ISSN 0031-1480
VOL. 51, NO 1-2, MARCH-JUNE 2008
Medical Society of Papua New Guinea
Executive 2008
President: Mathias Sapuri Vice-President: Nicholas Mann Secretary: Sylvester Lahe Treasurer: Harry Aigeeleng Executive Member: Osborne Liko
ACKNOWLEDGEMENT
We are grateful to the Government of Australia through AusAID for providing funding for the publication of this issue of the Journal.
The Editors Published quarterly by the Medical Society of Papua New Guinea
Papua New Guinea Medical Journal
ISSN 0031-1480
March-June 2008, Volume 51, Number 1-2
EDITORS: PETER M. SIBA, NAKAPI TEFUARANI
Editorial Committee
F. Hombhanje G. Mola A. Saweri J. Vince
Assistant Editor: 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 Moore 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 51, Number 1-2, March-June 2008
CONTENTS
EDITORIAL Treatment of uncomplicated malaria in Papua New Guinea: what should be done? T.M.E. Davis 1
ORIGINAL ARTICLES Knowledge and attitudes about infant feeding among nulliparous and parous women in Port Moresby: a comparative study D. Frank, P. Ripa, J.D. Vince and N. Tefuarani 5
A review of the current state of malaria among pregnant women in Papua New Guinea I. Mueller, S. Rogerson, G.D.L. Mola and J.C. Reeder 12
A case-control study of VDRL-positive antenatal clinic attenders at the Port Moresby General Hospital Antenatal Clinic and Labour Ward to determine outcomes, sociodemographic features and associated risk factors G.D.L. Mola, A. Golpak and A.B. Amoa 17
Variations of the middle thyroid vein in Papua New Guinean Melanesians D.J. Hasola, G. Gende and O. Liko 27
A focused ethnography about treatment-seeking behaviour and traditional medicine in the Nasioi area of Bougainville J.E. Macfarlane 29
Primary repair of colonic injuries at the Kundiawa and Madang General Hospitals, Papua New Guinea J. Kuzma and J. Jaworski 43
Barriers to the delivery of the hepatitis B birth dose: a study of five Papua New Guinean hospitals in 2007 S.G Downing, W. Lagani, R. Guy and M. Hellard 47
The prevalence of HIV infection in women attending antenatal clinics in Fiji C.H. Washington, L.M. Singer, T. McCaig, L. Tikoduadua, S.T. Ali, J. Fong, J. Luveni, T.O. Kyaw-Myint, S. Watson and F. Russell 56
MEDLARS BIBLIOGRAPHY 60
PNG Med J 2008 Mar-Jun;51(1-2): 1-4
EDITORIAL
Treatment of uncomplicated malaria in Papua New Guinea: what should be done?
In a recently published document, to determine the value of alternative ‘Guidelines for the Treatment of Malaria’, the regimens in advance of need. This assumes World Health Organization (WHO) (1) that funding, facilities and staff will be in place recommends that assessment of antimalarial to carry out what are usually quite technically treatment policy should be based on: i) and logistically demanding studies. Aspects efficacy assessments in which the of study design including selection of study parasitological cure rate is defined over ≥28 sites, sample sizes, and appropriate inclusion days of follow-up with polymerase chain and exclusion criteria need to be considered reaction (PCR) genotyping to distinguish carefully. For example, a 28-day treatment recrudescent parasites from new infections, failure rate of 86% for existing treatment ii) initiation of change of the antimalarial might be associated with an upper 95% treatment when the cure rate with current confidence limit that is >90% (and thus not therapy falls to <90%, and iii) introduction of significantly different from a failure rate the new antimalarial therapy with an average WHO considers acceptable) if the sample cure rate of ≥95% in clinical trials. In the size is <225 patients. In remote areas, the same document, the WHO recommends attrition rate is usually relatively high, which artemisinin-based combination therapy can increase the sample size required and (ACT) for uncomplicated falciparum malaria, prolong the duration of the study. In areas in specifically artemether-lumefantrine, which pharmaceuticals including antimalarial artesunate-amodiaquine, artesunate- drugs are relatively freely available in the mefloquine or artesunate-sulfadoxine- community, prior treatment can itself bias the pyrimethamine. The choice between these results or, if pre-treated patients are ACTs should be based on the local level of rigorously excluded, impair efficient parasite resistance to the partner drug. If recruitment. ACTs are unavailable, non-artemisinin- based combinations such as amodiaquine- The recently published results of a study sulfadoxine-pyrimethamine may be used as from Karimui (Simbu Province), South an interim measure provided that the Wosera (East Sepik Province) and the North component therapies retain efficacy. Coast (Madang Province) are illustrative (3). Despite variation between the three study These recommendations were published areas, the overall 28-day PCR-confirmed in the aftermath of strong criticism of the cure rate for chloroquine plus sulfadoxine- WHO that appeared in the Lancet, including pyrimethamine in 128 children with a mean the statement that “WHO should publish age close to 6 years was 83% with a 95% malaria treatment guidelines that countries confidence interval of 75-89%. For a larger can depend on as authoritative norms” (2). group of 521 younger children (mean age They seem simple and evidence-based, and approximately 4 years) treated with capable of being put into place in a country amodiaquine-sulfadoxine-pyrimethamine, such as Papua New Guinea (PNG). But is it there was an 81% (78-84%) cure rate. In all that easy? While the WHO should be both cases, the upper limit of the confidence applauded for getting to grips with this interval was below 90%, consistent with important subject, there are issues with the WHO recommendations that alternative recommendations that could have treatment should be introduced for both age consequences for their implementation. In groups. this article, each of the key WHO recommendations will be assessed critically. Prospective testing of candidate replacement regimens should follow the The three WHO recommendations relating same principles. A cure rate of 95% needs to antimalarial treatment policy guidelines at least 160 subjects so that the lower 95% assume that valid data will be collected at confidence limit is ≥90% and thus regular intervals for existing therapies and significantly better than the failure rate the that efficacy assessments will also be done WHO considers unacceptable. In addition,
1 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 the use of the word ‘average’ in respect of adherence and pharmacokinetic variability be the ≥95% recommended cure rate in the implicated rather than parasite resistance? WHO guidelines (1) implies that there should Efficacy studies should use drug formulations be more than one study performed (or manufactured under good manufacturing perhaps one study with multiple sites or arms) practice (GMP) specifications and stored before a decision on replacement therapy is under optimal conditions to ensure stability made. This may mean ‘borrowing’ data from during the trial, but this does not always another country or region in which the happen. Even when treatment is directly parasite sensitivity and patient immunity may observed and vomiting has not occurred, be quite different. In the case of PNG, the there are occasions, especially in children, results of a large-scale comparison between when unmeasurable drug concentrations in chloroquine-sulfadoxine-pyrimethamine and subsequent blood samples suggest that the three ACTs (artesunate-sulfadoxine- patient has managed to expel the dose pyrimethamine, artemether-lumfantrine and without this being detected – this may be dihydroartemisinin-piperaquine) will soon be particularly so if a drug has an unpleasant available. This study, conducted by the PNG taste or is administered rectally to a young Institute of Medical Research and the child unable to tolerate oral therapy (5). University of Western Australia School of Medicine and Pharmacology, was co- Between-subject variability in drug sponsored by the WHO and the National disposition is also a factor. In the case of Health and Medical Research Council of artemether-lumefantrine, this is marked in Australia. The data should be useful in part due to the need for the lipid-soluble informing treatment policy for uncomplicated lumefantrine component to be taken with fat malaria in PNG, especially if they are (usually a biscuit or milk) so that consistent with other studies in the region or bioavailability is enhanced. In one recent in other geographical areas with a similar study (6), day 7 plasma lumefantrine epidemiology and history of antimalarial drug concentrations were close to undetectable in deployment. some patients despite directly supervised treatment. Unfortunately, measurement of Expert microscopy and laboratory facilities plasma drug concentrations is not part of for PCR differentiation of recrudescence from WHO efficacy assessment despite being a re-infection in efficacy studies are essential useful way of identifying important non- for successful efficacy studies, and are well parasite causes of treatment failure. established in PNG. ‘Routine’ PCR cannot determine whether parasite DNA in a sample While there are a number of issues with is from sexual or asexual forms, and so efficacy studies of both contemporary and microscopy for gametocytes becomes candidate replacement regimens that are an important. Despite this, there are still patients important part of the WHO’s antimalarial who become PCR positive during follow-up treatment policy, usual care throws up its own without symptoms or asexual/sexual parasite problems. The effectiveness of unsupervised forms on blood smear, even after ACT (4). treatment is what determines treatment Are these true treatment failures or the failure in the community outside the confines harbinger of clinically significant resistance? of a clinical trial. Complex regimens with Unfortunately, the presence of parasite frequent side-effects, such as the 7 days of mutations that are known to be associated quinine and tetracycline that was used in with resistance such as those involving pfcrt, countries such as Thailand as recently as the dhfr and dhps may not accurately predict 1990s, are associated with poor adherence. clinical response (4) and they require the Even 3 days of artemether-lumefantrine, availability of relatively sophisticated which is relatively well tolerated, gives laboratory techniques. In addition, in vitro significantly lower plasma lumefantrine parasite culture of local strains to generate concentrations when given in an concentrations that inhibit parasite growth by unsupervised than in a supervised treatment pre-specified amounts (typically 50% or 90%) setting, with the risk of increased do not correlate well with the in vivo response recrudescence or reinfection (6). Apparent either. treatment failures may even reflect the local appearance of counterfeit or poor quality In these cases, and even when the antimalarial drugs (7). recrudescence is relatively clear-cut, could factors related to drug quality, treatment Ideally, efficacy studies in the WHO mould
