea pigs the toxic filtrate in the gut lumen was not destroyed. In the Papua New Guinea Highlands the has to be directly injected into the small 22 bowel at laparotomy. The lesions could normal diet is very low in protein. Al- not be produced by intragastric dosing, though clinical malnutrition is not com- presumably because of destruction of mon, subnutrition is; children are small the toxin in the stomach and upper for age and will grow faster if their diet intestine by proteases. But the demon- is supplemented.23 stration that â toxin, without organisms, can cause full thickness necrosis of the intestine, Experimental work with guinea pigs is important when considering the showed that pig-bel-like lesions could be pathogenesis of pig-bel. produced if animals were given growing cultures of CWC intragastrica1ly with raw In recent experiments in Goroka, a soy-bean flour, which contains inhibitors
PAPUA NEW GUINEA MEDICAL JOURNAL 41
41 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 of trypsin, but not if the soy-bean flour of dietary inadequacy and reduced stimul- was first autoclaved destroying the inhi- ation due to low levels of protein in the bitor. Protein deficient guinea pigs had gut lumen, this is probably not possible. growing cultures of CWC introduced into the stomach with no effect. But, if the Using the gelatin film test I have shown culture was mixed with raw sweet-potato low levels of protease activity in the stools flour first, lesions very similar to those in of PNG children in Goroka Hospital, who pig-bel were produced and the animals had been on a normal village diet before died. If the sweet-potato flour was auto- admission compared with people on a claved first no lesions were produced.24 European diet.
Sweet potato also contains inhibitors The explanation for the non-appear- of trypsin which are stable at the temper- ance of pig-bel in Europeans in Goroka atures reached in normal cooking in the may be that their customary good or Highlands. The protection of â toxin over-nutrition leads to an abundance of from proteolysis afforded by the com- proteases in the upper small intestine, bination of low levels of trypsin and the easily able to destroy any â toxin being presence of trypsin inhibitors in the made there. The disappearance of Darm- staple diet probably explains the persis- brand in Germany would then be attribu- tence of pig-bel in Papua New Guinea25. table to improving nutrition and the re- Eating sporadic meals of meat provides turn of high levels of tryptic activity. a good medium for rapid growth of the organism as well as introducing sub- The case of pig-bel that occurred here strate excess to compete for any trypsin in a European, was said to have been in a present. Another contributing factor may heavy drinker, who had not eaten meat be the presence of trypsin inhibitors in for some time prior to taking part in a the secretions of Ascaris lumbricoides, village pig-feast. His nutrition may have which is commonly present in patients been poor and it is possible that his pan- with pig-bel. In the chick, infestation with creatic secretory capacity was low as a a similar parasite has been shown to delay result of chronic pancreatic disease. Other absorption of dietary protein somewhat sporadic cases of necrotising enteritis in suggesting that protein digestion was the Western world have occurred in peo- delayed, possibly due to the presence of ple malnourished because of illness or gut inhibitors. disease.29 The condition is well recogn- ised following gastric surgery where there Sugiura et al.26 have described heat is fasting and often physical by-pass of stable inhibitors of trypsin in sweet pot- the upper duodenum with reduced stim- ato. The inhibitor inhibits trypsin only, ulus to pancreatic secretion and poor but not chymotrypsin, the other major mixing of enzymes and gut contents. protease in pancreatic juice. In monkeys it has been shown that protein deficiency Guinea pig experiments have shown leads to a much more marked decrease in the importance of multiplication of org- chymotrypsin levels than in those of try- anisms in the upper gut. Animals did not psin.27 In fact, chymotrypsin secretion get pig-bel when dosed in tragastrically with ceased in the animals studied while tryp- a growing 4 hour culture containing high sin was still being secreted at lowered levels of toxin and a protease inhibitor levels. So, in Highlanders the sweet pot- aprotinin (Trasylol, Bayer), but in which ato inhibitor might effectively inhibit a the potential for further growth was limit- large proportion of their pancreatic pro- ed due to a lack of carbohydrate substr- teolytic activity. Animals on a high pro- ate. If the same procedure was followed, tein diet can compensate for the presence but the dose mixture contained autoclaved of dietary inhibitors by over-secretion of sweet potato and meat, lesions occur- proteases.28 But in New Guinea, because red. The presence of substrate for growth
42 VOLUME 22, NO. 1, MARCH, 1979
42 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 and inhibitor to protect â toxin was been a resident or transient in the upper necessary. CWC culture with autoclaved small intestine which then overgrew with sweet potato (inhibitors destroyed) alone a change in environment. One patient seen did not produce lesions after intragastric had eaten pig a week before becoming ill. dosing. The pig had been kept in the roof of his father’s house for some weeks and was Studies from Veterinary Diseases putrid. He showed no ill effects from eat- ing the pig, but a week later ate a very Bullen and Scarisbrick31 showed that large meal of marita soup made from the feeding of enormous numbers of pathog- pandanus palm, which is very high in fat enic organisms to sheep did not reproduce and presumably rich in protein. The next Type D enterotoxaemia except when the day he became ill with pig-bel. animal was overfed at the same time. It appears that those ingested organisms Malnutrition has been shown to in- that survive passage through the stomach crease the number of organisms present multiply for a limited period in the upper in the usually lightly contaminated up- small bowel. In the presence of a sudden per small intestine.33 Enteric parasites change of diet conditions are such that may harbour organisms in their intestinal more extensive proliferation and toxin tract, increasing the contamination in formation with subsequent damage can the host gut lumen. occur. There are two basic models for CWC enterotoxaemia in the veterinary Experiments Relating to Attachment of diseases: one where the organism is al- Clostridia to Villi ready present in the gut, the other where it is introduced on contaminated food. With many enteric pathogens such as In the sheep the CWC or Type B is a entero-pathogenic strains of E. coli, at- rumen commensal until a change in cond- tachment of the organism to the villous itions leads to sudden overgrowth with surface has been considered essential in increased toxin formation and the initi- the production of disease.34 In the case ation of disease. In the neonatal pig the of these organisms there is an area on the organism is eaten with milk because of organism that specifically binds to a part faecal contamination of the sow’s teats. of the villous surface. Proximity obvious- These organisms can grow rapidly in the ly enhances any effect secreted material piglet’s intestine where gastric and will have on the villi. Arbuckle35 has pancreatic secretion is very low. In addi- emphasised the importance of attachment tion, pigs obtain their maternal anti- in CWC enteritis in piglets. Examination bodies in colostrum rather than trans- of sections of gut in pig-bel and experi- placentally as in man. The colostral anti- mentally produced pig-bel in guinea-pigs bodies are protected from tryptic des- shows the villi covered with gram positive truction by the presence of colostral rods. However, Arbuckle has not shown trypsin inhibitors. Griner32 suggested that attachment precedes damage to the that this inhibition with subsequent villus tip by toxin, and it seems likely protection of â toxin was the basic cause that toxin damage precedes attachment of necrotising enteritis in piglets. by the organisms. In a series of experi- ments in Type A enteritis of chicks, Al- In pig-bel in man it seems likely that Sheikly36 showed that damage to the in many cases the CWC is a contaminant villus tip occurred within 15 minutes of on the protein food that precedes the instillation of a bacteria free toxin filtrate; onset of illness. However, we have seen the subsequent changes and mucosal cases where the food was most unlikely shedding examined by electron micro- to have been heavily contaminated, such scopy look very similar to the early as tinned meat or fish. In these cases and changes Arbuckle described. After toxin probably in others the CWC may have damage organisms can attach to the sur-
PAPUA NEW GUINEA MEDICAL JOURNAL 43
43 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 face and if proliferating there, the effects groups having a larger proportion with of the toxin being made will be maximal. circulating â antitoxin. In Germany the age distribution was different. Good Shutz37 working in Germany on nutrition and high trypsin levels protected Darmbrand was able to produce lesions the population until the war when priv- in experimental animals when given grow- ation with malnutrition removed their ing cultures if a damaging material, was enzymatic protection. Cases of Darmbr- given at the same time. This suggests that and occurred in all ages.21 mucosal damage, from toxin, parasites or physical causes, is necessary to begin pig- Obviously a large factor in the natural bel. antibody protection against pig-bel must be IgA immunity, both anti-toxin and Further work using washed organisms antibody activity directed against surface would be of great interest. To see if at- determinants on the organism. In tachment can occur without prior toxin veterinary practice immunisation with damage, and to see if more pathogenic toxoid and bacterins - material that raises organisms are characterised by the ability antibodies to the bacterial surface as well to attach to the mucosa. has been shown to give improved protecti- on. IgA antibody to gut pathogens has Immunity been shown to prevent specific attach- ment. Naturally induced immunity is The arguments regarding attachment probably directed against the toxin and have some significance in relation to the organism.29 immunity against CWC enterotoxaemia. Pathology IgG antibodies, produced after the in- jection of toxoids provide protection The pathology of pig-bel is striking. against the veterinary diseases and in Patchy necrotic areas are seen in the in- experimentally produced pig-bel. IgG testine with a strikingly banded appear- antibodies pass up the villus and leak ance- “Tiger-bel” (Smith). The banded out into the lumen playing a part in gas- appearance suggests a vascular basis for trointestinal immunity.38 Small amounts the lesion, but that the lesions are the of IgG present might prevent initial result of absorbed toxin seems more like- damage by protecting the villus in the ly. The typical antimesenteric necrotic vulnerable short period of rapid organism patches are full thickness necrosis of the growth and high levels of toxin producti- gut wall. Smaller spotty lesions which on following the initiating meal. If the are also full thickness necrotic areas are initial damage necessarily precedes attach- seen; these spots are tiny, only 2-3 mm ment followed by further multiplication in diameter, yet the necrosis extends right and toxin damage, then small amounts of through the gut wall.40 circulating IgG might be protective. We have not seen second episodes of pig-bel It does not seem possible that vascular in Goroka apart from one recent case blockage could cause these lesions. It where Mr. F. Smith operated on a patient seems more likely that they are the result with old resolving pig-bel who had a of tissue necrosis along the route of toxin couple of fresh looking necrotic areas absorption, perhaps from an area of dam- near the obstructive lesion caused by the aged mucosa or an area of local damage first attack. and organism attachment. Much of the descriptive pathology has referred to the That most pig-bel occurs in children patchy areas of necrosis in pig-bel as in- suggests a disease in which immunity is farction. But Roth6 noted in the begin- acquired. The serological findings of ning that the mesenteric arteries were Murrell2 support this, with the oldest age pulsatile right up to the gut. In areas of
44 VOLUME 22, NO. 1, MARCH, 1979
44 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 patchy necrosis thrombosis of vessels is must be caused by non uniform uptake seen in the area of separation between of toxin. Local damage to mucosa by necrotic and living tissue, blocked vessels parasites might play some part. Recent are found in the zone between complete work on the processing of antigens in necrosis and less heavily damaged tissues. the gut suggests another possibility. In general changes are more marked in Peyer’s patches are lymphoid tissue veins, the draining vessels where Cooke aggregates that take up gut antigens thro- considers endothelial proliferation to be ugh a specialised mucosa after which a hallmark of pig-bel. Although these lymphoid cells are stimulated to produce changes could cause the demonstrated IgA antibodies. These cells divide and necrosis it seems more likely that they travel up the thoracic duct and then home are changes occurring after toxin damage. on the gut wall where they populate the In guinea pig experimental lesions full lamina propria and produce surface anti- thickness necrosis very similar to the bodies.38 It has been shown in small human disease can be demonstrated in bowel biopsies in man that similar small animals killed 24 hours after exposure to lymphoid bodies occur in the upper growing cultures and trypsin inhibitors. jejunum.41 It is possible that lymphoid The vascular changes are not seen, sug- tissue rapidly takes up the very necrotis- gesting that, at least in the guinea pig, ing â toxin which then damages the tissue the necrotic areas are not the result of and surrounding areas allowing CWC to infarction following vascular obstruction. attach to the mucosa, leading to further toxin absorption and setting in train the The other aspect of the pathology chain of events leading to pig-bel, and that has not received enough attention is perhaps explaining the patchy distributi- the gross thickening and oedema of the on. gut wall in areas where full thickness damage has not occurred. This thickening Time course of Pig-bel lesions is at least partially a result of lymphatic obstruction. In pig-bel the mesenteric The temporal occurrence of the lesions glands draining the damaged area are in- in pig-bel has been discussed previously. 42 variably greatly enlarged and often hae- The changes in pig-bel occur in a re- morrhagic. Dilated tortuous lymphatics latively short period of time after the can be seen running over the serosal sur- offending high protein meal. The path- face of the gut, cystic fluid filled spaces ological chain of events occurs in a short are sometimes seen in the gut wall and time with overgrowth in the upper small in most cases when a specimen is tran- sected clear fluid oozes from the cut sur- gut, toxin damage, stasis, and attachment face. The lymphatic obstruction, which to damaged mucosa. The luminal con- also contributes to the nutritional pro- tents containing growing organisms, in- blems that develop in pig-bel, is probably flammatory exudate and tissue break- related to the effects of absorbed toxin down products could not be expected to along the course of the lymphatics. contain high levels of â toxin capable of initiating disease further down for long The patchy segmental distribution of periods of time when the time course of the lesions is very striking. On examinati- â toxin levels in culture is considered. In on of operative specimens, small ridges vitro cultures of CWC result in maximal of surviving tissue can be found, often levels of â toxin about 3½ to 4 hours surrounded by areas in which the mucosa after inoculation.43 The subsequent de- is totally necrotic. It is difficult to ascri- cline in â toxin has been attributed, apart be this distribution to vascular changes from exhaustion of carbohydrate for followed by infarction. If the initial lesion growth, to production of proteolytic en- seen in pig-bel is necrosis caused by absor- zymes in the culture by the organism bed toxin, then the patchy distribution itself.
PAPUA NEW GUINEA MEDICAL JOURNAL 46
45 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
â antitoxin given intravenously before, people, been present for a long time. The at the same time as, or two hours after brother of a present Sina Sina leader intragastric dosing with growing culture probably died of it about fifty years ago. and soybean flour protects the guinea The reasons for its persistence lie in part pig from induced pig-bel. Two of five in the dietary practices in the Highlands. guinea pigs given â antitoxin seven hours after dosing developed typical intestin- The diet is very low in protein. Growth al lesions of experimental pig-bel, even in Highland children is retarded, at least though the clinical disease in this situati- partly because of inadequate protein in on does not become apparent for at least the diet, although energy is low as well. 24 hours after dosing. Sweet potato (Ipomoea batatas) is the staple. It provides a very high proportion In most human cases the lesions are of the calories eaten. Sweet potato is localised and the subsequent history of eaten every day. It is tasty, bulky, high the case is the result of efforts to repair yielding and relatively easy to grow. The this damage. In some cases, where the trouble with sweet potato is that it forms small bowel is affected in its entire such a large proportion of the diet. length, the situation may be more analog- People have a steady intake of trypsin ous to gas gangrene with tissue invasion inhibitor with a very low protein diet. and multiplication of the clostridia. This The normal response to consumed inhi- is not common. After the initial damage bitors is pancreatic hypertrophy and other organisms in the gut may become compensatory over secretion, which may involved in invading the necrotic areas. be difficult with a low protein diet. Type A organisms from the faecal flora may play a part. The strain of CWC used Another factor possibly increasing the in producing experimental pig-bel, alth- effects of inhibitors is suggested by the ough it produces gross gut lesions, is not recent discovery of the recirculation of very pathogenic when a culture is injected trypsin and chymotrypsin.44 It has been into the thighs of mice. Progressive clos- shown in the rabbit that trypsin and tridial myonecrosis does not result, just chymotrypsin are absorbed whole in local gangrene of the leg without progres- the upper small intestine and quite rap- sive disease. Variation in toxin production idly secreted again. If this pertains to man and pathogenicity is well known in Clos- and is of reasonable magnitude, then the tridia. Some of the differences in clinical results of eating inhibitors coupled with cases must undoubtedly be due to dif- a very low protein diet may be even more ferences in the organism strains. marked. If the stored pancreatic enzymes available for secretion at the beginning Dietary aspects of a meal normally are recycled a number of times during that meal then binding As discussed earlier, Darmbrand occur- by an inhibitor early would prevent this red in Germany in a very malnourished and the subject would depend on enzy- population recovering from the hardships mes being made de novo when the capa- of war. The onset of the disease was city was reduced because of a very low often related to eating unusually rich food protein diet. In growing children the before the onset of symptoms. Meat and effects would be increased. high carbohydrate meals were often im- plicated. After about three years Darmb- All this is of importance because of rand suddenly became much less common the sporadic way in which meat is eaten and then virtually disappeared. The ex- in the Highlands. Pigs are large and they planation for this probably lay in improv- are not killed and eaten very often. Even ed nutrition with more proteolytic acti- in Tari Smith4 has found that eating pig vity in the gut. In Papua New Guinea pig- meat is for most of the population about bel remains, and has, according to local a once a week occasion. Infrequent meat-
46 VOLUME 22, NO. 1, MARCH, 1979
46 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 containing meals would contribute less organisms leads to further toxin damage nutritionally, than the same total amount and the necrotic lesions typical of pig-bel. if eaten more frequently over a period. They would also introduce substrate REFERENCES excess to the already inadequate proteoly- tic capacity. Most cases of pig-bel are the result of eating pig from small local 1. Hansen, K., Jeckeln, E., Jochims, J., Letzius, celebrations such as bridal exchanges A., Meyer-Burgdorff, H. and Schu”tz, F. rather than the large widely spaced pig- (1949) Darmbrand, Enteritis Necroticans. feasts. George Thiem Verlag, Stuttgart. 2. Murrell, T.G.C., Egerton,J.R., Rampling, A., If sweet potato formed a smaller pro- Samels, J. and Walker, P.D. (1966) The portion of the diet, other crops contri- ecology and epidemiology of the pig-bel buting more, I think pig-bel would be syndrome in man in New Guinea. J.Hyg, much less common. Camb., 64, 375-396. It is obvious that changes in cropping 3. Lawrence, G.W. Shann, F.A. Freestone, D. and eating habits are long-term matters. S. and Walker, P.D. (1979) The prevention The other possible dietary change, that of enteritis necroticans (pig-bel) in Papua of killing and eating of the pigs more New Guinea by active immunisation. Lancet frequently may actually be dangerous. in press. Smith has reported that pig is eaten at 4. Smith, D. (1979) Mortality from Pig-bel least once per week in a lot of Tari (Enteritis Necroticans) in children in Tari households. I think this may be more 1971-1976. P.N.G. Med. J. 22. often than is usual in Sina Sina. Yet, in Tari, the mortality from pig-bel is much 5. Shepherd. A. (1979) Clinical features and higher than in Sina Sina. More frequent operative treatment of pig-bel - enteritis eating of meat, if not enough to result in necroticans P.N.G. Med. J. 22. improved protein nutrition, and coupled 6. Roth, L. (1966) Diagnosis and management with persistence of the sweet potato of Pig-bel. P.N.G. Med. J. 9,51-54. staple and of CWC in the environment, may lead to more frequent pig-bel. 7. Egerton, J.R. (1966) Bacteriology of enter- itis necroticans in New Guinea Highlanders. Advice on changing dietary habits to P.N.G. Med. J. 9,55-59. eliminate pig-bel should be given with caution. 8. Sterne M. and Warrack, G.H. (1964) The types of Clostridium Perfringens. J. Path. Conclusion Bact. 88, 279-283. 9. Oakley, C.L. and Warrack, G.H. (1953) Pig-bel persists in the Papua New Routine typing of Clostridium welchii. Guinea Highlands because of a peculiar J. Hyg., 51,102-107. group of circumstances. CWC is common- ly present in the environment. Undernu- 10. Willis, A,T. (1969) Clostridia of wound in- trition, the consumption of inhibitors in fection. Butterworths, London. Page 41. the dietary staple and intestinal parasites 11. Yamagishi, T., Yoshizawa, J., Kawai, M., all tend to limit the levels of proteolytic Seo, N. and Nishida, S. (1971) Appl. Micro- activity in the upper small gut when CWC, biol. 21,787-793. either contaminating food, or already present in the intestine, multiplies in the 12. Lawrence, G.W., Walker, P.D., Garap, J. changed environment caused by a high and Avusi, M.G. (1979) The occurrence of protein meal. â toxin produced by the Clostridium welchii Type C in Papua New organism damages the surface of the gut. Guinea. P.N.G. Med. J. 22. Motility is impaired and attachment of 13. Zeissler, J. and Rassfeld-Sternberg, L.
PAPUA NEW GUINEA MEDICAL JOURNAL 47
47 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
(1949) Enteritis necroticans due to Clostri- Biophys. Acta. 328,407-417. dium welchii Type F. B.M.J. 1, 267-269. 27. Gyr, K., Wolf, R.H., Imondi, A.R. and 14. Hobbs. B.C., Smith, M.E., Oakley, C.L., Felsenfeld, O. (1975) Exocrine pancreatic Warrack, G.H. and Cruickshank, J.C. function in protein-deficient Patas monkeys (1953) Clostridium welchii food poisoning. studied by means of a test meal and an in- J. Hyg., Camb., 51,75. direct pancreatic function test. Gastroenter- ology, 68,488-494. 15. Dunclm, C.L. (1973) Time of enterotoxin formation and release during sporulations of 28. Liener, I.E. and Kakade, L.M. (1969). Clostridium perfringens Type A. J. Bact, 113, Protease inhibitors. In Toxic constituents 932-936. of plant food stuffs. Liener, I.E. ed. Academic Press, New York and London. 16. Skjelkvale, R. and Duncan, C. L. (1975) pp 7-68. Characterisation of enterotoxin purified from Clostridium perfringens type C. Inf. 29. Patterson, M. and Rosenbaum, H.D. (1952) Immun. 11, 1061-1068. Enteritis Necroticans. Gastroenterology, 21, 110-118. 17. Field, H.I. and Goodwin, R.F.W. (1959) J. Hyg. 57,81. 30. Williams, M.R. and Pullan, J.M. (1953) Necrotising enteritis following gastric surg- 18. Willis, A.T. (1969) Clostridia of wound ery. Lancet, ii, 1013. infection. Butterworths London. p 88. 31. Bullen, J.J. and Scarisbrick, R. (1957) 19. Rooney, J., Personal communication. Enterotoxaemia of sheep: experimental 20. Parnas, J. (1976) The effects of Clostridium reproduction of the disease. J. Path. Bact, perfringens beta toxin (Type C) on the 73,495-509. motility of intestinal segments in vitro. 32. Griner, L.A. (1963) Some factors influenc- Zentralb. Bact. Hyg., I. Abt. 234,243-246. ing the incidence ot enterotoxaemia in 21. Kloos, K. and Brummund, W. (1951) domestic animals. Bull. Off. int. Epiz, 59, Epidemiologische Studien uber die Enterit- 1443-1451. is gravis in Nordostholstein (ausser Lubeck) 33. Gracey, M., Subarjono, Sunoto and Stone, Zeit. Hyg. 132, 64-96. D.E. (1973) Microbial contamination of the 22. Bailey., K.V. and Whiteman, J. (1963) gut: another feature of malnutrition. Am. J. Dietary studies in the Chimbu (New Guinea Clin. Nutr, 26, 1170-1774. Highlands). Trop. Geogr. Med., 15,377-388. 34. Jones, G.W. and Rutter, J.M. (1972) Role 23. Malcolm, L.A. (1970) Growth and Develop- of the K88 antigen in the pathogenesis of ment in New Guinea. A Study of the Bundi neonatal diarrhoea caused by Escherichia People of the Madang District. Institute of coli in piglets. Inf and Imm 6, 918-927. Human Biology, Papua New Guinea. 35. Arbuckle, J.B.R. (1972) The attachment of Monograph Series No. 1, page 86. Clostridium welchii (Cl. perfringens) Type 24. Lawrence, G. (1975) Recent advances in C to intestinal villi of pigs. J. Path, 106,65- pig-bel. Med. Soc. of P.N.G. 11th Sympos- 72. ium, 27-33. 36. Al-Sheikly, F. and Truscott, R.B. (1977) 25. Lawrence, G.W. and Walker, P.D. (1976) The pathology of necrotic enteritis of chic- Pathogenesis of Enteritis Necroticans in kens following infusion of crude toxins of Papua New Guinea. Lancet, 1, 125. Clostridium perfringens into the duodenum. Avian Dis, 21,241-247. 26. Sugiura, M., Ogiso, T., Kazuyuki, T., Sumi- 37. Shutz, F.(1949) Atiologie und Pathogenese. hiro, T. and Akira, I. (1973) Studies on in: Darmbrand. George Thiem, Stuttgart. trypsin inhibitors in sweet potato, I. Puri- fication and some properties. Biochim 38. Walker, W.A. and Isselbacher, K. J. (1977).
48 VOLUME 22, NO. 1, MARCH, 1979
48 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Intestinal Antibodies, NEJM. 297,767-773. on pig-bel. P.N.G. Med. J. 21, 113-116. 39. Pierce, N.F. (1978) Intestinal antibodies. J. 43. Pivnick, H., Habeeb, A.F.SA., Gorenstein, Inf. Dis. 137,661-662. B., Stuart, P.F. and Hauschild, A.H.W. (1964) effect of pH on toxinogenesis by 40. Cooke, R.A. (1979) The pathology of pig- Clostridium perfringens Type C. Can. J bel. P.N.G. Med. J. 22. Microbiol, 10, 329-344. 41. Phillips, A.D., Rice S.J., France, N.E. and 44. Gotze, H. and Rothman, S. (1975) Enter- Walker-Smith, J.A. (1978) Lancet, i,454. opancreatic circulation of digestive enzyme Letter. as a conservation mechanism. Nature, 257, 42. Lawrence, G. (1978) Further speculation 607 -609.
PAPUA NEW GUINEA MEDICAL JOURNAL 49
49 PNG Med J 2005 Mar-Jun;48(1-2):50-57
STRONGYLOIDES SPECIES INFESTATION IN YOUNG INFANTS OF PAPUA NEW GUINEA: ASSOCIATION WITH GENERALIZED OEDEMA John D. Vince, Richard W. Ashford, Michael J. Gratten and Joseph Bana-Koiri
SUMMARY.
Clinical and laboratory details of thirteen infants with “Swollen Belly Sickness” who had abdominal distension, respiratory distress, generalised oedema, and variable disturbance of gastrointestinal function were analysed. The intimate connection between the syndrome and infection with Strongy- loides species closely resembling Strongyloides fulleborni - is discussed, and a standard treatment using thiabendazole, plasma transfusion and antibiotics, is proposed.
The Kamea people, known also as the tems were performed were stated as Kukukuku clan, inhabit difficult moun- pneumonia in one, and peritonitis in the tainous terrain some fifty miles north of other. Heavy infestation of the stools of Kerema, in the Gulf Province. The inac- affected infants with hookworm was cessibility of their territory, combined noted, but its relationship to the swollen with their reputation as fierce warriors, belly remained unclear. resulted in their relative insulation from the influence of exploration until the As a result of the persistence of the early 1960s, when Roman Catholic mis- Kanabea Mission staff, and their convic- sion stations were established in Kaintiba tion that the “Swollen Belly Sickness” and Kanabea. The opening of Menyamya, was indeed a specific entity, it was de- Kaintiba, and Kanabea airstrips had led to cided, early in 1977, to renew attempts rapidly-increasing Mission and Govern- to determine its aetiology and to define ment influence in the area. Since the intro- the syndrome in more detail. Several duction of health services, health workers affected infants were therefore transferred have consistently reported the presence to Port Moresby Hospital for full invest- of an apparently previously unknown gation. The discovery of extremely heavy sickness affecting young babies, which infection of the stool of one such infant failed to respond to treatment, and with a Strongyloides species led to the which was almost invariably fatal. The reappraisal of past and subsequent cases sickness, for which the name “Swollen which forms the basis of this paper. Belly” was used, consisted of respiratory distress, abdominal distension, and gen- Case: This male infant was the parents’ eralised oedema. It was both common, fourth child, the previous three having and well-recognised by the Kamea died, apparently with the “Swollen Belly people, who regarded the appearance of Sickness”. He presented at Kanabea the swollen abdomen as the end of all clinic at the age of two to three months, hope of the infant’s survival. Early having been unwell for approximately attempts to find the cause of the sick- two weeks with intermittent diarrhoea. ness were inconclusive. The causes of On examination he weighed 4.3 kg. death in two infants on whom post-mor- (approximately 80% standard weight for age), and appeared well nourished and Port Moresby General Hospital. Department of Child Health and Pathology, University of well hydrated, but ill. There was no aden- Papua New Guinea Faculty of Medicine. opathy, cyanosis, pallor or pyrexia, but marked pitting oedema of the limbs and
120 VOLUME 22, NO. 2, JUNE, 1979
50 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 back was noted. The abdomen was grossly weak, he was otherwise well. distended but not tympanitic, there was no splenomegaly, but the liver edge was Methods. palpable 3 cm. below the costal margin. Shifting dullness was present, and bowel All thirteen patients studied were said sounds were high-pitched but not ob- by the Kanabea Mission staff to be typ- structive. Respiratory rate was raised and ical of the Swollen Belly Sickness. Six there was mild intercostal indrawing. Only of these patients were seen personally a few fine crepitations were heard at the (JV) - four at Port Moresby General bases. Pulse rate was normal. The anterior Hospital and two at Kanabea Mission. The fontanelle was normotensive, but the infant Hospital records of five patients referred had a staring expression reported by the to Kerema Hospital were studied, as Kanabea staff to be typical of such infants. were the record and post mortem report Haemoglobin was 10.5 gm/100 ml and of one patient referred to Port Moresby eosinophil count was 24%. Blood culture was Hospital in 1975, and the post mortem negative and the cerebrospinal fluid was report of one baby who died in Kanabea, crystal clear, with no cells or organisms on whose body was subsequently transfer- microscopy. The serum was noted to be red to Port Moresby. An analysis was extremely turbid. Total serum protein was made of the presenting symptoms and subsequently shown to be 3.5 mg/100ml, signs, duration of illness, laboratory data, with albumen of 0.2 mg/100 ml. Abdominal outcome and, where applicable, post paracentesis produced opalescent fluid. mortem findings of these babies. Sepa- Examination of the greenish stool revealed rate analyses were made of those babies masses of ova, (404,000 eggs/ml) many in who were, and who were not, treated with chain-like formation, diagnosed as those of thiabendazole. Histological sections from a Strongyloides species. three post mortem examinations were reviewed by Dr. G. Aitken, Pathologist, Treatment with thiabendazole (approx- Port Moresby General Hospital. imately 25 mg/kg/twice daily for 3 days) was commenced, and antibiotics with- Results. held. There seemed to be some improve- ment in the child’s condition initially but 1. Patients and Outcome. the abdominal distension and oedema persisted, and because of subsequent There were six male babies and seven slight deterioration in condition antibiotic females. Of the nine for whom weights therapy was commenced four days after were available, one was below the 60th admission. On the sixth day, although percentile, three below and five above the the respiratory signs had improved, the 80th percentile of standard weight for baby looked very unwell, the abdominal age. The median age of presentation was distension had increased slightly, and approximately eight weeks, with a range there was no improvement in the oedema. of two weeks to approximately six months. The total duration of illness for nine of There was no hepatomegaly, and there the cases with available data varied from were no other signs of cardiac failure. six days to seven weeks. Six out of seven 150 ml of plasma was given. At this stage cases who presented prior to the intro- the baby appeared to be in a terminal duction of specific treatment with thia- condition and, in accordance with the bendazole died. All those who presented parents’ wishes, treatment was discon- after the introduction of thiabendazole tinued and he was allowed home, presum- treatment survived, although there was ably to die at home. Eleven days later the no record of one case having received this child was brought back to the clinic: there drug. was no abdominal distension and no oedema and, though he appeared a little 2. Symptoms and Signs
PAPUA NEW GUINEA MEDICAL JOURNAL 121
51 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
122 VOLUME 22, NO. 2, JUNE, 1979
52 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
3. Laboratory Investigations. able. In three patients abdominal X-rays were reported as normal but in one of (a) Haematological. Details are shown these there was progression to non speci- in Table III. fic changes of slightly distended bowel loops with some small fluid levels, and (b) Stool. Five babies were reported one showed distension of the large bowel. as having ova of Strongyloides species in the stool, and five were reported to have 4. Patients not treated with thiabendazole. hookworm ova (in one case the report was of hookworm or Strongyloides). Seven patients presented prior to the Results of specimens were not available discovery of Strongyloides described in from the two babies who died in 1974 case 1 and did not receive thiabendazole and 1975. therapy. There was no record of one patient, who presented after the discovery, (c) Serum Proteins. Three patients in having received treatment. Of those eight whom oedema and ascites were promi- patients, six died, five between the ages nent had total serum proteins of 3.2,3.5, of two and three months. The sixth, who and 2.7 gm/100 ml with albumens of 0.8, was also exceptional by virtue of lack of 0.2 and 0.2 gm/100 ml, while a fourth patient respiratory signs, died at the approximate in whom oedema was not a feature had age of seven months. All eight had a dis- values of 5.4 and 1.4 gm/100 ml. tended abdomen, and seven had signs of respiratory distress. Peripheral oedema (d) Cerebrospinal Fluid. Results were was a feature of all six who died, but was available from five patients. In three there absent in at least one of the survivors. were no leucocytes. In two the fluid was Ascites was known to be present in five bloodstained, but there appeared to be a patients, of whom four, who also had relative lymphocytosis the results being:- pleural effusions, died. Pericardial effu- sion was found in two fatal cases. Hook- Polymorphs Lymphocytes Red cells worm was reported in the stool of three, 1 22 “numerous” and Strongyloides in two. Reports for 7 18 69 two of the fatal cases were not available, and the stool of one of the survivors (who Cultures of these fluids showed no growth. had ascites) was initially reported as negative for ova. All eight patients received (e) X-Ray Findings. Signs of consoli- treatment with antibiotics, and two dation were present in five of the six received blood. patients with Chest X-ray reports avail- 5. Patients treated with thiabendazole.