2 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 can be done reliably and efficiently, and on a may still have a role where they are shown regular basis, so that trends in treatment to be at least as effective as ACT in well- failure rate can be detected early, and designed studies (12). replacement regimens assessed and implemented in a timely fashion. However, What do the WHO guidelines mean for there may also be the opportunity to use PNG? There should soon be evidence from routine surveillance mechanisms to identify properly designed and conducted efficacy evidence of developing parasite resistance. studies that ACT should replace chloroquine- In countries with even simple malaria sulfadoxine-pyrimethamine and, perhaps, an notification schemes, epidemiological indication of which of the three candidate markers such as the percentage of patients ACTs should be used. The epidemiological presenting again with malaria requiring context of these studies (3,13) differs greatly treatment within 28 days could provide a from those of previous studies in South-East signal that parasite resistance is on the Asia and even Indonesian Papua (9), and increase. resembles the hyperendemic transmission seen in Africa, where relatively few Given that the WHO recommendation is adequately powered trials of ACT have been for ACT to be first-line therapy for conducted to date. Ideally the choice uncomplicated malaria, which one should be between ACTs will be made on the basis of used? The WHO considers artemether- cost-effectiveness and considerations such lumefantrine to be the only ACT that can be as adherence in a real-world vs clinical trial used both in areas of multidrug resistance setting. For oral therapy, non-GMP-standard (South-East Asia) and in Africa (1). It is also treatment might be considered (as it is in a an ACT that is co-formulated (thus simplifying number of other tropical countries) provided administration compared to separate drug that tablet content is within acceptable dosing) and satisfies the WHO requirement bounds, especially as the supply of GMP- that the artemisinin derivative component manufactured artemether-lumefantrine has must be given for at least three days for been problematic in the recent past (14). optimum effect (1). However, it is relatively However, when treatment is supplied by the expensive unless heavily subsidized or Global Fund or other sponsor, GMP- donated, has to be given in 6 doses, manufactured product is usually mandatory. preferably with fat-containing food, and the lumefantrine component has a shorter half- Above all, there needs to be a plan for life than other partner drugs such as regular efficacy assessments so that, as mefloquine and piperaquine (8). This latter implied by the WHO recommendations (1), characteristic means that there is a greater failing conventional therapy can be identified risk of re-infection in the aftermath of early and an assessment of candidate treatment (9). replacement regimen(s) can be implemented. This will need government Other co-formulated ACTs such as sponsorship and support, involvement of key dihydroartemisinin-piperaquine and local agencies such as the PNG Institute of artemisinin-naphthoquine are not yet on the Medical Research and their international WHO-recommended list, perhaps because partners, and assistance from external they may not be GMP-manufactured, are agencies such as WHO. The alternative is given over less than 3 days and still require to risk the consequences of increasing evidence of efficacy, but they may prove to antimalarial treatment failures, which has be as effective as artemether-lumefantrine. been interpreted by some authorities as Although there is a view that using an equivalent to ‘medical malpractice’ (2). artemisinin drug with a partner drug with failing efficacy is not to be recommended (10), artesunate-amodiaquine and Timothy M.E. Davis artesunate-sulfadoxine-pyrimethamine remain on the WHO list and they have the University of Western Australia advantage that artesunate, primarily because School of Medicine and Pharmacology of its water solubility, is the artemisinin Fremantle Hospital derivative with the most favourable PO Box 480 pharmacological profile (11). In addition, Fremantle despite the WHO stance on ACT, non- Western Australia 6959 artemisinin-based combination therapies Australia
3 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
findings from a clinical trial in Uganda. Malar J [email protected] 2006;5:59. 7 Gaudiano MC, Di Maggio A, Cocchieri E, Antoniella E, Bertocchi P, Alimonti S, Valvo L. REFERENCES Medicines informal market in Congo, Burundi and Angola: counterfeit and sub-standard antimalarials. 1 World Health Organization. Guidelines for the Malar J 2007;6:22. Treatment of Malaria. Geneva: World Health 8 Davis TME, Karunajeewa HA, Ilett KF. Artemisinin- Organization, 2006. based combination therapies for uncomplicated 2 Attaran A, Barnes KI, Curtis C, d’Alessandro U, malaria. Med J Aust 2005;182:181-185. Fanello CI, Galinski MR, Kokwaro G, 9 Ratcliff A, Siswantoro H, Kenangalem E, Looareesuwan S, Makanga M, Mutabingwa TK, Maristela R, Wuwung RM, Laihad F, Ebsworth Talisuna A, Trape JF, Watkins WM. WHO, the EP, Anstey NM, Tjitra E, Price RN. Two fixed-dose Global Fund, and medical malpractice in malaria artemisinin combinations for drug-resistant treatment. Lancet 2004;363:237-240. falciparum and vivax malaria in Papua, Indonesia: 3 Marfurt J, Mueller I, Sie A, Maku P, Goroti M, an open-label randomised comparison. Lancet Reeder JC, Beck HP, Genton B. Low efficacy of 2007;369:757-765. amodiaquine or chloroquine plus sulfadoxine- 10 Adjuik M, Babiker A, Garner P, Olliaro P, Taylor pyrimethamine against Plasmodium falciparum and W, White N, International Artemisinin Study P. vivax malaria in Papua New Guinea. Am J Trop Group. Artesunate combinations for treatment of Med Hyg 2007;77:947-954. malaria: meta-analysis. Lancet 2004;363:9-17. 4 Thapa S, Hollander J, Linehan M, Cox-Singh J, 11 Ilett KF, Batty KT. Artemisinin and its derivatives. Bista MB, Thakur GD, Davis WA, Davis TME. In: Yu VL, Edwards G, McKinnon PS, Peloquin C, Comparison of artemether-lumefantrine with eds. Antimicrobial Therapy and Vaccines: Volume sulfadoxine-pyrimethamine for the treatment of II. Antimicrobial Drugs. London: ESun Technologies uncomplicated falciparum malaria in eastern Nepal. LLC, 2004:957-978. Am J Trop Med Hyg 2007;77:423-430. 12 Zongo I, Dorsey G, Rouamba N, Tinto H, 5 Karunajeewa HA, Ilett KF, Dufall K, Kemiki A, Dokomajilar C, Guiguemde RT, Rosenthal PJ, Bockarie M, Alpers MP, Barrett PH, Vicini P, Davis Ouedraogo JB. Artemether-lumefantrine versus TME. Disposition of artesunate and amodiaquine plus sulfadoxine-pyrimethamine for dihydroartemisinin after administration of artesunate uncomplicated falciparum malaria in Burkina Faso: suppositories in children from Papua New Guinea a randomised non-inferiority trial. Lancet with uncomplicated malaria. Antimicrob Agents 2007;369:491-498. Chemother 2004;48:2966-2972. 13 Moir JS, Garner PA, Heywood PF, Alpers MP. 6 Checchi F, Piola P, Fogg C, Bajunirwe F, Biraro Mortality in a rural area of Madang Province, Papua S, Grandesso F, Ruzagira E, Babigumira J, Kigozi I, Kiguli J, Kyomuhendo J, Ferradini L, Taylor WR, New Guinea. Ann Trop Med Parasitol 1989;83:305- Guthmann JP. Supervised versus unsupervised 319. antimalarial treatment with six-dose artemether- 14 Tilley L, Davis TME, Bray PG. Prospects for the lumefantrine: pharmacokinetic and dosage-related treatment of drug-resistant malaria parasites. Future Microbiol 2006;1:127-141.
4 PNG Med J 2008 Mar-Jun;51(1-2): 5-11
Knowledge and attitudes about infant feeding among nulliparous and parous women in Port Moresby: a comparative study
D. FRANK1, P. RIPA2, J.D. VINCE2,3 AND N. TEFUARANI2
Port Moresby General Hospital, Papua New Guinea and School of Medicine and Health Sciences, University of Papua New Guinea, Port Moresby
SUMMARY
Knowledge of the advantages of breastfeeding, the disadvantages of bottle feeding and the Papua New Guinean legislation to protect breastfeeding was determined in a comparative study of nulliparous and parous women attending the Antenatal Clinic of Port Moresby General Hospital. A high proportion (40%) of both groups had had experience of bottle feeding. Whilst the large majority (94%) of the women indicated that breastfeeding was the best way to feed babies, knowledge of the reasons for its superiority over bottle feeding and of the dangers of bottle feeding was poor in both groups. Knowledge of the legislation to protect breastfeeding was also poor and was the only area in which there was a significant difference between the groups, nulliparous women having poorer knowledge (p = 0.015). The level of education did not appear to be associated with knowledge about feeding. There is an urgent need to review the legislation, to find ways of enforcing it, and to improve the education of young people on issues of infant feeding.
Introduction prevalence of bottle feeding in Port Moresby with rates falling from 33% in 1976 (with 2 Exclusive breastfeeding for the first 6 out of 3 artificially fed infants being months of life is recommended as the best malnourished) to 11% three years later (8,9). feeding option for the vast majority of the Unfortunately the impact was short-lived, and world’s children, and exclusive breastfeeding bottle feeding has flourished in the city and for the first 4-6 months of life is the foundation in other parts of Papua New Guinea (10). A of the Papua New Guinea (PNG) Infant large infant feeding survey in 1995 showed Feeding Policy (1,2). The association of that 20% of 1822 mothers in different parts bottle feeding with high infant mortality and of the country had used bottle feeding (11). morbidity from infection and malnutrition in It has become clear that although there is a resource-poor countries has been well law to protect breastfeeding and control bottle documented (3,4). Papua New Guinea was feeding, it is not being implemented (12), and the first country to legislate to protect several attempts by the Paediatric Society breastfeeding, with the Baby Foods (Control) of Papua New Guinea to amend the law to Act in 1977 (5) (subsequently amended to make its implementation more practical and include control of sales of baby feeding cups effective have been unsuccessful. in 1984 (6)), some 4 years before the International Code of Marketing of Breast- Having a law to control the sale of bottles milk Substitutes was introduced in 1981 (7). is, however, only one aspect of the matter. The introduction of the legislation appeared It would seem reasonable to assume that to have an immediate impact on the mothers wish the best for their babies and
1 Port Moresby General Hospital, Free Mail Bag, Boroko, National Capital District 111, Papua New Guinea
Present address: Goroka General Hospital, PO Box 392, Goroka, Eastern Highlands Province 441, Papua New Guinea
2 School of Medicine and Health Sciences, University of Papua New Guinea, PO Box 5623, Boroko, National Capital District 111, Papua New Guinea
3 Corresponding author [email protected]
5 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 children. Of equal – if not greater – criteria. Women who were currently or importance to legislation, therefore, is previously employed, trained or training as mothers’ knowledge about best infant feeding health workers, those who had previously practices and the dangers associated with attended another antenatal clinic during the bottle feeding. Relatively little is known about current pregnancy and those unable to this, and the present study was designed to communicate in Tok Pisin or English and who determine the extent of knowledge on this had no interpreter were excluded. issue among pregnant women. All women were interviewed by the first The aims of the study were: author in a quiet and confidential setting. Interviews usually lasted about 10 minutes 1. To assess and compare knowledge of and an interpreter was used for only two the benefits of breastfeeding and the women. Data were collected on a pretested disadvantages of bottle feeding in and questionnaire and entered and analyzed between nulliparous (those who have using SPSS v10 software. not borne any children but who may have been pregnant previously) and Results parous (those who have borne at least one child) pregnant women. Study participants
2. To assess and compare knowledge 50 nulliparous women with a median age about the legislation relating to the of 22 years and 100 parous women with a control of bottle feeding in and between median age of 26 years were interviewed nulliparous and parous women. (Table 1). 3 of the nulliparous women were unmarried. There were no statistical Methods differences between the groups in education or employment status. A higher, but not The comparative study was carried out significantly different, proportion of the parous between May and August 2004 at the women had received some formal education Antenatal Clinic of the Port Moresby General on infant feeding. Of these, 81% of the total Hospital (PMGH), which enrols around 4000 study group indicated that they had received of the 10,000 women who deliver in the formal education on infant feeding from hospital each year (13). Patients were school. selected for the study at their first antenatal visit, and before any health education Past and proposed feeding practices instruction by the antenatal staff. For every nulliparous woman two parous but otherwise 70 of the parous women had exclusively unmatched women who had previously breastfed their previous children, 2 had totally delivered a live baby were invited to bottle fed, and 26 had used both breast and participate. No woman declined to be bottle feeding. Cup and spoon feeding and interviewed, and all women gave informed adoption accounted for the remaining 2. Of consent. There were no specific inclusion the women who had ever breastfed, 14% had
1ELBAT
CHARACTERISTICS OF NULLIPAROUS AND PAROUS WOMEN
suorapilluN suoraP )05=N( )001=N(
A))egnarelitrauqretni(naidem:)sraey(eg 2)42-02(2 13-32(62
N4noitacudeo 10
G7ylnonoitacude5-1edar 15
U5deyolpmen 377
F5gnideeftnafninonoitacudelamro 193
6 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 experienced difficulties, mainly due to some mothers had not given a great deal of insufficient milk. thought to the matter. The mean age at which parous women reported weaning their Details of previous feeding practices and children was 3.6 months. Parous women proposed feeding for the current pregnancy planned to wean the current child earlier than are shown in Table 2 . There were no the nulliparous women. differences between the proportions of nulliparous and parous women having Advantages of breastfeeding previously bottle fed. For the nulliparous women, their experience had in nearly all 94% of the total study population cases been gained by babysitting. Mothers responded that breastmilk was the best way returning to work was the main reason for to feed the baby, whilst 3% responded that bottle feeding. When asked specifically if bottle feeding was best and 3% responded there were medical reasons for bottle that both bottle and breast, or other methods, feeding, none of the mothers gave a positive were best. response. The knowledge about the main There were no differences in the feeding advantages of breastfeeding for the child, plans for the current pregnancy, although mother and family is shown in Table 3.
2ELBAT
PREVIOUS AND PROPOSED FEEDING PRACTICES
suorapilluN suoraP )05=N( )001=N(
elttobdesuylsuoiverP
O0dlihcnw 32
B0gnittisyba 26
O2reht 2
latoT 202 4
gnideefelttobrofsnosaeR
W3rehtomgnikro 102
M1tnangerprehto 5
B1smelborpgnideeftsaer 6
O7sreht 9
ycnangerptneserprofsnalP
E6gnideeftsaerbevisulcx 358
M0gnideeftsaerbylnia 19
T2gnideefelttoblato 2
M1gnideefelttobylnia 2
A1tpod 2
7 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
Although there were no significant Problems associated with bottle differences between the nulliparous and feeding parous women’s knowledge in each of the three categories, the proportions of ‘Don’t A similar proportion of both groups (32 know’ responses was consistently higher in nulliparous and 71 parous) were aware of the nulliparous group. 75% of the total study problems associated with bottle feeding. 42 population responded that they would be (70%) of 60 women who had previously bottle unlikely to become pregnant again whilst fed had been advised of the advantages of breastfeeding, whilst 23% either did not know breastfeeding and of how to care for the bottle about the contraceptive effect of and properly mix the milk. 30 (36%) of the breastfeeding, or thought they were more 84 parous women who indicated that bottle likely to become pregnant again if they had feeding was not good for the baby had bottle sex whilst they were still breastfeeding. fed.
3ELBAT
KNOWLEDGE OF THE ADVANTAGES OF BREASTFEEDING
suorapilluN suoraP )05=N( )001=N(
dlihcroF
B5noitirtuntse 123
H6rettebsworg/reihtlae 164
D4wonkt'no 131
O5sreht 9
rehtomroF
L0smelborpmutraptsopsse 8
E4rettebsta 122
M1elbaliavaemitero 14
D2wonkt'no 374
O3sreht 9
ylimafroF
S9secruoser/yenomseva 26
H5ylimafreihtlae 4
S2ylimafevitroppu 9
D3wonkt'no 365
O1sreht 5
8 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
Knowledge of legislation In this study relatively fewer women educated beyond primary school than those Significantly fewer of the nulliparous than with no or primary education responded that the parous women (4 vs 25, Fisher’s exact feeding other than breastfeeding was best test, p = 0.015) were aware of the legislation for babies – but the numbers were small and about bottle feeding, and only 19% overall, the differences were not considered although 69% of the total study population significant. Whilst the large majority of the knew that it was not possible to get a feeding women responded that breastfeeding was bottle from the pharmacy without a the best way to feed the child, 40% of the prescription. parous women had previously bottle fed. This might suggest that knowledge of the Relationship between education status reasons for the superiority of breast over and knowledge of best feeding bottle feeding was poor. Knowledge of the practices advantages for the mother and family appeared to be particularly poor in both Within the total study sample of 149 groups of women and knowledge of women for whom information was available, problems associated with bottle feeding was 1 of 13 with no formal education, 5 of 54 with also poor in both groups. education up to grade 6 (primary school) and 2 of 82 with education up to grade 12 30 (36%) of the 84 parous women who indicated bottle feeding, both bottle and indicated that bottle feeding was not good breast feeding, or other forms of feeding as for babies had bottle fed. 20 of the 40 parous being best for the baby (÷2 test, p = 0.2) women, and 13 of the 22 nulliparous women who had previously bottle fed indicated the Discussion mother’s return to employment as the reason. There is little doubt that the return to work The study has highlighted some worrying influences the working mother’s decision to facts. In this opportunity-based but otherwise introduce bottle feeding even though she may unselected sample, the experience of bottle be aware that it is not the best way of feeding. feeding was widespread – 40% of parous In a study of urban clinic nursing staff carried women and a similar proportion of nulliparous out 4 years after the PNG Baby Feed Supply women had previously bottle fed either their Control Act was passed, those who returned own or another person’s children. This to work within the first 6 months postpartum indicates that, in spite of the efforts to protect were more likely than those who returned and encourage breastfeeding and limit bottle later to have used bottle feeding (14). The feeding, the latter practice is well established same study, however, suggested that where in the community and accepted as an facilities for breastfeeding at work were alternative to breastfeeding and that bottles provided or where the workplace was near are easily accessible. Factors influencing the enough for the babysitter to bring the child widespread community acceptance of bottle for breastfeeding the large majority of the feeding may include the increasing numbers nurses exclusively breastfed. The current of working mothers, the acceptance of the situation is that although PNG subscribes to working mother as an important income the International Labour Organization earner for the family, and increased recommendations for working mothers, awareness of bottle feeding as a feeding which includes the provision of breastfeeding option from exposure to magazines, breaks, these recommendations are rarely television and other media sources. Bottles implemented in either the public or private are readily and legally accessible from sector. registered pharmacies on prescription, and None of the women asked to participate prescriptions are easily obtained from in the study declined, and we think the medical personnel in the public and women were a representative sample of especially the private sector. Bottles can also those attending the PMGH Antenatal Clinic. be obtained without prescription. In the PNG Whether or not they are representative of all Infant Feeding Survey of 1995 just over half antenatal mothers in Port Moresby is perhaps of the bottle-feeding mothers had obtained open to question. It is possible that the ready the bottle from a pharmacy or store without availability of doctors at the PMGH Antenatal prescription and others had been given the Clinic may have introduced some bias. bottle by a relative or friend (12). Although many of the more educated and
9 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 affluent women choose to attend the private the general population and the support of system it is still possible that PMGH may breastfeeding for individual women, HIV- attract more highly educated women than the infected or not, are tasks which should not urban clinics, where those less well educated be left to government agencies alone. may feel more comfortable. The fact that 82 Organizations such as Susu Mamas have (55%) of the women had had secondary experience and expertise in this area and education above Grade 6, with 70 (47%) they should receive financial and logistical reaching Grade 10 suggests that this might support from and appropriate recognition by be so. Nevertheless there was a broad and the government. similar spectrum of educational attainment in both nulliparous and parous women. Studies in Malaysia, Africa and other resource-poor countries including Papua There is clearly an urgent need to improve New Guinea have shown that legislation or the knowledge of the population about the the introduction of guidelines on feeding can advantages of breastfeeding and the positively affect feeding practices (17-20). disadvantages of bottle feeding. Even For these gains to be maintained, however, though more than half of the women constant reminding of the population surveyed had been educated to Grade 8 or concerned, together with means of enforcing higher, their knowledge of the advantages the regulations are required. Knowledge of of breastfeeding, the disadvantages of bottle the legislation in PNG is poor. In the present feeding and the existence of the study, although 69% of the women reported breastfeeding protection legislation was poor. that a prescription was necessary to get a This is disappointing since infant feeding is bottle from a pharmacy, only 19% knew about supposedly taught to Grade 8 girls in Home the legislation protecting breastfeeding. Economics, and currently to both girls and Sadly, and for a number of reasons, boys in Grade 8 Design and Technology enforcement of the PNG legislation is lacking. classes in the education reform curriculum (15). Conclusions
The HIV (human immunodeficiency virus) Whilst there appears to be general epidemic has added a further dimension to recognition that breastfeeding is the best way the importance of knowledge about infant to feed babies, knowledge of its advantages feeding. For most HIV-infected women in and of the disadvantages of bottle feeding is PNG exclusive breastfeeding for the first 6 poor. Knowledge of the legislation to protect months of the infant’s life offers the lowest breastfeeding is also poor. Experience of risk of transmission whilst non-exclusive, bottle feeding is widespread in the community mixed feeding is likely to be associated with and appears to be as common among young significantly increased transmission (16). female babysitters as it is among the parous Mixed feeding in the first 6 months of life is population. common in PNG (11) and in the present study 22 of the 150 mothers interviewed planned Recommendations to use some degree of mixed feeding. Clear and consistent messages need to be • Attention needs to be focused on provided about the benefits of exclusive revitalizing and sustaining the education breastfeeding and the risks of bottle and of young people about the importance mixed feeding for the infants of all women, of breastfeeding for themselves, their irrespective of their HIV status. children and their families. Such education should begin in primary The World Health Organization (WHO) school and be reinforced at subsequent has recently stated (16) that, “Governments levels of the education system. and other stakeholders should revitalize breastfeeding protection, promotion and • More active, innovative and repeated support in the general population. They education on infant feeding at antenatal should also actively support HIV-infected clinics is required. mothers who choose to exclusively breastfeed and take measures to make • The Papua New Guinean legislation to replacement feeding safer for HIV-infected protect breastfeeding and control bottle women who chose that option.” The and baby cup feeding urgently needs promotion and support of breastfeeding in updating.