PAPUA NEW GUINEA MEDICAL JOURNAL 123
53 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Of the five patients who received thia- number of ova which superficially resem- bendazole, none died. All had a distended ble those of hookworm, but which, on abdomen, although in one case this was closer examination (see below), are found described as minimal at the time of admis- to be those of Strongyloides. Treatment sion, and all had respiratory distress. with antibiotics may produce initial Gross peripheral oedema and ascites were improvement, and the oedema may be present in two cases - both seen person- temporarily relieved following blood or ally. All were reported to have either plasma transfusion: but in the majority, hookworm or Strongyloides in the stool. in the absence of anti Strongyloides treat- All patients received antibiotics. The two ment the course is that of deterioration patients with oedema received plasma, to death. and one also received blood. It seems that abdominal distension is 6. Post mortem findings. the initial sign, and that the appearance of peripheral oedema is - in the absence Post mortem reports on three babies of adequate treatment - a pre-terminal gave the cause of death as pneumonia in event. Other features reported but not two cases and peritonitis in one. Small adequately documented are the presence and large bowels appeared macroscopic- of a staring expression, and a rather high - ally normal, although in one case the pitched cry. mucosa appeared pale. A review of the histological sections from all three post The features thus described are indivi- mortems failed to reveal evidence of dually of an unspecific nature, with systemic strongyloidosis. Unfortunately, diverse aetiologies: viewed together, how- sections of small and large bowel were ever and occurring at this age, they form available from only one, and in this case a specific and highly unusual clinical no evidence of mucosal involvement was syndrome. Of particular interest and of seen. almost certain aetiological significance is the finding of nematode ova in the stool Discussion. of ten of the eleven patients examined. The distinction between the ova of hook- From the present analysis, it is possible worm and Strongyloides is, to the un- to describe a typical patient with fully- trained eye, not easy, and it seems likely developed “Swollen Belly Sickness” as that the ova seen in the stools of all ten an infant of either sex, aged around two patients were in fact those of Strongy- months, presenting with symptoms of loides. This seems all the more likely in abdominal distension - which apparently view of the finding - fully described in a may, at least in the early stages, be inter- Separate communication (Ashford, R.W., mittent - and respiratory distress, and et al 1) - of a very high prevalence of probably with mild diarrhoea and occa- Strongyloides infection in infants from sional vomiting. Examination will confirm the area inhabited by the Kamea people. the presence of abdominal distension, with evidence of both gaseous distension The reported absence of ova in one and shifting dullness, and there will be affected case who survived in spite of the obvious signs of respiratory distress. The presence of ascites and the absence of child may or may not be febrile, and peri- thiabendazole treatment, is puzzling, pheral oedema is present. The haemo- though it is pertinent to point out that globin level is probably in the low normal the presence of Strongyloides ova in old range, and there is probably an eosino- samples may be surprisingly easy to over- philia. Chest X-ray may show minor look. pneumonic changes. Microscopic examin- ation of the rather loose greenish stool The Strongyloides species has been will reveal the presence of a very large characterised previously (Vince et al 2).
124 VOLUME 22, NO. 2, JUNE, 1979
54 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
It is distinguished from Strongyloides loides stercoralis. stercoralis by the absence of the larval form in fresh stool, and by certain mor- The precise aetiology of the respiratory phological differences which render it signs and symptoms remains speculative. very similar to Strongyloides fulleborni Such features, in conjunction with the and to the species found in the Kiunga very heavy intestinal infection with the district and described by Kelly (3). et al Kanabea Strongyloides is suggestive of It is suggested that infection with this autoinfection, but we have no evidence Strongyloides species is intimately asso- that this process occurred in our patients. ciated with the Swollen Belly Sickness. Whilst abdominal distension may in itself cause respiratory distress in infants the The pathological relationship between clinical and radiological signs indicated the “Kanabea Strongyloides” infection definite respiratory pathology, and the clinical features, however, remains at present speculative. One indication Larvae of Strongyloides stercoralis have probably lies in the presence of oedema, been found in the meninges in association an important, and previously almost with bacterial meningitis (Owor et al 10) always pre-terminal, feature. Examination and it is tempting but purely speculative of the available data excludes anaemia and to ascribe the cerebrospinal fluid changes heart failure as its prime cause. The very found in two of our patients to larval low total serum protein and serum al- migration. bumin values found in three babies indi- cate that the oedema is hypoproteinaemic As far as we are aware, there are no in origin. Deficient protein intake in this reports of Strongyloides infections occur- age group in this situation is highly un- ring in patients as young as ours. The likely. Renal protein loss is unlikely, appearance of extremely heavy intestinal renal histology being normal in three infection with signs and symptoms com- cases, and proteinuria being absent in one patible with larval migration in babies as grossly oedematous baby. young as two and three weeks of age sug- gests to us a method of transmission other Malabsorption has however been asso- than that of skin penetration commonly ciated with Strongyloides stercoralis in- found in nematode infections. Trans- fections with in some cases, alterations mammary passage of nematode larvae of small bowel histology (Yoeli et al 4, from adult milk to suckling young in Bartholemew et al 5, and O’Brien 6) and other mammals has been amply docu- oedema with hypoproteinaemia was a mented (Stone W. and Smith F.W.11) and, feature in some of the cases described by in the case of Strongyloides ransomi Huchton and Horn 7. infection in pigs, results in a fatal disease of piglets characterised by many of the We have, at present, no specific histo- features we have described (Moncol D.J. logical evidence of bowel involvement in and Batte E.G.12). Brown and Girarde- our patients, but it is difficult to escape au13 recently reported finding larvae of the conclusion that intestinal function is Strongyloides fulleborni in one of 113 significantly affected by the extremely samples of milk from twenty-five mothers heavy infection found in our patients; followed from birth, suggesting at least and although a malabsorptive state is the possibility of human trans-mammary possible, it is equally likely that a protein- infection with this nematode. Other losing state results. Such a state has been possibilities - such as transplacental pass- shown to occur in pigs infected with a age - exist, and these, together with the Strongyloides species (Giese et al. 8), more conventional mode are discussed and Laudanna et al 9 have similarly elsewhere (Ashford et al1), Obviously demonstrated enteric loss of Cr. 51 albu- the determination of the mode of trans- men in one patient infected with Strongy- mission is crucial to the institution of
PAPUA NEW GUINEA MEDICAL JOURNAL 125
55 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 preventive measures. 1. Thiabendazole 25 mgs./kg./dose twice daily for three days. Of the six patients who died, there was post mortem evidence of bacterial infec- 2. Plasma, whole blood, or stabilised tion in three. These findings, and the plasma protein solution, by intra- transient improvement of several patients venous infusion at 40 mls./kg., with following treatment with antibiotics indi- frusemide 0.1 mls./kg. imi. cate a definite but probably secondary role for infecting organisms in the natural 3. Broad spectrum antibiotics. history of the illness. It is pertinent that a) Crystalline Penicillin 50,000units/ most of the babies who have died from kg six hourly IVI or imi. the swollen belly syndrome have done so at the age of two to three months, a time b) Chloramphenicol 25 mgs/kg six when humoral antibody defences are low. hourly IVI or orally. If there is indeed a protein-losing state as we have postulated, the host defences ACKNOWLEDGEMENTS. may be further reduced by loss of both immunoglobulins and lymphocytes from This work has its origin in the concern of the bowel. Sister Gwen Daw and members of the staff of Kanabea Mission for the Kamea people, and in In view of the high prevalence of the their persistence that the Swollen Belly Sickness Kanabea Strongyloides infection found was a definite entity. We gratefully acknowledge in infants from the Kamea area to which the help and encouragement given to us by Sister allusion has already been made (Ashford Gwen and the Mission staff under the leadership 1 of Father Cyril Blake. Several of the Medical et al ), there remains another problem, Officers based in Kerema have greatly assisted namely, the discovery of the factors deter- us, and in particular we thank Dr. Bill Miles, Dr. mining which babies will, and which will Andy Hall and Dr. Ron Anderson. Many of not, develop the Swollen Belly Sickness. our colleagues have contributed by virtue of All our patients were apparently in good encouragement, discussion, and direction, and health prior to the illness, and the possi- we specially thank Professor Biddulph, Depart- bility of underlying debilitative states is ment of Child Health, and Professor Wyatt, unlikely. It was not possible to obtain an Department of Community Medicine, Medical adequate family history in many cases, Faculty, University of Papua New Guinea. We and although one patient had one, one are indebted to Dr. Graeme Aitken, Pathologist at Port Moresby General Hospital, for his review four, and one three affected siblings all of the post mortem histology, and to the Path- of whom had died, this cannot be taken ology Department and Dr. Andrew Buck, Medi- as evidence of a familial predisposition. cal Faculty of the University of Papua New The possibility that “Swollen Belly” Guinea for the Protein Estimations. We are babies are for some reason more heavily grateful to the Papua New Guinea Public Health infected than their compatriots is Department and the University Medical Faculty discussed elsewhere1. At present, the for allowing us to undertake this work. Finally problem remains unsolved. we thank Mrs. Raka O. Natera for her secretarial help. In conclusion, we believe that the “Swollen Belly Sickness”, a hitherto unde- REFERENCES scribed clinical entity, is intimately asso- ciated with infection with a species of 1. ASHFORD, R.W., VINCE, J.D.,GRATTEN, Strongyloides closely resembling Strongy- M.J. and BANA-KOIRI, J.: Strongyloides loides fulleborni. As a result of our experi- Infection in a Mid Mountain Community ence, we recommend that the standard Papua New Guinea Med. J. 22: 128, 1979. treatment of the patients should consist 2. VINCE, J.D., GRATTEN, M.J., BANA- of KOIRI, J. and ASHFORD, R.W. The
126 VOLUME 22, NO. 2, JUNE, 1979
56 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Kanabea Story: Swollen Baby Syndrome 8. GIESE, W., DEYHAZRA, A and ENIGK, and Strongyloides infection. First Report K.: Enteric loss of plasma proteins in to Papua New Guinea Public Health Depart- Strongyloides infection of pigs. 1nt. J. ment. March 1978. Parasitol. 3,631-639, 1973. 3. KELLY, A, LITTLE, M.D. and VOGE, M.: 9. LAUDANNA, A.A, POLACK, M., BETA- Strongyloides fulleborni - like infections RELLO, A and KIEFFER,J.: Evidence of in Man in Papua New Guinea. Am. J. Trop. Protein losing enteropathy in Strongyloid- Med. Hyg. 25 (5), 694-699,1976. iasis Rev. 1nst. Med. Trop. Sao Paulo 15, 222-226, 1973. 4. YOELI, M., MOST, H., BERMAN, H.H. and SCHEINESSON, G.P.: The clinical picture 10. OWOR, R. and WAMOKUTA, W.M.: A and pathology of a massive Strongyloides fatal case of Strongyloidiasis with Strongy- infection in a child. Trans. R. Soc. Trop. loides larvae in the meninges. Trans. R. Soc. Med. Hyg., 57 No 5. 346-352, 1963. Trop. Med. Hyg., 70 (5-6), 497-499, 1977. 5. BARTHOLEMEW, C., BUTLER, AK., 11. STONE, W. and SMITH, F.W.: Parasito- BHASKAR, A.G. and JANKEY, N.: Pseudo logical Review. Infection of Mammalian obstruction and a sprue - like syndrome hosts by milk-borne Nematode larvae: a from strongyloidiasis. Post graduate Medi- review. Experimental Parasitology, 34,306- cal Journal 53, 139-142, 1977. 312, 1973. 6. O’BRIEN, W.: Intestinal Malabsorption in 12. MONCOL, D.J. and BATTE, E.G.: Trans- colostral infection of newborn pigs with acute infection with Strongyloides Ster- Strongyloides ransomi Veterinary Medicine/ coralis. Trans. R. Soc. Trop. Med. Hyg., 69, Small Animal Clinician 61,583-586, 1966. No 1. 69-77, 1975. 13. BROWN, R.C. and GIRARDEAU, M.H.F.: 7. HUCHTON, P. and HORN, R.: Strongy- Transmammary passage of Strongyloides sp. loidiasis. Journal of Pediatrics, 602 - 608, larvae in the human host. Am. J. Trop. 1977. Med. Hyg., 26, No 2. 215-219, 1977.
PAPUA NEW GUINEA MEDICAL JOURNAL 127
57 PNG Med J 2005 Mar-Jun;48(1-2):58-65
STRONGYLOIDES INFECTION IN A MID—MOUNTAIN PAPUA NEW GUINEA COMMUNITY
Results of an Epidemiological Survey
R.W. Ashford,* J.D. Vince,** M.J. Gratten** and J. Bana–Koiri**
SUMMARY Following our finding that an acute, fatal disease of infants is associated with massive Strongyloides infection, studies are reported here on the status of the infection in the community concerned. The parasite was found to be abundant in children from 3 weeks to 5 years, but apparently rare in adults. Correlations were sought, but not found,between egg output and serum pro- tein levels, albumin–globulin ratio and eosinophil counts. Very heavy infec- tions were seen without apparent disease. Despite the examination of milk and placentae, the mode of transmission to infants remains unknown.
Strongyloides spp. has been found to In Africa, S. fulleborni is thought usually be associated with generalized oedema in to be a monkey based zoonosis4, though young infants in Papua New Guinea1. it also occurs as an anthroponosis5. Brown and Girardeau6 showed it to be Epidemiological surveys of many highly prevalent in infants in Zaire, but types have been conducted over a large also to infect adults freely. They discovered proportion of Papua New Guinea. How- a small number of larvae in breast milk, ever, there are many geographically strongly suggesting this as a means of isolated communities and the results of a transmission to infants. Pathogenicity survey in one place may not be applic- has only been described by Pampiglione able quite close by. The most significant and Ricciardi7 who infected a volunteer, recent stool surveys are those of Kelly2, who suffered a variety of non specific, in which more than 4000 stools were though rather uncomfortable symptoms. examined from 19 localities throughout the country. Egg-producing Strongyloides The survey described here was designed were only found at 3 localities, close to to investigate the prevalence of Strongy- sea level, in Western Province. The nearest loides in the Kamea people, and to look of these is 500 km from Kanabea, which for other information bearing on the is in a quite different climatic zone, at transmission of the parasite and its path- 1350 m above sea level. Both free living ogenicity. and parasitic female worms were collec- ted, and these were described by Kelly, MATERIALS AND METHODS Little and Voge3 who showed them to be similar, though not identical to S. Fulle- Kanabea Mission was visited between borni. No studies were carried out on November 28 and December 5, 1977, pathogenicity, and infants were not and patrols were carried out to Paina and included in Kelly’s survey. Dr. R. L. Manimango, 8 and 4 hours walk distant. Muller (pers. comm.) has found similar eggs in stool samples from Irian Jaya. Most of the people attending the clinics at these centres were known to the mission * Dept. of Pathology, U.P.N.G., P.O. Box staff, and detailed histories were available. 5623, Boroko, Papua New Guinea. ** Port Thus, infants and children below 5 years Moresby General Hospital, Free Bag, Boroko, could be aged accurately, and for women, Papua New Guinea. details were recorded of their reproductive
128 VOLUME 22, NO. 2, JUNE, 1979
58 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 history and the survival of their offspring. people from whom serum was collected (Spencer method). In response to the reluctance of many people to donate faeces, no attempt was Placentae were collected by the mission made to select donors. As many specimens nurses from some of the few births occur- as possible were collected from people ring at the clinic. They were preserved in of all ages. Faeces were collected in con- 10% formol saline and forwarded to Port tainers which had been distributed the Moresby for examination of H. & E. previous day; a thick and thin blood film stained sections. was taken from each faeces donor. Many faeces samples were examined fresh, in RESULTS order to identify people with heavy in- fections for treatment, and for culture of Relevant Aspects of the Area and People the worm. All samples were also preserved in 10% formalin. Helminth eggs were Kanabea is at 1350 m; the surrounding counted in saturated NaCl, in a 0.5 ml area is steeply hilly, from 1000 m to 2500 counting chamber, but this was unsatis- m, and is completely forested except near factory for Strongyloides eggs, as they the few settlements and tracks. The few were attached together in strings, and rock outcrops are mostly Miocene lime- were also too heavy to float. Counts were stone. The annual rainfall is around 4500 therefore repeated after thorough emulsi- mm and is not distinctly seasonal. Cloud fication of the sample in 10% NaOH; then and mist are often continuous for weeks. either a 0.5 ml sample, diluted 20 or 100 No temperature records are available. times was counted on the floor of the chamber, or a 0.1 ml sample diluted 10 The most conspicuous feature of the times was counted under a 22mm2 cover- people is their small stature. Their super- glass. The highest count for any species ficial good health is apparently belied by in any sample was recorded. their extremely poor rating in nutritional surveys8. No detailed nutritional obser- Faeces cultures were set up on damp- vations were made, but the formal diet filter paper, and also on charcoal and consisted entirely of vegetables: bananas sterile soil. Cultured worms were fixed Musa spp, taro Colocasia esculenta, sweet - with hot 10% formalin. Parasitic female potato Ipomoea batatas, pumpkin Cucur- worms were expelled in the stool of treated bita spp, and a wide variety of “bush patients, and were subsequently collected cabbage”, mostly young leaves of Cucur- by searching with a dissecting microscope, bitaceae and Chenopodiaceae in small and transferred to 70% ethanol. quantities. High protein foods are very rarely eaten, though pigs and dogs are Milk samples were collected by nursing present in small numbers, as well as a few mothers following careful swabbing of semi-feral cats. Wild animals are hunted the breast, and were preserved by adding avidly, and are very sparse. Small birds, formalin to approximately 10%. rats and frogs, often eaten raw, provide most of the animal protein. Serum was collected from heavily in- fected infants, and from various other Recent innovations include the intro- people. So many infants were infected duction of cows, beans, a good walking with Strongyloides that matched controls track joining villages 4 days’ walk apart, were unobtainable. Sera were screened cotton clothes and law and order. These for total protein (biuret method) and for changes have met with limited acceptance, globulin and albumin levels by electro- and only materially affect people in the phoresis. immediate vicinity of the main centres. Improved access has allowed a greater Haemoglobin was estimated in those number of people to leave in search of
PAPUA NEW GUINEA MEDICAL JOURNAL 129
59 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 work, and a growing awareness of the rest and was almost universal in people age of the world is enhanced by the increas- above 12 months. The egg counts ing number of strangers who visit and were, however, rarely very high; the come to work in the area. highest count was 12,000 eggs/ml, and only 3 were above 5,000. The high counts While no demographic studies were were all in children aged below 5 years. made, some indication is given by the No attempt was made to identify the fact that the oldest women seen were still hookworm eggs specifically, though they bearing children, and by the following varied considerably, and in 2 cases resem- data taken from the clinic records: 108 bled Trichostrongylus. youngest children, selected at random from Kanabea and 4 other villages had a Ascaris infection was almost as pre- total of 337 siblings, of whom 140 were valent as that of hookworm, but heavy dead. Mothers with 4 or 5 dead babies loads were again infrequent. The pre- were commonplace. Pneumonia, malaria valence in young children was less than and swollen belly syndrome were thought that of hookworm, the prevalence of 80% to be responsible for a large proportion being reached around 5 years. of this infant mortality. Trichuris infection was surprisingly Helminth survey rare, and only very light infections were seen. The youngest infected child was One hundred and forty nine stool sam- aged 2-3 years. ples were examined. Thirty eight of these were from adults, and many of the rest Enterobius eggs were unusually abun- from infants. Three samples were anomal- dant in stools. Twenty six infections were ous in consistency, and were discarded as seen, and many of these were quite heavy. being probably incorrectly aged. In 3 cases there was insufficient material for Strongyloides eggs were passed in enor- the second, accurate count of Strongy- mous numbers (up to 106/ml) by nearly loides eggs. The results of the egg counts, all infants. In the faeces of infants aged and prevalence rates are shown in tables less than 12 months, these were usually 1 and 2. held together in membranous tubes con- taining as many as 50 eggs in a single or Hookworm infection was abundant, double strand, and often making up a
Table 1. Kanabea, Decemeber 1977: Stool Survey: Age Graded Helminth Infection Prevalences (%)
130 VOLUME 22, NO. 2, JUNE, 1979
60
Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 Table 2. Kanabea, December 1977: Stool survey: Summary of Egg Loads of Main Helminth Infections. Helminth Main of Loads Egg of Summary survey: Stool 1977: December Kanabea, Table 2.
PAPUA NEW GUINEA MEDICAL JOURNAL 131
61 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 large proportion of the faecal matter neither with active nor passive case detec- (Fig. 1). Faeces of 2 babies seen with tion procedures, no conventional epidemi- “Swollen Belly Sickness” had 404,000 ological conclusions can be drawn except and 200,000 eggs/ml. While these were that both P. falciparum and P. vivax are not the heaviest infections seen, they were present in the area. There was no correl- the heaviest in their age group. Infected ation detectable between malaria and children in all age groups up to 10 months Strongyloides infection or the other para- had a mean load above 20,000 eggs/ml. meters measured. The overall mean for children below 2 years, including those uninfected, was Eosinophils 45,400 eggs/ ml. In older children both the prevalence and load decreased dram- Counts were made on 116 blood films. atically and, with a single exception, in As these were not related to total white those older than 5 years, only light infec- cell counts their interpretation is difficult. tions were seen. Only 5 adults passed Eosinophils varied between < 1% of white eggs, and these had very light loads. cells and 60%. Although 2 adults had counts above 20%, most of the high Malaria counts were in children. As shown in table 3, there is no clear age correlation One hundred and twenty nine films with eosinophilia, though the youngest were examined. Six showed malaria para- children, <2 months of age had high counts: sites, and of these, 5 were from 33 films mean = 21.5%. Within the age grades from Minimango, the lowest of the 3 there was no correlation between Strongy- villages studied, at 1000 m above sea level. loides egg counts and eosinophilia, though As the blood film collections conformed the picture may be blurred by the paucity
Plate 1. Faeces sample from infant aged 2 months with swollen belly syndrome. Strings of eggs of Kanabea Strongyloides constitute a large proportion of the solid material.
132 VOLUME 22, NO. 2, JUNE 1979
62 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Table 3. Kanabea, December 1977: Age Graded Milk Eosinophiol Counts The sediment from 400ml of milk from 40 mothers was examined. Although vari- ous contaminating organisms, an ant and a mite were found, no helminths were seen. Very few samples were available from mothers of new babies, and no colo- strum has been examined. Placentae Sections of 4 placentae have been examined, but no worms have been found. Treatment and Reinfection Longitudinal studies in these circum- stances are very difficult, and must largely rely on opportunism. The results to date, shown in table 4 indicate that treatment of light infections in infants, and the with thiabendozole drastically reduces, abundance of other helminths in adults. and in most cases eliminates Strongy- loides egg output within 48h. Long term studies have been started to see if treated Sera babies become reinfected. The small number of such cases studied to date do Serum protein estimates were obtained not give a reliable indication either way. for 35 individuals, including seven adults and 23 children aged less than two years. The Parasite Three of the infants were suffering from swollen belly syndrome and had levels of The Strongyloides parasite involved will 3.5, 5.4 and 2.7 gm/100ml. One other be fully described when specimens are infant, with a moderate Strongyloides available from a wider geographical area. infection, had a very low level, 3.7 gm/ Those seen so far indicate that the para- 100ml; this child also had a heavy Plas- sitic females are indistinguishable from modium falciparum infection. S. fulleborni, but that the ‘free living adults differ slightly from those of this species Among the remaining 31 “healthy” as well as from the free living females subjects, the protein values ranged from described by Kelly et al. 3 For the present 6.2 to 9.1gm/100ml. There was no signi- they cannot be identified, and should be ficant difference between the values for referred to as “Kanabea Strongyloides”. subjects with Strongyloides (x = 7.8gm/ 100ml, n = 20) and those without (x = DISCUSSION 7.74gm/100ml, n = 9). Nor was there any indication of correlation between the The most significant findings to date are serum protein levels and the faecal egg that infection with Kanabea Strongy- output (r = 0.24, t = 1.05, d.f. = 18:p > loides is universal in infants in the area 0.1). studied, commonly in massive, apparently asymptomatic infections, and that patent Albumin globulin ratios were measured infection in adults is rare and light. Efforts for 22 sera. The ratios varied greatly from to indicate the transmission mechanism 0.57 — 2.59. but as with the total serum are inconclusive so far. protein there was no correlation with Strongyloides egg output. The high prevalence of the infection
PAPUA NEW GUINEA MEDICAL JOURNAL 133
63 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Table 4. Karabea Strongyloides: Egg Output Post Treatment*
*Expressed as % of pre—treatment value.
raises the question of why only a small the response of the syndrome to thiabenda- proportion of infants become acutely ill, zole treatment. Although the sick infants and indeed, whether the worms are res- did not have the heaviest egg loads, they ponsible for the Swollen Belly Sickness. did have the heaviest for their age. Possibly It is reasonable to postulate that such it is a heavy infection combined with the heavy infections must be in some way in- early age which promotes pathogenesis. herited from the mother, despite negative There is also the question of migrating findings in milk and placentae, and not- larvae and autoinfection. Lung aspirates withstanding in rarity of eggs in adult were not attempted, and infant sputum faeces. The pattern of hookworm infec- was not available. Small quantities of tion is so different that soil—borne per- peritoneal fluid did not contain worms, cutaneous infection is most unlikely to and no worms were found in l0ml of be the only route of Strongyloides trans- blood from the mother of a swollen baby. mission. Clearly further studies are indicated, The lack of correlation between serum and will be carried out, on the geograph- protein levels or albumin globulin ratios ical distribution of the infection, its and Strongyloides egg output in “healthy” transmission and pathogenesis, and on the subjects may indicate that the syndrome taxonomy of the parasite. is precipitated when some threshold is crossed, or that indeed, the syndrome is ACKNOWLEDGEMENTS unrelated to intestinal infection. The strongest evidence for Strongyloides being We would like to acknowledge the responsible for the disease, remains the unstinting, and continuing cooperation
134 VOLUME 22, NO. 2, JUNE, 1979
64 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 of the Kanabea Mission staff, both medi- M. Strongyloides fulleborni-like infec- cal and non medical. In particular, Sister tions in man in Papua New Guinea. Am Gwen Daw and Rosemary Byrne were J Trop. Med Hyg . 25:694-699, 1976. responsible for the success of our visit. Dr. Andrew Buck kindly repeated the (4) PAMPIGLIONE, S. and RICCIARDI, M.L. The presence of Strongyloides fullebomi; serum protein estimates by a micro-biuret von Linstow, 1905, in man in Central and technique. The Research Committee of East Africa Parassitologia, 13:257-269, University of Papua New Guinea and the 1971. Department of Public Health provided funds. (5) HIRA, P.R. and PATEL, B.G. Strongy- loides fulleborni infections in man in REFERENCES Zambia. Am J Trop Med Hyg, 6:640- 643,1977. (1) VINCE, J. D., ASHFORD, R. W. GRAT- TEN, M. and BANA-KOIRI J. Strongy- (6) BROWN R.C. and GlRARDEAU M.H.F. Transmammary passage of Strongyloides loides species infestation in young infants sp. larvae in the human host. Am J Trop of Papua New Guinea Association with Med Hyg, 26:215-219,1977. generalized Oedema Papua New Guinea Med. J. 22:120-127,1979 (7) PAMPIGLIONE, S. and RICCIARDI M.L. Infezione sperimentale nell’uomo con (2) KELLY, A. Alimentary Parasites of Man Strongyloides fulleborni da ceppo umano. in Papua New Guinea, Papua New Guinea Parassitologia, 13:505-511,1971. Institute of Medical Research, 56pp. (mimeo), undated. (8) LAMBERT, J.M. National Nutrition Sur- vey. P.N.G. Government National Planning (3) KELLY, A., LITTLE, M. D. and VOGE, Office, 9pp. (mimeo), 1978.
PAPUA NEW GUINEA MEDICAL JOURNAL 135
65 PNG Med J 2005 Mar-Jun;48(1-2):66-72
BY JOHN BIDDULPH AND PETER PANGKATANA* Weanling Diarrhoea GASTROENTERITIS is a disease of major Lawson and Curtis (1967) described 280 importance for young children in Papua and children admitted with gastroenteritis to the Port New Guinea. It was the second commonest Moresby General Hospital over a seven-month cause of admission to Administration hospitals period (March to October 1964). There were in Papua and New Guinea of children aged 1 to eight deaths in this series, giving a mortality 11 months for the year ended 31.3.1967, and rate of 2.8 per cent. A presumed causative or- comprised 24 per cent of the hospital admis- ganism was only isolated in 15 per cent of sions in this age group. Children hospitalised cases; Shigella flexneri type 2 was the most with gastroenteritis in this age group had a common organism isolated. Approximately 70 mortality rate of 3.4 per cent. It was the second per cent of the Shigellae isolated were resistant commonest cause of death for children in the 1 to sulphonamides. Severe malnutrition was said to 11 month age group who were admitted to to be present in one-seventh of the cases, while hospital, and accounted for 17 per cent of the many others were said to have probable growth hospital deaths in this age group. retardation due to malnutrition.
Gastroenteritis was the commonest cause of Biddulph (1966) described an epidemic of admission to Administration hospitals in Papua gastroenteritis seen at the Port Moresby General and New Guinea of children aged 1 to 4 years Hospital that involved hospitalisation of 305 for the year ended 31.3.1967, and comprised 17 children during a two-month period (June to per cent of the hospital admissions in this age July 1965). There were two deaths in this group. The mortality rate in the 1 to 4 year age series, giving a mortality rate of under one per group of children hospitalised with gastroenteri- cent. Stool cultures were mainly negative bac- tis was 2.3 per cent. It was the second com- teriologically, and virus studies carried out by monest cause of death for children in the 1 to 4 the Queensland Institute of Medical Research year age group who were admitted to hospital, were also unrewarding. . and accounted for 17 per cent of the hospital deaths in this age group. (Hospital Disease All three of the above studies discussed the Statistics 1964-1967.) low level of isolation of etiological agents des- pite intensive bacteriological investigation. The Unfortunately the age groups arbitrarily association of malnutrition with many of the selected for statistical purposes—under one cases of gastroenteritis was commented on. This month, 1 to 11 months and 1 to 4 years—mask theme was developed by Ryan and Murrell the relative importance of gastroenteritis in the (1964) who stated that the vast majority of age group 6 months to 18 months, during which their cases of marasmus presented as cases of age period gastroenteritis is probably the com- gastroenteritis. Lawson and Biddulph both monest cause of admission to hospital; and prob- stressed the importance of a standardised treat- ably accounts for at least one third of the hos- ment schedule using 2.5 per cent Dextrose in pital admissions in this age group. half strength Darrow’s solution as the intra- venous rehydration fluid of choice to keep the There have been several previous descriptions mortality rate to a minimum. of gastroenteritis in young children in Papua and New Guinea. The purpose of the present paper is to discuss the concept of the epidemiological entity, wean- Ryan (1962) described the clinical features of ling diarrhoea. Weanling diarrhoea is the gastro- 506 children admitted with gastroenteritis to the enteritis that occurs with the transition of babies Port Moresby General Hospital over a 13- from a completely breast fed existence to that of month period (May 1960 to June 1961). There a mixed diet. Gordon (1969) says ‘Acute were 22 deaths in his series, giving a mortality diarrhoeal disease in developing countries seem- rate of 4.5 per cent. Definite malnutrition was ingly owes its importance, in clinical effect and stated to be present in one-fifth of the cases. as a cause of death, to the particular form of the It was noted that gastroenteritis was a more disease called weanling diarrhoea, a clear severe disease in the malnourshed and that mal- synergism of nutrition and infection.’ It is a nutrition was often precipitated by recurrent syndrome associated with a particular age group gastroenteritis. and determined epidemiologically by the nutri- tional and environmental stresses attendant on *Port Moresby General Hospital. dietary change (Gordon 1963 et al.).
VOLUME 14, NO. 1, MARCH, 1971 7
66 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PRESENT SURVEY After satisfactory rehydration the child’s weight was plotted on the WHO Western Pacific During the 12 months May 1969 to April Regional Office (WPRO) weight chart. These 1970, 654 children were admitted to the Port weight charts have been especially prepared for Moresby General Hospital with gastroenteritis. assessing the nutrition of children in developing There were five deaths, giving a mortality rate countries of the Western Pacific Region (WHO, of under one per cent. One hundred consecutive 1969). It was found that 48 per cent of the cases of gastroenteritis seen during February- children had weights that fell below the mal- March 1970 were assessed for nutritional status. nutrition line (Figures 1 and 2).
8 PAPUA AND NEW GUINEA MEDICAL JOURNAL
67 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Analysis of these 100 cases showed some than females are admitted with the disease. Yet notable differences between those adequately more females than males are affected in the nourished and those malnourished. malnourished group who have gastroenteritis. Age and Sex This is a reflection of the greater prevalence of malnutrition among girls than among boys. Hospital disease statistics and previous surveys Fifty-eight per cent of the cases of gastroenteri- of children with gastroenteritis at Port Moresby tis occurred in the 12-month age group, 6 to 17 General Hospital have shown that more males months (Figure 3). As previously mentioned
VOLUME 14, NO. 1, MARCH, 1971 9
68 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Figure 3.- Age and sex distribution of 100 consecutive cases admitted with gastroenteritis, Port Moresby General Hospital, February-March, 1970 this concentration of cases in the age group 6 to one day, while the majority of the malnourished 17 months is hidden when the usual statistical group required intravenous rehydration for age groupings are used. longer than one day.
Duration of Rehydration
All the children in this series required intra- venous rehydration. An approximate assessment of the length of time and recurrence of the diarrhoea is given by the number of days each case required intravenous rehydration (Figure 4). The one adequately nourished child who required intravenous rehydration for four days was intolerant to lactose and had watery diarrhoea whenever he ingested milk containing lactose.
Figure 4.- Numbers requiring intravenous rehydra- The majority of the adequately nourished tion according to nutritional status and duration group required intravenous rehydration for only of rehydration
10 PAPUA AND NEW GUINEA MEDICAL JOURNAL
69 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Lactose Intolerance
Only one of the adequately nourished group, approximately 2 per cent, was intolerant to lac- tose. Among the malnourished group 12, i.e., 25 per cent were intolerant to lactose. Associated Conditions Other illnesses were more frequently present among the malnourished group (Table 1). DISCUSSION
Duration of Hospitalisation These differences in gastroenteritis according to the nutritional state of the child support The number of days spent in hospital was Hardy’s statement (1959) that in general there different for the two groups (Figure 5). The are two major types of gastroenteritis in develop- one adequately nourished child who remained ing countries. One type is the acute episode in the previously normal, well-nourished child, in which complete recovery usually occurs after a few days. The second type of gastroenteritis is seen in children with malnutrition. The diarrhoea is not confined to a single episode, but persists or recurs, often leading to a progres- sive downward course.
Historical Perspective
At the beginning of this century weanling diarrhoea was prevalent in countries that today comprise the technologically advanced nations in hospital for more than a week was the child (Gordon et al. 1963-Table 3). The death rate who was intolerant to lactose. Over one-third from diarrhoea in infancy in New York in 1961 of the malnourished children absconded from was over 100 times lower than it was in 1900. hospital, so that many of them stayed in hospital Similarly the death rate from diarrhoea in a shorter time than considered necessary. children aged 1 to 4 years in 1961 was more than 150 times lower than it was in 1900. Many factors such as improved nutrition, better Type of Malnutrition environmental sanitation, a favourable economy and more rational therapy have contributed to The type of malnutrition seen in the mal- this dramatic fall in the death rates from nourished group is shown in Table 2. diarrhoea among young children in today’s tech- nologically advanced countries. It can be seen from this table on age specific death rates that the figures for New York in 1900 were just as bad, and probably worse, than the present day figures for developing countries.