10 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
• Effective ways of enforcing the law to Moresby, 1977. protect breastfeeding need to be 6 Papua New Guinea Government. Baby Feed Supplies (Control) (Amendment) Act 1984. The devised. Independent State of Papua New Guinea Government Gazette, Port Moresby, 1985:53. • Effective ways of supporting 7 World Health Organization. International Code of breastfeeding in the workplace are Marketing of Breast-milk Substitutes. Geneva: World Health Organization, 1981. needed. 8 Lambert J , Bashford J. Port Moresby infant feeding survey . PNG Med J 1977;20:175-179. • More support for organizations such as 9 Benjamin A , Biddulph J. Port Moresby infant Susu Mamas that promote and support feeding survey, 1979. PNG Med J 1980;23:92-96. breastfeeding in the community is 10 Primhak RA, Chojnowska E, Rongap A. Trends in infant feeding in Port Moresby. PNG Med J needed. 1989;32:113-116. 11 Friesen H, Vince J, Boas P, Danaya R, Mokela D, ACKNOWLEDGEMENTS Ogle G, Asuo P, Kemiki A, Lagani W, Rongap T, Varughese M, Saweri W. Infant feeding practices in Papua New Guinea. Ann Trop Paediatr We gratefully acknowledge the 1998;18:209-215. constructive comments of Professor A.B. 12 Friesen H, Vince J, Boas P, Danaya R. Protection Amoa and Dr Grace Kariwiga on the design of breastfeeding in Papua New Guinea. Bull World of the study and questionnaire, and the Health Organ 1999;77:271-274. 13 Amoa AB. Annual Report of the Obstetrics and assistance of the nursing staff of the Port Gynaecology Unit, Port Moresby General Hospital, Moresby General Hospital Antenatal Clinic. Papua New Guinea, 2003. 14 Marshall LB. Infant feeding practices among clinical REFERENCES nursing staff in urban Papua New Guinea. Int J Nurs Stud 1983;20:63-74. 15 Papua New Guinea Department of Education. 1 Department of Nutrition for Health and Grade 8 Technology School Syllabus. Papua New Development and Department of Child and Guinea Department of Education, Port Moresby, Adolescent Health and Development, World 2004. Health Organization. The Optimal Duration of 16 World Health Organization. WHO HIV and Infant Exclusive Breastfeeding. Report of an Expert Feeding Technical Consultation – Consensus Consultation. Geneva: World Health Organization, Statement. Held on behalf of the Inter-agency Task 2001:1-2. Team (IATT) on Prevention of HIV Infections in 2 Papua New Guinea Department of Health. Pregnant Women, Mothers and Their Infants. National Policy on Infant Feeding Practices. Port Geneva: World Health Organization, 2007. Moresby: Department of Health, 1994. 17 MCH News PAC. Breastfeeding Act will aid working 3 Brown KH, Black RE, Lopez de Romaña G, Creed mothers. Commonwealth of the Northern Mariana de Kanashiro H. Infant-feeding practices and their Islands. MCH News PAC 1987;2(4):9,12. relationship with diarrhoeal and other diseases in 18 Kader HA. Early breast-feeding patterns in a Huascar (Lima), Peru. Pediatrics 1989;83:31-40. Malaysian maternity hospital 1980-1983. Malays J 4 Almeida RM, De Marins VM, Valle J. Reprod Health 1984;2:105-110. Breastfeeding, socio-economic conditions and 19 Sall MG, Kuakuvi N, Sow HD, Ngom A, Sanokho nutritional status of children younger than 12 months A, Wade B. Strategies for maintaining the practice in Brazil. Ann Trop Paediatr 1999;19:257-262. of breastfeeding in urban Senegal. [Fr] Afr Med 5 Papua New Guinea Government. Baby Feed 1986;25:479-480. Supplies (Control) Act 1977. The Independent State 20 Biddulph J. Legislation to protect breastfeeding. of Papua New Guinea Government Gazette, Port PNG Med J 1983;26:9-12.
11 PNG Med J 2008 Mar-Jun;51(1-2):12-16
A review of the current state of malaria among pregnant women in Papua New Guinea
IVO MUELLER1, STEPHEN ROGERSON2, GLEN D.L. MOLA3 AND JOHN C. REEDER1
Papua New Guinea Institute of Medical Research, Goroka, Department of Medicine, Royal Melbourne Hospital, University of Melbourne, Australia and School of Medicine and Health Sciences, University of Papua New Guinea, Port Moresby
SUMMARY
Besides young children, pregnant women are at high risk of malaria in highly endemic countries. This paper reviews evidence from studies conducted in Papua New Guinea (PNG) in the last 20 years on the burden and prevention of malaria in pregnancy and highlights gaps in our knowledge of malaria in pregnancy in PNG. Overall, primigravidae were found to be at higher risk than multigravidae, with up to 40% of primigravidae but only 10-25% of multigravidae infected with Plasmodium falciparum at delivery. Such infections were found to be associated with a 128-145 g decrease in birthweight. Mean birthweights reported between 1980 and 2003 range from 2.58 to 2.72 kg in primigravidae and 2.84 to 3.09 kg in multigravidae, with 21% to 48% and 9% to 19% of babies born to primigravidae and multigravidae, respectively, of low birthweight (<2500 g). The negative impact of malaria in pregnancy is compounded by relatively low rates of antenatal coverage. The current PNG national treatment policy which prescribes a treatment course of first-line antimalarial treatment (currently chloroquine and sulphadoxine- pyrimethamine) at first antenatal clinic contact, followed by weekly chloroquine prophylaxis and iron and folate supplementation, may no longer be effective given the high levels of resistance to chloroquine in PNG and poor compliance. In order to reduce the burden of malaria in pregnancy in PNG, alternative methods of control such as insecticide-treated nets and intermittent preventive treatment in pregnancy (IPTp), as well as improved modes of delivery of maternal health interventions, are urgently needed.
Introduction fifth leading cause of lost DALYs. With a maternal mortality ratio of 370/100,000 Malaria in Papua New Guinea (PNG) is childbirth is still among the highest mortality the leading cause of outpatient attendances risks for women of childbearing age in PNG. nationally, the third commonest cause of hospital admission and the second Malaria in pregnancy and maternal commonest cause of death, and causes the health greatest burden of lost disability-adjusted life- years (DALYs) at 4894/100,000 per year. Precise estimates of the burden of malaria Accordingly, malaria is a top priority of the in pregnancy are not available, but regional national health response. As in highly patterns of birthweight indicate that alongside endemic areas elsewhere in the world, in maternal nutrition and socioeconomic factors PNG both prevalence of malaria infection and malaria is a major causative factor for the incidence of morbidity are highest in young high prevalence of low birthweight (LBW) in children (1,2) and pregnant women (3). lowland and coastal parts of PNG (4). In Maternal deaths (to which malaria in Madang and Maprik, where most in-depth pregnancy contributes significantly) are the studies on malaria in pregnancy have been
1 Papua New Guinea Institute of Medical Research, PO Box 60, Goroka, Eastern Highlands Province 441, Papua New Guinea
2 Department of Medicine, Royal Melbourne Hospital, University of Melbourne, Parkville, Victoria 3052, Australia
3 School of Medicine and Health Sciences, University of Papua New Guinea, PO Box 5623, Boroko, National Capital District 111, Papua New Guinea
12 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 done, mean birthweights reported between prevalence of malaria infection at delivery. 1980 and 2003 range from 2.58 to 2.72 kg in Infection rates were significantly higher in primigravidae (PG) and 2.84 to 3.09 kg in primigravidae, reaching over 40% both in multigravidae (MG), with 21% to 48% and peripheral and placental blood in some 9% to 19% of babies born to PG and MG, studies (Table 1). respectively, of low birthweight (<2500 g) (3,5-9). Comparative studies between Placental pathology was not investigated malaria-endemic coastal PNG and the in early studies, but a recent study in Madang malaria-free highlands suggest that malaria found that 42% of women delivering at in pregnancy is responsible for up to 11% of Alexishafen Health Centre showed evidence anaemia and 40% of low birthweight in of active, chronic or past chronic malaria coastal areas (6). infection on placental histology (10).
In the only published study looking at the In a pooled sample that contained roughly history of infections during pregnancy, Brabin equal numbers of PG, SG and MG, Allen et et al. (3) found that the prevalence of malaria al. (9) found that peripheral and placental infections at first antenatal clinic (ANC) visit infection at delivery were associated with a in primigravidae peaked at 9-16 weeks of 128 g and 145 g decrease in birthweight gestation (55%). No similar trend was respectively. Other PNG studies lacked observed for multigravidae. Despite sufficiently large sample sizes to find receiving chloroquine prophylaxis at all ANC significant differences in birthweight in visits, the average prevalence of infections relation to malaria infection status. at any ANC visit was 34% in PG, 30% in secundigravidae (SG) and 19% in MG. Anaemia is a very common feature in Several published studies looked at pregnant women in many parts of lowland
1ELBAT
PREVALENCE OF MALARIAL INFECTIONS AT DELIVERY
Penoitcefniairalamrofevitiso cnerefeR
PGG SGSGMdnaG M
larehpireP
M%7891-6891,gnada 4%4 2)6 5(
M%6991-4991,gnada 2%5 2%0 1)3 9(
M%3002-2002,gnada 2%6 1%1 1)1 01(
latnecalP
M%8891-6891,kirpa 4%1 2%3 8)7(
M%6991-4991,gnada 3%4 2%6 1)4 9(
M%3002-2002,gnada 2%4 1%3 1)3 01(
ygolotsihlatnecalP
M%3002-2002,gnada 6%3 5%0 3)3 01(
eadivargimirp=GP eadivargidnuces=GS eadivargitlum=GM
13 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
PNG. In 1986-1988, 89% of women in pregnancy. However, further in-depth delivering at the major referral hospital in studies are needed to assess the contribution Madang had a haemoglobin level (Hb) of <11 of non-falciparum malaria to malarial disease g/dl and 19% of PG and 17% of MG had in pregnancy in PNG. severe anaemia, ie, Hb <7 g/dl (6). Haemoglobin values measured at first There have been only a few studies on booking at a rural health centre in Madang congenital malaria, but the limited data were significantly lower in both PG (8.6 g/dl) indicate that transplacental transfer of P. and MG (8.7 g/dl) than in non-pregnancy falciparum parasites is common. In a small controls (10.4 and 10.2 g/dl respectively), study in Madang umbilical cord infection was despite similar levels of iron deficiency, with found in 7 of 15 children (47%) born to a tendency for haemoglobin levels to women with parasitaemia at time of delivery decrease with increasing length of gestation (12); 4 of these children also had detectable and to be lower in PG with concurrent malaria peripheral parasitaemia. Little is known infection (–0.7 g/dl, p = 0.15) (5). How far about transplacental transfer of other malaria these drops are directly related to malaria or species, although at least one case of a simply due to physiological dilution in symptomatic P. vivax infection acquired in pregnancy, most of which will have occurred utero has been described (13). in PNG women prior to booking, is unclear. However, the fact that malaria control Very little is known about the problem of interventions such as intermittent preventive malaria in non-immune women living in areas treatment in pregnancy (IPTp) result in of low endemicity such as the highlands. The significantly increased haemoglobin levels overall burden of malaria in pregnancy is (11), underlines the importance of malaria likely to be low as indicated by the as a cause of anaemia in pregnancy. Up to substantially higher haemoglobin levels and 40% of women showed signs of iron lower rates of LBW in highland areas (6). deficiency. With chloroquine (CQ) However, due to low immune status acquired prophylaxis and treatment for severe infections are more likely to be severe and anaemia, Hb in these women recovered to mortality rates in pregnant mothers with 9.6 g/dl in PG and 9.3 g/dl in MG at delivery severe malaria can be as high as 50% (14). with no significant difference between women positive and negative for malaria. Perhaps The detrimental effects of malaria in related to increasing coverage of iron/folate pregnancy are compounded by low rates of supplementation following their introduction antenatal coverage and supervised into the national treatment guidelines in 1985, deliveries. Nationally, only 33% of women haemoglobin values during pregnancy and receive any antenatal care during their at delivery have been higher in recent studies pregnancy and 44% of deliveries are (9), and the proportion of women severely supervised; however, there are large regional anaemic (Hb <7 g/dl) at delivery decreased variations. Antenatal coverage can drop to 13% in 2002-2003 (10). below 50% and the proportion of supervised deliveries falls to as low as 10-15% in some Anaemia (Hb <8 g/dl) was significantly rural districts of the country (15). associated with a decrease in birthweight in PG (–281 g) but not in MG (–84 g) (5). Two factors contribute to this low antenatal However, in-depth studies showed that coverage and low rate of supervised anaemia is mainly linked to an increased risk deliveries: limited access to health care and of preterm delivery rather than a decrease in strong customary beliefs surrounding birthweight in term infants (9). childbirth. While the number of women delivering at provincial hospitals has greatly Although non-falciparum malaria is increased over the past two decades, rates common in PNG, little is known about its of supervised deliveries in rural areas have effect on pregnant mothers and their babies. been decreasing due to a range of factors. In a study in the mid-1980s the prevalence In many rural parts of PNG, women will have of P. vivax was lower in pregnant mothers to walk for several hours through often attending ANC (3) (and under CQ difficult terrain to reach the nearest health prophylaxis) than in the postnatal period, but centre. With a decline in mobile ANC clinic this is likely to be more a reflection of the coverage, access to both ANC and delivery high effectiveness of CQ prophylaxis against services is therefore severely limited. Unless non-falciparum malaria than of reduced risk a delivery plan for supervised birth is worked
14 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 out with the woman and her husband during had little effect on malaria infection rates at the ANC period, many women will not be able delivery (5,9,17) although it was associated to reach the health centre once labour has with increased haemoglobin levels (5,9) and started, regardless of complications. Staffing decreased risk of preterm delivery (9). Thus, levels and morale are other important in the absence of information regarding obstacles to good clinical care. Women often alternative approaches, the policy is state that they do not deliver at the health continuing today. centres because they do not want to be attended to by male nurses, or because they The use of insecticide-treated bed nets are not sure if the health centre will be open (ITNs) during pregnancy is also advised as or they will be able to find a nurse if they arrive part of the national guidelines, but to date no after hours. special bed net distribution for pregnant mothers is in place. In many areas ITNs can In addition, childbirth is still the focus of be bought from health centres but supplies many customary beliefs and restrictions in are unreliable and prices often a deterrent. some PNG cultures. Childbirth, like This situation is expected to change in the menstruation, is often believed to have a near future, as PNG has secured a grant from ‘polluting’ influence, in particular on men, and the Global Fund for AIDS, Tuberculosis and assistance to women in labour is often Malaria that will allow the provision of long- limited. In some places women in labour will lasting ITNs to all people living in malarious go to the bush and deliver their babies areas in PNG. Monitoring the impact of this completely unattended. There may also be program on adverse pregnancy outcomes strong beliefs associated with the disposal will be an important part of assessing its of placentas that inhibit women from effectiveness. delivering at a health facility. Current treatment guidelines for malaria Policy, prevention and treatment in pregnancy indicate the use of CQ and SP for uncomplicated disease, oral quinine with Both malaria and safe motherhood have SP for treatment failure, and parenteral been identified as priority areas in the 2001- quinine for severe malaria in pregnancy. The 2010 PNG National Health Plan (15). The clinical efficacy of CQ plus SP against P. plan calls for a reduction of maternal mortality falciparum is still high (93%) (18); however, to 260/100,000 and LBW to <10%, while at the high levels of parasitological failure (up the same time aiming to increase ANC to 15%) indicate that these drugs will reach coverage to 90% and the proportion of the end of their life span sooner rather than supervised deliveries to 70%. The goals for later. Although not yet part of the official maternal mortality and LBW will not be treatment guidelines for pregnancy malaria, reached without effective control of the artesunate and IM artemether are second- detrimental effects of malaria in pregnancy. line treatments for malaria in non-pregnant people, and are regularly used to treat Given the problems with access to women in their third trimester admitted to adequate health care and the reluctance of hospital with a presumptive diagnosis of mothers to deliver at health centres, malaria. preventive interventions have to be the main approach to improving the health of pregnant Future research needs mothers and their babies. The high levels of morbidity and mortality The current PNG national treatment policy indicate that the current policies for prescribes a treatment course of first-line antimalarial treatment (currently chloroquine prevention and treatment of malaria in and sulphadoxine-pyrimethamine (SP)) at pregnancy are inadequate and new or first ANC contact followed by weekly improved approaches are needed. Research chloroquine prophylaxis and iron and folate into new options for the prevention of malaria supplementation. However, the usefulness in pregnancy such as intermittent preventive of chloroquine prophylaxis is questionable, treatment (IPTp), the better integration of given the high levels of resistance to ITNs and/or new and improved forms of chloroquine (16) in PNG and well-known prophylaxis is of high priority. In the medium problems of compliance. Even in the mid- term, new drugs for the treatment of malaria 1980s and 1990s chloroquine prophylaxis in pregnancy will be needed.