Age of Maximum Prevalence
Gastroenteritis mainly occurs in the age period 6 to 18 months, the time during which the Figure 5.- Duration of hospitalization according child is introduced to food other than breast to nutritional status milk. Weaning begins within the same general
VOLUME 14, NO. 1, MARCH, 1971 11
70 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 period that gastroenteritis appears–6 to 18 nourished child. The converse also occurs, an months–although the exact timing of the two episode of diarrhoea often precipitates kwashior- events differs from one community to another. kor. Both infection and malnutrition interact with each other, and the interaction is syner- Gordon et al. (1963) showed in his longitu- gistic (Scrimshaw et al. 1968). dinal study of young children in 11 Punjab villages that the age group with the highest Growth of the Weanling diarrhoea death rate was the age group 6 to 17 months (Table 4). Breast fed babies in developing countries grow as well as babies born into the privileged societies of U.S.A., Europe or Australia during the first months or so of life. Thereafter breast milk by itself is no longer sufficient for continued adequate growth, and marked retarda- tion of growth occurs.
A recent survey of children under the age of five years living in an urban squatter settle- Value of Breast Feeding ment in Port Moresby demonstrated this very clearly (Biddulph 1970). In the under six- The mortality from gastroenteritis among month age group approximately half the babies young children of developing countries would weighed more than the Boston median. The be much higher than it is, if it were not for the three malnourished babies in this age group practically universal practice of breast feeding. were all artificially fed. After the age of six In Gordon’s longitudinal study in the Punjab, months, retardation of growth occurred; pro- 20 infants were artificially fed from birth and gressively fewer babies weighed more than the only one survived past the age of one year. An Boston median, and progressively more babies infant mortality rate among the artificially fed became malnourished, until in the two-year age of 950 per 1000, compared to an infant mor- group no child weighed more than the Boston tality fate of 120 per 1000 for those who were median, and over half the children had weights breast fed. below the WPRO malnutrition line (Figure 6).
The increasing trend of urbanized mothers to Synergism between Infection and Malnutrition exchange the breast for a bottle means that gastroenteritis and malnutrition will continue to rise in the urban areas as more babies are fed The introduction of contaminated foods in dilute mixtures of concentrated bacterial content the weaning period explains the increased inci- in feeding bottles. dence of gastroenteritis during this period, but does not explain the high mortality. This is only explainable in terms of the synergistic interaction Risks to the Weanling between malnutrition and infection. As malnu- trition develops because of the poor diet in the The first introduction to food other than weaning period gastroenteritis becomes more breast milk brings two risks. However careful likely to lead to death. At the same time, gastro- the mother is, the family diet lacks the sanitary enteritis worsens the child’s nutritional state by quality of breast milk. Usually, however, envir- reducing appetite, increasing metabolic loss of onmental sanitation is such that many infectious nitrogen and frequently by ill-advised therapeu- agents are introduced to the child through con- tic restrictions on diet. The stage is thus set for tamination of the food. The second risk is the the vicious cycle of diarrhoea-malnutrition. usual practice in developing countries for the child to be fed foods of poorer dietary quality It is frequently impossible to say which came than breast milk; often he is restricted by cus- first, the diarrhoea or the malnutrition. Each has tom to bulky low protein foods. a deleterious effect on the other, and may be caused by or, in turn, cause the other. Because This transition to adult food is of no conse- these two conditions are inextricably wedded quence among well nourished groups who during the weaning period it is both foolish and understand and practice sound principles of in- misleading for statisticians to assign one cause fant nutrition. In communities where dietary only as the primary diagnosis for hospital ad- education is lacking and suitable protein food mission or death. Malnutrition is frequently ig- is not fed to young children the results are nored in national statistics because of this absurd serious. The malnourished child reacts more statistical convention. Respiratory infections or severely to gastroenteritis than does the well gastroenteritis are more obvious and dramatic
12 PAPUA AND NEW GUINEA MEDICAL JOURNAL
71 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
First, the introduction of foods other than breast milk exposes the child to potential patho- gens. The low standard of environmental sanita- tion in developing countries ensures that the young child receives a large dose of these poten- tial pathogens.
Second, the weaning period coincides with the period of growth retardation brought about in developing countries by the dietary inadequacy of what young children are fed, mainly due to ignorance and prejudice. This state of nutritional deficiency leaves the child more susceptible to infection.
Third, gastroenteritis produced by contact with contaminated foods itself results in further deterioration of nutritional status. It is the synergistic interaction between infection and poor nutrition during the weaning period that forms the epidemiological entity termed wean- ling diarrhoea. It is weanling diarrhoea which makes gastroenteritis so important an illness of young children in Papua and New Guinea as in other developing countries.
Control of the condition requires attention to be paid to both nutrition and infection, recog- nizing that each interacts with the other with profound effects on weanlings.
Figure 6. - Per cent of preschool children in Port REFERENCES Moresby urban squatter settlement by age group and nutritional status BIDDULPH, J. (1966). Report on An Epidemic of Gas- troenteritis in Port Moresby, Medical Society of Papua and New Guinea Symposium, Mount Hagen. than malnutrition, so that these diagnoses ap- BIDDULPH, J. (1970). Survey of Nutrition among Pre- pear at the top of the list for hospital admissions school Children in an Urban Squatter Settlement and deaths, while malnutrition is frequently (unpublished). GORDON, J. E., CHITKARA, 1. D. and WYON, J. B. overlooked. Yet if the child had not been mal- (1963). Weanling Diarrhoea, Amer. J. Med. Sci., nourished it is unlikely that his attack of pneu- 245 : 345. monia or gastroenteritis would have ended fatally. GORDON, J. E . (1969). Social Implications of Nutri- tion and Disease. Arch. Environ. Health, 18: 216. The situation has been well put by Jelliffe HARDY, A. V. (1959). Diarrhoeal Diseases of Infants (1970), ‘Death is due to an accumulation of and Children: Mortality and Epidemiology, WHO disease rather than to any single entity. The im- Bull., 21 : 209. mature, anaemic baby, living in overcrowded un- HOSPITAL DISEASE STATISTICS, 1964-1967. Department hygienic surroundings, becomes relatively mal- of Public Health, T.P.N.G. JELLIFFE, D. B. (1970). Diseases of Children in the nourished in the second semester of life. His Subtropics and Tropics, 2nd edition. Page 19, Ed- anaemia and nutrition deteriorate still further as ward Arnold, London. a result of persistent malaria, leaving an attack LAWSON, J. S. and CURTIS, P. G. (1967). Diarrhoeal of bronchopneumonia or gastroenteritis to add Disease in Papuan Infants and Children, Med. J. the last straw to his pathological burden’. Aust., 1: 101. RYAN, B. P. (1962). Gastroenteritis in New Guinea, SUMMARY Med. J. Aust., 2 : 658. RYAN, B. and MURRELL, T. G. C. (1964). Nutrition and Infection in Papuan Children, Med. J. Aust., Gastroenteritis can occur at any age, but is 1 : 556. most frequent and severe in developing coun- SCRIMSHAW, N. S., TAYLOR, C. E. and GORDON, J. E. tries during the weaning period, which in most (1968). Interactions of Nutrition and Infection, communities is between 6 to 18 months. There WHO Monograph No. 57, WHO, Geneva. are several reasons for this association of gastro- WHO (1969). The Health Aspects of Food and Nut- enteritis with the weaning period. rition: A Manual for Developing Countries in the Western Pacific Region, WHO, Manila.
VOLUME 14, NO 1, MARCH, 1971 13
72 PNG Med J 2005 Mar-Jun;48(1-2):73-86
PNG Med J 34:185-198, 1991
The bacteriology of acute pneumonia and meningitis in children in Papua New Guinea: assumptions, facts and technical strategies
MIKE GRATTEN1 AND JANET MONTGOMERY1 Queensland Institute of Medical Research, Brisbane, Australia and Papua New Guinea Institute of Medical Research, Goroka
SUMMARY
Acute respiratory infections in children aged less than 5 years in the Eastern Highlands of Papua New Guinea were investigated bacteriologically for 10 years from November 1978. Haemophilus influenzae and Streptococcus pneumoniae were responsible for 73% of all bacteria cultured from lung aspirate (83 samples), 85.5% from blood (1024 samples) and 92% from cerebrospinal fluid (155 samples). Nonencapsulated H. influenzae was carried by up to 90% of children and was the predominant haemophilus type cultured from lung tissue. Mixed infections of the lung with two types of H. influenzae (8 cases) and both H. influenzae and S. pneumoniae (18 cases), commonly together with other organisms of questionable pathogenicity, reflected the proximity of this organ to the upper respiratory tract. Serotype b accounted for 62% and 82% of H. influenzae isolated from bacteraemic pneumonia and meningitis cases, respectively. Polymicrobic bacteraemic pneumonia occurred in 16 children. Both H. influenzae and S. pneumoniae establish dense, unregulated long-term colonization in the nasopharynx during the neonatal period. Each inhibit autochthonous microflora by mechanisms that are currently unclear. Infections with two or more types occur in 30% (S. pneumoniae) and 60% (H. influenzae) of carriage-positive children. 70-75 % of H. influenzae and S. pneumoniae isolates from blood concomitantly colonize the upper respiratory tract. Intense exposure of Papua New Guinean children to penicillin at all levels of health care since the 1940s has resulted in widespread relative resistance among pneumococci to this antibiotic. Resistant strains are now found in 32 serotypes, and in children penicillin resistance is present in 75% of all carriage strains and 52% and 22% of blood and cerebrospinal fluid isolates, respectively. Penicillin-susceptible and resistant pneumococcal serotypes commonly coexist in multiply populated carriage sites. Resistance to betalactam antibiotics is rare among H. influenzae strains and resistance has not been detected in either H. influenzae or S. pneumoniae to chloramphenicol, erythromycin, tetracycline or cotrimoxazole. It should not be assumed that the technology of respiratory bacteriology as it is practised in developed countries can be transferred to the third world for utilization in paediatric aetiology and carriage studies. Respiratory bacteriology strategies as they evolved in Goroka were subject to diverse influences. The type distribution of the major causative agents defied fashionable beliefs, generated the need for more precise epidemiological differentiation and, by virtue of their carriage density, cultural properties and response to commonly used antibiotics, required the introduction or development of compatible diagnostic procedures.
Introduction Respiratory Infection (ARI) Unit in 1979 a platform was provided from which a compre- The Pneumonia Research Program of the hensive investigation into the aetiology, Papua New Guinea Institute of Medical epidemiology and prevention of acute Research (PNGIMR) was initiated in 1977; respiratory disease in young Papua New with the full establishment of an Acute Guinean children could be conducted. In
1 Queensland Institute of Medical Research, Bramston Terrace, Herston, Qld 4006, Australia Formerly Papua New Guinea Institute of Medical Research, PO Box 60, Goroka, EHP, Papua New Guinea 185
73 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal particular, the support offered by both field Sp or both on 9 occasions (3). In all, 88 and hospital-based activities created a unique organisms were recovered from lung cultures opportunity to develop technical strategies for and 73% were either Hi (40 isolates) or Sp (24 the isolation and characterization of respira- isolates). The need to mix lung aspirate with tory bacteria from the upper respiratory tract Hanks buffered saline (for virus isolation) (URT) and normally sterile sites such as lung before cultures were set up may have com- tissue, blood and cerebrospinal fluid (CSF). promised the recovery of small numbers of This paper summarizes the bacteriological bacteria. However, primary culture plates findings of these studies and the establishment were inoculated generously at the bedside of laboratory methods that were compatible with diluted aspirate and incubated without with research objectives. delay in a 5% CO2 atmosphere. Invasive Disease Consistently good isolation rates from blood cultures were gained in all studies Studies undertaken in the decade from involving both community-based children November 1978 have shown conclusively with mild pneumonia (7% significant that Haemophilus influenzae (Hi) and isolation rate) (4) and those hospitalized with Streptococcus pneumoniae (Sp) are the major moderate or severe illness (18.0-25.0%) (5) as causative agents of acute bacterial pneumonia defined by standard criteria (6). Isolation rates and meningitis in children aged 5 years or less of 33% were seen when children with severe in Papua New Guinea (PNG). disease were selected on the basis of a chest X-ray that showed an opacity accessible to Meningitis lung aspiration (3). Of 1024 children investigated by blood culture, 184 (18.0%) 155 children with purulent culture-positive grew 200 significant organisms. 85.5% (171 meningitis were studied between March 1980 strains) were either Hi (47.0%) or Sp (38.5%). and September 1984 (1). 92% were infected Others included Salmonella choleraesuis with Hi (49%) or Sp (43%). Other agents (9 strains), other enterobacteria (7 strains), encountered included Neisseria meningitidis Streptococcus pyogenes (7 strains), S. aureus (8 isolations), Streptococcus pyogenes (2 (5 strains) and N. meningitidis (1 strain). isolations) and Streptococcus agalactiae and Klebsiella pneumoniae (1 each). The selection of a blood culture broth that would support the growth of exacting If culture facilities are not immediately respiratory pathogens such as H. influenzae available, survival of Hi and Sp strains in was essential. Many formulations have been purulent CSF is satisfactory if samples are held advocated and several were used with mixed at 5-1O°C. Viability limits for 9 strains of Hi success in the early Goroka studies. After were 2-11 days (mean 5.5) and 2-24 days (mean 10.3) for 8 Sp strains (2). some experimentation, tryptose phosphate broth (Difco) was supplemented domestically Pneumonia with gelatin (1% , Difco), agar (0.1% , Difco Noble) and sodium polyanetholsulphonate 1105 children with pneumonia were studied (0.025%, Roche). This inexpensive medium in several projects (Table 1). 1025 had either has been evaluated successfully on separate lung aspirate or blood or both cultured. Of occasions in parallel with supplemented 83 children who had lung culture, Hi peptone broth-II (BBL) in 1983 (Gratten and (including one unspeciated strain) and/or Sp Montgomery, unpublished data) and Trans- were recovered from 39 (47%) and Isolate medium (7) in 1987 (Montgomery, Staphylococcus aureus from 1. Isolations of unpublished data) and has been used questionable significance and unaccompanied exclusively at the PNGIMR since July 1983. by primary pathogens were made from a further 8 children and included viridans Antecedent antibiotic usage should be streptococci, Staphylococcus epidermidis and determined in children from whom blood is diphtheroid species. Branhamella catarrhalis cultured. It has been found in PNG that more was isolated in conjunction with either Hi, children who received antibiotics within 48
186 74 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Vo lume 34, No 3, September 1991
hours before blood culture fail to grow an environment in the PNG highlands and the organism than children who had not received close association between people and pigs. an antibiotic or whose last dose was 48 hours to 7 days before (3). Mothers may either deny In order to isolate Hi on a solid medium, earlier therapy or be unaware that an antibiotic chocolate agar is required. Oxoid blood agar has been administered. A simple qualitative base No 2 supplemented with 5% horse blood test utilizing the Oxford staphylococcus and heated with care produces a highly (NCTC 6571) and 6.0mm Schleicher and supportive substrate. Chocolate agar should Schuell discs impregnated with either serum be stored at refrigeration temperature (4-6°C) or urine from the child will identify for 24-48 hours before use. Failure to use antimicrobial residues (8). The diameter of the chocolate agar has severely compromised zone of inhibition is recorded. several earlier lung aspiration studies (10).
Bacteraemic strains of S. choleraesuis were Multiple Invasion isolated with consistent frequency, which highlights the vigilance required for the Of 48 children with positive lung cultures recognition of uncommon pathogens (9). 22 had mixed infections, 18 of which were While nonhuman isolates of most salmonella due to Hi and Sp (Table 2). B. catarrhalis was serotypes have a variety of animal hosts, S. associated with these organisms in 9 cases. choleraesuis strains are isolated almost Two populations of Hi were cultured from 8 exclusively from pigs. The organism is highly children and in 7 were also accompanied by invasive, is commonly associated with Sp. Multiple isolations of Hi invariably bacteraemia and has a predilection for extra- consisted of encapsulated and nonencapsulated intestinal foci such as lung tissue. Its presence organisms with the former contributing to 5 in PNG children reflects the domestic of 7 episodes of bacteraemia recorded in these
187
75 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal
8 children. Mixed infection of the lung by Hi transmission reservoir. If properly performed, and Sp was also responsible for the 4 cases of such studies will monitor the acquisition and polymicrobic bacteraemia seen in this study. duration of carriage of target bacteria in a Multiple invasion which occurred in 9% of susceptible population and determine all episodes of bacteraemic pneumonia seen frequency and type distribution, antibiotic throughout the decade was caused by Hi and sensitivity patterns and the emergence of Sp in 14 of 16 cases and was associated with resistant mutants in the community. a 31% case fatality rate. All children were less than 12 months of age. Because carriage studies may involve village communities some distance from Mixed isolations of respiratory bacteria laboratory facilities, a reliable specimen from lung cultures have been reported by transport system that will ensure the survival others (11-14) but multiple infections due to of respiratory bacteria such as Hi and Sp for more than one type of Hi have not been up to 48 hours is essential. In PNG a recognized. The differentiation of encapsulat- commercially acquired transport system ed and nonencapsulated colonies of Hi in (Transtube, Medical Wire and Equipment Co. primary cultures requires experience. An eye Australasia Pty Ltd) employing Amies’ lens (3-4 x magnification) or dissecting charcoal medium (17) was used. Recent work microscope is desirable. Pneumonia associated by one of us (JM) has reconfirmed that nose with multiple invasion of lung tissue with swabs are preferable to throat swabs for haemophili was not suspected before the start determining URT colonization rates of Hi and of ARI research in Goroka and early cases Sp (Table 4). The anterior nares of each therefore may not have been recognized. nostril was sampled and the swab placed in Uncommon isolates or those with unusual transport medium. An insulated container was properties should be forwarded to an used to protect clinical material from adverse appropriate reference laboratory for confirma- environmental temperatures in transit. tion. This ensures that such information, when published or presented, is accepted as valid. Community carriage studies (18,19) have shown that Hi and Sp together colonize the Experience in Goroka has shown that blind URT of more than 95% of PNG children subcultures of blood culture broths within 24 aged less than 5 years. Colonization by each hours of receipt and their incubation in a organism is dense and persistent and is
CO2-rich atmosphere will optimize detection associated with a copious often purulent nasal of paediatric bacteraemia. In supplemented discharge. One-third of children carry tryptose phosphate broth, Hi characteristically encapsulated Hi and 70-90% have non- produces no visible sign of growth. This serotypable strains. 6-8% harbour type b property has been noted elsewhere (15). Direct organisms. All serotypes except type c are Gram stains are of little value for the detec- evenly distributed. The commonest tion of small gram-negative organisms in pneumococcal serotypes are 6, 19 and 23 and blood culture broths. account for 44-64% of carriage isolates. Among carriage-positive children 30% and Austrian and Collins (16) clarified the 50-60% harbour more than one type of Sp importance of carbon dioxide for the isolation and Hi, respectively (19,20). Both Hi and Sp of pneumococci on solid media and also are acquired within the neonatal period by showed that the phenomenon was type related 60% of infants and all children are colonized and, in their study, confined largely to by each within the first 3 months of life (21). serotypes 1, 3, 16, 28 and 33. In Goroka all type 5 pneumococci from all sites were Several important technical manoeuvres obligate capnophiles. Carbon dioxide depend- govern the successful outcome of carriage ence was observed less frequently for types projects. The selective culture of each target 7, 19 and 46 (Table 3). organism gives a true estimate of the carriage rate by excluding autochthonous microflora, Upper Respiratory Tract Carriage provides a semiquantitative evaluation of carriage density and ensures the isolation of The culture of URT secretions defines a discrete colonies of Hi and Sp in primary 188
76 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Vo lume 34, No 3, September 1991
189
77 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal cultures. Blood agar with gentamicin sulphate, Type Distribution, Frequency and Differen- 5 µg/ml final concentration (22) and chocolate tiation of H. influenzae and S. pneumoniae agar with bacitracin, 300 ìg/ml (23) characteristically yielded pure or near pure Serotyping provides a means of epidemio- cultures of Sp and Hi, respectively, from URT logically differentiating encapsulated bacteria secretions. This enables the examination of and, in circumstances where vaccination is colony profiles in order to determine the envisaged, enables a formulation of vaccine presence of multiple types and facilitates compatible with invasive serotypes to be single colony subcultures for tests of constructed. Encapsulated strains of Hi and Sp differentiation. However, different types of cultured at the ARI Unit in Goroka were both Hi and Sp commonly produce colonies serotyped with antisera produced by which are indistinguishable. Thus from early Wellcome Reagents, Beckenham, England, 1982, when multiple carriage populations and Statens Seruminstitut, Copenhagen, were initially sought and the relationship Denmark, respectively. between carriage and invasion was becoming apparent, a minimum of four colonies each of No attempt had been made to serotype Hi Hi and Sp were subcultured for typing. isolates from aetiology studies involving lung aspiration before that of Shann and others in Interbacterial Antagonism in the Upper Goroka (3). In the lung aspirate study (3) the Respiratory Tract frequency of nonserotypable Hi which were commonly accompanied by non-b serotypes Observations suggesting that interbact- created a technical dilemma which was not erial antagonism occurred in the URT were resolved until the accuracy of the typing made by one of us (MG) in the course of a antisera had been established. Valuable lessons carriage study in infants. Here, nasal cultures were learnt. The ability to confidently occasionally displayed areas of marked distinguish mucoid (encapsulated) and smooth regional inhibition effected by Sp at the (nonencapsulated) variants of Hi in primary expense of autochthonous microflora. In vitro culture was essential. During the learning experiments confirmed the ability of both Sp phase, negative staining of capsules using and Hi either to inhibit or prevent the growth India ink was helpful. The property of of other carriage bacteria including B. iridescence by which capsular material is catarrhalis, Corynebacterium hofmannii and identified was exploited using brain heart Corynebacterium xerosis. Staphylococci were infusion agar supplemented with Fildes less susceptible. No antagonism could be extract (Oxoid). demonstrated between strains of Hi and Sp. Kinetic studies using associative cultures (24) When Hi serotyping is undertaken, a indicated that the inhibition was bactericidal. complete set of monovalent antisera should This ability of Hi and Sp to antagonize other be available to ensure that types other than b nasopharyngeal inhabitants may explain, at are detected (26). Slide agglutination results least in part, their frequency and persistence should be interpreted within 30 seconds of of carriage in the URT. mixing the reactants in order to avoid misclassification of nonserotypable Studies of bacterial interference in the (nonencapsulated) strains because of oropharynx have suggested that interbacterial nonspecific aggregation. An alternative antagonism is one mechanism by which typing procedure such as counterimmuno- organisms such as viridans streptococci and electrophoresis or coagglutination is certain staphylococci maintain the balance of desirable in order to clarify equivocal slide pharyngeal microflora and prevent over- agglutination results. Occasional strains of growth or colonization by potential invaders Hi may be nontypable on initial isolation such as enterobacteria, S. aureus, N. menin- because of glycocalyx formation (27). gitidis and S. pyogenes (24,25). No organism Glycocalyces are lost on serial subculture capable of regulating the uncontrolled and antibody binding sites are re-exposed. colonization by Hi and Sp appears to exist Multiple subcultures, however, may among the aerobic nasal microflora in young encourage the loss of capsular material and PNG children. hence type-specificity.
190 78 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Vo lume 34, No 3, September 1991
Lung aspirate bacteriology disclosed the pneumococcal serotypes as ‘paediatric’ or important aetiological role of nonencapsulated ‘adult’ is based on findings in developed Hi in paediatric pneumonia. Biochemical countries. Studies in Goroka have shown that typing (Minitek System, BBL) was introduced ‘adult’ types are more invasive than to differentiate nonserotypable strains and to ‘paediatric’ types (30) and the predominance clarify their association with upper airway of the former in children with acute meningitis colonization. Biotyping further demonstrated supports this finding. that nonserotypable Hi were unrelated to encapsulated Hi when both were cultured The proximity of the lung to the densely together from lung tissue. A study of 505 Hi colonized URT exposes lung tissue to isolated from the URT of children and adults opportunistic infection by organisms of revealed close relationships between biotypes modest pathogenicity. Viral infections, and serotypes. Of the serotypable strains, 97% impaired nutrition and defective immune belonged to biotypes I, II and IV. A function heighten the susceptibility of young comparison of serotypes a to f with biotypes children to acute respiratory disease. Infection I, II and IV exposed a significant association of more remote foci such as the blood and of all serotypes except c with one or two meninges is restricted to more highly invasive biotypes, viz., serotype f strains were biotype organisms such as type b haemophili and I, 98% of type b strains belonged to biotypes ‘adult’ pneumococcal types, and polymicrobic I or II, biotype IV contained 89% of invasion becomes less frequent (Table 2). serotype d isolates, serotype e organisms were distributed exclusively in biotypes I and IV Relationship between Invasion and Carriage and 58% of serotype a strains were found in of H. influenzae and S. pneumoniae biotype II. No encapsulated biotype V, VI or VII organisms were isolated (28). The single The relationship between invasive strains of biotype VII strain reported in this study had Hi and Sp and URT colonization is close. only recently been described (29). 70-75% of bacteraemic Hi and Sp cultured during 1981-1987 were concomitantly carried Accurate biotyping results are dependent on in the nasopharynx (Tables 8 and 9). While all the preparation and use of a dense inoculum children at the time of blood culture had nose and the layering of seeded substrate discs with swabs cultured, URT bacteriology during sterile mineral oil to ensure that rapidly much of the period was completed only for developing reactions such as urease and indole children shown to be bacteraemic. Data for the are not lost during the overnight incubation specificity of the association between carriage which is necessary to detect ornithine and invasion of Hi and Sp are therefore not decarboxylase activity. The inoculum is considered in this paper. Pneumococcal produced by harvesting surface growth from serotypes have been grouped into 3 classes on a chocolate agar plate and mixing with 0.6ml the basis of several studies in PNG (3-5) as of peptone water broth. follows: less commonly carried, highly invasive types (Class I), ‘paediatric’ types The distribution and frequency of Hi types which frequently invade from a heavily isolated from the URT, lung, blood and CSF colonized carriage site (Class II), and less are summarized in Table 5. 80% of 35 lung invasive types with a variable carriage isolates were types other than b. 54% were frequency (Class III). Fewer bacteraemic Class nonencapsulated. Significantly more type b I pneumococci were recovered from nasal haemophili were isolated from blood (62% of cultures, as one would expect by definition of 92 strains) and CSF (82% of 73 strains). this class (Table 8). Among Hi the weaker association between invasion and carriage for ‘Paediatric’ pneumococcal serotypes 6,14,19 nonserotypable strains may imply direct and 23 were responsible for 49% of 88 person-to-person transmission or that a more isolates from lung tissue and blood (Table 6) detailed colony assessment of primary but only 18% of 67 CSF isolates (Table 7) carriage cultures is warranted. while ‘adult’ types 1-5,7-9,12,45 and 46 contributed 28% and 66%, respectively. The The association of invasion with classification of commonly invasive nasopharyngeal colonization may depend on 191
79 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal
192
80 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Vo lume 34, No 3, September 1991
the persistence of the bacteriologist. In Goroka The presence of selective agents such as the failure to confirm the carriage of a gentamicin sulphate and bacitracin in primary bacteraemic isolate of Hi or Sp when four isolation media may increase the cell division colonies of each were initially examined times of some Hi and Sp types. In this event, necessitated the reappraisal of the primary the detection of more slowly multiplying URT cultures. Typing of an additional 5-10 organisms currently considered to be in- colonies not uncommonly detected the frequently carried may further strengthen the bacteraemic strain. relationship between carriage and invasion.
193
81 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal
Antibiotic Susceptibility of H. influenzae forms at various levels of health care was used and S. pneumoniae intensively. It has been calculated that in the 10 years from 1961 the amount of penicillin Penicillin became available in PNG after used by the PNG Health Department was World War II and was widely distributed equivalent to 10 670 000 5-day courses (31). when the aid post system of primary health In such circumstances the development of care was established. Penicillin was one of penicillin resistance (PR) among pneumococci several medical agents provided to aid post in PNG was not surprising. Neither oily orderlies and was used mostly indiscriminately procaine penicillin, the first penicillin at village level. At about the same time a formulation to be used in PNG, nor its campaign, based on penicillin usage, was replacement, aqueous procaine penicillin, initiated throughout PNG to eradicate yaws. produce high blood levels. Even with massive From the late 1940s penicillin in different penicillin therapy, the effective range of 194 82 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Vo lume 34, No 3, September 1991
195
83 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal
penicillin activity attained in the URT may values of common carriage types isolated in only reach 0.06-0.3 µg/ml (25). In the upper 1980-1982 are compared with those cultured airways pneumococci have therefore been in 1985-1987 (Table 11). exposed to sublethal concentrations of penicillin perfusing across densely colonized Of 73 invasive pneumococci cultured from surfaces and resistant mutants have emerged. lung and blood during 1978-1987 38 (52%) In PNG, pneumococcal resistance is confined were relatively resistant to penicillin (Table 6). to minimum inhibitory concentration (MIC) However, less frequently carried but highly levels in the range 0.1-1.0 ìg/ml of penicillin invasive ‘adult’ types were responsible for the and is regarded as ‘intermediate’ or ‘relative’. lower resistance –15/67 (22%) – seen in CSF isolates (Table 7). Since the first observation in 1969 the acquisition of penicillin resistance among Resistance to betalactam antibiotics among pneumococcal serotypes in PNG has increased Hi is rare in PNG (36,37) and there is no steadily. Between 1969 and 1978 resistance in evidence of any significant increase. 15 capsular types was recorded (32-35) and Chloramphenicol resistance has not been resistant strains within a further 17 serotypes observed in either Hi or Sp. None of 655 were isolated during 1980-1986 (Gratten and pneumococcal strains tested during Montgomery, unpublished data) (Table 10). 1985-1987 were resistant to erythromycin, By 1987 nearly 75% of paediatric carriage tetracycline or cotrimoxazole (19). Standard- pneumococci showed PR (19). Among 956 ized disk diffusion procedures such as those carriage pneumococci isolated from children of Kirby-Bauer (38) and Stokes (39) provide in 1980-1982 (18) PR was detected in 63%. a practical, accurate and less expensive means 79.5% of the most frequently carried types of screening antibiotic responses of normally (serotypes 6,19,23) were resistant to penicil- susceptible bacteria. Quantitative technology lin while only 27% of all those strains which is required for commonly resistant organisms individually contributed less than 2% of in order to monitor further increases in isolates were resistant. High resistance rates resistance in community and hospital strains. were also present in several infrequently carried types. However, as PNG children ACKNOWLEDGEMENTS commonly harbour coexisting populations of both penicillin-resistant and penicillin- susceptible types, it has been postulated that Many people contributed to the projects in such circumstances genetic transfer of PR described in this paper. The assistance of the could occur (20). Although pneumococci in following is acknowledged: Michael Alpers, PNG continue to remain only relatively Jane Barker, Bronwyn Best, Emmanuelle resistant to penicillin, a significant increase is Carrad, Chris and Keith Coakley, Gairo apparent within this area when the MIC Gerega, Helen Gratten, Benetty Kadivaion, 196 84 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Vo lume 34, No 3, September 1991
Makit Kajoi, George Koki, Deborah 12 Hughes JR, Sinha DP, Cooper MR, Shah KV, Lehmann, Sebeya and Tony Lupiwa, Audrey Bose SK. Lung tap in childhood: bacteria, viruses Michael, Alwyn Poli, Marilyn Raymer, Ian and mycoplasmas in acute lower respiratory tract infections. Pediatrics 1969; 44:477-485. Riley, Frank Shann, Henry Siwi, Tom Smith, 13 Mimica I, Donoso E, Howard JE, Ledermann Ray Spark and Veronica Spooner. We are GW. Lung puncture in the aetiological diagnosis grateful to Joy Chapple for typing the of pneumonia. Am J Dis Child 1971; 122:278-282. manuscript. 14 Wall RA, Corrah PT, Mabey DCW, Greenwood BM. The aetiology of lobar pneumonia in The Gambia. Bull WHO 1986; 64:553-558. REFERENCES 15 Szymczak EG, Barr JT, Durbin WA, Goldmann DA. Evaluation of blood culture procedures in 1 Gratten M, Barker J, Shann F, Gerega G, a pediatric hospital. J Clin Microbiol 1979; Montgomery J, Kajoi M, Lupiwa T. The 9:88-92. aetiology of purulent meningitis in highland 16 Austrian R, Collins P. Importance of carbon children: a bacteriological study. PNG Med J dioxide in the isolation of pneumococci. J Bact 1985; 28:233-240. 1966; 92:1281-1284. 2 Gratten M, Theo L, Linnemann V, Lupiwa T. 17 Amies CR. A modified formula for the prepar- Viability of bacteria in cerebrospinal fluid at low ation of Stuart’s Transport Medium. Can J temperature. Proceedings of the Sixteenth Annual Public Health 1967; 58:296-300. Symposium of the Medical Society of Papua New 18 Gratten M. Carriage and invasion of respiratory Guinea, Madang, 19-20 Sep 1980:122-126. bacterial pathogens in Melanesian children. MSc 3 Shann F, Gratten M, Germer S, Linnemann V, Thesis, University of Papua New Guinea, Port Hazlett D, Payne R. Aetiology of pneumonia in Moresby, 1984. children in Goroka Hospital, Papua New Guinea. 19 Montgomery JM. A longitudinal study of upper Lancet 1984; 2:537-541. respiratory tract bacterial carriage and its 4 Montgomery J, Lehmann D, Smith T, Michael association with acute lower respiratory infections A, Joseph B, Lupiwa T, Coakley C, Spooner V, in children in the highlands of Papua New Guinea. Best B, Riley ID, Alpers MP. Bacterial MSc Thesis, University of Papua New Guinea, colonization of the upper respiratory tract and its Port Moresby, 1989. association with acute lower respiratory tract 20 Gratten M, Montgomery J, Gerega G, infections in highland children of Papua New Gratten H, Siwi H, Poli A, Koki G. Multiple Guinea. Rev Infect Dis 1990; 12 (Suppl 8): colonization of the upper respiratory tract of SI006-S1016. Papua New Guinea children with Haemophilus 5 Barker J, Gratten M, Riley I, Lehmann D, influenzae and Streptococcus pneumoniae. Montgomery J, Kajoi M, Gratten H, Smith D, Southeast Asian J Trop Med Public Health 1989; Marshall TFdeC, Alpers MP. Pneumonia in 20:501-509. children in the Eastern Highlands of Papua New 21 Gratten M, Gratten H, Poli A, Carrad E, Raymer Guinea: a bacteriologic study of patients selected M, Koki G. Colonisation of Haemophilus by standard clinical criteria. J Infect Dis 1989; influenzae and Streptococcus pneumoniae in the 159:348-352. upper respiratory tract of neonates in Papua New 6 Papua New Guinea Department of Health. Guinea: primary acquisition, duration of carriage, Standard Treatment for Common Illnesses of and relationship to carriage in mothers. BioI Children in Papua New Guinea. A Manual for Neonate 1986; 50:114-120. Nurses, Health Extension Officers and Doctors, 22 Converse GM 3d, Dillon HC Jr. Epidemio- 4th edition. Port Moresby: Department of Health, logical studies of Streptococcus pneumoniae in 1984. infants: methods of isolating pneumococci. J Clin 7 Ajello GW, Feeley JC, Hayes PS, Reingold AL, Microbiol 1977; 5:293-296. Bolan G, Broome CV, Phillips CJ. Trans-Isolate 23 Hovig B, Aandahl EH. A selective method for medium: a new medium for primary culturing and the isolation of Haemophilus in material from the transport of Neisseria meningitidis, Streptococcus respiratory tract. Acta Pathol Microbiol Scand pneumoniae and Haemophilus influenzae. J Clin 1969; 77:676-684. Microbiol 1984; 20:55-58. 24 Sanders E. Bacterial interference. 1. Its occur- 8 World Health Organization. Acute Respiratory rence among the respiratory tract flora and charac- Infections. Laboratory Manual of Bacteriological terization of inhibition of Group A streptococci Procedures. Manila: WHO Regional Office for by viridans streptococci. J Infect Dis 1969; the Western Pacific, 1986. 120:698-707. 9 Gratten M, Barker J, Rongap A. Pneumonia due 25 Sprunt K, Redman W. Evidence suggesting to Salmonella choleraesuis in infants in Papua New importance of role of interbacterial inhibition in Guinea. Lancet 1983; 2:580-581. maintaining balance of normal flora. Ann Intern 10 Shann F. Etiology of severe pneumonia in Med 1968; 68:579-590. children in developing countries. Pediatr Infect 26 Wallace RJ Jr, Musher DM, Septimus EJ, Dis J 1986; 5:247-252. McGowan JE Jr, Quinones FJ, Wiss K, Vance 11 Abdel-Khalik AK, Askar AM, Ali M. The causa- PH, Trier PA. Haemophilus infIuenzae tive organisms of broncho-pneumonia in infants infections in adults: characterization of strains by in Egypt. Arch Dis Child 1938; 13:333-342. serotypes, biotypes and â-lactamase production. J Infect Dis 1981; 144:101-106.