15 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
The above interventions will, however, environment and socio-economic factors. Ann Hum only be successful if they can fit into local Biol 2002;29:74-88. 5 Brabin BJ, Ginny M, Sapau J, Galme K, Paino J. circumstances, customs and beliefs. Consequences of maternal anaemia on outcome of Operational research studies into modes of pregnancy in a malaria endemic area in Papua New delivery of maternal health interventions, in Guinea. Ann Trop Med Parasitol 1990;84:11-24. particular on ways of increasing coverage of 6 Brabin B, Piper C. Anaemia- and malaria- attributable low birthweight in two populations in ANC and supervised deliveries, as well as a Papua New Guinea. Ann Hum Biol 1997;24:547- better understanding of women’s perceptions 555. of their own health, are thus needed if the 7 Desowitz RS, Alpers MP. Placental Plasmodium high levels of maternal mortality, severe falciparum parasitaemia in East Sepik (Papua New Guinea) women of different parity: the apparent maternal anaemia and low birthweight are absence of acute effects on mother and foetus. Ann to be reduced. Trop Med Parasitol 1992;86:95-102. 8 Oppenheimer SJ, Macfarlane SB, Moody JB, The special epidemiology of malaria, and Harrison C. Total dose iron infusion, malaria and the genetic and cultural diversity, as well as pregnancy in Papua New Guinea. Trans R Soc Trop Med Hyg 1986;80:818-822. imminent changes to malaria control policies, 9 Allen SJ, Raiko A, O’Donnell A, Alexander NDE, make PNG an ideal location to conduct in- Clegg JB. Causes of preterm delivery and depth studies into different aspects of malaria intrauterine growth retardation in a malaria endemic in pregnancy. Of particular interest are the region of Papua New Guinea. Arch Dis Child Fetal Neonatal Ed 1998;79:F135-F140. aetiology and pathology of non-falciparum 10 Benet A, Khong TY, Ura A, Samen R, Lorry K, malaria in pregnancy or the effect of PNG- Mellombo M, Tavul L, Baea K, Rogerson SJ, specific host genetic protective traits such as Cortés A. Placental malaria in women with South- Southeast Asian ovalocytosis, alpha- East Asian ovalocytosis. Am J Trop Med Hyg 2006;75:597-604. thalassaemia or Gerbich blood group 11 Garner P, Gülmezoglu A. Drugs for preventing negativity on the risk and effects of malaria malaria-related illness in pregnant women and death in pregnancy. in the newborn. Cochrane Database Syst Rev 2002:CD000169. Building on earlier work the PNG Institute 12 Lehner PJ, Andrews CJA. Congenital malaria in Papua New Guinea. Trans R Soc Trop Med Hyg of Medical Research is committed to tackling 1988;82:822-826. the challenges posed by the high levels of 13 Schuurkamp GJ, Paika RL, Spicer PE, Kereu RK. malaria and maternal mortality in the country Congenital malaria due to Plasmodium vivax: a case and thereby contributing to a better and study in Papua New Guinea. PNG Med J healthier future for all PNG women. 1986;29:309-312. 14 Lalloo DG, Trevett AJ, Paul M, Korinhona A, Laurenson IF, Mapao J, Nwokolo N, Danga- REFERENCES Christian B, Black J, Saweri A, Naraqi S, Warrell DA. Severe and complicated falciparum malaria in 1 Genton B, Al-Yaman F, Beck HP, Hii J, Mellor S, Melanesian adults in Papua New Guinea. Am J Trop Narara A, Gibson N, Smith T, Alpers MP. The Med Hyg 1996;55:119-124. epidemiology of malaria in the Wosera area, East 15 Papua New Guinea Department of Health. Papua Sepik Province, Papua New Guinea, in preparation New Guinea National Health Plan 2001-2010. Port for vaccine trials. I. Malariometric indices and Moresby: Department of Health, Aug 2000. immunity. Ann Trop Med Parasitol 1995;89:359-376. 16 Müller I, Bockarie M, Alpers M, Smith T. The 2 Cattani JA, Tulloch JL, Vrbova H, Jolley D, Gibson epidemiology of malaria in Papua New Guinea. FD, Moir JS, Heywood PF, Alpers MP, Stevenson Trends Parasitol 2003;19:253-259. A, Clancy R. The epidemiology of malaria in a population surrounding Madang, Papua New Guinea. 17 Mola GL, Wanganapi A. Failure of chloroquine Am J Trop Med Hyg 1986;35:3-15. malaria prophylaxis in pregnancy. Aust NZ J Obstet 3 Brabin BJ, Ginny M, Alpers M, Brabin L, Eggelte Gynaecol 1987;27:24-26. T, Van der Kaay HJ. Failure of chloroquine 18 Marfurt J, Müller I, Sie A, Maku P, Goroti M, prophylaxis for falciparum malaria in pregnant Reeder JC, Beck HP, Genton B. Low efficacy of women in Madang, Papua New Guinea. Ann Trop amodiaquine or chloroquine plus sulfadoxine- Med Parasitol 1990;84:1-9. pyrimethamine against falciparum and vivax malaria 4 Müller I, Betuela I, Hide R. Regional patterns of in Papua New Guinea. Am J Trop Med Hyg birthweights in Papua New Guinea in relation to diet, 2007;77:947-954.
16 PNG Med J 2008 Mar-Jun;51(1-2):17-26
A case-control study of VDRL-positive antenatal clinic attenders at the Port Moresby General Hospital Antenatal Clinic and Labour Ward to determine outcomes, sociodemographic features and associated risk factors
GLEN D.L. MOLA1, ALEX GOLPAK2,3 AND A.B. AMOA2
School of Medicine and Health Sciences, University of Papua New Guinea, Port Moresby and Division of Obstetrics and Gynaecology, Port Moresby General Hospital, Papua New Guinea
SUMMARY
Between June 2001 and December 2002, 152 antenatal patients at Port Moresby General Hospital who were Venereal Disease Research Laboratory (VDRL) serology positive and 150 unselected antenatal patients who tested negative were studied to determine the gestational age at which the tests were performed, the time it took for results to become available, the proportion of patients who received treatment, the sociodemographic characteristics associated with VDRL positivity and the effect of VDRL positivity on maternal and perinatal outcomes. The prevalence rate of VDRL positive among antenatal clinic attenders in Port Moresby at that time was 4.4%. Of the 152 VDRL-positive patients in this study 97% were also Treponema pallidum haemagglutination (TPHA) positive. Significantly more of the positive patients were of highlands origin, lived in settlements, had previous marriages, had lower parities, delivered preterm babies, had stillbirths, had growth-restricted babies and had babies with lower Apgar scores at both 1 and 5 minutes. The mean birthweight was significantly lower among the positive patients. Significantly more of the positive patients were married to spouses with occupations which were regarded as ‘risky’ for sexually transmitted infections. There was no difference between the two groups with respect to patient’s education, marital status, husband’s education, gestational age at delivery and the number of days the baby spent in the Special Care Unit. The study concluded that the current antenatal screening does not provide adequate coverage for our patients. If the current availability of clinic-based strip tests provided by a non-government organization can be continued by the Ministry of Health we should be able to overcome this problem.
Introduction screening for syphilis is a routine. This public health measure is particularly rewarding as The importance of maternal syphilis as a three patients – the mother, her sexual cause of abortion, perinatal death and partner and her unborn child – may be congenital syphilis is well known. The protected from the ravages of this disease. detection of syphilis in antenatal populations In the USA from the 1940s to the late 1980s, and its proper management has been one of the number of infant deaths from congenital the major success stories of antenatal care. syphilis fell by 99%; and rates of clinically In most antenatal clinics, in both developed apparent congenital syphilis were reduced and developing countries, serological almost 100-fold (1). In the National Capital
1 Discipline of Obstetrics and Gynaecology, School of Medicine and Health Sciences, University of Papua New Guinea, PO Box 5623, Boroko, NCD 111, Papua New Guinea
2 Division of Obstetrics and Gynaecology, Port Moresby General Hospital, Free Mail Bag, Boroko, NCD 111, Papua New Guinea
3 Current address: Division of Obstetrics and Gynaecology, Kimbe General Hospital, PO Box 736, Kimbe, WNB 621, Papua New Guinea
17 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
District (NCD) of Papua New Guinea (PNG), • To determine if the positive patients where some 12,000 women attend public have any sociodemographic antenatal clinics each year, about 4-5% are characteristics which may be used to Venereal Disease Research Laboratory identify them. (VDRL) positive. Prevalence rates among asymptomatic pregnant women have been Patients and methods found to be 1% in The Gambia (2), 12.5% in Zambia (3) and 17.5% in Ethiopia (4); in the Study design and study population highlands provinces of Papua New Guinea the prevalence rate ranges from 10% to 20% This was a case-control study conducted (5). For the devastating effects of syphilis to between June 2001 and December 2002. be obviated, it is important to find out the The study population consisted of all current situation of routine testing and the pregnant patients attending the Antenatal management of patients who test positive. Clinic of the PMGH and patients delivering It is also important to determine the at the Labour Ward of the PMGH, and their sociodemographic characteristics of those babies. The cases were VDRL-positive who test positive so that, where necessary, mothers attending the Antenatal Clinic or awareness and educational programs can delivering at the PMGH Labour Ward. usefully and effectively be put in place for primary prevention. Controls were as follows:
Aims of the study • The next VDRL-negative mother after a VDRL-positive case who booked in To determine: the same trimester as the case at the PMGH Antenatal Clinic, or • The prevalence of syphilis in the Port Moresby General Hospital (PMGH) • The next VDRL-negative mother Antenatal Clinic population and the delivering at the Labour Ward of the gestational age (GA) at diagnosis PMGH who booked at the same gestational age as the case. • Risk factors associated with VDRL positivity Data sources
• The perinatal outcomes in those who 1. Antenatal Clinic VDRL register. are syphilis serology positive. 2. Central Public Health Laboratory VDRL Measurable objectives: register.
• The proportion of subjects who are 3. Labour Ward birth registers. VDRL reactive 4. Patients’ hospital records.
• The proportion of the reactive VDRLs 5. A structured interviewer-administered who are Treponema pallidum questionnaire. haemagglutination (TPHA) positive Definition of terms • The gestational age at which the syphilis serology diagnosis was made Husbands’ occupation ‘risky’ – this was defined as a husband who worked with the • The time interval between sending the police, army or prison service, was a taxi or specimen and linking the result/report public motor vehicle (PMV) driver or was a with the patient freelance business man or politician. Previous studies (6) have shown that these • How many of the positive patients occupations have been associated with complete treatment before delivery increased sexually transmitted infection (STI) risk. • The proportion of partners treated Intrauterine growth restriction (IUGR) – the • The stillbirth and neonatal death rates diagnosis of IUGR was based mostly on
18 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 clinical grounds by paediatricians after birth. OR 6.59 (2.94 -15.18), p <0.00000). Only a few cases were diagnosed antenatally. In order to accurately detect Pregnancy characteristics are shown in IUGR one must have a reliable menstrual Table 2. The proportion with a previous bad date or an early ultrasound scan. Many of perinatal outcome was not different in the two our antenatal patients do not keep accurate groups. The mean parity of the cases was records of their last menses and we are not significantly lower than in the control group, able to do routine early pregnancy ultrasound (1.4 vs 2.8, p = 0.0109); the cases had blood scans to assess or confirm gestation. collected at a significantly later GA than the Moreover the majority of our antenatal controls (25 weeks vs 23 weeks, p = 0.0195). patients book after 20 weeks gestation. The mean interval between test and linkage of the result with the patient was 3 weeks. Prematurity – as with IUGR, the prenatal diagnosis of this relies largely on an accurate In the period under study, the VDRL menstrual date or an early ultrasound scan positivity rate amongst all the antenatal date. Most of our cases were diagnosed by bookings in the National Capital District was paediatricians by neonatal assessment using 4.4% and, of these, 86% were also TPHA the Dubowitz system. positive (5). Among the cases in this study, 97% were also TPHA positive; and 92% were Results treated (Table 2). The mean gestational age at which the patients were treated was 27.6 Between June 2001 and December 2002 weeks. Only 5% were treated before 19 there were 15,232 deliveries at the PMGH weeks gestation, 84% were treated between Labour Ward. A total of 356 (178 cases and 19 and 35 weeks gestation and 3% were 178 controls) antenatal attenders were treated after 35 weeks. 74% of partners of recruited for study. Unfortunately 26 (15%) the cases were also treated. of the cases were unaccounted for by the completion of the study. Most of the cases Labour and perinatal outcomes are shown lost to follow-up did not attend the PMGH for in Table 3. There were 12 stillbirths among delivery. The corresponding controls were the cases and none among the controls (p = automatically excluded from the study. 0.00045). There were significantly more Records were available for 152 cases and preterm deliveries among the cases (OR 2.9 150 controls. (1.1-7.9), p = 0.015). Significantly more babies with IUGR were found among the The mean gestational age at booking for cases (OR 5.3 (2.3-12.4), p <0.00000). The the cases in the study was 24 weeks; this mean birthweight was significantly lower was similar in both groups. among the cases (2835 g vs 3069 g, p = 0.043) and the mean Apgar scores at both 1 Sociodemographic characteristics are and 5 minutes were significantly lower among shown in Table 1. Significantly more of the the cases (p = 0.0365 and p = 0.0027 cases were of highland region origin (36% respectively). However, the mean GA at vs 25%, p = 0.040) and significantly more of delivery and the mean number of days the them lived in settlements (46% vs 27%, p = baby spent in the Special Care Unit were not 0.003). There was no difference between significantly different for cases and controls. the two groups as far as age, village residence, patient education, occupation and 84% of the babies of the cases were given marital status were concerned. Husband’s treatment with benzathine penicillin shortly education and employment status were also after birth. not significantly different between cases and Discussion controls. However, significantly more of the cases had husbands who had ‘risky’ Serological testing for syphilis in occupations as defined above (31% vs 14%, pregnancy has been done routinely in Britain p = 0.0004). for over 45 years, in Norway and France by law since 1948 and 1932 respectively, as a Marital status and number of years married routine in Hong Kong for 50 years, and in were not different between cases and PNG for the last 4 decades. controls; however, significantly more of the cases had previous marriages (30% vs 6%, This study and others in recent times have
19 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008
1ELBAT
SOCIODEMOGRAPHIC CHARACTERISTICS
)%(sesaC )%(slortnoC selbairaV Op)IC%59(R H-M 251=N 051=N
selbairavlacirogetaC
nigirofonoigeR
H4sdnalhgi 573
I96sdnals
M1esamo 141
S8nrehtuo 739
H)noigeRnrehtuoSsusrevsdnalhgi 5)5.53(251/4 3)7.42(051/7 1410.3-10.1(47. 30.0
H)tserehtsusrevsdnalhgi 5)5.53(251/4 3)7.42(051/7 1168.2-99.0(26. 040.0
02 tniopecnereferehtsaecnedisernabru:ecnediseR V)egalli 2)5.41(251/2 3)3.32(051/5 1025.2-56.0(72. 554.0 S)tnemeltte 7)1.64(251/0 4)7.62(051/0 2397.3-62.1(91. 00.0 noitacudes'tneitaP D)sraey6-0noitaru 8)2.75(251/7 3.94(051/47 D)sraey6>noitaru 6)8.24(251/5 7)7.05(051/6 1322.2-58.0(73. 961.0 noitapuccos'tneitaP U)deyolpmen 1)0.37(251/11 67(051/411 S)dellik 1)5.21(251/9 7.61(051/52 S)delliksime 1)9.7(251/2 4(051/6 U)delliksn 1)6.6(251/0 3.3(051/5 sutatslatiraM M3deirra 144 41 S2detarape 0 W1wodi 5 S6elgni 1 sutatselgnissusrevdeirraM S)elgni 6)9.3(251/ 1)7.0(051/ 618.431-17.0(0. 60.0 U)deirramn 9)9.5(251/ 6)4(051/ 1429.4-84.0(5. 4.0 20 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 rentraptneserpotdeirramsraeyforebmuN 1)sraey5- 4)3.82(251/3 7.63(051/55 >)sraey5 1)7.17(251/90 9)3.36(051/5 0741.1-14.0(86. 0.0 segairramsuoiverP 4))6.92(251/5 9)6(051/ 6081.51-49.2(95. 0000.0< noitacudes'dnabsuH D)sraey6-0noitaru 4)0.72(251/1 5.13(941/74 D)sraey6>noitaru 1)0.37(251/11 1)5.86(941/20 1831.2-37.0(52. 3.0 sutatstnemyolpmes'dnabsuH V9regalli 17 E2deyolpm 190 9 U1deyolpmen 433 T2lato 195 41 noitapuccos'dnabsuh'yksiR' Y)se 4)9.03(251/7 2)41(051/1 2401.5-94.1(57. 000.0 Csselbairavsuounitno esaC slortnoC eulavpHW-K )DS(naeM )DS(naeM A))sraey(eg 2)31.5(5.4 2167.4(5.5 10.0 P)detelpmocsraeyloohcss'tneita 6)24.3(5. 7655.3(3. 40.0 Y)rentraptneserphtiwsrae 4)23.4(5. 5801.4(3. 10.0 H)loohcstasraeys'dnabsu 8)65.3(7. 9858.5(3. 26.0 oitarsddo=RO oitarsddofolavretniecnedifnoc%59=IC%59 eulavpderauqsihClezsneaH-letnaM=pP-M noitaiveddradnats=DS eulavptsetsillaW-laksurK=pHW-K shown that syphilis is still one of the major higher rates of STIs in the highlands than in causes of perinatal morbidity and mortality. the rest of the country. For example, Garner It is well known that infection with syphilis et al. in 1972 showed a very high incidence leads to late spontaneous abortion, preterm (28.9%) of syphilis in adults in residences delivery, intrauterine growth restriction, along the Highlands Highway (9) whilst a stillbirth and early neonatal death (3,7). A more recent study showed the incidence of study done at the Goroka Base Hospital in syphilis in antenatal clinic attenders at the 2000, where the babies were diagnosed with Goroka Base Hospital to be 7.1% (8). This syphilis on clinical and serological findings, rate was higher than the 4.4% found in the showed that congenital syphilis accounted for antenatal clinics of Port Moresby in 2002 (5). 5.5% of neonatal admissions and 22% of all neonatal deaths (8). Women in urban settlements were significantly more likely to be seropositive In this study we identified some major risk than their village and urban metropolitan factors associated with this infection in our counterparts. This is not surprising. In the antenatal mothers. One of them was United States and some developing countries highlands ethnicity. Past studies have shown syphilis is associated with poverty, drug 21 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 2ELBAT PREGNANCY CHARACTERISTICS C)selbairavlacirogeta %(sesaC )%(slortnoC Op)IC%59(R H-M 251=N 051=N P)shtaedlatanirepsuoiver 1)6.8(251/3 7)7.4(941/ 1044.5-86.0(09. 81.0 T)evitisopAHP 7.69(251/741 T)detaer 1.29(251/041 evitisop-LRDVhcihwtaAGnaeM p))skeew(detaerterewstneita 3.6DS(6.72 G)skeew91 G)skeew53-91tnemtaerttaA 6.38(251/721 G)skeew53>tnemtaerttaA 3.3(251/5 sesaC slortnoC selbairavsuounitnoC pHW-K )DS(naeM )DS(naeM P)ytira 1)722.1(73. 29723.1(57. 010.0 G))skeew(gnikoobtaA 2)26.6(7.3 2270.6(2.3 934.0 G))skeew(noitcellocdoolbtaA 2)81.6(9.4 2579.5(2.3 910.0 tluserdnatsetneewteblavretnI ())skeew 3)805.1(70. 29588.0(19. 48.0 oitarsddo=RO oitarsddofolavretniecnedifnoc%59=IC%59 eulavpderauqsihClezsneaH-letnaM=pH-M =AHPT mudillapamenoperT noitanitulggameah egalanoitatseg=AG yrotarobaLhcraeseResaesiDlaereneV=LRDV noitaiveddradnats=DS eulavptsetsillaW-laksurK=pHW-K abuse and social instability (7,10,11), all of likely to have positive VDRL serology. which are not uncommon in the Port Moresby Husbands with these occupations have the urban squatter settlements. opportunity for sexual encounters with multiple sexual partners. Members of the STIs including syphilis are said to be disciplined forces (police and the defence associated with the single marital status and force) for example, who usually go on night adolescence (10). While our study failed to patrols or travel to other provinces or places demonstrate this, patients with these on duty, where they are taken away from the sociodemographic characteristics are more family setting for long periods of time, are likely to be exposed to multiple sexual more likely to have the opportunity to engage partners and perhaps a larger sample may in other sexual relationships. Politicians and show this trend. We did not demonstrate any business men often go on domestic and significant difference in age, education, international trips and spend time in hotels marital status or socioeconomic status. This and places where they can seek extramarital is in agreement with a study in Zambia (3). female companions. As Cunningham et al. An interesting finding was that women whose put it, “among the fighting armies… and husbands had ‘risky’ occupations were more prostitutes syphilis continues rampant” (11). 