197
85 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Medical Journal
27 Costerton JW, Irwin RT, Cheng K-J. The role Pneumococci relatively insensitive to penicillin in of bacterial surface structures in pathogenesis. Australia and New Guinea. Med J Aust 1974; CRC Crit Rev Microbiol 1981; 8:303-338. 2:353-356. 28 Gratten M, Lupiwa T, Montgomery J, Gerega 34 Hansman D. Penicillin-insensitive pneumococci G. Distribution and relationship to serotype of in Australia and New Guinea: a progress report. Haemophilus influenzae biotypes isolated from Med J Aust 1978; 2:295-296. upper respiratory tracts of children and adults in 35 Gratten M, Naraqi S, Hansman D. High Papua New Guinea. J Clin Microbiol 1984; prevalence of penicillin-insensitive pneumococci 19:526-528. in Port Moresby, Papua New Guinea. Lancet 29 Gratten M. Haemophilus influenzae, biotype 1980; 2:192-195. VII. J Clin Microbiol 1983; 18:1015-1016. 36 Shann F, Gratten M, Montgomery J, Lupiwa T, 30 Riley ID, Lehmann D, Alpers MP. Pneumo- Polume H. Haemophilus influenzae resistant to coccal vaccine trials in Papua New Guinea: penicillin in Goroka. PNG Med J 1982; 25:23-25. relationships between the epidemiology of 37 Montgomery J, West B, Michael A, Kadivaion pneumococcal infection and vaccine efficacy. B. Bacterial resistance in the Eastern Highlands Rev Infect Dis 1991; 13(Suppl 6):S535-S541. Province. PNG Med J 1987; 30:11-19. 31 Riley ID. Pneumonia in Papua New Guinea. 38 National Committee for Clinical Laboratory MD Thesis, University of Sydney, Sydney, 1979: Standards. Performance Standards for Anti- 369p. microbial Disk Susceptibility Tests. Approved 32 Hansman D, Glasgow H, Sturt J, Devitt L, Standard M2-A4, 4th edition. Villanova, Douglas R. Increased resistance to penicillin of Pennsylvania: National Committee for Clinical pneumococci isolated from man. N Engl J Med Laboratory Standards, 1990. 1971; 284:175-177. 39 Stokes EJ, Ridgeway GL. Clinical Bacteriology, 33 Hansman D, Devitt L, Miles H, Riley I. 5th edition. London: Edward Arnold, 1980.
198
86 PNG Med J 2005 Mar-Jun;48(1-2):87-94
Papua New Guinea Med J 31:109-116, 1988
Nutrition and Morbidity: Acute Lower Respiratory Tract Infections, Diarrhoea and Malaria
DEBORAH LEHMANN1, PETER HOWARD1 AND PETER HEYWOOD2
Papua New Guinea Institute of Medical Research, Goroka and Madang
SUMMARY
The three most important infectious diseases of young children in Papua New Guinea are acute lower respiratory tract infections, diarrhoea and malaria, each of which has been shown to have a negative effect on growth. Low nutritional status is associated with increased risk and severity of acute lower respiratory tract infections and with increased severity of diarrhoea. There is no evidence to indicate that malnutrition is associated with increased risk of malaria. Adequate control and prompt treatment of infectious diseases will improve nutritional status. At the same time, improve- ment in nutritional status will reduce morbidity and mortality due to infectious disease, particularly acute lower respiratory tract infections and diarrhoea.
INTRODUCTION Province (SHP) show that on average there are 2.5 episodes of ALRI per child under one The proximate causes of growth year of age and 1.6 episodes per child aged retardation are generally agreed to include low 12-23 months (2). Children under one year of birthweight, deficient nutrient intake and age suffer more severe disease than older repeated episodes of infectious disease, the children: 20% of ALRI episodes in infants are importance of each factor varying with the classified as moderate-severe disease (the specific situation. children have chest indrawing in addition to cough and breathlessness with or without The most important infectious diseases of fever), and one-tenth of those with chest young children in Papua New Guinea (PNG) are indrawing have signs of more severe disease acute lower respiratory tract infections, (difficulty in feeding and/or cyanosis and/or diarrhoea and malaria. The interaction between heart failure). In children aged 1-4 years, these infections and nutrition, particularly 10% of ALRI illnesses are classified as protein-energy malnutrition, is reviewed in this moderate-severe cases. paper with particular emphasis on studies carried out in PNG. Data from Tari on children with known birthweight show that children weighing less than 2.5kg at birth have four times the risk of NUTRITION AND ACUTE LOWER dying of ALRI in the first year of life compared RESPIRATORY INFECTIONS to children who are born heavier (D. Lehmann and P. Heywood, unpublished data). Further- Acute lower respiratory tract infections more, among children who were weighed (ALRI) are the commonest cause of admission between 4 and 59 months of age and then and death among children in PNG. In the followed in Tari (SHP) and Asaro, Eastern highlands, 47% of admissions under the age of Highlands Province (EHP) it was found that 5 years are for pneumonia while in the lowlands those who were less than 70% weight-for-age pneumonia accounts for approximately one- (compared to the Harvard median) had eight quarter of admissions in the same age group times the risk of dying of ALRI of heavier (1). Data from Tari, Southern Highlands children (D. Lehmann, unpublished data).
1 Papua New Guinea Institute of Medical Research, PO Box 60, Goroka, EHP, Papua New Guinea 2 Papua New Guinea Institute of Medical Research, PO Box 378, Madang, Papua New Guinea
109
87 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J
Effect of ALRI on Nutrition suggesting, firstly, that highland children experience more infections than expatriate Mata has shown growth faltering (weight children and, secondly, that their immune loss and height arrest) when children suffer function may be depressed. from measles, whooping cough or ALRI, which may continue for weeks or months (3). James (7) has reported that malnourished Children who are malnourished before children have similar numbers of attacks of developing measles are likely to have more respiratory disease as well-nourished children marked growth faltering and are more likely but that malnourished children are sick for to suffer prolonged respiratory or gastro- longer and are three times more likely to intestinal symptoms (4). develop bronchitis and nineteen times more likely to develop pneumonia. Mata (4) noted Under the stress of illness, children go into that children who grow well are likely to suffer negative nitrogen balance. Tomkins et al. (5) fewer days of respiratory symptoms than those report that during infection there is an increase who do not grow well. By contrast, in an in protein breakdown which is greater than the intervention study in India, episodes of lower rise in synthesis of protein. If malnutrition is respiratory infection were of shorter duration present at the onset of infection, these where medical services were provided, while responses to infection are reduced. nutrition supplementation in addition to medical services did not have a summation The high incidence of ALRI in young effect (8). However, the investigators comment children in PNG contributes to the deter- that it was financially beneficial to have the ioration in nutritional status over the first two two programs together and that this would years of life. Children who suffer from ALRI provide all the benefits of a medical care become anorexic and have difficulty feeding program and a nutrition care program. because of breathlessness, and therefore their food intake falls. Data from Goroka Base Hospital, EHP show that children under two years of age who Effect of Nutrition on ALRI are malnourished are four times more likely to be admitted with pneumonia and, if admitted, Malnourished children have depressed are four times more likely to die of the disease immune responses which make them more (9). Figure 1 shows the mean percent weight vulnerable to infection. Delayed cutaneous for age (compared to the Harvard median) by hypersensitivity (DCH) to various antigens (an age, in healthy children from the Asaro Valley expression of cell-mediated immunity) is near Goroka, in children admitted to hospital impaired in malnourished children; the response with moderate or severe disease who survived correlates directly with the nutritional status and in children who died of severe disease. The and improves with nutrition therapy. Serum relationship between malnutrition and risk of antibody responses may be normal or developing ALRI and severity of ALRI is main- decreased, while salivary IgA, which protects tained irrespective of the duration of illness against invasion by microorganisms at mucosal before admission. surfaces, has been shown to be low in malnourished children. Bactericidal activity of Data concerning the role of iron in polymorphonuclear leukocytes is reduced in infection are conflicting. Weinberg (10) has malnourished children and can be corrected with reported that microorganisms require iron for improved nutrition. Malnutrition also results in growth and that during microbial infections a reduction of complement components, in hosts attempt to withhold iron from invading particular C3 (6). In highland children in PNG, organisms. He also reports that in conditions antibody titres to respiratory pathogens may be which result in hyperferraemia or hypotrans- lower in young children suffering from ALRI. ferrinaemia there is an increased risk of Moreover, highland children have higher levels infection. Papua New Guinean highland of immunoglobulin and depressed cell-mediated children under 2 years of age are iron deficient immunity compared to expatriate children of by western standards as determined by free the same age (C. Witt, personal communication), erythrocyte protoporphyrin (FEP) (C. Witt,
110
88 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Volume 31, No 2, June 1988
personal communication). However, in a pro- malnutrition: spective study of children from the Asaro 1. Prompt appropriate antibiotic therapy Valley, children who were iron deficient, as for a minimum of five days will reduce measured by levels of FEP and mean the number of days of illness and the corpuscular haemoglobin concentration associated anorexia and difficulty in (MCHC), were at no greater risk of developing feeding. Parents must be taught that ALRI than children who were not iron children with cough and breathlessness deficient (C. Witt, personal communication). require a course of penicillin at an aid post and that inpatient care is required Interventions for those children who in addition have difficulty feeding. The following measures aimed at controll- 2. Aid post orderlies must treat children ing ALRI morbidity will assist in preventing suffering from cough and breathlessness
111
89 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J
with the correct dose of penicillin. They improvement of the nutritional status of must be able to recognize chest indraw- children. ing and difficulty in feeding and refer these children for inpatient care. During the last thirty years the interactions 3. Immunization against measles and between infections, particularly diarrhoeal whooping cough as near the diseases, and nutrition have been extensively recommended age as possible will studied and keenly debated. The pioneering prevent children from developing these work of Scrimshaw and coworkers in Central diseases and their chronic sequelae. America led to the hypothesis that a 4. Pneumococcal vaccine has been shown bidirectional causal relationship exists between to reduce mortality due to ALRI as the diarrhoea and nutritional status. Malnutrition sole cause of death by 50% in young predisposes a child to diarrhoea and, children in the highlands (11). conversely, diarrhoea causes growth retardation Preliminary analysis of morbidity data and precipitates severe malnutrition (12). from Tari shows that there is a reduction in the number of episodes of This synergistic relationship (diarrhoea moderate and severe ALRI in the first causing acute weight loss, arrest in linear year after being immunized with growth and malnutrition) has been demon- pneumococcal vaccine (T. Marshall, strated in detailed prospective studies in personal communication). There is no Guatemala (4, 13), The Gambia (14,15), difference in efficacy of pneumococcal Uganda (15) and Bangladesh (16). The evidence vaccine in preventing ALRI deaths for the converse, that malnutrition predisposes among malnourished and well-nourished to diarrhoea, is not so conclusive and is con- children (D. Lehmann, unpublished troversial. Tomkins in Nigeria (17) and Trow- data). If pneumococcal vaccine were bridge in El Salvador (18) found that impaired delivered through routine health services nutritional status was associated with an in the highlands in the future we would increased prevalence of diarrhoea. One study almost certainly see a reduction in in Guatemala (19) showed an increased morbidity and mortality from ALRI in incidence of diarrhoea in lighter children (lower children. weight for age) 1-4 years old. Two studies did not establish a relationship between three Measures taken to prevent malnutrition parameters of nutritional status (low weight, discussed elsewhere in this issue will reduce stunting and wasting) and incidence (7, 20), the risk of developing ALRI and also the risk whereas Tomkins (17) showed wasting, but not of dying of the disease. low weight or stunting, to be a predictor of incidence. The only study to disentangle the NUTRITION AND DIARRHOEA component factors of prevalence (incidence and duration of episodes) is that of Black et al. The evidence for the strong association (21). They found that Bangladeshi children less between diarrhoea and malnutrition in many than 2 years old with low weight for height had parts of the developing world is compelling. episodes of longer duration but similar The peak age-specific prevalence rates for both incidence rates when compared to better- diarrhoeal diseases and malnutrition coincide in nourished children. infants and children under five years of age. Both are significant problems where pre- In practice, where severe malnutrition and disposing conditions coexist, namely diarrhoea coexist at high levels of endemicity inadequate and polluted water supplies, poor few could argue that there is not a close environmental sanitation, low levels of personal relationship between them, whatever its precise and domestic hygiene, poverty, overcrowding nature. Because of their multiple and often and low education levels. In addition, historical shared determinants preventive and therapeutic evidence from developed countries suggests that interventions will be directed at reducing the the correction of these adverse conditions burden of both problems. In areas where severe preceeded, or at least coincided with, a fall in malnutrition (low weight for age) and diarrhoea the incidence of diarrhoea and a general are present at low to moderate endemic levels,
112
90 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Volume 31, No 2, June 1988 such as PNG, it is probably difficult to define 2. PNG is a relatively underpopulated country the precise nature of the interaction because and with the exception of urban and a few rural of the logistic problems in studying the large areas population densities are low and many sample sizes required to demonstrate the small communities relatively isolated. This probably differences between risk groups. acts to reduce the amount of inter-household and inter-community transmission. However, There are two studies which have addressed this situation is changing as road commun- the question of the diarrhoea-malnutrition ications lead to increased local population con- interaction in PNG. Biddulph and Pangkatana centrations. (22) showed that malnourished children (less than 75% weight for age) with severe gastro- 3. Prolonged breastfeeding is almost universal enteritis required longer periods of rehydration in PNG societies. This practice has been pro- than well-nourished children, suggesting tected by the Baby Feed Supplies (Control) Act increased severity in the former group. In of 1977, which has probably stemmed an children between the ages of one and four years increase in the use of bottle feeding, especially in Enga Province, Binns (23) found significantly in urban areas, although unfortunately there is higher incidence rates of diarrhoea in lighter no hard evidence to demonstrate this. children (less than 80% weight for age) than in their heavier peers. In addition, following 4. In many parts of the country the episodes of diarrhoea 83% of children failed to nutritional status in the first five years of life gain weight or lost weight during the month is reasonably good. after a diarrhoea episode. The measure of incidence in this study was based on a one- 5. In most rural societies, except after a large month recall by mothers, which is recognized as mumu (feast), when unconsumed food may be unreliable by most experienced field workers, kept for some days, it appears that food is not and the methodological description does not stored overnight or for long periods during the allow assessment of the validity of the study day. Thus transmission via contaminated food for other areas of the country. may be uncommon, though this needs verification. In the light of present knowledge it is not possible to draw firm conclusions about the 6. In the highlands at least, traditionally nature of the interaction between diarrhoea little water is drunk. Compared to some low- and malnutrition in PNG children. However, land areas water is plentiful and reasonably if, as the available data suggest, the incidence accessible although utilization for all purposes of diarrhoea is low compared to many develop- is reportedly low. From what evidence is ing countries (estimated at about one episode available, it is not grossly polluted. per child per year in children less than five years old), it is perhaps worth considering why Whatever may be the community incidence this should be so. A number of factors related of diarrhoea and malnutrition in PNG, they are to transmission and severity may combine to still major public health problems as measured produce the low force of infection. by the burden on curative health services. Preventive and curative interventions to reduce 1. It appears that most cultural groups have this burden have been given high priority in a natural appreciation of the polluting effects the government’s current five-year health plan of human faecal material. This is reflected (1986-90) (24). The cornerstone of the in an awareness not to defaecate near water diarrhoeal disease control (CDD) program is supplies and the no-touch or indirect methods based on correct case management – the assess- used for anal cleansing. However, it is not ment of dehydration status, appropriate oral known whether an infant’s faeces or diarrhoea rehydration therapy (ORT) and improved stools are regarded as being more or less nutritional management. Current efforts focus polluting. In spite of low levels of personal on improved management at village level hygiene, this suggests that the level of con- including home-based recognition, the use of tamination of the environment by faecal home-made fluids, the maintenance of energy enteropathogens is low. intake during episodes and increased feeding
113
91 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J during the recovery phase to ensure catch-up frequent in primiparae. Mean singleton birth- growth. weight was depressed by 170g in the presence of malaria but the difference was only sign- Preventive CDD interventions which have ificant for primiparae. been given a high priority are: 1. measles immunization to reduce the The effect of malaria on postnatal growth morbidity and mortality from measles- has been investigated in two different types of associated diarrhoea; studies. In the first type, examples of which 2. the provision and increased use of safe are the studies in The Gambia of Marsden (29) and adequate water supplies in rural and Rowland et a1. (14), the effect of individual areas; episodes of illness on weight gain during the 3. improved facilities for the safe disposal period in which the illness occurred was of faeces (pit latrines); estimated. Both studies indicated that an 4. improved personal hygiene; and episode of malaria had a negative effect on 5. improved weaning practices (more weight gain in the period immediately following energy introduced at a younger age) and the episode. food hygiene. Whilst an episode of malaria may have a It is recognized that the vital component short-term effect on growth, through depressed to successful implementation of these inter- food intake and negative energy and nitrogen ventions is the mobilization of resources at the balance, the net effect of malaria on growth village level. Emphasis is to be directed at the will depend on the extent to which the capacity integrated primary health care approach by for catch-up growth is subsequently realized. motivation and education of individuals and communities through schools, village-level The second type of study has addressed health services (aid posts, MCH clinics and this question by estimating the net effect of health inspectors) and the use of the mass malaria on growth over a much longer period media. of time.
Recent publications edited by Gracey The most rigorous study is that of (25) and Chen and Scrimshaw (26) provide McGregor et a1. (30). Two groups of 20 new- excellent reviews of the relationship between borns were enrolled in a 3-year prospective diarrhoea and malnutrition. study. One group was given weekly chloroquine and the other a placebo. Unprotected children NUTRITION AND MALARIA initially gained weight more slowly but sub- sequently gained ground and by 36 months Effect of Malaria on Growth were just as heavy as the protected group. However, the mean height of the protected Malaria may affect growth both ante- and children at three years of age was more than post-natally. 3cm greater than that of the unprotected group. Birthweight is an important determinant of postnatal growth (27) and the effect of Thus, the evidence from the one well- malaria during pregnancy on birthweight has designed study of the longer-term effects of been investigated in a number of studies. The malaria on growth is for a negative effect on most convincing of these is that of McGregor height but not weight. However, short-term et al. in The Gambia (28). The effect of studies of the effect of individual episodes placental malaria infection on birthweight was indicate the reverse — a negative effect of investigated in more than 6000 births. Pregnant malaria on weight but not on height. Part of women were more likely to have parasites in this conflict can be explained by the greater the platental than in the peripheral blood, short-term variability in weight than in height. malarious placentae were more likely in primiparous than multiparous women, and Nevertheless, although malaria may be an dense placental infections were also more important factor affecting growth the fact that
114 92 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Volume 31, No 2, June 1988 both the shortest and the tallest children in DISCUSSION PNG are found in districts in which malaria is highly endemic (Maprik and Morehead districts, Each of the important infectious diseases respectively) indicates that other factors are of children has been shown to have a negative also very important. effect on growth, at least in the short term. However, if catch-up growth after each episode All of the studies quoted above have been is complete it is possible for there to be a in areas where malaria is highly endemic. In negative short-term effect and no effect in the areas where malaria is episodic rather than long term. One of the most important factors endemic, such as the highlands of PNG, and determining the net long-term effect is the immunity to malaria is less, an individual extent to which catch-up growth occurs. episode may be more severe and have greater effects. Thus, in Enga Province, Sharp and Under favourable conditions the rate of Harvey (31) found splenomegaly, used as an growth following a period of illness can be very index of recent experience of malaria, to be high. However, most evidence indicates that associated with stunting in young children. although only modest increases in food intake are needed to support catch-up growth during Effects of Nutrition on Malaria the convalescent period, they seldom occur (35). The cultural context in which a disease The few studies which have addressed this occurs is likely to be an important variable question - see Heywood and Harvey (32) - influencing catch-up growth (32). The cultural have serious design and analysis flaws and it is context will influence not only the actual foods not possible at the moment to draw firm produced and considered food, but also who conclusions. receives food and when, as well as the availability of food during an illness episode. Iron The synergism between nutrition and Malarial parasitaemia usually results in infection means that there are considerable some degree of anaemia in which red cell health gains to be made by a combined attack morphology is consistent with iron deficiency. on both malnutrition and infection. However, Successful intervention studies have resulted in because the control measures for infection will increased haemoglobin levels. At the same time often be very different from those used for there is a rise in serum iron and serum ferritin. malnutrition the synergism between the control Thus, understanding of the haematological programs will often not be apparent. It is picture in malaria is complicated since, important that health staff appreciate the although red cell morphology is consistent with relationship between malnutrition and infection iron deficiency, other measures indicate so that control programs, including prompt and effective treatment of episodes of infectious adequate iron status (32). disease, may be coordinated and the maximum improvement in the health of young children The effect of iron status on malaria is obtained. controversial. Recent studies in PNG indicate that the effect in children may depend on their REFERENCES immune status. In infants Oppenheirner et al. (33) showed that those subjects injected with 1 Papua New Guinea Department of Health. Hand- iron supplement at 2 months of age showed book of Health Statistics, Papua New Guinea, significantly greater malarial parasitaemia and 1984. Port Moresby: Department of Health, 1985. palpable spleens at 6 and 12 months of age 2 Lehmann D. Tari Research Unit: Final Report than children who received a placebo. In for the Southern Highlands Rural Development contrast, Harvey (34) working with school- Project. Mendi: Media Unit, 1984. 3 Mata LJ, Kronmal RA, Urrutia JJ, Garcia B. children showed no effect of an oral iron Effect of infection on food intake and the supplement, and improved iron status, on nutritional state: perspective as viewed from the malaria. village. Am J Clin Nutr 1977; 30: 1215-1227.
115
93 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J
4 Mata LJ. The Children of Santa María Cauqué: A NS. Acute diarrhoeal disease in less developed Prospective Field Study of Health and Growth. countries. 2. Patterns of epidemiological Cambridge: MIT Press, 1978. behaviour in rural Guatemalan villages. Bull 5 Tomkins AM, Garlick PJ, Schofield WN, Water- WHO 1964; 31:9-20. low JC. The combined effect of infection and 20 Chen LC, Huq E, Huffman SL. A prospective malnutrition on protein metabolism in children. study of the risk of diarrheal diseases according Clin Sci 1983;65:313-324. to the nutritional status of children. Am J 6 Puri S, Chandra RK. Nutritional regulation of Epidemiol 1981; 114:284-292. host resistance and predictive value of immun- 21 Black RE, Brown KH, Becker S. Malnutrition ologic tests in assessment of outcome. Pediatr is a determining factor in diarrheal duration, but Clin North Am 1985; 32:499-516. not incidence, among young children in a 7 James JW. Longitudinal study of the morbidity longitudinal study in rural Bangladesh. Am J of diarrheal and respiratory infections in Clin Nutr 1984; 37:87-94. malnourished children. Am J Clin Nutr 1972; 22 Biddulph J, Pangkatana P. Weanling diarrhoea. 25:690-694. Papua New Guinea Med J 1971; 14:7-13. 8 Kielman AA, Taylor CE, Parker RL. The 23 Binns CW. Food, sickness and death in children Narangwal Nutrition Study: a summary review. of the highlands of Papua New Guinea. J Trop Am J Clin Nutr 1978; 31:2040-2052. Pediatr 1976;22:9-11. 9 Barker J, Gratten M, Riley I, Lehmann D, Phillips 24 Papua New Guinea Department of Health. Papua P, Montgomery J, Kajoi M, Pratt D, Canil K, New Guinea National Health Plan, 1986-1990. Gratten H, Smith D, Marshall TF de C, Alpers M. Port Moresby: Department of Health, 1986. Pneumonia in children in the Eastern Highlands 25 Gracey M, ed. Diarrhoeal Disease and of Papua New Guinea: an aetiological study of Malnutrition: a Clinical Update. London: patients selected by standard clinical criteria. Churchill Livingstone, 1985. Submitted manuscript. 26 Chen LC, Sctrimshaw NS, eds. Diarrhea and 10 Weinberg ED. Infection and iron metabolism. Malnutrition: Interactions, Mechanisms and Am J Clin Nutr 1977; 30:1485-1490. Interventions. New York: Plenum Press, 1983. 11 Riley ID, Lehmann D, Alpers MP, Marshall TF 27 Mata LJ, Urrutia JJ, Kronmal RA, Joplin C. de C, Gratten H, Smith D. Pneumococcal vaccine Survival and physical growth in infancy and prevents death from acute lower-respiratory-tract early childhood. Am J Dis Child 1975; 129:561- infections in Papua New Guinean children. 566. Lancet 1986; 2:877-881. 28 McGregor IA, Wilson ME, Billewicz WZ. Malarial 12 Scrimshaw NS, Taylor CE, Cordon JE. Inter- infection of the placenta in The Gambia, West actions of Nutrition and Infection. World Health Africa: its incidence and relationship to stillbirth, Organization Monograph Series No 57. Geneva: birthweight and placental weight. Trans R Soc World Health Organization, 1968. Trop Med Hyg 1983; 77:232-244. 13 Martorell R, Habicht JP, Yarbrough C et al. 29 Marsden PD. The Sukuta Project. A longitudinal Acute morbidity and physical growth in rural study of health in Gambian children from birth Guatemalan children. Am J Dis Child 1975; 129: to 18 months of age. Trans R Soc Trop Med Hyg 1296-1301. 1964; 58:455-489. 14 Rowland MGM, Cole TJ, Whitehead RG. A 30 McGregor lA, Gilles HM, Walters JB, Davies AH, quantitative study into the role of infection in Pearson FA. Effects of heavy and repeated determining the nutritional status in Gambian malarial infections on Gambian infants and village children. Br J Nutr 1977; 37:441-450. children. Br Med J 1956; 2:686-692. 15 Cole TJ, Parkin JM. Infection and its effects on 31 Sharp PT, Harvey P. Malaria and growth stunting the growth of young children: a comparison of in young children of the highlands of Papua New The Gambia and Uganda. Trans R Soc Trop Med Guinea. Papua New Guinea Med J 1980; 23:132- Hyg 1977; 71:196-198. 140. 16 Black RE, Brown KH, Becker S. Effects of 32 Heywood P, Harvey PWJ. Protein-energy diarrhoea associated with specific pathogens in the malnutrition, iron status and malaria. Papua New growth of children in rural Bangladesh. Pediatrics Guinea Med J 1986; 29:45-52. 1984; 73:799-805. 33 Oppenheimer SJ, Gibson FD, Macfarlane SB, 17 Tomkins A. Nutritional status and severity of Moody JB, Hendrickse RG. Iron supplementation diarrhoea among pre-school children in rural and malaria. Lancet 1984; 1:389-390. Nigeria. Lancet 1981; 1:860-862. 34 Harvey PWJ. Iron repletion and susceptibility to 18 Trowbridge FL, Newton LH. Seasonal changes malarial infection. Ph.D. Thesis, Cornell in malnutrition and diarrheal disease among University, 1987. preschool children in El Salvador. Am J Trop 35 Calloway DH. Nutritional requirements in Med Hyg 1979; 28:136-141. parasitic diseases. Rev Infect Dis 1982; 4:891- 19 Gordon JE, Guzman MA, Ascoli W, Scrimshaw 895.
116
94 PNG Med J 2005 Mar-Jun;48(1-2):95-101
Papua New Guinea Med J 29:11-17, 1986
Small-Area Variations in the Epidemiology of Malaria in Madang Province
JACQUELINE A. CATTANI 1 2, JAMES S. MOIR 1, F. DAVID GIBSON 1, MEZA GINNY 1, JOSEPH PAINO1, WENDY DAVIDSON 1 AND MICHAEL P. A LPERS 1
Papua New Guinea Institute of Medical Research, Madang, Papua New Guinea
SUMMARY
Small-area variations in the epidemiology of malaria can have important consequences in the evaluation of malaria control or intervention programs. Epidemiological features of malaria were studied in six villages in Madang Province through six-monthly surveys over a two-year period and three-monthly surveys for a single year. Data on parasite prevalence, infection density, splenic enlargement and morbidity were analyzed to determine the existence of variation within a previously defined study area. While average parasite rates did not differ greatly among villages, rates in the 1-4 years age group in the dry season were significantly higher (60.2%) in the villages north of Madang town compared with those located to the south (37.7%) (α = 0.05). There were also differences in the parasite species ratio (i.e., 21.7% of infections in one village were Plasmodium vivax compared with 12.6% in another), in spleen rates and in degree of splenic enlargement. Utilization data from village aides were used to estimate fever attacks over a 12-month period, but interpretation of these rates was complicated by the alternative sources of treatment available to the village populations.
INTRODUCTION on exploring this limited type of intervention.
The principal reason for studying the The Malaria Research Program of the epidemiology of malaria today is to facilitate Papua New Guinea Institute of Medical Research the design and evaluation of interventions (PNGIMR) has been investigating the epidem- directed towards reducing the prevalence of iology of malaria in a coastal area of Madang malaria infection and occurrence of disease. Province since 1981. The objectives of the Studies in the past have sometimes accepted project include: measurement of malario- results of one cross-sectional malariometrical metrical, serological and other health indices in survey as defining the endemicity of malaria in the study population; examination of the a given geographical area. The possible over- or relationship between these indices and the under-estimation of baseline prevalence based impact of environmental and socioeconomic on a single survey may not be critical in evalua- factors; and provision of baseline data on a ting interventions designed to have a major defined population for the possible field test impact on parasite prevalence and transmission in a highly endemic area. Evaluation of inter- of an antimalarial vaccine. ventions with more modest objectives such as a 20 to 50% reduction in parasite rates in infants The purpose of this paper is to present and young children, however, would require results from the study of small-area variations greater precision in baseline estimates and in the epidemiology of malaria in six villages careful characterization of any variation within from the Madang study area. The parameters the study area. For the most part efforts to investigated longitudinally in these villages control malaria are now realistically focussed include parasite rates, spleen rates, degrees of
1 Papua New Guinea Institute of Medical Research, P.O. Box 378, Madang, Papua New Guinea 2 Present address: Department of Tropical Public Health, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115, USA 11 95 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J
12
96 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Volume 29, No. 1, March 1986 splenic enlargement, parasite densities and bed-net utilization by household were conducted, morbidity. Results of selected entomological and data on genetic polymorphisms were investigations are reported by Charlwood et al. collected. These data will be reported separately. (1) in this issue.
METHODS RESULTS
The study area of the PNGIMR Malaria Average participation among village Research Program is contained within a 22km inhabitants over the four surveys was as radius of Madang town and includes a popula- follows:84.3% in the 1-4 year age group; 85.6% tion of approximately 16,500. The area and the in 5-15 year olds; and 68.6% among adults general epidemiology of malaria within it is older than 15 years. Five of the six villages had described elsewhere (2). Six villages were an overall compliance rate of greater than 78% selected for inclusion in the study. Two are for all ages (all surveys combined). The remain- located on the north coast and two on the ing village had an overall participation rate of south coast, at sea level . One is situated inland 53.7%; however, compliance in the 5-15 year in the north at 350m and one inland in the age group over all surveys was 71.3% and in the south at 150m (Figure1). The villages are 1-4 years group 60%. As these are the age characterized by hyperendemic malaria; how- groups of greatest interest, participation was ever, based upon results of previous epidem- considered adequate to retain this village in iological studies we had evidence for differing the age-group-specific analyses. levels of hyperendemicity in the area. The present study was designed and the six villages Average parasite rates did not differ chosen to explore the differences in greater greatly among the villages; however, more detail. short-term variation was observed in the inland villages and those on the north coast than in Malariometrical surveys were conducted at those on the south coast. There were consistent approximately three-monthly intervals from differences in the range of variation in parasite July of 1983. Four surveys were conducted in rates in 1-9 year olds between villages north each of the six villages. Species-specific parasite of Madang, whether inland or on the coast, rates, densities and spleen grades were recorded and those south of Madang town. Rates in this for all individuals in each survey. The survey age group varied in the north from 40.7% methodology, laboratory techniques and quality (lowest rate among the three villages) to 88% control procedures are described elsewhere (2). (highest rate). Villages in the south ranged from a low of 21.7% to a high of 72%. Figure 2 Morbidity surveys were conducted con- shows the parasite rates for all species in 1-4 currently with the parasitological surveys to year olds over time in two groups of villages. ascertain episodes of fever over a preceding Combining the rates from all surveys carried two-week period for each individual. The out in the dry season gives an average rate of methodology and results of these surveys are 60.2% for the three villages north of Madang; reported by Moir et al. (3). A voluntary village the corresponding average rate for villages south aide was trained from each village to dispense of Madang was 37.7%. This difference was chloroquine and amodiaquine to individuals statistically significant (α = 0.05). Respective presenting with fever and to record for each rates for the wet season were 60.2 and 48.9%, episode the individual’s name, house number, which were not statistically different. However, date and treatment given. the difference between rates in the dry and wet seasons for the villages in the south was statist- A questionnaire was used to collect data on ically significant, suggesting a greater increase socioeconomic status. Each head of household in transmission in the wet season in the south. was interviewed in all villages. In addition to Differences in the 1-9 years age group, while providing information on selected socio- smaller, were also observed. economic indicators, data from the question- naire permitted enumeration of the domestic The predominant species of parasite in the animal populations of each village. Surveys of study area is Plasmodium falciparum (2). This
13
97 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J
14
98 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Volume 29, No. 1, March 1986 was consistent in the six villages included in the intense transmission at that time. Maraga study. In the village with the highest proportion appeared to show a more gradual rise in spleen of Plasmodium vivax infections (Maraga), 48 of rates, again with the steepest rise in the wet 221 (21.7%) infections were of this species. In season of 1984. Only one data point was the village with the lowest proportion (Butelgut) available for Dogia before the three-monthly 26 of 206 (12.6%) infections were positive for surveys; however, based on analysis of other P. vivax. The difference was significant (α =0.05). parameters it is considered likely to demonstrate Plasmodium malariae and Plasmodium ovale are a pattern similar to that observed in Maraga. found in the study area, but at a prevalence too Spleen rates in adults were high (> 60%) in low for meaningful analysis at the village level Butelgut, Mebat, Budip and Dogia, while in Sah (2). With few exceptions parasite densities for the average rate in adults was 52.5%. Participa- P. falciparum (asexual forms) were highest in tion in Maraga was too low among adults to be the wet season for all age groups. Densities able to report a valid rate. The variation in the reached the highest level in 1-4 year olds with proportion of enlarged spleens grade 2 averages in two villages during the wet season (Hackett) or greater was significant between the higher than 3400 parasites/mm3 blood. Adults in wet and dry seasons in all age groups in all villages had the lowest densities. Average Butelgut and in 2-9 year olds in Maraga. Table 1 parasite densities were lower than 600 shows the proportion of enlarged spleens grade parasites/mm3, with one exception in which a 2 or greater by season for two age groups. The slightly greater average density was observed in average enlarged spleen (AES) (4) in the 1-4 one village at the end of the wet season. years age group varied over the four surveys from a low of 1.7 (Maraga and Dogia) to 3.1 Spleen rates in children 2-9 years have (Mebat). The AES in this age group was con- consistently been above 75% in Butelgut, Mebat sistently higher in the villages of Butelgut, and Budip since 1981. In the village of Sah they Mebat and Budip than in Maraga, Dogia and were significantly lower but showed a steep rise Sah, with the exception of Sah in the wet during the wet season of 1984, which suggests season of 1984. In both groups of villages and
15
99 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Papua New Guinea Med J
all age groups the observed AES was higher in available to serve as alternative hosts would also the wet season. appear to account for some variation in parasite rates as indicated in preliminary multivariate Utilization data from voluntary village analyses (Cattani, unpublished data). We believe aides were used to estimate fever attacks over a that the relative degrees of infection observed 12-month period. Table 2 shows the number of in the study area are important in terms of future fever cases by age group standardized by the evaluation of malaria control measures, population in each age group for each village. including possibly an antimalarial vaccine. The Butelgut shows the highest utilization of the efficacy of intervention has been measured in village aide, but the interpretation of these data the past by differences in estimated parasite or is not entirely straightforward. spleen rates. If intervention and control areas are selected geographically without regard to DISCUSSION moderate but consistent differences in malario- metric indices the effects of an intervention This study suggests significant variation could be significantly under- or over-stated. among villages in a relatively small geographical area that is not readily apparent in the standard Morbidity as measured by attacks of fever malariometrical indices from one or even two presenting to the village aide was significantly cross-sectional surveys. The hypothesis that higher in Butelgut but comparable among the differing levels of endemicity exist in the study other villages. This finding was a1so reported by area was supported by comparisons of parasite Kass (5) in the same study area. Interpretation rates in 1-4 year olds and of proportions of of morbidity data from village aide utilization is enlarged spleens grade 2 or greater in the 2-9 complicated by the alternative sources of care years age group for both seasons. These available to inhabitants in these villages. In differences would appear to be influenced both by entomological factors, such as vector density, addition to the permanent sources of health sporozoite rates and average bites by sporozoite- care identified in Figure 1, maternal and child infected mosquitoes per night (T. Burkot, health mobile clinics visit most villages on a P.M. Graves and I.D. Charlwood, unpublished monthly basis and there is a large outpatient data), and the use of antimalarial drugs and bed clinic at the hospital in Madang. Antimalarial nets in villages. The number of pigs and dogs drugs are also dispensed in schools and are
16
100 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Volume 29, No. 1, March 1986 available commercially in Madang town. The REFERENCES number and diversity of sources of antimalarials 1 Charlwood JD, Graves PM, Alpers MP. The precluded attempts to follow up all fever ecology of the Anopheles punctulatus group of episodes in the study villages. Moir et al. (3) mosquitoes from Papua New Guinea: a review of reported the following figures from a follow-up recent work. Papua New Guinea Med J 1986; of 106 fever cases from villages with high 29:19-26. 2 Cattani JA, Tulloch JL, Vrbova H, Jolley D, utilization of village aides: 50% received Gibson FD, Moir JS, Heywood PF, Alpers MP, chloroquine or amodiaquine from the village Stevenson A , Clancy R. The epidemiology of aide; 19.6% sought antimalarials from another malaria in a population surrounding Madang, source; 4.9% received treatment not related to Papua New Guinea. Am J Trop Med Hyg 1986; 35:3-15. malaria; and 25.5% received no treatment. 3 Moir JS, Tulloch JL, Vrbova H, Jolley DJ, Four cases were excluded because the fever had Heywood PF, Alpers MP. The role of voluntary developed within 24 hours of the time of the village aides in the control of malaria by survey. These figures may be used as a guide for presumptive treatment of fever. 2. Impact on village health. Papua New Guinea Med J 1985; interpreting village data, but hypotheses 28:267-278. regarding differences in fever attack rates 4 Bruce-Chwatt LJ. Essential Malariology. between villages could not be tested from these London: William Heinemann Medical Books, data. The difficulties of obtaining good data on 1980: 354 p. malaria morbidity on a village basis should be 5 Kass R. Malaria and primary health care: a study of village level workers used in the control of recognized in attempts to design and evaluate malaria in rural Papua New Guinea. MSc. Disserta- interventions. tion, University of London, 1983: 170p.
17
101 PNG Med J 2005 Mar-Jun;48(1-2):102-108 Hereditary Ovalocytosis in Melanesians D. AMATO AND P.B. BOOTH
A distinctive type of hereditary ovalocytosis has been found in Papua New Guinea and a few areas of Southeast Asia. Its main features include a high incidence among tropical lowland dwellers, autosomal recessive inheritance, specific depression of a number of red cell antigens, a characteristic morphology in blood films, and an effect on the erythrocyte sedimentation rate.
Speculation has occurred as to whether the high incidence of ovalocytosis in malarious areas may be related to a selective advantage possessed by ovalocytics with regard to severe malaria. Preliminary data tend to support this hypothesis, but the evidence is not conclusive and much further work is needed.
PRELIMINARY NOTE ON systematic study of high-frequency ovalocy- NOMENCLATURE tosis was reported by Lie-Injo in 1965; this article mentions a frequency of 12.3% The terms ‘ovalocytosis’ and ‘elliptocy- among 440 healthy and hospitalized abori- tosis’ have long been used interchangeably ginal Malayans. for the sporadic type of hereditary ellip- tocytosis which is inherited in an autosomal Other studies have revealed high inci- dominant manner and which has been dences of ovalocytosis among several popu- known for over 70 years. Since the condi- lations: 6.6 to 20.9% in several groups of tion which is the subject of this review, Malayan aborigines (Lie-Injo, Fix, Bolton although also hereditary, differs in several and Gilman, 1972); 12.7% among the land important ways from the sporadic type, we Dayaks and 9.0% among the sea Dayaks of shall not use the terms interchangeably. Sarawak (Ganesan, Lie-Injo and Ong, Rather, ‘ovalocytosis’ will refer to the high- 1975); and 1.7% among the Batak: and frequency type discussed in detail below, 7.2% among the Minangkabau in North and ‘elliptocytosis’ will denote the sporadic Sumatra (Sembiring, Siregar and Kosasih, type. 1975) . INTRODUCTION Although the condition has been known for some years in Papua New Guinea A distinctive type of hereditary ovalocy- (P.N.G.), its first mention in the literature tosis is common among Papua New Gui- did not occur until 1975 (Isbister, Amato neans of coastal and insular origin. In this and Woodfield, 1975). In that same year, article, we review the salient features of a study of 1,020 blood films of in-patients this condition: its incidence, geographic and out-patients at the Port Moresby distribution, inheritance, appearance in General Hospital (P.M.G.H.) found oval- blood films, clinical significance, serological ocytosis in 11.2%; analysis of smaller findings, and the question of whether the numbers of films from Pari and Hisiu abnormality confers a selective advantage; villages in the Central Province showed in addition, the chief differences between frequencies of 22.4% and 16.8%, respec- this type of ovalocytosis and the sporadic tively (Amato, 1975). type of hereditary elliptocytosis will be summarized. Its incidence among 334 Waskia Kar Kar Islanders (Madang Province) was 13.8% INCIDENCE (Booth, Serjeantson, Woodfield and Amato, 1977); and that among 583 residents of The first mention of a high frequency of Kikori, Malalaua and Kerema in the Gulf ovalocytosis in this part of the world was made almost 40 years ago in a Dutch language journal by Bonne and Sandground - D. Amato, M.D., F.R.C.P. (C.), Haematologist, Port Moresby General Hospital. Boroko, P.N.G. (1939); the phenomenon was mentioned - P.B. Booth, F.R.C. Path, Director, Red Cross Blood incidentally in an article on echinostomiasis Transfusion Service, Boroko, P.N.G. in Celebes. To our knowledge, the first
26 VOLUME 20, NO. 1, MARCH, 1977
102 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Province was 12.9% (Amato, 1976) . It is It has been stated (Lie-Injo, L.E., per- also seen frequently at Wewak Hospital in sonal communication) or implied (Baer, the East Sepik Province, and at the Angau Lie-Injo, Welch and Lewis , 1976) that the Memorial Hospital in the Morobe Province ovalocytosis found in Malaysia and Indo- (Spark, R., and Crane, G.G., personal com- nesia is autosomal dominant; however, the munications). type of rigorous statistical study of families necessary to distinguish a dominant from a Thus its incidence among coastal popula- high-frequency recessive character has not tions in P.N.G. appears to be between 5 yet been reported from those populations. and 20 per cent, i.e., of a similar order to the frequencies reported from the Malay- sian and Indonesian populations mentioned It is clear from family studies in P.N.G. above. This is several hundred times more (Booth et al., 1977) that individuals hete- frequent than the sporadic type of ellipto- rozygous for ovalocytosis usually display cytosis, the incidence of which is about normal red cell morphology. There is evi- 0.02 - 0.05 per cent (Cooper and Jandl, dence (Serjeantson, S., personal communi- 1972). cation) that occasionally the red cells of an obligate heterozygote assume the oval GEOGRAPHIC DISTRIBUTION shape after storage for a few days in EDTA. In the P.M.G.H. study (Amato, 1975), of MORPHOLOGY 97 patients with ovalocytosis whose home province was ascertained, 96 came from 10 coastal or island provinces. The one high- The high-frequency type of ovalocytosis lander was from the Tari Sub-province of is readily recognized on a well prepared the Southern Highlands Province. Serologi- thin blood film. Under low power (100 x), cal studies ( see below) provide further examination of the thicker area of the film evidence for a coastal distribution of the reveals that the red cells are ‘piled to- trait, with the exception of parts of the gether’; ie., they do not form rouleaux as Southern Highlands (Booth, 1972; Booth seen in the thick areas of non-ovalocytic films. and Homabrook, 1972b). Moreover, there is evidence, from other genetic markers, of coastal affinities of some Southern High- At higher power (400 x), in the thin lands populations, especially those from part of the film the great majority of cells the Lake Kutubu area (Booth and Horna- are seen to be oval in shape, some of them brook, 1972a). To date, ovalocytosis and/or only mildly so (Figure 1). A few oval antigenic depression has not been found in macrocytes are usually present, as are a any person from the other highland pro- small number of elongated cells with blunt vinces (Booth, 1972; Booth and Horna- or squared-off ends (‘bacillary’ forms). brook, 1972b; Bashir, H., personal commu- Stomatocytes (red cells with a slit-shaped nication; Amato, D., unpublished observ- central pallor) and knizocytes (cells with ations). a double pallor separated by a well hae- moglobinized narrow area) are commonly There is little doubt, then, that the type seen; their numbers range from few to of ovalocytosis under discussion is found many. only among groups originally of coastal or island (i.e., lowland) origin. The autosomal dominant type of ellipto- cytosis is occasionally seen in P.N.G. (Pryor INHERITANCE and Pitney, 1967; Crane, G.G., personal communication; Amato, D., unpublished It has long been known that the sporadic observations). In the families studied with type of elliptocytosis is inherited as an this type, the morphology has been differ- autosomal dominant (Cooper and Jandl, ent from the high-frequency type: the red 1972). cells of the former have a greater length- to-width ratio, i.e., they are more cigar- In contrast, data from Kar Kar Island shaped. In addition, rouleaux can be seen (Booth et al., 1977) are consistent with an in the thick part of the film; and stomato- autosomal recessive inheritance for the high cytes, knizocytes and oval macrocytes do frequency type of ovalocytosis seen in not form part of the picture. P.N.G.
PAPUA NEW GUINEA MEDICAL JOURNAL 27
103 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
Figure 1. Appearance of ovalocytosis on blood film. Most of the red cells are oval in shape. There are a number of stomatocytes (slitshaped central pallor). Arrow indicates a knizocyte (Bessis, 1974). Original magnification x 400.
Occasionally, a film is seen with a smal- CLINICAL SIGNIFICANCE ler proportion of oval cells (i.e., < 50%), and these may show some of the charac- It has long been known that sporadic teristics of the high-frequency type men- hereditary elliptocytosis may be associated tioned above. Whilst it is possible that the with haemolysis. While the majority of gene for ovalocytosis may show variable cases show no evidence of haemolysis, a penetrance, it is our opinion that such minority (10-15%) have a compensated films should not be classified as ovalocytic, haemolytic state; i.e., there is a mild degree especially since there are many other con- of haemolysis, but this does not exceed the ditions which can be associated with small erythropoietic capacity of the marrow and numbers of oval or elliptical cells. thus there is no anaemia. In an even smal- ler minority, there is overt haemolytic anae- The presence of a few oval macrocytes mia which is usually improved or corrected raised the question of megaloblastosis. by splenectomy (Cooper arid Jandl, 1972). However, among 26 patients with ovalocy- tosis who underwent marrow aspiration in In contrast, high-frequency ovalocytosis a 2½ year period, only 4 had definite does not appear to be associated with hae- megaloblastic changes. This proportion did molysis. In the survey of P.M.G.H. patients not differ significantly from that seen in (Amato, 1975), the frequency distribution non-ovalocytics (chi-square = 0.07, p < of haemoglobin concentrations was essen- 0.75) (Amato, D., unpublished observa- tially similar when comparing ovalocytic tions) . and non-ovalocytic patients of either sex; the modal range of haemoglobin was the The megaloblastic ovalocytics did not same (9.1 - 11.0 g/ dl) in all four groups. show greater numbers of macrocytes in the peripheral film than those ovalocytics with- Thus, while patients with ovalocytosis out megaloblastosis. However, other chang- may certainly be anaemic, they are anaemic es were seen in the former which can be for the same reasons that other Papua New associated with megaloblastosis: hyperseg- Guineans may be anaemic, e.g., iron defi- mented neutrophils in all 4, Howell-Jolly ciency, infection, folate deficiency, malaria, bodies in 3, basophilic stippling in 1, beta-thalassaemia minor, tropical spleno- megaly syndrome, etc. Even when there is leucopenia in 1 and thrombocytopenia in evidence of haemolysis, causes other than 1. ovalocytosis can be found.
28 VOLUME 20, NO. 1, MARCH, 1977
104 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
We have yet to see a patient with the IT weakness was not artefactual (anti-H, D high-frequency type of ovalocytosis who is -N, -P1 and -I ) were not among those haemolyzing on the basis of the red cell directed at antigens subsequently shown membrane defect. A similar situation ap- to be of the depressed series, and so the pears to obtain for the high-frequency more widespread nature of the antigenic ovalocytosis seen in Malaysia and Indonesia depression remained unrecognized until in (Lie-Injo et al., 1972). 1974 investigations in Christchurch, New Zealand, of warm type autoimmune hae- The main significance of this type of molytic anaemia antibodies of anti-LW ovalocytosis for the clinician would appear specificity, demonstrated that LW was to be its effect on the erythrocyte sedimen- T tation rate (ESR). The ESR depends on a depressed on I weak cells. Further investi- number of factors, one of the most impor- gations have shown that the depressed series of antigens include IT, IF, LW, D, C, e, S, s, tant being the degree to which the red b a b a b a b cells form rouleaux in the sedimentation U, Kp , Jk , Jk , Xg , Wr , Sc1, En , and Di , tube (Dacie and Lewis, 1975). This pro- which are simultaneously affected when D perty is in turn enhanced by several factors, present. Reactivity of H, A, A1, B, I , i, b a a a such as high levels of serum globulins, P1, M, N, Lu , k, Fy , Co , Vel, Ge and Jr especially fibrinogen (an acute phase reac- appeared to be within the normal range (Booth, tant) and gamma-globulins, and an increas- 1975). ed ratio of plasma to red cells (anaemia). It was eventually realized that the work Yet patients with the high-frequency on ovalocytosis, and that on antigenic de- type of ovalocytosis usually do not develop pression, represented complementary inves- an increased ESR in association with many tigations of the same hereditary condition. conditions where such an increase would be A series of 235 bloods previously screened expected. Thus we have seen 3 ovalocytic for antigenic weakness was examined, patients with acute rheumatic fever, 2 with blind, for ovalocytosis, and the statistical multiple myeloma, and numerous others test for association of antigenic status and with acute infections and/or anaemia, in red cell morphology gave chi-square > whom the ESR was within normal limits. 179.0, thus removing all doubt about the This phenomenon is most likely due to the correlation. inability of these ovalocytes to form rou- leaux. The antigens of the depressed series include those against which haemolytic Thus, if a patient has a normal ESR in a auto-antibodies act, which suggests that situation where one would have expected another common factor affecting these anti- an elevated rate, the blood film should be gens may be a propensity to undergo checked. If it is typically ovalocytic, then changes resulting in the body failing to the ESR is of no value in diagnosing or recognize them as part of itself, and treat- following the patient’s illness. ing them as foreign antigens.
SEROLOGICAL CORRELATIONS The antigenic status of the Indonesian and Malaysiau ovalocytic bloods is not yet In 1966, Booth, Jenkins and Marsh re- known, but when ascertained should im- ported that a cold-acting auto-antibody commonly encountered in Papua New mediately either confirm or rebut the hypo- Guinean sera, recognized a ‘new’ blood thesis that these arise from the same here- group antigen, related to Ii and designated ditary condition as exits in P.N.G. T I . It was soon apparent that selective T depression of IT occurred in some 15% of The I strength of cells of individuals coastal Melanesians (Booth, 1972), and this from 6 different families with (sporadic) was shown by family studies to be inherited elliptocytosis, both haemolytic and non-hae- as an autosomal recessive characteristic molytic, has been found to be normal (Booth and Hornabrook, 1972b). (Booth et al., 1977; Woodfield, D.G., Ama- to, D., unpublished data), thus providing Unfortunately, during the early work, a further distinguishing feature between other antibodies selected to show that the the two conditions.
PAPUA NEW GUINEA MEDICAL JOURNAL 29
105 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
THE MALARIAL HYPOTHESIS children aged 6 months to 5 years present- ing at Madang Hospital with fever (> 38° Geographical location rather than ethnic C). In this way, it was hoped to secure a or linguistic group appears to determine greater proportion of malarious subjects. the occurrence of ovalocytosis/antigenic However, a possible disadvantage emerged, depression in P.N.G. Also, the incidence of as only 14 ovalocytics (5.5%) were found the condition is notably constant from in the series, while among 84 afebrile population to population, which is in con- children (controls) the rate was 8.3%. trast to the very variable distribution of Even so, 119 of 241 normocytic children blood group alleles in the same populations. had malarial parasitaemia, compared with These considerations suggested (Booth, 3 of the 14 ovalocytics, which is a signific- 1975; Amato and Andrews, 1975) that antly lower rate (p < 0.05). When the some selective mechanism might apply, analysis was restricted to those with falci- though as the incidence of ovalocytosis parum malaria, there was a lower, but non- seemed not to exceed 20%, it might be significant rate of parasitaemia among the that the condition conferred advantage at ovalocytics. some period of life, and disadvantage at another, supposing that equilibrium gene Baer et al. (1976) have reported resis- frequencies for the hereditary condition tance to higher grades of malarial parasi- have been attained in coastal regions and taemia among ovalocytic Temuan people in that ovalocytosis is not neutral in effect. If Malaysia (two-thirds of malaria seen in the ovalocytics were advantaged during child- region is due to P. falciparum). They also hood and disadvantaged in utero or in found a higher frequency of ovalocytosis adulthood, a balanced polymorphism could among adults than among children, which result. would be expected if ovalocytics possessed a selective advantage. However, Serjeant- The congruent distribution of ovalocyto- son et al. (1977) found no significant dif- sis and malaria in P.N.G. has prompted ference in the incidence of ovalocytosis some preliminary investigations, which between adults and children in P.N.G. have provided results tantalizingly sugges- tive, but not, in general, statistically signi- It is noteworthy that the protection af- ficant of an increased resistance to malaria forded by the sickle-cell trait (Hb S trait) in ovalocytic children. The main difficulty against malaria is most obvious when the arises from the relative paucity of ovalocy- deficiency of sicklers among children dying tic compared to normal subjects in any of malaria is treated statistically (Motulsky, random series, and, of course, this is com- 1964). Whether an analogous study would pounded by the need to sub-divide the prove feasible in regard to ovalocytosis and subjects into those with and without mala- malaria in P.N.G. is doubtful. rial parasites present. Clearly, an apparently relatively minor The data of Serjeantson, Bryson, Amato, change in red cell membrane composition and Babona ( 1977), on children aged can profoundly affect susceptibility to ma- 2-14 years from Kar Kar Island and laria. In the absence of the Duffy blood Gogol Valley, show that the likelihood of group antigens (Fya and Fyb), as occurs infection in ovalocytic children is about in 80-90% of West Africans, P. vivax can- 75% that in normocytic children for Plas- not gain entry to the erythrocytes (Miller, modium falciparum, P. vivax and for all Mason, Clyde, and McGinniss, 1976), pre- species combined. Of 397 children exami- sumably because in this situation the re- ned, 52 were ovalocytic, and of these only ceptor normally activated by this parasite 15 had demonstrable parasites. These work- is also absent. Fy (a-b-) cells have normal ers emphasize the very large sample size morphology and survival. needed to demonstrate significant differen- ces in parasitaemia rates between normocy- If ovalocytosis confers any resistance to tics and ovalocytics. malaria, then the mechanism for this pro- tection seems likely to occur after entry In an attempt to overcome this problem, into the red cells by the parasites, as Babona and Amato (1976) studied 255 ovalocytic subjects can be seen with para-
30 VOLUME 20, NO. 1, MARCH, 1977
106 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 parasitaemia, and more specifically, parasites ADDENDUM can be seen in oval cells. Professor Sir John Dacie has recently COMPARISON OF ‘ELLIPTOCYTOSIS’ brought to our attention an article (Honig, AND ‘OVALOCYTOSIS’ G.R., Lacson, P.S., and Maurer, H.S. (1971) A new familial disorder with ab- The main features which help to differ- normal erythrocyte morphology and in- entiate these two conditions are summari- creased permeability of the erythrocytes to zed in Table 1. sodium and potassium. Pediatric Research
NATURE OF THE MEMBRANE DEFECT 5:159), in which the authors describe ovalocytosis without anaemia or evidence Very little is known about the nature of of haemolysis in several members of a the red cell membrane defect( s) in either Filipino family. From their description and of these two abnormalities. No consistent photomicrographs, it appears that the con- differences between patients with the hae- dition is identical with the one seen in molytic and non-haemolytic forms of here- P.N.G. The significance of the report with ditary elliptocytosis have been found. respect to this type of ovalocytosis is twofold: It is probable that the defect in ovalocy- tosis associated with antigenic depression 1. Several in vitro abnormalities (without involves the portions of the membrane apparent clinical significance) are identi- where the depressed antigens are normally fied in this condition, including decreased expressed. One might assume either that osmotic fragility, increased autohaemolysis the antigens subject to depression all de- after 48 hr incubation, increased glucose pend, for their full expression, upon the consumption by red cells, and decreased same membrane component(s), or alter- intracellular potassium; natively that the membrane anomaly in some way physically obscures the antigenic 2. The disorder is documented in yet sites. The series of depressed determinants another country of the Southeast Asia-Me- would thus be associated by structure and/ lanesia region (though we are not aware or spatial orientation on the red cell mem- of any reports bearing on its incidence in brane. the Philippines).
PAPUA NEW GUINEA MEDICAL JOURNAL 31
107 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
ACKNOWLEDGEMENT blood group system occurring in certain Melanesian sera. Brit . J. Haematol. 12: 341. We are grateful to Mrs D. Skelton for secretarial assistance. BOOTH, P.B., SERJEANTSON, S., WOODFIELD, D. G., AND AMATO, D. (1977) Selective de- REFERENCES pression of blood group antigens associ- ated with hereditary ovalocytosis among AMATO, D. (1975) Elliptocytosis in Papua Melanesians. Vox Sanguinis 32: 99. New Guinea. Proc. 10th Ann. Symp., Med. Soc. P.N.G., Port Moresby, pp. 140 - 154. COOPER, R.A, AND JANDL, J.H. (1972) Here- ditary elliptocytosis. In Hematology (ed AMATO, D. (1976) Falciparum malaria in W.J. WILLIAMS) , New York, MCGRAW- Melanesians with hereditary ovalocytosis HILL, pp. 387-389. and selective antigenic depression. Proc. 16th Int. Congr. HematoI., Kyoto, p. 362. DACIE, J.V. AND LEWIS, S.M. (1975) Practic- (Abstract) al Haematology. Edinburgh, Churchill Liv- ingstone, pp. 580-582. AMATO, D., AND ANDREWS, S. (1975) Ellipto- cytosis in Papua New Guinea. Proc. 3rd GANESAN, J., LIE-INJO, L.E., AND ONG, B.P. Meeting Asian Pacific Div., Int. Soc. He- (1975) Abnormal haemoglobins, glucose- 6- matol., Jakarta, p. 124. (Abstract) phosphate dehydrogenase deficiency and hereditary ovalocytosis in the Dayaks of BABONA, D., AND AMATO, D. (1976) Heredi- Sarawak. Proc. 3rd Meeting Asian Pacific tary ovalocytosis and malaria in children. Div., Int. Soc. Hematol., Jakarta, p. 123. Proc. 12th Ann. Symp. Med. Soc. of P.N.G., (Abstract) Lae. ISBISTER, J.P., AMATO, D., AND WOODFIELD, BAER, A., LIE-INJO, L.E., WELCH, Q.B., AND D.G. (1975) The blood film in the investi- LEWIS, AN. (1976) Genetic factors and gation of anaemia. Papua New Guinea Me- malaria in the Temuan. American Journal dical Journal 18: 47. of Human Genetics 28: 179. LIE-INJO, L.E. (1965) Hereditary ovalocyto- BESSIS, M. (1974) Corpuscles - Atlas of Red sis and haemoglobin E-ovalocytosis in Ma- Blood Cell Shapes. Berlin, Springer- Ver- layan aborigines. Nature 208: 1329. lag. LIE-INJO, L. E., FIX, A., BOLTON, J.M., AND BONNE, C., AND SANDGROUND, J.H. (1939) GILMAN, R.H. (1972) Haemoglobin E-here- Echinostomiasis in Celebes veroorzaakt ditary elliptocytosis in Malayan aborigines. door het eten van zoetwatermossellen. Acta Haemat. 47:210. Geneesk. Tijdschr. v Nederl. - Indie 79: 3016. MILLER, L.H., MASON, S.J., CLYDE, D.F., AND MCGINNISS, M.H. (1976) The resistance BOOTH, P.B. (1972) The occurrence of weak factor to Plasmodium vivax in blacks: the IT red cell antigen among Melanesians. Duffy-blood-group genotype, FyFy. New Vox Sanguinis 22: 64. Engl. J. Med. 295: 302.
BOOTH, P.B. (1975) Selective depression of MOTULSKY, A.G. (1964) Hereditary red cell blood group antigens among Melanesians. traits and malaria. Am. J. Trop. Med. Hyg. 13: Proc. 3rd Meeting Asian Pacific Div., Int. 147. Soc. Hematol., Jakarta, p. 128. (Abstract) PRYOR, D.S., AND PITNEY, W.R. (1967) Here- BOOTH, P.B., AND HORNABROOK, R.W. (1972a) ditary elliptocytosis: a report of two fami- Blood group genetic data from the Lake lies from New Guinea. Brit. J. HaematoI. 13: 126. Kutubu region, Southern Highlands, Pa- pua. Human Biol. Oceania 1: 229. SEMBIRING, P., SIREGAR, A., AND KOSASIH, E.N. (1975) Frequency of ovalocytosis in Me- BOOTH, P.B., AND HORNABROOK, R.W. (1972b) dan (Nort Sumatera). Proc. 3rd Meeting Weak IT red cell antigen in Melanesians: Asian Pacific Div., Int. Soc. Hematol., Ja- family and population studies. Human Biol. karta, p. 125. (Abstract) Oceania 1: 306. SERJEANTSON, S., BRYSON, K., AMATO, D., AND BOOTH, P.B., JENKINS, W.J., AND MARSH, W.L. BABONA, D. (1977) Malaria and hereditary (1966) Anti-IT: a new antibody of the I ovalocytosis. Human Genet. (in press).
32 VOLUME 20, NO. 1, MARCH, 1977
108 PNG Med J 2005 Mar-Jun;48(1-2):109-121
TRENDS IN SEXUALLY TRANSMITTED DISEASE INCIDENCE IN P APUA NEW GUINEA
Candy K. Lombange
Papua New Guinea Institute of Medical Research
Dr Candy K. Lombange Papua New Guinea Institute of Medical Research P. O. Box 60 Goroka Papua New Guinea
SUMMARY Ten years ago the five-year National Health Plan was introduced. The objectives of the A retrospective study of gonorrhoea and 1974-1978 National Health Plan (5) in respect syphilis from Health Department records was to sexually transmitted disease control were to carried out in Papua New Guinea. During the reduce the incidence of the disease through: ten-year period (1974-1983) 101,636 new cases of gonorrhoea and 34,422 of syphilis were 1. readily available, free and confidential reported among the general population of treatment; Papua New Guinea. The incidence of both 2. contact tracing and treating sexual part- sexually transmitted diseases have significantly ners; (P < 0.005) increased over the decade despite the introduction and implementation of the 3. training of health personnel; National Sexually Transmitted Disease (STD) 4. health education; Control Programme. Some which contribute 5. coordinated efforts of all agencies to the present increase in sexually transmitted important in the control of the disease. diseases are segregation of health and non-health services, insufficient staff training and increased The targets of the plan were to have by 1978: immigration to urban centres. A. 50% of STD patients naming and locat- INTRODUCTION ing sexual partners and 50% of these contacts being treated; Venereal diseases did not exist in Papua New B. 95% of all patients coming for treatment Guinea before contact with Western civilization. notified and reported to National In 1874 early in the medical history of this Disease Control Headquarters; country Rev. Lawes reported that no syphilis C. established STD clinics staffed by HEOs was seen among Papuans (1). Similarly, in the or nurses at most centres; same century there was no record of reported D. at national level a specialist appointed cases of venereal disease in German New in venereal diseases; and Guinea (2). E. the teaching of venereology to all health personnel. In 1902, Wenland first reported sexually transmitted disease in New Guinea (3) while in The proposals of the plan recommended 1908 Bellamy reported a 5% prevalence of STD some important changes to improve the manage- in Papua (4). After World War II sexually transmitted disease was noticed as prevalent in ment and control programme of venereal disease. the Trust Territory of Papua and New Guinea. Among these proposed changes was the need The first case of syphilis in Papua New Guinea to direct and evaluate the control measures and was reported in 1960. Subsequently, an the recommendation that the venereologist epidemic of syphilis was reported in the New co-ordinate action at national and provincial Guinea highlands region in 1969 (4). By levels. 1970-1972 gonorrhoea was prevalent through- out the country whereas syphilis was more pre- Evidently the STD Control Programme valent along the highlands highway. which ended five years ago had not been
145 109 PNG Med J 2005 Mar-Jun;48(1-2):
PAPUA NEW GUINEA MEDICAL JOURNAL evaluated. It is important that health planners establishment units into the provincial total. and responsible organizations know the out- These provincial total case morbidity figures come of the implemented programme. are submitted to the National Disease Control statistics section for final reference. MATERIALS AND METHODS There was a total of 120 EPINT records The method used was a retrospective record studied from the National Disease Control research of gonorrhoea and syphilis morbidities statistics unit. The case morbidities for in Papua New Guinea. The reported morbidities gonorrhoea and syphilis were analysed over a of both diseases and contact-tracing activities ten-year period. Specific information regard- throughout the country for the last ten years ing the operation of the sexually transmitted were studied. The incidences of both gonorrhoea disease control programme was obtained from and syphilis are reported in the EPINT system the national STD control coordinator by which monitors monthly morbidities of the interviewing him. The sister-in-charge of the diseases at the provincial level. The provincial PMGH STD clinic was also interviewed for disease surveillance agents collate statistics from supplementary information in regard to the special STD clinics and peripheral health operation of the STD programme.
146
110 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
VOLUME 27, Nos. 3-4, SEPTEMBER - DECEMBER 1984
RESULTS Both the diseases have comparatively in- creased over the decade despite the implemen- National Morbidity tation of the STD Control Programme (Table 1 In the period 1974-1983 gonorrhoea and and Figure 1). syphilis surveillance reported the following. Regional Distribution Gonorrhoea Tables 2 and 3 show the regional morbidi- There was a total of 101,638 new cases ties for both the diseases and the distribution of reported among the general population of the incidences between the period of implementa- country, giving an incidence over 10 years of tion of the National Health Plan and the sub- 31.76/1000. Of these, 42,930 cases were sequent five years (6). from 1974-1978 with a five-year incidence of 15.17/1000 and 58,708 cases for 1979-1983 It can be seen from the results that there with a five-year incidence of 18.34/1000. has been a significant increase in the incidences Between the two five-year intervals the disease of both diseases in recent years. For the high- increased by 26.8%. lands region the incidence of gonorrhoea has come down, which is partly due to poor report- Syphilis ing, especially no reporting from the Eastern A total of 34,422 new cases was reported Highlands Province since 1982. from the general population of the country with an incidence over 10 years of 10.75/1000. There have been very late or even no monthly From this there were 11,666 cases in 1974-1978 reports sent in from some provinces. This is at a five-year incidence of 4.12/1000 and in indicated in Table 3 and Figure 2, where there 1979-1983 22,756 cases, with a five-year are very low numbers in certain provinces incidence of 7.11/1000. Between these two compared to their populations. Despite the periods syphilis increased by 48.7%. inconstant reporting, there is no doubt that
147
111 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PAPUA NEW GUINEA MEDICAL JOURNAL
148
112 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
VOLUME 27, Nos. 3-4, SEPTEMBER - DECEMBER 1984
149
113 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PAPUA NEW GUINEA MEDICAL JOURNAL
both diseases are becoming almost out of C. cases being treated privately and not control in all regions. reported to the public disease surveillance.
Urban Distribution Age and Sex Distribution The first National Seminar on STD in 1980 Distributions of gonorrhoea and syphilis by requested the operating STD Clinics to submit age and sex for PNG (1979-1983) are presented full details of diseases, particularly on sex, age in Table 6. These results show that those most and laboratory diagnosis. Before this no clinic commonly affected for both diseases, irrespec- tabulations had been attempted. The reported tive of sex, are 15-24 years old. Male patients cases for the two PNG cities are tabulated from present to the clinic more often than females, EPINT (6) for 1981-1983 in Tables 4 and 5 and because more females are asymptomatic than graphed in Figure 3. their partners.
There has been an increase in the incidence Contact-Tracing Activities of gonorrhoea in both cities. The incidence of The contact-tracing activities for 1979-1983 syphilis has considerably increased in the Port are shown in Table 7. For the whole country Moresby clinic while there had been a significant 11 % of the sexual contacts have been traced decline (P < 0.005) in the incidence of syphilis and 86.5% of those cases found were treated. in the Lae clinic. This discrepancy is probably This is a very small portion of case-detection related to: activities compared to the National STD Plan target of 50% naming and locating of sexual A. underreporting; partners. Each region has done very poorly B. uncertainty of diagnosis of genital ulcers; or with contact-tracing activities (Table 8).
150
114 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
VOLUME 27, Nos. 3-4, SEPTEMBER - DECEMBER 1984
Secondary Syphilis Crude Birth Rate. Incidences of gonorrhoea From 1979 to 1983, a total of 848 cases of and syphilis in 1971-1972 were 2.1/1000 and secondary syphilis have been reported and this 0.7/1000, respectively (4). In 1974-1983 incid- accounts for about 3.7% of the known syphilis ences of gonorrhoea and syphilis were 32/1000 cases in that period. In Port Moresby General and 11/1000, respectively. This is an eighteen- Hospital STD Clinic during 1979-1983 more fold increase compared with 1972 (9). than half of the secondary syphilis cases have been reported (466 cases), which is 19.2% of Despite the introduction and implementation the reported syphilis cases in that five-year of the National Sexually Transmitted Diseases period from the clinic. Control Programme from 1974 to 1978, gonorrhoea and syphilis have become almost There were no cases of late syphilis reported out of control. This is not only evident from in the 1979-1983 period but two cases had the significant increase in the national STD been reported in 1971-1982 (7,8). incidences but also in all regions the disease control programme has become ineffective. DISCUSSION It must be noted that reported incidences The crude rate of gonorrhoea and syphilis represent only a proportion of the general (STD) for the country from 1974 to 1983 is population who seek treatment through these 42.5/1000, which is almost identical to the PNG special STD clinics. There are many patients
151
115 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PAPUA NEW GUINEA MEDICAL JOURNAL
who are missed out in the disease surveillance The transition to a Western socioeconomic system of the country. For instance, asympto- lifestyle has been a contributing factor to the matic female patients with gonorrhoea, those spread of venereal diseases. This is quite evident attending general practitioners’ clinics, and all over the world. The Director-General of patients either being helped out by wantok WHO in 1964 described the movement of health workers or those who come for treat- population as one of the most important factors ment through public outpatient clinics are in the spread of venereal diseases (10). In evidently not reported. With these facts in England in 1967, 39%’of treated patients with mind it is apparent that less than the national syphilis were immigrants (11). This is true in target of cases is being reported to the National PNG with increased trends of STD incidences STD Control Programme. It is quite evident in the two cities and most other towns. It is from the study that sexual transmitted disease obvious that gonorrhoea and syphilis are urban is becoming a serious clinical and psychosocial diseases and their increase is associated with health problem in Papua New Guinea. the rapid growth rate of urban population drift.
The diseases commonly affect the early This is not only a public health problem of reproductive age groups, 15-29 years, especially PNG but is a leading disease morbidity problem males. This is closely correlated with the steady in the world today. WHO reported that 25% of migration of this age group into urban centres the world’s population are affected by sexually associated with socioeconomic development. transmitted diseases. The incidences have been
152
116 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
VOLUME 27, Nos. 3-4, SEPTEMBER - DECEMBER 1984
more or less steady in developed countries, gonococcal endemicity (11). In Sri Lanka both while in developing countries they are increas- syphilis and gonorrhoea increased by 25% from ing rapidly. This can be seen in Africa where 1972 to 1973 (11) and there is further evidence 20-40% of females of childbearing age are of increase in relation to the present socio- suffering from STD and there is a close correla- economic situation. Studies in PNG have tion between high rates of infertility and shown that 15% of all gynaecology admissions
153
117 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PAPUA NEW GUINEA MEDICAL JOURNAL
are for pelvic inflammatory disease (PID) and been 11%, which is well below the target of of these 43% have gonococcus present (12). the introduced programme. This whole issue of contact-tracing activities seems to be a world- It is clear that large population movements wide problem. The reasons for these failures caused by tourism, training schemes and labour are multifactorial, the main obvious ones being migration as well as socio-economic changes social, cultural, legal, economic and human tend to favour the transmission of sexually habit, which tend to hinder the contact-tracing transmitted diseases. activities. Direct contact tracing of sexual partners is becoming very complicated. Contact-tracing activities have gone down throughout the country. Overall activities have However, developed countries have begun a
154
118 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
VOLUME 27, Nos. 3-4, SEPTEMBER - DECEMBER 1984
155
119 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
PAPUA NEW GUINEA MEDICAL JOURNAL control program based on providing information tions. Some of these factors have caused low for increased public awareness of STD. This staff morale which reflects the present problem certainly has had some effect leading to an of STD in the whole country. increase of incidence reporting. For instance United Kingdom and Singapore have reported C. Lack of Teaching Interest that increased incidences of STD reporting The lack of teaching interest in STD at all have resulted from mass health education levels of medical training in the country is campaigns about the disease (10). With this becoming evident. This is shown by the lack health education method, more patients report of professional interest and public awareness to clinics for medical check-ups and seek early of the disease by all health personnel. Training treatment as soon as symptoms start to develop. and teaching of venereology must be properly This could be the appropriate method that aimed at all levels of health training institu- PNG must seriously consider adopting since the tions, from the medical school down to medical direct contact-tracing method has failed. aid training.
There are some operational problems which D. Health Resource Constraints contribute to the present difficulties with the Constraints of trained manpower, financial control of sexually transmitted diseases in resources and specialized technical services are Papua New Guinea. These are: problems everywhere in developing countries. These problems are more political than directly A. Segregation of Services related to health. Therefore health workers can Sexually transmitted disease clinics have do little to solve them, though one way to help apparently been isolated from the main health overcome them would be to integrate the STD care system. For example, in many instances control services with the general health care medical superintendents and hospital doctors system. have completely ignored the proper care and management of STD cases. The author has There are limitations of funding for transport experienced this in the country’s largest city and other services for special STD activities but STD clinics. Doctors carry out no regular this can be solved by proper planning and supervision of sisters and HEOs engaged in STD management of the hospital and provincial management activities. These paramedical resources within its day-to-day services. The workers can do what is required for diagnosis current Government NPEP funding must be and treatment of this serious clinical and fully used by the provinces for extension of psychosocial disease but must be backed up health education activities rather than contact by doctors in the hospital clinics. The best tracing because it is a misuse of resources to way to integrate and provide effective service is spend them on a method known to be unsuccess- to bring STD services under the general package ful. of health care services. This has been adopted in USA and other developed countries where Special funding is not warranted at this stage disease control programmes have thereby of the country’s economy, but rather the become easier in practice and less costly (10). coordination of inter-departmental activities. It has been recommended from a study done in B. Low Staff Morale the highlands that the Department of Educa- From the interview results regarding staff tion can teach sexual and STD education morale, participating staff feel bored by the through primary schools, secondary schools and repetition of duties with no future for advance- teacher training institutions so that the message ment in the field. They are further discouraged can then be disseminated to the community by by the lack of interest shown by hospital them (13). There are other Government depart- doctors. There is no stimulation or encourage- ments such as the Department of Information ment by the provision of inservice or other short Services for public campaigns through the courses for the staff who are actually involved media and distribution of publications, and the in the delivery of the programme by the Department of Youth and Religion for counsel- Department of Health and its teaching institu- ling and public seminars at rallies, etc. which are
156
120 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
VOLUME 27, Nos. 3-4, SEPTEMBER - DECEMBER 1984 needed to help combat this increasing public thank the Sister-in-charge of the PMGH STD health problem. clinic for contributing her views on current operational problems with STD management CONCLUSIONS and the control programme.
Sexually transmitted diseases, especially REFERENCES gonorrhoea and syphilis, are becoming major health morbidity problems comparable to 1. Lawes W G. Australian Medical Gazette. 1887. pneumonia and gastroenteritis in this country. Quoted by Blackburn CRB. op. cit.(3). The vector and host of the diseases are both 2. Calov WL, Weir H. Gonorrhoea in natives of New human and therefore the basic goal of the STD Guinea. Med J Aust 1925; 2: 720-724. control programme can only be achieved 3. Blackburn CRB. Medicine in New Guinea: three and a half centuries of change. Postgrad Med J through developing human awareness and 1970;46: 250. modifying human behaviour. This can be 4. Bell C. The Diseases and Health Services of Papua done by: New Guinea. Port Moresby: Department of 1. Campaign for STD health education Public Health, 1973. - at all levels of schools 5. Department of Health. National Health Plan (1974-1978). Department of Health, Konedobu, - through local and national media Papua New Guinea, 1974. -through all other non-government co- 6. EPINT. Papua New Guinea National Health ordinating agencies, e.g. churches Statistics. Konedobu: Department of Health, - national and local newspapers. 1974-1983. 7. Wilkey IS, Sutherland G. A case of congenital 2. Integration of STD care in the general syphilis in New Guinea. Papua New Guinea Med health care system e.g. STD control J 1971; 14(2): 50-51. becomes part of the overall communicable 8. Adler R, Price M. Late syphilis in a Papuan male. disease control programme. Papua New Guinea Med J 1972; 15(3): 157-158. 3. Permitting the optimal use of existing 9. Pataki-Schweizer KJ, Tabua TW. Psychosomatics of venereal disease: toward a programme for the infrastructures at central, provincial or 1980’s. Papua New Guinea Med J 1980; 23(1): local health units for prevention of STD. 46-50. 4. Establishment of a central unit for training 10. Catterall RD, Nigol S. Sexually Transmitted and evaluation of the control programme Diseases. London; Academic Press 1976. 11. Catterall RD. Sexually Transmitted Diseases: for communicable diseases. Third African Regional Conference. Br J Vener Dis 1983; 59(5): 337. ACKNOWLEDGEMENTS 12. Rooke P. Pelvic inflammatory disease in Port Moresby General Hospital. Papua New Guinea Med J 1982; 25(1): 29-32. I thank Mr John Tawai, the National STD 13. Ree H. Knowledge and attitudes of young people Control Co-ordinator, for making the informa- regarding sexually transmitted diseases. Papua tion available for this project. Secondly, I New Guinea Med J 1982; 25(1): 33-36.
157
121 PNG Med J 2005 Mar-Jun;48(1-2):122-126
A NEW GUINEA POPULATION IN WHICH BLOOD PRESSURE APPEARS TO FALL AS AGE ADVANCES
BY IAN MADDOCKS *, AND LUKE ROVIN†
LARGE surveys of blood pressure among measure complete village communities, as has populations in Europe and the U.S.A. have been done in other surveys in the Western shown that blood pressure tends to rise with age Pacific, because the Chimbu are grouped in small (Master, Garfield and WaIters, 1952 ; Hamilton, family hamlets. In each centre, an attempt was Pickering, Roberts and Sowry, 1954; Boe, made to gather all the members of several Humerfelt and Wedervang, 1957). The prob- “lines”, and it is thought that these were fairly lem of whether this rise can be regarded as completely represented. A “line” is a clan or “normal”, or whether it represents the operation census division, the members of which have of an abnormal condition associated with elevated common ties of family and of land ownership. blood pressure, continues to exercise workers in Older members of each line were particularly this field (Pickering, 1961). sought, and were traced where possible, to their houses or gardens if they had not come to the Blood pressure tends to rise with age even in survey centre. The three different populations populations among whom the disease which we did not differ significantly in any of the para- call “essential hypertension” is rare or virtually meters measured, and they are therefore pre- absent (Maddocks, 1961a). There have been sented as a single population, comprising 268 reports of populations in which no appreciable male and 161 female adults. change in blood pressure with age was recorded: The Samburu people of Kenya (Shaper, Williams METHODS and Spencer, 1961); Ceylonese (Bibile, Cullumbine, Kurtsinghe, Watson, and Wikama- The technique for the measurement of blood nayake, 1949); Caroline Islanders (Murrill, pressure has previously been described (Lovell, 1949) and Kalahari Bushmen (Kaminer and Maddocks and Rogerson, 1960). Subjects were Lutz, 1960); but either because they were some- seated at a table and had been rested for at least what exotic populations, or because techniques 10 minutes before the reading was taken. Height of sampling or measurement were not always was measured to the nearest centimetre, and standard, these reports have not received much weight to the nearest pound on an accurate beam prominence. scale. Skinfold thickness measurements were made below the angle of the left scapula using a This paper reports findings among a popula- Harpenden caliper. tion in which blood pressure appears to fall with age. Age estimation presented some difficulties. It POPULATION. could be done only by arranging the members of The Chimbu district lies in the Eastern High- each line in generations and in order of seniority, lands of New Guinea, and contains an estimated and then estimating the age of certain key population of 160,000. The Chimbu people persons in the line by reference to a single have been described in detail by Brown and memorable event—the first European expedition Brookfield (1959). The populations represented to the area in 1932. Persons not born at that in this survey were from the areas of Mintima, time were now (December, 1962) under 30 Wandi and Gumine. It was not possible to years of age, men who had carried for the expedition were now probably over 50 years of age, and men who were already fathers of several * Lecturer in Medicine, Papuan Medical College. † Student, Papuan Medical College. children at the time of the expedition were now probably over 60 years of age.
VOLUME 8, NO. 1, MARCH, 1965 17
122 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
RESULTS Mean blood pressure = Diastolic + Systolic – Diastolic 3 Findings are grouped in decades of age, and It is close to the physiological mean of blood recorded as means for each decade in Table 1. pressure and provides a way of examining For both males and females, means for height, systolic and diastolic pressures together. weight, skinfold thickness, arm circumference and blood pressure tend to fall in groups of The change in mean blood pressure with age increasing age. For systolic pressures of females, may be represented by simple regression however, mean values are highest in the middle equations: age-range, falling only in the oldest age group. Males—Mean Blood Pressure (the pattern of blood pressures is illustrated in (mm. Hg.) = 95.5 — 0.2024 age (years). Figure 1.) “ Mean” blood pressure is calcu- Females—Mean Blood Pressure lated from the formula: (mm. Hg.) = 93.5 — 0.1023 age (years). For both sexes, the regression coefficient differs significantly from zero as shown by the standard error. S.E. (Males) = ±0.02 ; S.E. (Females) = ± 0.014.
DISCUSSION
The Chimbu people may be unique in having a pattern of blood pressures which falls with increasing age. Almost identical findings reported by Whyte (1958) and Barnes (1964) in other populations from the same region suggest that this pattern is not an artefact of case selection or technique. A number of factors may be con- sidered in attempting to assess why the Chimbu have such an unusual pattern.
1. Selective Mortality and Disease Pattern
The average life expectancy among the Chimbu is probably less than 40 years, though no accurate figures are available. The pattern of blood FIGURE 1. BLOOD PRESSURES OF A CHIMBU POPULATION pressure found among them might be explained SHOWING TENDENCY TO FALL AS AGE ADVANCES. if persons with higher blood pressures were dying young. But the common causes of death in the New Guinea Highlands (acute respiratory
18 PAPUA AND NEW GUINEA MEDICAL JOURNAL
123 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005 infections, diarrhoeal disease and other infec- adult life. Fat is almost absent from the diet tions) are not associated with high blood pres- of the Chimbu. The women of the area carry a sure, and selective mortality does not appear to particularly heavy burden of physical labour and influence blood pressure. a continuous cycle of reproduction, synthesizing foetal and placental protein or producing breast 2. Body Build milk over a continuous period of up to 15 years. This has been described as “the maternal deple- A steady fall in the mean values for all tion syndrome” (Jelliffe and Maddocks, 1964). measurements concerned with body build is a However, Indonesian populations on even lower striking finding for both sexes after the third dietary intakes of protein have a pattern of decade of age. This is not the usual finding in blood pressure which rises with age (Bailey, Western populations, where body weight, arm 1963) and protein deficiency does not seem to circumference, etc., are greatest in the age range 40-60 years (Maddocks, 1964). This suggests that be a sufficient explanation of the Chimbu falling body build may explain falling blood pressures. pressures among the Chimbu, since body build has a definite effect on blood pressure (Boe et 4. Dietary Factors—Salt al., 1957 ; Miall and Oldham, 1958). The import- Salt has long been a valuable trade item in ance of body build can be assessed by calculating the Chimbu area. A small sodium spring in partial regression equations in which the separate the Wahgi Gorge near Gumine was the site of influence of age and body build on blood pres- a flourishing salt-making industry until about sure are represented. In the following equa- 10 years ago. Local missionaries remember when tions, body build is represented by arm circum- nearly 100 huts clung to the sides of the gorge ference, an index which closely parallels body weight and weight/height. near the spring, and here the leaves were soaked in spring water, burned, and the salt extracted Males from the ashes by leaching through bark funnels. Finally the solution was evaporated to dryness to Mean Blood Pressure = form thin flat cakes about 12 inches across. 64.4 — 0.116 age + 1.14 arm circum- This salt, being almost the only good source ference (cm.). of sodium in the whole region, was highly prized, and one cake is said to have been worth a pig. Females Now, however, trade stores sell imported salt throughout the Eastern Highlands, and the old Mean Blood Pressure = 59.9 — 0.042 age + 1.49 arm circum- salt industry has collapsed. ference (cm.). The Chimbu were apparently a population with a low intake of salt, and this was probably true Comparing these with the simple regression of the whole New Guinea Highlands. In West equations presented above, it is apparent that even New Guinea, Oomen (1961) reported the mean after allowing for the fall in body build with age, 24-hour urinary sodium excretion (an accepted blood pressure still falls with age, though the rate measure of sodium intake) as less than 70 mg., of fall is reduced. Some other factor as well as lower than most salt-free diets! Even today, body build must be operating to make pressures with trade-store salt freely available, salt intake fall with age. is low. In 10 of the adult male Chimbu seen on this survey, the average 24-hour sodium 3. Dietary Factors—Protein excretion was 1.5g. about one fifth of the usual European value. The diet of the Chimbu people has been studied by Oomen and Malcolm (1958), Salt intake is relevant to this discussion because Venkatachalem (1962) and in two communities there is a considerable amount of evidence that by Whiteman (1962). The results of their blood pressure is affected by dietary salt. The surveys are similar. They describe an average evidence falls under three main headings: daily intake of between 25 g. and 30 g. protein, and from 1,850 to 2,883 calories. Such a low (i) Epidemiological. Dahl (1961) has sug- protein intake, most of it from vegetable protein, gested that in societies or groups habitually amounts to chronic protein malnutrition, and consuming less than 5g. of salt daily, essen- probably explains the falling body bulk during tial hypertension will be rare, and in societies
VOLUME 8, NO. 1, MARCH, 1965 19
124 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
or groups habitually consuming more than members of “primitive” communities move 10-15g. of salt daily hypertension will be away from the simple life to take up a com- common; petitive wage-earning existence in the cities, hypertension begins to appear. It has not so far (ii) Therapeutic. Severe salt restriction had been possible, however, to definitely incriminate an accepted place in the therapy of hyper- any specific environmental factors (Scotch, 1960; tension before potent hypotensive drugs Maddocks, 1961b; Cruz-Coke, 1963). became available; and We may hope that repeated careful observation (iii) Experimental. Certain strains of rats of populations like the Chimbu will enable us can be rendered hypertensive with an exces- to record such a change if it is going to occur. sive sodium intake, the degree of hyper- tension running parallel to the amount of Unfortunately, even in such isolated communities, sodium ingested. Young rats are more so many other changes in diet, social organiza- prone to this hypertension than old rats, and tion and pattern of activity will occur simul- excess potassium in the diet has a protective taneously that it is unlikely that the operation of effect (Meneely and Dahl, 1961 ; Lemley- specific environmental factors will be apparent. Stone, Darby and Meneely, 1961). The answer to the problem of the aetiology of essential hypertension will probably not come If men behave at all like rats, the Chimbu are from such studies. Nevertheless it would be favoured by both their low sodium intake and foolish to neglect the study of populations which by high potassium content of the sweet have no hypertension, because in a few genera- potato which is their staple. The mean potas- tions they may have ceased to exist. sium/sodium ratio in the urine (usually approxi- mately one in Europeans), was 24.2 in the ten adult males mentioned above. Further, experi- SUMMARY mental hypertension in rats suggests the possi- bility that young men have been more suscep- The Chimbu people of the New Guinea High- tible to the recent increase in salt consumption lands have a pattern of blood pressure in which than have old men, and this may explain the there is an apparent fall in both systolic and higher levels of pressure in the younger age diastolic pressures as age increases. This is not groups. We may expect that salt intake will explained by changes in body build with age. continue to rise in the Chimbu people, and it It may be influenced by a low dietary intake of will be important to observe whether their blood salt, and by psychological factors which are pressures will show a parallel increase. difficult to define. Changes may occur in the blood presure pattern of the Chimbu over the RELEVANCE OF THE CHIMBU next 10-20 years, and it is desirable that these FINDINGS TO THE PROBLEM OF should be accurately recorded and associated, if ESSENTIAL HYPERTENSION possible, with other changes in their way of life.
These findings add a further population to ACKNOWLEDGEMENTS those now reported among whom blood pressure We wish to acknowledge with thanks the valuable rises little if at all with age. Most of the assistance provided in the field by Dr. R. Barnes, populations with this pattern of blood pressure Mr. J. Ross, Mr. Noah Temgue and Mr. Toule Tarabi. are small and relatively isolated communities in Mr. MichaeI Pemberton did all the basic tedious which a subsistence economy is still the rule, statistical analysis. and traditional family authority and security is undisturbed. One is tempted to wonder whether REFERENCES the blood pressures of Stone-Age or Neolithic man were not similar to these, and that in BAILEY, K. V. (1963), “Blood Pressure in Under- these “primitive” communities we see the nourished Javanese”, Brit. Med. J., 2 : 775. survival of the “natural” or “original” pattern BARNES, R. (1964), Personal Communication. of human blood pressure. In organized, sophisti- BIBILE, S . W., CULLUMBINE, H., KURTSINGHE, C., cated, urbanized Western (and Eastern) com- WATSON, R. S., AND WIKAMANYAKE, T. (1949), munities this pattern has been modified, and at “The Variation with Age and Sex of Blood Pres- the same time a disease “essential hypertension” sure and Pulse Rate for Ceylonese Subjects” , has appeared. There is some evidence that as Ceylon J. Med. Sci., 6: 79.
20 PAPUA AND NEW GUINEA MEDICAL JOURNAL
125 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
BOE, J., HUMERFELT, S., AND WEDERVANG, F. (1957), MASTER, A. M., GARFIELD, C. I. AND WALTERS, M. “The Blood Pressure of a Population. Blood B. (1952), “Normal Blood Pressure and Hyper- Pressure Readings and Height and Weight Deter- tension” , Lea and Febiger, Philadelphia. minations in the Adult Population of the City of MENEELEY, G. R. AND DAHL, L. K. (1961), “Electro- Bergen”, Acta Med. Scand. Supp.: 321. lytes in Hypertension. The effects of Sodium BROWN, P. AND BROOKFIELD, H.C. (1959), “Chimbu Chloride”, Med. Clin. N. Amer., 45: 271. Land and Society”, Oceania, 30: 1. MIALL, W. E. AND OLDHAM, P. D. (1958), “Factors CRUZ-COKE, R. (1960), “Environmental Influences Influencing Arterial Blood Pressure in the General and Arterial Blood Pressure”, Lancet, 2: 885. Population”, Clin. Sci., 17: 409.
DAHL, L. K. (1961), “Possible Role of Chronic MURRILL, R. J. (1949), “A Blood Pressure Study of Excess Salt Consumption in the Pathogenesis of the Natives of Ponape Island, Eastern Carolines”, Essential Hypertension”, Amer. J. Cardiol., 8: 571. Hum. BioI., 21: 47.
HAMILTON, M., PICKERING, G. W., ROBERTS, J. A. F. OOMEN, H. A. P. C. AND MALCOLM, S. H. (1958), AND SOWRY, G. S. C. (1954), “The Aetiology of “Nutrition and the Papuan Child”, S. P. C. Tech- Essential Hypertension. 1. The Arterial Pressure in the nical Paper: 118; Noumea. General Population”, Clin. Sci., 13: 11. OOMEN, H. A. P. C. (1961), “The Nutrition Situation JELLIFFE, D. B. AND MADDOCKS, I. (1964), “Ecologic in Western New Guinea”, Trop. Geogr. Med., 13: Malnutrition in the New Guinea Highlands”, Clin. 321. Pediatr. (Phila.) 3: 432. PICKERING, G. (1961), “The Nature of Essential KAMINER, B. AND LUTZ, N. P. W. (1960), “Blood Hypertension”, J. & A. Churchill, London. Pressure in Bushmen of the Kalahari Desert”, Circulation, 22: 289. SCOTCH, N. A. (1960), “A Preliminary Report of the LEMLEY-STONE, J., DARBY, W. J. AND MENEELY, Relation of Sociocultural Factors to Hypertension
G. R. (1961), “Effect of Dietary Sodium: Potas- among the Zulu”, Ann. N. Y. Acad. Sci., 84; 1,000. sium Ratio on Body Content of Sodium and Potas- SHAPER, A. G., WILLIAMS, A. W. AND SPENCER, P. sium in Rats”, Amer. J. Cardiol., 8: 748. (1961), “Blood Pressure and Body Build in an LOVELL, R .R. H., MADDOCKS, I., AND ROGERSON, G. African Tribe living on a Diet of Milk and Blood”, W. (1960), “The Casual Arterial Pressure of E. Afr. Med. J., 38: 569. Fijians and Indians in Fiji”, Aust. Ann. Med., 9 :4. VENKATACHALEM, P. S. (1962), “A Study of the MADDOCKS, I. (1961a), “Possible Absence of Hyper- Diet, Nutrition and Health of the People of the tension in Two Complete Pacific Islands Popula- Chimbu Area”, Department of Public Health Mono- tions”, Lancet, 2: 396. graph, No. 4, Port Moresby. MADDOCKS, I, (1961b), “The Influence of Standard WHITEMAN, J. (1962), “Diet and Nutrition—Chimbu of Living on Blood Pressure in Fiji”, Circulation, 24: 1220. People”, File 54–14–6. Public Health Depart- ment, Port Moresby. MADDOCKS, I. (1964), “Dietary Factors in the Genesis of Hypertension” in “Nutrition”, Proceedings of WHYTE, H. M. (1958), “Body Fat and Blood Pres- the Sixth International Congress, Edinburgh 1963. sure in Natives of New Guinea: Reflections on Livingstone, Edinburgh. Essential Hypertension”, Aust. Ann. Med., 7: 36.
VOLUME 8, NO. 1, MARCH, 1965
126 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
MEDLARS BIBLIOGRAPHY
PUBLICATIONS OF RELEVANCE TO PAPUA NEW GUINEA AND MELANESIA
Bibliographic Citation List generated from MEDLARS
1 Aghanwa HS. significantly lower in the blanket group (P = 0.031). Attitude toward and knowledge about mental illness No difference was found between the contact and in Fiji islands. the control groups (P = 0.93). Whether this apparent Int J Soc Psychiatry 2004 Dec;50(4):361-375. reduced leprosy incidence in the first three years in BACKGROUNDS: There is a dearth of the blanket group is due to a delayed development information on the extent of knowledge about mental of leprosy or a complete clearance of infection needs illness and attitudes toward the mentally ill in Fiji. to be determined. AIMS: This study aimed to explore these aspects, and also to determine the factors influencing them. 5 Bakker MI, Hatta M, Kwenang A, Faber WR, van METHOD: Market vendors, peri-urban dwellers, Beers SM, Klatser PR, Oskam L. white-collar and health workers from Greater Suva Population survey to determine risk factors for were interviewed. The interview schedule used Mycobacterium leprae transmission and infection. elicited socio-demographic characteristics, Int J Epidemiol 2004 Dec;33(6):1329-1336. Epub knowledge of, and attitudes toward, mental illness. 2004 Jul 15. RESULTS: A majority of the subjects attributed the BACKGROUND: Not every leprosy patient is cause of mental illness to substance abuse, believed equally effective in transmitting Mycobacterium in the diversity of mental illness, considered hospital leprae. We studied the spatial distribution of infection as an important source of help and acknowledged (using seropositivity as a marker) in the population the effectiveness of medication. Less than one-fifth to identify which disease characteristics of leprosy of the subjects were willing to marry or employ patients are important in transmission. METHODS: mentally ill persons. About 42% of the sample would Clinical data and blood samples for anti-M. leprae be deterred by embarrassment from seeking help. ELISA were collected during a cross-sectional survey Educational attainment was correlated with on five Indonesian islands highly endemic for leprosy. knowledge about mental illness, except with A geographic information system (GIS) was used to knowledge about early mental symptoms (p < 0.01). define contacts of patients. We investigated spatial Prestigious occupation, single marital status, female clustering of patients and seropositive people and gender, younger age and urban dwelling were used logistic regression to determine risk factors for associated with positive disposition toward the seropositivity. RESULTS: Of the 3986 people mentally ill (p < 0.01). Race was not significantly examined for leprosy, 3271 gave blood. influential on almost all attitudinal variables. Seroprevalence varied between islands (1.7-8.7%) CONCLUSION: Health education is capable of and correlated significantly with leprosy prevalence. positively influencing knowledge about, and attitudes Five clusters of patients and two clusters of toward, mental illness in Fiji. seropositives were detected. In multivariate analysis, seropositivity significantly differed by 2 Allen L, Shrimpton R. leprosy status, age, sex, and island. Serological The International Research on Infant status of patients appeared to be the best Supplementation Study: implications for programs discriminator of contact groups with higher and further research. seroprevalence: contacts of seropositive patients had J Nutr 2005 Mar;135(3):666S-669S. an adjusted odds ratio (aOR) of 1.75 (95% CI 0.922- 3.31). This increased seroprevalence was strongest 3 Andrade AL, Martelli CM. for contact groups living < or =75 m of two Globalisation of Hib vaccination – how far are we? seropositive patients (aOR = 3.07; 95% CI 1.74- Lancet 2005 Jan 1-7;365(9453):5-7. 5.42). CONCLUSIONS: In this highly endemic area for leprosy, not only household contacts of 4 Bakker MI, Hatta M, Kwenang A, Van Benthem seropositive patients, but also people living in the BH, Van Beers SM, Klatser PR, Oskam L. vicinity of a seropositive patient were more likely to Prevention of leprosy using rifampicin as harbour antibodies against M. leprae. Through chemoprophylaxis measuring the serological status of patients and Am J Trop Med Hyg 2005 Apr;72(4):443-448. using a broader definition of contacts, higher risk An intervention study was implemented on five groups can be more specifically identified. Indonesian islands highly endemic for leprosy to determine whether rifampicin can be used as 6 Baravilala WR, Moulds RF. chemoprophylaxis to prevent leprosy. The A Fijian perspective on providing a medical population was actively screened before the workforce. intervention and subsequently once a year for three Med J Aust 2004 Dec 6-20;181(11-12):602. years. In the control group, no chemoprophylaxis was given. In the contact group, chemoprophylaxis 7 Beadell JS, Gering E, Austin J, Dumbacher JP, was only given to contacts of leprosy patients and in Peirce MA, Pratt TK, Atkinson CT, Fleischer RC. the blanket group to all eligible persons. The cohort Prevalence and differential host-specificity of two consisted of 3,965 persons. The yearly incidence avian blood parasite genera in the Australo-Papuan rate in the control group was 39/10,000; the region. cumulative incidence after three years was Mol Ecol 2004 Dec;13(12):3829-3844.
127 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
The degree to which widespread avian blood 9 Berlioz-Arthaud A, Barr IG. parasites in the genera Plasmodium and Laboratory-based influenza surveillance in New Haemoproteus pose a threat to novel hosts depends Caledonia, 1999-2003. in part on the degree to which they are constrained Trans R Soc Trop Med Hyg 2005 Apr;99(4):290-300. to a particular host or host family. We examined the We aimed to evaluate the annual incidence of host distribution and host-specificity of these influenza in New Caledonia and to identify the parasites in birds from two relatively understudied circulating viral types and subtypes in order to gather and isolated locations: Australia and Papua New information for the local vaccination programme and Guinea. Using polymerase chain reaction (PCR), regional influenza surveillance. A surveillance we detected infection in 69 of 105 species, network was set up in 1999; it included sentinel representing 44% of individuals surveyed (n = 428). practitioners in Noumea and the virology department Across host families, prevalence of Haemoproteus of the Pasteur Institute. Influenza circulated in New ranged from 13% (Acanthizidae) to 56% Caledonia every year, regularly during the southern (Petroicidae) while prevalence of Plasmodium hemisphere winter and occasionally during March- ranged from 3% (Petroicidae) to 47% May. Isolates were generally consistent with world (Ptilonorhynchidae). We recovered 78 unique surveillance, except in 1999, when a new A/H1N1 mitochondrial lineages from 155 sequences. variant was identified. This study emphasises the Related lineages of Haemoproteus were more likely need for regular influenza surveillance, even when to derive from the same host family than predicted performed on a limited scale. Importantly the optimal by chance at shallow (average LogDet genetic time for local vaccination was found to be in distance = 0, n = 12, P = 0.001) and greater depths December or January each year. (average distance = 0.014, n = 11, P < 0.001) within the parasite phylogeny. Within two major 10 Billo M, Fournet A, Cabalion P, Waikedre J, Bories Haemoproteus subclades identified in a maximum C, Loiseau P, Prina E, deArias AR, Yaluff G, likelihood phylogeny, host-specificity was evident up Fourneau C, Hocquemiller R. to parasite genetic distances of 0.029 and 0.007 Screening of New Caledonian and Vanuatu based on logistic regression. We found no significant medicinal plants for antiprotozoal activity. host relationship among lineages of Plasmodium by J Ethnopharmacol 2005 Jan 15;96(3):569-575. Epub any method of analysis. These results support 2004 Dec 1. previous evidence of strong host-family specificity Sixty-seven extracts of 30 medicinal plants in Haemoproteus and suggest that lineages of traditionally used in New Caledonia or Vanuatu by Plasmodium are more likely to form evolutionarily- healers to treat inflammation, fever and in cicatrizing stable associations with novel hosts. remedies were evaluated in vitro for their antiprotozoal activity against Leishmania donovani, 8 Becker AE, Gilman SE, Burwell RA. Leishmania amazonensis and Trypanosoma cruzi. Changes in prevalence of overweight and in body Among the selected plants, Pagiantha cerifera was image among Fijian women between 1989 and 1998. the most active against both Leishmania species; Obes Res 2005 Jan;13(1):110-117. four extracts were active against promastigotes of OBJECTIVE: To investigate changes in Leishmania donovani at EC(50) values inferior to 5 prevalence of overweight and obesity and in body microg/ml. Garcinia pedicillata extract had an image among ethnic Fijian women in Fiji during a EC(50) value of 12.5 microg/ml against intracellular period of rapid social change and the relationship amastigotes of Leishmania amazonensis. Alone between changes in body image and BMI. Amborella trichopoda reduced by more of 80% the RESEARCH METHODS AND PROCEDURES: The trypomastigotes of Trypanosoma cruzi in the blood. study design was a multiwave cohort study of BMI in a traditional Fijian village over a 9.5-year period 11 Billo M, Cabalion P, Waikedre J, Fourneau C, Bouttier S, Hocquemiller R, Fournet A. from 1989 to 1998. Cohorts were identified in 1989 Screening of some New Caledonian and Vanuatu (n=53) and in 1998 (n=50). Selection criteria medicinal plants for antimycobacterial activity. included Fijian ethnicity, female gender, age of at J Ethnopharmacol 2005 Jan 4;96(1-2):195-200. least 18 years, and residence in a specific coastal Twenty plants, belonging to sixteen families, Fijian village in 1989 and 1998, respectively. used in traditional New Caledonian and Vanuatu Assessments consisted of measurement of height medicine for treatment of symptoms potentially and weight, collection of demographic data by written related to tuberculosis (cough, fever or inflammation) survey, and administration of the Nadroga Language were screened for antimycobacterial activity. We Body Image Questionnaire. RESULTS: The also screened an original endemic plant, Amborella prevalence of overweight and obesity was trichopoda, only member of the monogeneric family significantly different between the cohorts, increasing Amborellaceae and considered the most primitive from 60% in 1989 to 84% in 1998 (p=0.014). In living angiosperm. In total, 55 extracts were addition, the age-adjusted mean BMI was evaluated for inhibitory activity against significantly higher in 1998 compared with 1989 Mycobacterium bovis BCG strain at a concentration (p=0.011). Finally, there were significant between- of 100 microg/ml. Methanolic and dichloromethane cohort differences in multiple measures of body extracts of Amborella trichopoda, Codiaeum image, which were mostly independent of BMI. peltatum, Myristica fatua, and essential oils Myoporum crassifolium showed an activity at this DISCUSSION: At 84%, the prevalence of overweight concentration. Methanolic extract of Amborella and obesity in this community sample of Fijian trichopoda fruits presented a significant activity with women is among the highest in the world. The a minimal inhibitory concentration included between dramatically increased prevalence over the 9.5-year 1 and 2.5 microg/ml. In the same conditions, this period studied corresponds with rapid social change activity was comparable with those of the reference in Fiji and significant shifts in prevailing traditional drugs pyrazynamide and ethambutol, at 20 and 2.5 attitudes toward body shape. microg/ml, respectively.
128 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
12 Boutlis CS, Weinberg JB, Baker J, Bockarie MJ, available in the literature were analyzed. Mgone CS, Cheng Q, Anstey NM. Phylogenetic studies using different algorithms Nitric oxide production and nitric oxide synthase (minimum evolution, neighbour joining, maximum activity in malaria-exposed Papua New Guinean parsimony, and maximum likelihood) gave the same children and adults show longitudinal stability and clear-cut results. Newly sequenced HTLV-I isolates no association with parasitemia described in this report allocated in three well-defined Infect Immun 2004 Dec;72(12):6932-6938. subtypes: Cosmopolitan, Central African, and a new Individuals in areas of intense malaria distinct one that we termed ‘Maroni’ subtype (present transmission exhibit resistance (or tolerance) to in the Maroni Basin, French Guiana, and West levels of parasitemia in their blood that would Indies). Clearly, the most divergent PTLV-I strains normally be associated with febrile illness in malaria- present in Asia- Australo-Melanesia as well as naive subjects. The resulting level of parasitemia African and Asian STLV-I derived from the same associated with illness (the pyrogenic threshold) is node in the phylogenetic tree as isolates of the highest in childhood and lowest in adulthood. Clinical Central African subtype. In addition, we showed that parallels between malarial and bacterial endotoxin within each PTLV-I subtype, groups of isolates may tolerance have led to the supposition that both share be characterized by nonrandom and systematically common physiological processes, with nitric oxide associated mutations. (NO) proposed as a candidate mediator. The hypotheses that NO mediates tolerance and blood 16 Cassar O, Capuano C, Meertens L, Chungue E, stage parasite killing in vivo were tested by Gessain A. determining its relationship to age and parasitemia Human T-cell leukemia virus type 1 molecular cross-sectionally and longitudinally in a population variants, Vanuatu, Melanesia. of 195 children and adults from Papua New Guinea Emerg Infect Dis 2005 May;11(5):706-710. encountering intense malaria exposure. Despite Four of 391 Ni-Vanuatu women were infected pharmacological clearance of asymptomatic with variants of human T-cell leukemia virus type 1 parasitemia, NO production and mononuclear cell (HTLV-1) Melanesian subtype C. These strains had NO synthase (NOS) activity were remarkably stable env nucleotide sequences approximately 99% similar within individuals over time, were not influenced by to each other and diverging from the main molecular parasitemia, and varied little with age. These results subtypes of HTLV-1 by 6% to 9%. These strains contrast with previous smaller cross-sectional were likely introduced during ancient human studies. Baseline NO production and NOS activity population movements in Melanesia. did not protect against recurrent parasitemia, consistent with previous data suggesting that NO 17 Chen N, Wilson DW, Pasay C, Bell D, Martin LB, does not have antiparasitic effects against blood Kyle D, Cheng Q. stage infection in vivo. The NO indices studied were Origin and dissemination of chloroquine-resistant markedly higher in specimens from study subjects Plasmodium falciparum with mutant pfcrt alleles in than in samples from Australian controls, and NOS the Philippines. activity was significantly associated with plasma Antimicrob Agents Chemother 2005 May;49(5):2102- immunoglobulin E levels, consistent with induction 2105. of NO by chronic exposure to other infections and/ The pfcrt allelic type and adjacent microsatellite or host genetic factors. These results suggest that marker type were determined for 82 Plasmodium NO is unlikely to mediate killing of blood stage falciparum isolates from the Philippines. Mutant pfcrt parasites in this setting and is unlikely to be the allelic types P1a and P2a/P2b were dominant in primary mediator in the acquisition or maintenance different locations. Microsatellite analysis revealed of malarial tolerance. that P2a/P2b evolved independently in the Philippines, while P1a shared common ancestry with 13 Brown H. Papua New Guinea chloroquine-resistant parasites. Treating the injured and burying the dead. Lancet 2005 Jan 15-21;365(9455):204-205. 18 Cortés A. A chimeric Plasmodium falciparum Pfnbp2b/Pfnbp2a 14 Cameron J. gene originated during asexual growth. Caring for mama and pikinini in Papua New Guinea. Int J Parasitol 2005 Feb;35(2):125-130. Epub 2004 Pract Midwife 2005 Feb;8(2):14-16. Dec 15. The Plasmodium falciparum line 3D7-A has an 15 Capdepont S, Londos-Gagliardi D, Joubert M, unusual invasion phenotype, such that it can invade Correze P, Lafon ME, Guillemain B, Fleury HJ. enzyme-treated and mutant red blood cells that are New insights in HTLV-I phylogeny by sequencing and resistant to invasion by other parasite lines. 3D7-A analyzing the entire envelope gene. has a chimeric Pfnbp2b gene that contains part of AIDS Res Hum Retroviruses 2005 Jan;21(1):28-42. the repeat region of the paralogous gene Pfnbp2a. The HTLV-I envelope plays a major role in the This chimeric gene originated by spontaneous gene process of target cell infection. It is implied in the conversion during normal maintenance in culture, recognition of the viral receptor(s), penetration of the indicating that ectopic recombination and gene viral genetic material, and induction of host immunity conversion during asexual growth are potentially to the virus. It is thus important to study the genetic important mechanisms participating in the evolution variability of the viral env gene as well as its variation of paralogous genes in Plasmodium. However, the in terms of evolution. In a new approach to these presence of the chimeric Pfnbp2b gene in 3D7-A features, we sequenced the entire env gene of 65 was not associated with its peculiar invasion HTLV-I isolates originating from Gabon, French phenotype. Guiana, West Indies, and Iran, such isolates representing all major HTLVI phylums but the 19 Cortés A, Mellombo M, Masciantonio R, Murphy Australo-Melanesian one. The sequences obtained VJ, Reeder JC, Anders RF. and all PTLV-I (HTLV-I and STLV-I) env sequences Allele specificity of naturally acquired antibody
129 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
responses against Plasmodium falciparum apical to be the causative etiology, with serotype 1 membrane antigen 1. predominating. Infect Immun 2005 Jan;73(1):422-430. Antibody responses against proteins located on 22 d’Arcangues C, Piaggio G, Brache V, Aissa RB, the surface or in the apical organelles of merozoites Hazelden C, Massai R, Pinol A, Subakir SB, Su- are presumed to be important components of juan G; Study Group on Progestogen-induced naturally acquired protective immune responses Vaginal Bleeding Disturbances. against the malaria parasite Plasmodium falciparum. Effectiveness and acceptability of vitamin E and low- However, many merozoite antigens are highly dose aspirin, alone or in combination, on Norplant- polymorphic, and antibodies induced against one induced prolonged bleeding. particular allelic form might not be effective in Contraception 2004 Dec;70(6):451-462. controlling growth of parasites expressing alternative A study (ISRCTN 77665712) was undertaken to forms. The apical membrane antigen 1 (AMA1) is a test the effectiveness and the acceptability of vitamin polymorphic merozoite protein that is a target of E and low-dose aspirin, alone or in combination, as naturally acquired invasion-inhibitory antibodies and treatment for prolonged vaginal bleeding induced by is a leading asexual-stage vaccine candidate. We Norplant. A total of 486 Norplant users who were characterized the antibody responses against AMA1 requesting treatment for bleeding lasting longer than in 262 individuals from Papua New Guinea exposed 7 days were enrolled in five centers: Beijing, China; to malaria by using different allelic forms of the full Jakarta, Indonesia; Santiago, Chile; Santo Domingo, AMA1 ectodomain and some individual subdomains. Dominican Republic; and Tunis, Tunisia. They were The majority of individuals had very high levels of randomized to one of four different 10-day oral antibodies against AMA1. The prevalence and titer treatments: 200 mg vitamin E daily, 80 mg aspirin of these antibodies increased with age. Although daily, both or a placebo. Treatment packs were antibodies against conserved regions of the molecule designed to ensure blinding of both the subjects and were predominant in the majority of individuals, most the clinical staff. Neither vitamin E nor low-dose plasma samples also contained antibodies directed aspirin nor their combination was found to have any against polymorphic regions of the antigen. In a few effect on reducing the length of the bleeding episode individuals, predominantly from younger age groups, for which treatment was taken or on the vaginal the majority of antibodies against AMA1 were bleeding patterns these women experienced during directed against polymorphic epitopes. The D10 the year of follow-up. allelic form of AMA1 apparently contains most if not all of the epitopes present in the other allelic forms tested, which might argue for its inclusion in future 23 Dale P, Sipe N, Anto S, Hutajulu B, Ndoen E, AMA1-based vaccines to be tested. Some important Papayungan M, Saikhu A, Prabowa YT. epitopes in AMA1 involved residues located in Malaria in Indonesia: a summary of recent research domain II or III but depended on more than one into its environmental relationships. domain. Southeast Asian J Trop Med Public Health 2005 Jan;36(1):1-13. A review of the literature was carried out to 20 Cortés A, Mellombo M, Mgone CS, Beck HP, evaluate malaria and its environmental relationships. Reeder JC, Cooke BM. Research, in 6 parts of Indonesia, addressed the Adhesion of Plasmodium falciparum-infected red relationship between malaria incidence and physical blood cells to CD36 under flow is enhanced by the and socioeconomic environmental factors, using cerebral malaria-protective trait South-East Asian longitudinal and cross-sectional approaches. ovalocytosis. Physical factors, which are generally important for Mol Biochem Parasitol 2005 Aug;142(2):252- malaria, included rainfall, mosquito breeding and 257. Epub 2005 Apr 15. resting sites, their distance from human habitation, and elevation, though the latter was not statistically 21 Corwin AL, Subekti D, Sukri NC, Willy RJ, Master significant. Housing conditions were occasionally J, Priyanto E, Laras K. important. Social and economic factors of A large outbreak of probable rotavirus in Nusa importance were income, education, use of bednets Tenggara Timur, Indonesia. and pattern of outdoor activities, especially at night. Am J Trop Med Hyg 2005 Apr;72(4):488-494. Use of repellents, mosquito coils and sleeping An outbreak of acute diarrheal disease was arrangements were significant in some of the studies. reported in Kupang, Nusa Tenggara, Indonesia, in August 2002. An investigative team carried out a 24 De Boer MA, Peters LA, Aziz MF, Siregar B, retrospective historical review of records, and a case- Cornain S, Vrede MA, Jordanova ES, Fleuren GJ. control study involving data and specimen Human papillomavirus type 18 variants: collections. Etiologic determination involving stool histopathology and E6/E7 polymorphisms in three specimens was based on an enzyme-linked countries. immunosorbent assay, with a reverse transcriptase- Int J Cancer 2005 Apr 10;114(3):422-425. polymerase chain reaction performed for serotyping In cervical cancer, human papillomavirus type purposes. Two thousand six hundred probable cases 18 (HPV 18) and HPV 16 are predominantly related were identified from hospital records during the to adenocarcinomas (ADCs) and squamous cell outbreak months of June, July, August, and carcinomas (SCCs), respectively. Here, we studied September 2002. Previous enteric outbreaks were whether the geographically distributed HPV intratypic recognized from the same months in the preceding variants are also associated with histologically years and all annual outbreak episodes following a different tumors. A total of 44 HPV 18-positive and period of prolonged, low rainfall. In contrast to 91 HPV 16-positive cervical carcinomas from previous outbreaks discerned from trend analysis, Indonesian, Surinamese and Dutch patients were the overwhelming burden of disease fell upon the histologically classified using hematoxilin and eosin, pediatric population versus the young and old in periodic acid Schiff plus and Alcian Blue staining. previous outbreak instances. Rotavirus was found Samples were sequenced and intratypic variants
130 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
were classified into the known phylogenetic haplogroups P and Q. branches. The Asian Amerindian HPV 18 variant Mol Biol Evol 2005 Jun;22(6):1506-1517. Epub 2005 was observed in 56% of ADCs compared to 15% of Apr 6. SCCs (p < 0.006). The African HPV 18 variant was Modern humans have occupied New Guinea and exclusively found in SCCs. By sequencing the HPV the nearby Bismarck and Solomon archipelagos of 18 E6 and E7 open reading frames, we found Island Melanesia for at least 40,000 years. Previous predicted amino acid changes only in 8 samples. mitochondrial DNA (mtDNA) studies indicated that Two amino acid changes were consistent throughout two common lineages in this region, haplogroups P the African branch. In HPV 16-positive tumors, we and Q, were particularly diverse, with the did not find a specific linkage between intratypic coalescence for P considered significantly older than variants and histopathology. We conclude that HPV that for Q. In this study, we expand the definition of 18 intratypic variants are differentially associated with haplogroup Q so that it includes three major adenocarcinoma and squamous cell carcinoma of branches, each separated by multiple mutational the cervix. The findings described here stress the distinctions (Q1, equivalent to the earlier definition biologic significance of intratypic HPV variants and of Q, plus Q2 and Q3). We report three whole- might help explaining differences in the pathogenesis mtDNA genomes that establish Q2 as a major Q of cervical ADCs and SCCs. branch. In addition, we describe 314 control region sequences that belong to the expanded haplogroups 25 Duke T. P and Q from our Southwest Pacific collection. The Slow but steady progress in child health in Papua coalescence dates for the largest P and Q branches New Guinea. (P1 and Q1) are similar to each other (approximately J Paediatr Child Health 2004 Dec;40(12):659-663. 50,000 years old) and considerably older than prior estimates. Newly identified Q2, which was found in 26 Duke T, Oa O, Mokela D, Oswyn G, Hwaihwanje I, Island Melanesian samples just to the east, is Hawap J. somewhat younger by more than 10,000 years. Our The management of sick young infants at primary coalescence estimates should be more reliable than health centres in a rural developing country. prior ones because they were based on significantly Arch Dis Child 2005 Feb;90(2):200-205. larger samples as well as complete mtDNA-coding AIMS: To investigate the epidemiology of illness region sequencing. Our estimates are roughly in among young infants at remote health clinics in a accord with the current suggested dates for the first rural developing country and to determine risk factors settlement of New Guinea-Sahul. The for mortality that might be used as triggers for phylogeography of P and Q indicates almost total emergency treatment or referral. METHODS: Multi- (female) isolation of ancient New Guinea-Island site 12 month observational study of consecutive Melanesia from Australia that may have existed from presentations of infants less than 2 months, and an the time of the first settlement. While Q subsequently investigation of neonates who died in one district diversified extensively in New Guinea-Island without accessing health care. RESULTS: Forty per Melanesia, it has not been found in Australia. The cent of 511 young infant presentations occurred in only shared mtDNA haplogroup between Australia the first week of life and most of these in the first 24 and New Guinea identified to date remains one minor hours. Twenty five deaths were recorded: 18 in the branch of P. health facilities and seven in villages. In addition there were eight stillbirths. Clinical signs predicting 29 Gessner BD, Sutanto A, Linehan M, Djelantik IG, death were: not able to feed, fast respiratory rate, Fletcher T, Gerudug IK, Ingerani, Mercer D, apnoea, cyanosis, ‘too small’, ‘skin-cold’, and severe Moniaga V, Moulton LH, Moulton LH, Mulholland abdominal distension. Signs indicating severe K, Nelson C, Soemohardjo S, Steinhoff M, respiratory compromise were present in 25% of Widjaya A, Stoeckel P, Maynard J, Arjoso S. young infants; failure to give oxygen therapy was a Incidences of vaccine-preventable Haemophilus modifiable factor in 27% of deaths within health influenzae type b pneumonia and meningitis in facilities. A high proportion of seriously ill young Indonesian children: hamlet-randomised vaccine- infants were discharged from health facilities early probe trial. without adequate follow up. A common reason for not seeking care for fatally ill neonates was the Lancet 2005 Jan 1-7;365(9453):43-52. perception by parents that health staff would respond BACKGROUND: Most studies of Haemophilus negatively to their social circumstances. influenzae type b (Hib) disease in Asia have found CONCLUSIONS: Clinical signs with moderate low rates, and few Asian countries use Hib vaccine positive predictive value for death may be useful in routine immunisation programmes. Whether Hib triggers for emergency treatment and longer disease truly is rare or whether many cases remain observation or urgent referral. The results of this undetected is unclear. METHODS: To estimate study may be useful in planning strategies to address incidences of vaccine-preventable Hib pneumonia high neonatal mortality rates in developing countries. and meningitis among children younger than 2 years in Lombok, Indonesia, during 1998-2002, we 27 Duke T, Tefuarani N, Baravilala W. undertook a hamlet-randomised, controlled, double- Getting the most out of health education in Papua blind vaccine-probe study (818 hamlets). Children New Guinea. Report from the 40th Annual Papua were immunised (WHO schedule) with diphtheria, New Guinea Medical Symposium. tetanus, pertussis (DTP) or DTP-PRP-T (Hib conjugate) vaccine. Vaccine-preventable disease Med J Aust 2004 Dec 6-20;181(11-12):606-607. incidences were calculated as the difference in rates of clinical outcomes between DTP and DTP-PRP-T 28 Friedlaender J, Schurr T, Gentz F, Koki G, groups. Analyses included all children who received Friedlaender F, Horvat G, Babb P, Cerchio S, at least one vaccine dose. FINDINGS: We enrolled Kaestle F, Schanfield M, Deka R, Yanagihara R, 55073 children: 28147 were assigned DTP-PRP-T Merriwether DA. and 26926 DTP. The proportion of pneumonia Expanding southwest pacific mitochondrial outcomes prevented by vaccine ranged from less
131 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
than 0 to 4.8%. Calculated incidences of vaccine- 94.25%. The results of the statistical analysis of the preventable Hib disease (per 10(5) child-years of data obtained in this study revealed that the observation) for outcome categories were: diagnostic specificity, the diagnostic positive substantial alveolar consolidation or effusion, less predictive value and the diagnostic efficiency of the than zero (-43 [95% CI -185 to 98]); all severe TB-Dot test were significantly higher (p < 0.05) than pneumonia, 264 (95% CI less than zero to 629); all those of the ICT-TB test. However, the diagnostic clinical pneumonia, 1561 (270 to 2853); confirmed sensitivity and the negative predictive value of both Hib meningitis, 16 (1.4 to 31); meningitis with tests did not differ significantly (p > 0.05). Viewed cerebrospinal-fluid findings consistent with a from the point of their practicability, it can be justified bacterial aetiology, 67 (22 to 112); and admission that the ICT-TB test is a very practicable test, which for suspected meningitis or presenting to a clinic with needs only 15 minutes and does not require special convulsions, 158 (42 to 273). INTERPRETATION: instruments to perform the test, but is more Hib vaccine did not prevent the great majority of expensive than the TB-Dot test. On the other hand, pneumonia cases, including those with alveolar though the TB-Dot test is not very practicable and consolidation. These results do not support a major relatively time consuming, it has a significantly higher role for Hib vaccine in overall pneumonia-prevention degree of diagnostic value and is much cheaper programmes. Nevertheless, the study identified high when compared to the ICT-TB test. incidences of Hib meningitis and pneumonia; inclusion of Hib vaccine in routine infant 32 Hastings MD, Maguire JD, Bangs MJ, immunisation programmes in Asia deserves Zimmerman PA, Reeder JC, Baird JK, Sibley CH. consideration. Novel Plasmodium vivax dhfr alleles from the Indonesian Archipelago and Papua New Guinea: 30 Han ET, Song TE, Park JH, Shin EH, Guk SM, association with pyrimethamine resistance Kim TY, Chai JY. determined by a Saccharomyces cerevisiae Allelic dimorphism in the merozoite surface protein- expression system. 3alpha in Korean isolates of Plasmodium vivax. Antimicrob Agents Chemother 2005 Feb;49(2):733- Am J Trop Med Hyg 2004 Dec;71(6):745-749. 740. To study the genetic diversity of re-emerging In plasmodia, the dihydrofolate reductase Plasmodium vivax in the Republic of Korea, (DHFR) enzyme is the target of the pyrimethamine nucleotide sequence variations at the merozoite component of sulfadoxine-pyrimethamine (S/P). surface protein-3alpha (PvMSP-3alpha) locus were Plasmodium vivax infections are not treated analyzed using 24 re-emerging isolates and 4 intentionally with antifolates. However, outside isolates from imported cases. Compared with the Africa, coinfections with Plasmodium falciparum and well known Belem strain (Brazil), a large number of P. vivax are common, and P. vivax infections are amino acid substitutions, deletions, and insertions often exposed to S/P. Cloning of the P. vivax dhfr were found at the locus of the isolates examined. gene has allowed molecular comparisons of dhfr The Korean isolates were divided into two allelic alleles from different regions. Examination of the types; type I (15 isolates), similar to the Belem strain, dhfr locus from a few locations has identified a very and type II (9), similar to the Chesson strain (New diverse set of alleles and showed that mutant alleles Guinea). Isolates from imported cases were of the vivax dhfr gene are prevalent in Southeast classified into three types; type III (1 from Malaysia), Asia where S/P has been used extensively. We have similar to type B from western Thailand, type IV (1 surveyed patient isolates from six locations in each from Indonesia and India), and type V (1 from Indonesia and two locations in Papua New Guinea. Pakistan), both being new types. Our results have We sequenced P. vivax dhfr alleles from 114 patient shown that the MSP-3alpha locus of re-emerging samples and identified 24 different alleles that Korean P. vivax is dimorphic with two allelic types differed from the wild type by synonymous and coexisting in the endemic area. nonsynonymous point mutations, insertions, or deletions. Most importantly, five alleles that carried 31 Handojo I, Arifin MZ. four or more nonsynonymous mutations were The immunoserological diagnosis of tuberculosis: a identified. Only one of these highly mutant alleles comparison of two tests. had been previously observed, and all carried the Southeast Asian J Trop Med Public Health 2005 57L and 117T mutations. P. vivax cannot be cultured Jan;36(1):141-144. continuously, so we used a yeast assay system to A comparative study of the diagnostic value of determine in vitro sensitivity to pyrimethamine for a the ICT-TB test and the TB-Dot test, based on subset of the alleles. Alleles with four laboratory examination, was carried out in 39 patients nonsynonymous mutations conferred very high suffering from sputum positive pulmonary levels of resistance to pyrimethamine. This study tuberculosis (25 males and 14 females, aged 16-50 expands significantly the total number of novel dhfr years) and in 48 patients (27 males and 21 females, alleles now identified from P. vivax and provides a aged 17-55 years) suffering from non-tuberculosis foundation for understanding how antifolate pulmonary diseases, that had attended the resistance arises and spreads in natural P. vivax Tembagapura Hospital and the TB Control Health populations. Center Timika-Mimika, Papua. The diagnostic sensitivity of the ICT-TB test was 87.18%, the 33 Hiawalyer G. diagnostic specificity was 81.25%, the diagnostic A surveillance information system as a management positive predictive value was 79.07%, the negative tool: a report from Papua New Guinea. predictive value was 88.64%, and the diagnostic Soz Praventivmed 2005;50 Suppl 1:S31-S32. efficiency was 83.91%. The diagnostic sensitivity To have a very good surveillance system, it is of the TB-Dot test was 93.31%, the diagnostic paramount important to have a functional health specificity was 95.83%, the diagnostic positive information system that could be easily used for predictive value was 94.74%, the negative predictive monitoring and investigation of disease outbreaks. value was 93.85%, and the diagnostic efficiency was In Papua New Guinea (PNG) a national health
132 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
information system was developed, trialed and Jan 24. implemented nationwide. Furthermore to have the A field-applicable assay for testing anthelmintic system working linked to it must be the local health sensitivity is required to monitor for anthelmintic system for sustainability and control. A public health resistance. We undertook a study to evaluate the manual for disease surveillance in PNG was ability of three in vitro assay systems to define drug developed and is now being used for surveillance. sensitivity of clinical isolates of the human hookworm This paper describes how the health information parasite Necator americanus recovered from system, particularly surveillance system was children resident in a village in Madang Province, developed and implemented on the national scale, Papua New Guinea. The assays entailed how it was integrated with other management observation of drug effects on egg hatch (EHA), larval information systems and how information has been development (LDA), and motility of infective stage used to support management decision-making and larvae (LMA). The egg hatch assay proved the best informed policy decision. It will highlight some of method for assessing the response to benzimidazole the hurdles that it has encountered while trying to anthelmintics, while the larval motility assay was implement the system. PNG has one of the best suitable for assessing the response to ivermectin. national health information systems as compared The performance of the larval development assay to many developing countries but limited information was unsatisfactory on account of interference caused generated from the system. There was also less by contaminating bacteria. A simple protocol was feedback from all levels of the health sector. We developed whereby stool samples were subdivided need to improve surveillance on the basic principles and used for immediate egg recovery, as well as for of integration, focus, and sharing of work. There faecal culture, in order to provide eggs and infective must be an appropriate and timely response and larvae, respectively, for use in the egg hatch assay feedback. We need to improve on the current system and larval motility assay systems. While the assays rather than building a new one. proved effective in quantifying drug sensitivity in larvae of the drug-susceptible hookworms examined 34 Hombhanje FW, Hwaihwanje I, Tsukahara T, in this study, their ability to indicate drug resistance Saruwatari J, Nakagawa M, Osawa H, Paniu MM, in larval or adult hookworms remains to be Takahashi N, Lum JK, Aumora B, Masta A, Sapuri determined. M, Kobayakawa T, Kaneko A, Ishizaki T. The disposition of oral amodiaquine in Papua New 38 Laman M, Ripa P, Vince J, Tefuarani N. Guinean children with falciparum malaria. Can clinical signs predict hypoxaemia in Papua New Br J Clin Pharmacol 2005 Mar;59(3):298-301. Guinean children with moderate and severe AIMS: We assessed the disposition of oral pneumonia? amodiaquine (AQ) and CYP2C8 polymorphism in Ann Trop Paediatr 2005 Mar;25(1):23-27. 20 children with falciparum malaria. METHODS: AQ Pulse oximetry was performed on 77 children and DEAQ concentrations were determined with admitted with acute lower respiratory tract infections SPE-HPLC method. CYP2C8 genotypes were (ALRI) to the children’s ward in Port Moresby General assessed by PCR-RFLP method. RESULTS: AQ Hospital, Papua New Guinea over a 4-month period was not detectable beyond day 3 postdose. Cmax in 2002. Clinical findings were correlated with for DEAQ was reached in 3.0 days. The mean different levels of hypoxaemia, <93%, <90% and values for t1/2, MRT, and AUCtotal were 10.1 days, <85%. Cyanosis, head nodding and drowsiness 15.5 days and 4512.6 microg l(-1) day, respectively. were good predictors of hypoxia but lacked All the children were CYP2C8* homozygous. sensitivity. Decisions to use oxygen based on these CONCLUSION: Our data are consistent with those signs would therefore result in a significant number previously reported, and the AQ regimen seems of children with hypoxia not receiving oxygen. Pulse pharmacokinetically adequate in the absence of oximetry is the best indicator of hypoxaemia in CYP2C8 polymorphism. children with ALRI and, although relatively expensive, its use might be cost-effective in controlling oxygen 35 Kevau IH, Vince JD, McPherson JV. requirements. Tailoring medical education in Papua New Guinea to the needs of the country. 39 Lawrence G, Leafasia J, Sheridan J, Hills S, Wate Med J Aust 2004 Dec 6-20;181(11-12):608-610. J, Wate C, Montgomery J, Pandeya N, Purdie D. Control of scabies, skin sores and haematuria in 36 Klapsing P, MacLean JD, Glaze S, McClean KL, children in the Solomon Islands: another role for Drebot MA, Lanciotti RS, Campbell GL. ivermectin. Ross River virus disease reemergence, Fiji, 2003- Bull World Health Organ 2005 Jan;83(1):34-42. Epub 2004. 2005 Jan 21. Emerg Infect Dis 2005 Apr;11(4):613-615. OBJECTIVE: To assess the effects of a 3-year We report 2 clinically characteristic and programme aimed at controlling scabies on five small serologically positive cases of Ross River virus lagoon islands in the Solomon Islands by monitoring infection in Canadian tourists who visited Fiji in late scabies, skin sores, streptococcal skin 2003 and early 2004. This report suggests that Ross contamination, serology and haematuria in the island River virus is once again circulating in Fiji, where it children. METHODS: Control was achieved by apparently disappeared after causing an epidemic treating almost all residents of each island once or in 1979 to 1980. twice within 2 weeks with ivermectin (160-250 microg/kg), except for children who weighed less 37 Kotze AC, Coleman GT, Mai A, McCarthy JS. than 15 kg and pregnant women, for whom 5% Field evaluation of anthelmintic drug sensitivity using permethrin cream was used. Reintroduction of in vitro egg hatch and larval motility assays with scabies was controlled by treating returning residents Necator americanus recovered from human clinical and visitors, whether or not they had evident scabies. isolates. FINDINGS: Prevalence of scabies dropped from Int J Parasitol 2005 Apr 1;35(4):445-453. Epub 2005 25% to less than 1% (P < 0.001); prevalence of sores
133 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
from 40% to 21% (P < 0.001); streptococcal two-locus haplotypes were examined in 367 contamination of the fingers in those with and without unrelated Melanesians living on the islands of sores decreased significantly (P = 0.02 and 0.047, Vanuatu and New Caledonia. Diversity at all HLA respectively) and anti-DNase B levels decreased (P class-I and class-II loci was relatively limited. In = 0.002). Both the proportion of children with class-I loci, three HLA-A allelic groups (HLA-A*24, haematuria and its mean level fell (P = 0.002 and P HLA-A*34 and HLA-A*11), seven HLA-B alleles or < 0.001, respectively). No adverse effects of the allelic groups (HLA-B*1506, HLA-B*5602, HLA-B*13, treatments were seen. CONCLUSION: The results HLA-B*5601, HLA-B*4001, HLA-B*4002 and HLA- show that ivermectin is an effective and practical B*2704) and four HLA-C alleles or allelic groups agent in the control of scabies and that control (HLA-Cw*04, HLA-Cw*01, HLA-Cw*0702 and HLA- reduces the occurrence of streptococcal skin disease Cw*15) constituted more than 90% of the alleles and possible signs of renal damage in children. observed. In the class-II loci, four HLA-DRB1 alleles Integrating community-based control of scabies and (HLA-DRB1*15, HLA-DRB1*11, HLA-DRB1*04 and streptococcal skin disease with planned HLA-DRB1*16), three HLA-DRB3-5 alleles (HLA- programmes for controlling filariasis and intestinal DRB3*02, HLA-DRB4*01 and HLA-DRB5*01/02) nematodes could be both practical and produce great and five HLA-DQB1 alleles (HLA-DQB1*0301, HLA- health benefits. DQB1*04, HLA-DQB1*05, HLA-DQB1*0601 and HLA-DQB1*0602) constituted over 93, 97 and 98% 40 Lehmann D, Vail J, Firth MJ, de Klerk NH, Alpers of the alleles observed, respectively. Homozygosity MP. showed significant departures from expected levels Benefits of routine immunizations on childhood for neutrality based on allele frequency (i.e. excess survival in Tari, Southern Highlands Province, Papua diversity) at the HLA-B, HLA-Cw, HLA-DQB1 and New Guinea. HLA-DRB3/5 loci on some islands. The locus with Int J Epidemiol 2005 Feb;34(1):138-148. Epub 2004 the strongest departure from neutrality was HLA- Nov 23. DQB1, homozygosity being significantly lower than BACKGROUND: Non-specific beneficial as well expected on all islands except New Caledonia. No as deleterious effects of childhood immunizations consistent pattern was demonstrated for any HLA have been reported in areas of high mortality. This locus in relation to malaria endemicity. study aimed to determine the effects of diphtheria- tetanus-whole-cell-pertussis (DTP), BCG, hepatitis 42 McGain F, Limbo A, Williams DJ, Didei G, Winkel B, and measles vaccines on mortality in the KD. highlands of Papua New Guinea (PNG). METHODS: Snakebite mortality at Port Moresby General Demographic events for children born in 1989-1994 Hospital, Papua New Guinea, 1992-2001. who were under monthly demographic surveillance Med J Aust 2004 Dec 6-20;181(11-12):687-691. in Tari were recorded from birth until age 2 years, OBJECTIVE: Fatal snakebites at Port Moresby out-migration, death, or the end of the study period. General Hospital (PMGH), Papua New Guinea Data on BCG, hepatitis B, DTP, measles and (PNG), were examined to identify interventions that pneumococcal polysaccharide vaccination were may improve patient survival. DESIGN: collected monthly from clinic records. To allow for Retrospective case series. SUBJECTS AND different characteristics of immunized and non- SETTING: Inpatients at PMGH who presented with immunized children, analysis included conditioning snakebite, had evidence of envenomation, and died on a propensity score for vaccination, adjusting for as inpatients between 1 January 1992 and 31 differences in children’s background characteristics. December 2001. OUTCOME MEASURES: Number RESULTS: In all, 101/3502 children (3%) who had and cause of fatalities; ventilation bed-days; at least one vaccine died between ages 29 days and antivenom timing, dose and price. RESULTS: 87 24 months were compared to 112/546 (21%) who deaths occurred among 722 snakebite admissions had none. BCG was associated with lower mortality to the intensive care unit (ICU). Of these 722 in the 1-5 month age group (hazard ratio [HR] = 0.17, patients, 82.5% were ventilated, representing 45% 95% CI: 0.09, 0.34), measles vaccine with lower of all ventilated ICU patients and 60% (3430/5717) mortality at age 6-11 months (HR = 0.42, 95% CI: of all ICU ventilator bed-days. The median duration 0.17, 1.01), and pneumococcal polysaccharide of ventilation in fatal snakebite cases was vaccine with lower mortality at age 12-23 months significantly less than in non-fatal cases for children (HR = 0.42, 95% CI: 0.19, 0.93). One or more doses (3.0 v. 4.5 days) and adults (3.0 v. 5.0 days). The of DTP was associated with lower overall mortality case-fatality rate for children (14.6%) was (HR = 0.27, 95% CI: 0.16, 0.44), particularly in the significantly greater than that for adults (8.2%). Sixty 1-5 month age group (HR = 0.19, 95% CI: 0.10, fatalities were examined in detail: 75% received 0.34), and also in those who had had prior BCG (HR blood products; 53% received antivenom (mostly a = 0.45, 95% CI: 0.22, 0.91). CONCLUSION: Routine single ampoule of polyvalent), but only 5% received immunizations are effective in reducing overall antivenom < or = 4 hours post-bite. Major causes mortality in young children in an area of high of death included respiratory complications (50%), mortality. In particular, DTP, whether considered probable intracerebral haemorrhage (17%), and separately or in addition to BCG, was associated renal failure (10%). Antivenom unit costs increased with a lowering of overall mortality, in contrast to significantly over the decade; in 2000 an ampoule findings reported from Guinea-Bissau. of polyvalent antivenom was 40-fold more expensive in PNG than in Australia on a gross domestic product 41 Maitland K, Bunce M, Harding RM, Barnardo MC, (A dollars) per capita basis. CONCLUSIONS: Clegg JB, Welsh K, Bowden DK, Williams TN. Management of severe snakebite is a major HLA class-I and class-II allele frequencies and two- challenge for PMGH. Improved antivenom locus haplotypes in Melanesians of Vanuatu and New procurement and use policies (including increased Caledonia. use of appropriate monovalent antivenoms), Tissue Antigens 2004 Dec;64(6):678-686. combined with targeted snakebite education HLA class-I and class-II allele frequencies and interventions (community- and hospital-based), are
134 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
key interventions to reduce the ongoing toll from (hemoglobin level < 7.5 g/dL) from 0.0-3.3% to 3.8- snakebite. 18.4%. These massive increases in morbidity have devastating impact on the affected communities and 43 Mosley LM, Sharp DS, Singh S. highlight that malaria epidemics are a serious and Effects of a tropical cyclone on the drinking-water increasing public health problem in the highlands of quality of a remote Pacific island. Papua New Guinea. Disasters 2004 Dec;28(4):405-417. The effect of a cyclone (Ami, January 2003) on 45 Owen IL. drinking-water quality on the island of Vanua Levu, Parasitic zoonoses in Papua New Guinea. Fiji was investigated. Following the cyclone nearly J Helminthol 2005 Mar;79(1):1-14. three-quarters of the samples analysed did not Relatively few species of zoonotic parasites have conform to World Health Organisation (WHO) been recorded in humans in Papua New Guinea. A guideline values for safe drinking-water in terms of greater number of potentially zoonotic species, chlorine residual, total and faecal coliforms, and mostly nematodes, occur in animals but are yet to turbidity. Turbidity and total coliform levels be reported from humans. Protozoa is the best significantly increased (up 56 and 62 per cent, represented group of those infecting man, with respectively) from pre-cyclone levels, which was Giardia duodenalis, Cryptosporidium parvum, likely due to the large amounts of silt and debris Cyclospora cayetanesis, Toxoplasma gondii, entering water-supply sources during the cyclone. Sarcocystis spp., Entamoeba polecki, Balantidium The utility found it difficult to maintain a reliable coli and, possibly, Blastocystis hominis. The only supply of treated water in the aftermath of the zoonotic helminths infecting humans include the disaster. Communities were unaware they were trematode Paragonimus westermani, the cestodes drinking water that had not been adequately treated. Hymenolepis nana, H. diminuta and the sparganum Circumstances permitted this cyclone to be used as larva of Spirometra erinacea, and the nematodes a case study to assess whether a simple paper-strip Trichinella papuae and Angiostrongylus cantonensis water-quality test (the hydrogen sulphide, H(2)S) kit and, possibly, Ascaris suum. Other groups could be distributed and used for community-based represented are Acanthocephala monitoring following such a disaster event to better (Macracanthorhynchus hirudinaceus)), insects protect public health. The H(2)S test results (Chrysomya bezziana, Cimex sp., Ctenocephalides correlated well with faecal and total coliform results spp.), and mites (Leptotrombidium spp. and, possibly as found in previous studies. A small percentage of Sarcoptes scabiei, and Demodex sp.). One leech samples (about 10 per cent) tested positive for faecal (Phytobdella lineata) may also be considered as and total coliforms but did not test positive in the being zoonotic. The paucity of zoonotic parasite H(2)S test. It was concluded that the H(2)S test species can be attributed to long historical isolation would be well suited to wider use, especially in the of the island of New Guinea and its people, and the absence of water-quality monitoring capabilities for absence until recent times of large placental outer island groups as it is inexpensive and easy to mammals other than pig and dog. Some zoonotic use, thus enabling communities and community helminths have entered the country with recent health workers with minimal training to test their own importation of domestic animals, in spite of water supplies without outside assistance. The quarantine regulations, and a few more (two importance of public education before and after cestodes, one nematode and one tick) are poised natural disasters is also discussed. to enter from neighbouring countries, given the opportunity. Improvement in water supplies, human 44 Mueller I, Namuigi P, Kundi J, Ivivi R, Tandrapah hygiene and sanitation would reduce the prevalence T, Bjorge S, Reeder JC. of many of these parasites, and thorough cooking Epidemic malaria in the highlands of Papua New of meat would lessen the risk of infection by some Guinea. others. Am J Trop Med Hyg 2005 May;72(5):554-560. As part of a larger study into the epidemiology 46 Poka H. of malaria in the highlands of Papua New Guinea, Practising in rural Papua New Guinea. outbreak investigations were carried out at the end Med J Aust 2004 Dec 6-20;181(11-12):609. of the 2002 rainy season in 11 villages situated between 1,400 and 1,700 meters above sea level 47 Pozio E, Owen IL, Marucci G, La Rosa G. that had reported epidemics. Locations and timing Inappropriate feeding practice favors the of these epidemics corresponded largely to those transmission of Trichinella papuae from wild pigs to reported in the pre-control era of the 1960s and saltwater crocodiles in Papua New Guinea. 1970s. On average, 28.8% (range = 10.3-63.2%) Vet Parasitol 2005 Feb 28;127(3-4):245-251. Epub of people in each of the 11 villages were found to be 2004 Dec 16. infected with malaria. Plasmodium falciparum The recent discovery of Trichinella zimbabwensis accounted for 59% of all identified infections and P. in farmed crocodiles (Crocodilus niloticus) of vivax for 34%. The majority (53%) of infections were Zimbabwe and its ability to infect mammals, and the symptomatic. Although symptomatic infections were development of both T. zimbabwensis and Trichinella most common in children 2-9 years of age (36%), papuae in experimentally infected reptiles led to an even in adults a prevalence of 20% was observed. investigation of Trichinella infection in saltwater A comparison with earlier non-epidemic data in three crocodiles (Crocodylus porosus) and in wild pigs of the villages without easy access to health care (Sus scrofa) of Papua New Guinea, to see if T. showed markedly increased levels of morbidity, with papuae also, is present in both cold- and warm- 6-10-fold increases in parasite prevalence, a 3-fold blooded animals. Of 222 crocodiles examined, 47 increase in both measured and reported fevers, and animals (21.2%), all from Kikori, Gulf Province, were a 12-fold increase in enlarged spleens. The average positive for non-encapsulated larvae in the muscles. hemoglobin levels were reduced by 2.3-3.5 g/dL, with The greatest number of larvae was found usually in a concurrent increase in moderate to severe anemia the biceps, with an average of 7 larvae/g. One
135 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
isolate from a crocodile infected successfully both in Mozambique, 10 in Indonesia, and 9 in South laboratory rats and mice. Of 81 wild pigs examined, Africa. At all study sites, the majority of RSV cases 9 from Bensbach river area (Western Province) and occurred in infants. CONCLUSION: These studies 1 from Kikori area (Gulf Province) were positive for demonstrate that RSV contributes to a substantial non-encapsulated larvae in the muscles. Trichinella but quite variable burden of LRI in children aged < 5 larvae from both saltwater crocodiles and wild pigs years in four developing countries. The possible have been identified by multiplex-PCR analysis as explanations for this variation include social factors, T. papuae. The sequence analysis of the region such as family size and patterns of seeking health within the large subunit ribosomal DNA, known as care; the proportion of children infected by human the expansion segment V, has shown the presence immunodeficiency syndrome (HIV); and differences of a molecular marker distinguishing T. papuae in clinical definitions used for obtaining samples. The isolates of Bensbach river area from those of Kikori age distribution of cases indicates the need for an area. This marker could be useful to trace back the RSV vaccine that can protect children early in life. geographical origin of the infected animal. The epidemiological investigation carried out in the Kikori 51 Rodgers-Bird QM. area has shown that local people catch young Challenges of Solomon Island medical students: a crocodiles in the wild and keep them in holding pens graduation vote of thanks. Letter for several months, before sending them to the Aust J Rural Health 2005 Apr;13(2):128. crocodile farm in Lae (Morobe Province). They feed the crocodiles primarily with wild pig meat bought at the local market and also with fish. These results 52 Sa JM, Nomura T, Neves J, Baird JK, Wellems stress the importance of using artificial digestion for TE, del Portillo HA. routinely screening of swine and crocodiles, and of Plasmodium vivax: allele variants of the mdr1 gene adopting measures for preventing the spread of do not associate with chloroquine resistance among infection, such as the proper disposal of carcasses isolates from Brazil, Papua, and monkey-adapted and the adequate freezing of meat. strains. Exp Parasitol 2005 Apr;109(4):256-259. 48 Ratu Sade K. We describe here the sequence of the The status of emergency medicine in Makira Ulawa Plasmodium vivax mdr1 gene from 10 different Province, Solomon Islands. isolates differing in chloroquine sensitivity. The Emerg Med Australas 2005 Apr;17(2):167-169. deduced amino acid sequence of PvMDR1 shares This Perspective reports on the challenges that more than 70% similarity with other malarial MDR face a solo doctor in a provincial hospital in the proteins and it displays consensus motifs of an ABC Solomon Islands following the civil disturbances of family transporter including two transmembrane 1998-2003. The Health Service is seriously domains and two ATP binding cassettes. Similarity constrained by a paucity of funding, supplies and and dendrogram analyses revealed that sequences personnel. In spite of that, a rudimentary service could be grouped according to their geographical can be provided and lives can be saved using simple origin. Within each geographical group however, techniques and basic resources. Further training of no correlation was found between chloroquine nurses, midwives and doctors is required. resistance and specific mutations. Emergency medicine, as a generalist discipline, provides a foundation to improve the delivery of care 53 Schultz R. to the acutely ill and injured in these circumstances. A two-year placement in the Solomon Islands. Med J Aust 2004 Dec 6-20;181(11-12):604-605. 49 Reeder JC. Papua New Guinea: targeting research to things that 54 Suligoi B, Danaya RT, Sarmati L, Owen IL, Boros matter. S, Pozio E, Andreoni M, Rezza G. Med J Aust 2004 Dec 6-20;181(11-12):610-611. Infection with human immunodeficiency virus, herpes simplex virus type 2, and human herpes virus 8 in 50 Robertson SE, Roca A, Alonso P, Simoes EA, remote villages of southwestern Papua New Guinea. Kartasasmita CB, Olaleye DO, Odaibo GN, Am J Trop Med Hyg 2005 Jan;72(1):33-36. Collinson M, Venter M, Zhu Y, Wright PF. To investigate the spread of human Respiratory syncytial virus infection: denominator- immunodeficiency virus (HIV) and other sexually based studies in Indonesia, Mozambique, Nigeria transmitted viruses, two serosurveys (the first in 1999 and South Africa. among 56 adults and the second in 2001 among Bull World Health Organ 2004 Dec;82(12):914-922. 351 adults) were conducted in remote villages of Epub 2005 Jan 5. the southwestern part of Papua New Guinea. Only OBJECTIVE: To assess the burden of respiratory one individual was positive for antibodies to HIV. In syncytial virus (RSV)-associated lower respiratory 2001, the seroprevalence of human herpes virus 8 infections (LRI) in children in four developing (HHV-8) was 32.2%, and the seroprevalence of countries. METHODS: A WHO protocol for herpes simplex virus type 2 (HSV-2) was 27.4%. prospective population-based surveillance of acute Both prevalence rates increased with age, and were respiratory infections in children aged less than 5 lower in the villages near the Bensbach River. The years was used at sites in Indonesia, Mozambique, seropositivity of HSV-2 was independently correlated Nigeria and South Africa. RSV antigen was identified with HHV-8 infection. Our data show that the by enzyme-linked immunosorbent assay performed inhabitants of the southwestern region of Papua New on nasopharyngeal specimens from children Guinea currently experience an extremely low meeting clinical case definitions. FINDINGS: Among circulation of HIV. However, the high prevalence of children aged < 5 years, the incidence of RSV- infectious agents that can be sexually transmitted, associated LRI per 1000 child-years was 34 in such as HSV-2 and to a lesser extent HHV-8, Indonesia and 94 in Nigeria. The incidence of RSV- indicates the presence of behavioral patterns that associated severe LRI per 1000 child-years was 5 may facilitate the spread of HIV in this area of
136 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
currently low endemicity. discharge and 41 were lost to follow-up. Major neurological sequelae were found in 50 (63 per cent) 55 Temple V, Mapira P, Adeniyi K, Sims P. of surviving children, and 27 (34 percent) had multiple Iodine deficiency in Papua New Guinea (sub-clinical severe complications. In rural Papua New Guinea iodine deficiency and salt iodization in the highlands meningitis causes high rates of mortality and severe of Papua New Guinea). long-term disability in a high proportion of survivors. J Public Health (Oxf) 2005 Mar;27(1):45-48. Epub High-level resistance to chloramphenicol is likely to 2004 Nov 25. be part of the reason for this, but widespread Data on the status about iodine nutrition in availability of third-generation cephalosporins for the children in Papua New Guinea (PNG) are scarce. treatment of meningitis, although urgently required, This study attempts to determine the mean daily per will not overcome the other problems of delayed capita consumption of salt, the iodine content of salt presentation with established complications. There in the households and retail shops and the urinary is a need for the introduction of conjugate iodine concentration in children (6-12 years) in Hella Haemophilus influenzae vaccine, and affordable Region, Southern Highland Province (SHP), PNG. vaccination strategies against Streptococcus The mean daily consumption of salt was 2.62 +/- pneumoniae. Richer countries could sponsor these 1.29 g. The iodine content of salt was >30 p.p.m. in vaccines in developing countries, and apply pressure 95 per cent of households and 100 per cent of retail on vaccine producers to lower the costs. shops. The median urinary iodine concentration of 48.0 mg/l for all the children indicates moderate 58 Wood DM, Alsahaf H, Streete P, Dargan PI, Jones iodine deficiency. The median urinary iodine AL. concentrations for the male (67.0 mg/l) and female Fatality after deliberate ingestion of the pesticide (44.0 mg/l) children indicate mild and moderate rotenone: a case report. iodine deficiency, respectively. 68.42 per cent of the Crit Care 2005 Jun;9(3):R280-284. Epub 2005 Apr male and 81.82 per cent of the female children have 29. urinary iodine concentration <100 mg/l, indicating Rotenone is a pesticide derived from the roots that iodine deficiency is a potential public health of plants from the Leguminosae family. Poisoning problem in the Hella region. These results indicate following deliberate ingestion of these plant roots a need for further assessment of the implementation has commonly been reported in Papua New Guinea. of the universal salt iodization strategy for the However, poisoning with commercially available elimination of iodine deficiency in the SHP, PNG. rotenone in humans has been reported only once previously following accidental ingestion in a 3.5- 56 Tokudome S, Soeripto, Triningsih FX, Ananta I, year-old child. Therefore, the optimal management Suzuki S, Kuriki K, Akasaka S, Kosaka H, of rotenone poisoning is not known. After deliberate Ishikawa H, Azuma T, Moore MA. ingestion of up to 200 ml of a commercially available Rare Helicobacter pylori infection as a factor for the 0.8% rotenone solution, a 47-year-old female on very low stomach cancer incidence in Yogyakarta, regular metformin presented with a reduced level of Indonesia. consciousness, metabolic acidosis and respiratory Cancer Lett 2005 Feb 28;219(1):57-61. compromise. Metformin was not detected in To elucidate factors associated with the very low premortem blood samples obtained. Despite risk of gastric neoplasia in Yogyakarta, Indonesia, intensive supportive management, admission to an approximately 1/50 of the level in Japan, we recruited intensive care unit, and empirical use of N- 52 male and 39 female participants from the general acetylcysteine and antioxidant therapy, she did not populace in the city of Yogyakarta in October 2003. survive. Poisoning with rotenone is uncommon but Helicobacter pylori IgG antibodies were found in only is potentially fatal because this agent inhibits the 5% (0-13) (95% confidence interval) and 4% (0-9) mitochondrial respiratory chain. In vitro cell studies for Javanese males and females, respectively, and have shown that rotenone-induced toxicity is reduced were statistically lower than the 62% (58-65) and by the use of N-acetylcysteine, antioxidants and 57% (53-60), respectively, in Japanese. potassium channel openers. However, no animal Furthermore, positive findings of pepsinogen test studies have been reported that confirm these were only 0 and 2% (0-6) for males and females, in findings, and there are no previous reports of Yogyakarta, and were again significantly lower than attempted use of these agents in patients with acute the 23% (22-25) and 22% (20-23), in Japan. The rotenone-induced toxicity. very low incidence of stomach cancer in Yogyakarta may be due to a low prevalence of H. pylori infection 59 Wuster W, Dumbrell AJ, Hay C, Pook CE, Williams and chronic atrophic gastritis. DJ, Fry BG. Snakes across the Strait: trans-Torresian 57 Wandi F, Kiagi G, Duke T. phylogeographic relationships in three genera of Long-term outcome for children with bacterial Australasian snakes (Serpentes: Elapidae: meningitis in rural Papua New Guinea. Acanthophis, Oxyuranus, and Pseudechis). J Trop Pediatr 2005 Feb;51(1):51-53. Epub 2004 Dec Mol Phylogenet Evol 2005 Jan;34(1):1-14. 15. We analyze the phylogeny of three genera of This study was undertaken to evaluate the long- Australasian elapid snakes (Acanthophis-death term neurological outcome for survivors of bacterial adders; Oxyuranus-taipans; Pseudechis- meningitis in rural Papua New Guinea. Children who blacksnakes), using parsimony, maximum likelihood, were discharged from Nonga Base Hospital in and Bayesian analysis of sequences of the Rabaul with a diagnosis of bacterial meningitis mitochondrial cytochrome b and ND4 genes. In between 1992 and 2000 were evaluated in their Acanthophis and Pseudechis, we find evidence of home villages or on review at hospital. Neurological multiple trans-Torresian sister-group relationships. and developmental complications were documented. Analyses of the timing of cladogenic events suggest The outcomes for 80 of 121 eligible children were crossings of the Torres Strait on several occasions determined; eight had died following hospital between the late Miocene and the Pleistocene.
137 Papua New Guinea Medical Journal Volume 48, No 1-2, Mar-Jun 2005
These results support a hypothesis of repeated land kavalactone, kawain, in kava extracts as causing connections between Australia and New Guinea in strong antiproliferative and apoptotic effect in human the late Cenozoic. Additionally, our results reveal bladder cancer cells. Flavokawain A results in a undocumented genetic diversity in Acanthophis and significant loss of mitochondrial membrane potential Pseudechis, supporting the existence of more and release of cytochrome c into the cytosol in an species than previously believed, and provide a invasive bladder cancer cell line T24. These effects phylogenetic framework for a reinterpretation of the of flavokawain A are accompanied by a time- systematics of these genera. In contrast, our dependent decrease in Bcl-x(L), a decrease in the Oxyuranus scutellatus samples from Queensland association of Bcl-x(L) to Bax, and an increase in and two localities in New Guinea share a single the active form of Bax protein. Using the primary haplotype, suggesting very recent (late Pleistocene) mouse embryo fibroblasts Bax knockout and wild- genetic exchange between New Guinean and type cells as well as a Bax inhibitor peptide derived Australian populations. from the Bax-binding domain of Ku70, we showed that Bax protein was, at least in part, required for 60 Zi X, Simoneau AR. the apoptotic effect of flavokawain A. In addition, Flavokawain A, a novel chalcone from kava extract, flavokawain A down-regulates the expression of X- induces apoptosis in bladder cancer cells by linked inhibitor of apoptosis and survivin. Because involvement of Bax protein-dependent and both X-linked inhibitor of apoptosis and survivin are mitochondria-dependent apoptotic pathway and main factors for apoptosis resistance and are suppresses tumor growth in mice. overexpressed in bladder tumors, our data suggest Cancer Res 2005 Apr 15;65(8):3479-3486. that flavokawain A may have a dual efficacy in Consumption of the traditional kava preparation induction of apoptosis preferentially in bladder was reported to correlate with low and uncustomary tumors. Finally, the anticarcinogenic effect of gender ratios (more cancer in women than men) of flavokawain A was evident in its inhibitory growth of cancer incidences in three kava-drinking countries: bladder tumor cells in a nude mice model (57% of Fiji, Vanuatu, and Western Samoa. We have inhibition) and in soft agar. identified flavokawain A, B, and C but not the major
138 Papua New Guinea Institute of Medical Research Monograph Series
ISSN 0256 2901
1. Growth and Development in New Guinea. 9. The Health of Women in Papua New A Study of the Bundi People of the Guinea. Madang District. Joy E. Gillett. ISBN 9980 71 008 X, L.A. Malcolm. ISBN 9980 71 000 4, 1970, 1990, 180p. 105p. 10. National Study of Sexual and 2. Endemic Cretinism. Reproductive Knowledge and Behaviour B.S. Hetzel and P.O.D. Pharoah, Editors. in Papua New Guinea. ISBN 9980 71 001 2, 1971, 133p. The National Sex and Reproduction 3. Essays on Kuru. Research Team and Carol Jenkins. R.W. Hornabrook, Editor. ISBN 9980 71 ISBN 9980 71 009 8, 1994, 147p. 002 0 (also 0 900848 95 2), 1976, 150p. 4. The People of Murapin. Monographs 1-5 are case-bound, 6-10 are P.F. Sinnett. ISBN 9980 71 003 9 (also paperbacks. 0 900848 87 1), 1977, 208p. Monographs may be obtained from 5. A Bibliography of Medicine and Human The Librarian, Biology of Papua New Guinea. Papua New Guinea Institute of R.W. Hornabrook and G.H.F. Skeldon, Medical Research Editors. ISBN 9980 71 004 7, 1977, PO Box 60, Goroka, EHP 441, 335p. (with 1976 Supplement, 36p.) Papua New Guinea 6. Pigbel. Necrotising Enteritis in Papua New Guinea. Cost of each Monograph (see below for M.W. Davis, Editor. ISBN 9980 71 005 Postage and Handling): 5, 1984, 118p. 7. Cigarette Smoking in Papua New 1,2……………………………………K 5.00 Guinea. 3,4……………………………………K 8.00 D.E. Smith and M.P. Alpers, Editors. 5……………………………………...K 12.00 ISBN 9980 71 006 3, 1984, 83p. 6,7,8,9……………………………….K 6.00 8. Village Water Supplies in Papua New 10…………………………………….K 12.00 Guinea. D.E. Smith and M.P. Alpers, Editors, Applications for free copies of any ISBN 9980 71 007 1, 1985, 94p. monograph should be sent to the Director at the above address.
)aniKGNP(gnildnaHdnaegatsoP
SLLIAMECAFRU IAMRIA
nihtiW PGGN W1NPnihti Z4eno Z6/3eno enoZ
1001,2, 700. 100.0 200.0 600.0 0.57
305,4, 100.4 200.0 400.0 900.0 0.501
609,8,7, 305. 500. 105.0 105.7 5.71
tneiciffusddaesaelp,ycnerrucrehtoynaniedamsitnemyapfI.aniKnitnemyapekamesaelP.aniK=KGP=K .segrahcknabllarevocotsdnuf THE MEDICAL SOCIETY OF PAPUA NEW GUINEA
Society Membership and Journal Subscription
Membership of the Medical Society of Papua New Guinea is open to all health workers whether resident in Papua New Guinea or overseas. Members of the Society receive four issues of the Papua New Guinea Medical Journal each year. The Society organizes an annual symposium and other activities.
Membership dues are:-
Papua New Guinea residents: Members – K150 Associate (Student) Members – K20 Overseas residents: K200; AU$120; US$80
I wish to join the Medical Society of Papua I enclose my membership fee of New Guinea as a Full Member K…………..for the year(s)……………
Please indicate your category Name: …………………………………
Medical Officer [ ] Title: ………………………………….. Scientific Officer [ ] Pharmacist [ ] Address: ……………………………... Health Extension Officer [ ] Nursing Officer [ ] …………………………………………. Laboratory Technologist [ ] Radiographer [ ] …………………………………………. Social Health Worker [ ] Other (Please specify) [ ] Telephone: ………………………….. Fax: …………………………………..
OR a Student Member Email: ………………………………… (for full-time students) (Forward to the Membership Secretary, Medical Student [ ] Medical Society of Papua New Guinea, PO Other Student (Please specify) [ ] Box 60, Goroka, EHP 441, Papua New Guinea) INFORMATION FOR AUTHORS 3 Garner PA, Hill G. Brainwashing in tuberculosis management. PNG Med J The Papua New Guinea Medical Journal invites 1985;28:291-293. submission of original papers and reviews on all 4 Cochrane RG. A critical appraisal of the aspects of medicine. Priority will be given to present position of leprosy. In: Lincicome articles and subjects relevant to the practice of DP, ed. International Review of Tropical medicine in Papua New Guinea and other Medicine. New York: Academic Press, countries in the South Pacific. 1961:1-42.
Manuscripts are accepted for publication only ILLUSTRATIONS with the understanding that they have not been published nor submitted for publication Tables and figures should be prepared on elsewhere. All manuscripts will be sent out for separate pages. Photographs should be glossy referees’ comments as part of the peer review prints, either 7 cm or 14.5 cm in width. Graphs process. and charts should preferably be in Microsoft Word or Excel. Photomicrographs should have internal Original Articles: Reports of original and new scale markers. Each table should have a heading investigations or contributions. and footnotes which make it understandable without reference to the text. Each figure should Brief Communications and Case Reports: have a legend; figure legends should be typed Contents similar to that of original articles but together on a separate sheet. Indicate the top of text should be no more than a total of 4 Journal figures lightly in pencil on the back. pages including all figures and tables. Abbreviations: Standard abbreviations and Reviews: Critical analysis of previously units should be used. collected and published information. Drug Names: Generic names of drugs should Letters: Short reports of clinical experience or be used. topics of interest. Text should not exceed 2 pages of the Journal. Orthography: The Shorter Oxford English Dictionary is followed. Other types of manuscript may also be accepted for publication at the Editor’s discretion. EDITORIAL MAIL
Submitted manuscripts should conform to the Manuscripts and other editorial instructions set out below. Manuscripts not communications should be forwarded to: conforming to these instructions will be returned. The Editor, MANUSCRIPTS Papua New Guinea Medical Journal, PO Box 60, Goroka, EHP 441, Submit the original with a virus-free electronic Papua New Guinea copy on disk, and specify the software used. All Email: [email protected] sections of the manuscript, including text, references, tables and legends, should be in SUBSCRIPTIONS AND ADVERTISEMENTS double spacing. Manuscripts should not be right justified. Each paper should include an Communications relating to advertisements or informative Summary, Introduction, Patients/ subscriptions should be addressed to the Journal Materials and Methods, Results, Discussion and as above. Matters related to the Society should References. The title page should include the be addressed to the Medical Society of Papua title, full names of all authors, names and New Guinea, PO Box 6665, Boroko, NCD 111, addresses of institutions where the work has Papua New Guinea. been done and full present address of the first or corresponding author. Subscriptions: Members of the Medical Society of Papua New Guinea receive the Journal as part References should be kept to a minimum, and of their annual subscription. Others may must be in the Vancouver style. Authors should subscribe and should contact the subscription check all references against the original source. secretary for a price. Sample references are shown below. Papua New Guinea Medical Journal Volume 48, Number 1-2, March-June 2005
CONTENTS 50TH YEAR CELEBRATION
FOREWORD 50th year celebration 1
ORIGINAL ARTICLES The first 50 years of the Papua New Guinea Medical Journal: a celebration of its Golden Jubilee M.P. Alpers 2
CLASSIC PAPERS Some epidemiological features of pig-bel T.G.C. Murrell 27
The pathogenesis of pig-bel in Papua New Guinea G. Lawrence 39
Stronglyloides species infestation in young infants of Papua New Guinea: association with generalized oedema J.D. Vince, R.W. Ashford, M.J. Gratten and J. Bana-Koiri 50
Strongyloides infection in a mid-mountain Papua New Guinea community: results of an epidemiological survey R.W. Ashford, J.D. Vince, M.J. Gratten and J. Bana-Koiri 58
Weanling diarrhoea J. Biddulph and P. Pangkatana 66
The bacteriology of acute pneumonia and meningitis in children in Papua New Guinea: assumptions, facts and technical strategies M. Gratten and J. Montgomery 73
Nutrition and morbidity: acute lower respiratory tract infections, diarrhoea and malaria D. Lehmann, P. Howard and P. Heywood 87
Small-area variations in the epidemiology of malaria in Madang Province J.A. Cattani, J.S. Moir, F.D. Gibson, M. Ginny, J. Paino, W. Davidson and M.P. Alpers 95
Hereditary ovalocytosis in Melanesians D. Amato and P.B. Booth 102
Trends in sexually transmitted disease incidence in Papua New Guinea C.K. Lombange 109
A New Guinea population in which blood pressure appears to fall as age advances I. Maddocks and L. Rovin 122
MEDLARS BIBLIOGRAPHY 127