22 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 3ELBAT LABOUR AND PERINATAL OUTCOMES *sesaC slortnoC selbairaV Op)IC%59(R H-M 251=N 051=N selbairavlacirogetaC F2shtribllitshser 1051/ 0d51/ U2enifedn 3.0 M2shtribllitsdetareca 1051/1 0d51/ U0enifedn 200.0 T2shtribllitslato 1051/2 0d51/ U5enifedn 4000.0 P2yreviledmreter 1051/9 7)51/ 259.7-1.1(9. 10.0 I0)051=N(RGU 3051/8 9)51/ 504.21-3.2(3. 0000.0< Csselbairavsuounitno esaC slortnoC pHW-K )DS(naeM )DS(naeM G))skeew(yreviledtaegalanoitatse 3)9.2(3.8 323.2(5.8 507.0 B))g(thgiewhtri 2)927(3.538 33494(3.960 40.0 A)etunim1taerocsragp 8)6.2(61. 959.0(70. 630.0 A)setunim5taerocsragp 9)3.2(91. 974.0(58. 200.0 laicepSotdettimdasyadforebmuN C)tinUera 0)53.1(12. 056.0(70. 42.0 * tnemtaertnevigerewsesacehtfoseibabehtfo)%2.48(251fotuo821 oitarsddo=RO oitarsddofolavretniecnedifnoc%59=IC%59 eulavpderauqsihClezsneaH-letnaM=pH-M noitcirtserhtworgeniretuartni=RGUI noitaiveddradnats=DS eulavptsetsillaW-laksurK=pHW-K This view is given credence from interviews suggestion of syphilis may be a past history with sex workers in our city, which showed of stillbirth. The impact of syphilis on perinatal that high profile clients, including politicians, outcome has been well demonstrated in were regular visitors (12). Drivers (and studies elsewhere, where significantly more sailors likewise) all over the world have been seroreactive than seronegative women are known to have a ‘partner’ at every port of likely to have had a previous abortion or call. A study done in 1972 clearly showed stillbirth (3,13,14). At the PMGH syphilis high prevalence rates of syphilis along the contributed to 10% of stillbirths (15). Women Highlands Highway, and several speakers in with syphilis abort more than 60% of their the main scientific meeting at the 2003 pregnancies (13). Symposium on HIV/AIDS in Mt Hagen spoke of a ‘thriving sex industry’ along the Highlands It is not surprising therefore that our Highway. These simply reflect the activities seroreactive patients showed significantly and impact of drivers along the Highlands higher rates of preterm deliveries, low Highway (9). birthweights, IUGR and perinatal mortality. These are well known sequelae of maternal We expected women with previous history syphilis infection, documented by numerous of stillbirths to show a higher rate of studies all over the world. Even as early as seroreactivity. Often in our practice, the first 1923, syphilis was found to be the aetiological 23 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 agent in about 40% of premature stillborn have left hospital before staff had the babies and 80% of macerated babies (16). opportunity to discuss the serious nature of In Goroka the average birthweight in the their condition with them. When patients are infected group of infants was 2.4 kg, whilst aware of the seriousness of their condition, the average birthweight in the infants of our they are often willing to wait for necessary cases was 2.8 kg; both are significantly lower treatment. Unfortunately it may also be the than the birthweights of babies born of case that the hospital staff find themselves seronegative mothers (8). too busy to educate or provide awareness for our patients, or we lack the understanding The mean gestational age at which most and therefore fail to see the seriousness of of our antenatal mothers first attended the the problem; or worse still, we just do not clinic was 24 weeks and the mean GA at care. blood collection was 25 weeks (Table 2). Since the mean interval between blood As far as the mothers themselves were collection and the receiving of results and concerned, 140 of the 152 cases (92%) were therefore commencing treatment is about 3 treated for syphilis according to our protocol. weeks, the mean GA of our cases 8% of the cases and 26% of the husbands of commencing syphilis treatment would be the cases did not receive or complete the about 28 weeks. Most of them were treated treatment. The reasons are the same as for between 19 and 35 weeks gestation; only 5% failure to treat the babies. were treated at <19 weeks gestation and 3% after 35 weeks. It was previously widely At the antenatal clinics of the National accepted that syphilis is a major factor in fetal Capital District the rate of seroreactive loss in the latter half of pregnancy and mothers was 4.4%. However, 1265 women appreciable damage to the fetus results only (12%) either had no records of the VDRL test later in the pregnancy when an adequate or were not screened during the antenatal immune response has developed in the fetus period (5). Furthermore, of those screened, (13). The current belief is that fetal infection 20% did not have their VDRL test result by does occur as early as 9-10 weeks (14). With the time they came to labour. The proportion this in view, many of the babies would have of unscreened mothers at other centres is had significant damage caused to them by worse, as shown by the Goroka study where the time treatment was instituted at the mean the proportion of antenatal mothers not gestation of 28 weeks. In any case even screened was as high as 70% (8). Even our when treatment is given early, and a full once-only testing is not foolproof. Studies course of the correct treatment is given to have shown seroreactivity in women at the mother, there have been some treatment delivery who were initially seronegative. In failures, as were reported from Fiji (10). addition the presence of early active syphilis These are babies who go on to develop in the latter half of pregnancy indicates congenital syphilis despite the mother recently acquired infection (3). completing treatment. It is therefore reassuring to note that the standard treatment Though our rates are lower it is not for common illnesses of children in Papua surprising that the pattern of the New Guinea recommends treatment of all complications associated with syphilis seen ‘normal-looking’ babies from VDRL-positive in our population (high rates of stillbirth, mothers with a single dose of benzathine preterm delivery and low birthweight in the penicillin. seroreactive women) is similar to that found in Africa (3) and in the pre-antibiotic era (14). Of the 152 babies born of the seroreactive Most of the seroreactive women in our study mothers, 128 (84%) of them were treated; (87%) were treated after 18 weeks gestation, 8% of them were perinatal deaths and the when significant fetal damage may have other 8% did not receive any form of occurred. This fact is reflected in the high treatment. The main reason for failure to treat rate of stillbirths, 8% (12/152), despite 92% was that they left hospital before there was treatment coverage, in the serologically time to locate the serology result and treat positive women, compared to a low stillbirth them. Many patients leave our hospital rate of 1.7% overall at the PMGH maternity before the discharge procedure is complete, unit (5). especially patients from poor socioeconomic backgrounds. They often do this to avoid The treatment of syphilis itself is simple. paying hospital fees. Some patients may The main problem in PNG is getting the 24 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 women screened early and treated so as to (human immunodeficiency virus) infection in avoid the destructive effects of the disease. parts of Africa and the United States (11). Even when patients book early in pregnancy, Both infections are transmitted via the same the time it takes to receive the result and the route and the genital lesions associated with necessary treatment is too long. syphilis facilitate the transmission of HIV. Therefore, our increasing rates of syphilis and Further analysis of our data showed the HIV infection should give us cause for following: concern, as HIV alters the clinical pattern of syphilis, with more cases presenting with the • 36 women (24%) were either not secondary disease, slower resolution of treated or did not complete treatment chancres and other cutaneous before delivery. manifestations, and a higher incidence and earlier onset of neurosyphilis (10,18). • Of the 38 babies with IUGR 30 (79%) received treatment after 20 weeks Conclusions gestation. The current antenatal screening and • Of the perinatal deaths, 58% of them treatment of cases is not adequate. New either did not receive any treatment or ways should be looked at to get women had not completed treatment by booked early at the antenatal clinics, delivery; all the perinatal death cases preferably before 16 weeks when syphilis who were treated received treatment infection can start damaging the fetus. after 20 weeks gestation. Screening for syphilis should be a priority at the booking visit at all antenatal clinics Despite good treatment coverage of our throughout Papua New Guinea so that seropositive attenders, there are still treatment can start as early as possible. The unsatisfactory outcomes because many screening test should ideally be simple and women are treated too late. Penicillin is still fast, so as to allow the whole screening and the drug of choice for all stages of the treatment process to be completed in a single disease, both in the pregnant and the non- antenatal visit. A recent study by Angue et pregnant. It has been the treatment of choice al. (19) clearly showed that the Abbot since 1944 when it was first used in Determine® test is simple and fast, has pregnancy (10); and it is still the treatment of acceptable sensitivity and specificity rates, choice today because of its efficiency and requires less equipment and expertise to lack of toxicity to both the mother and her perform, and, overall, costs less. Their baby. The benefits outweigh the small recommendation that the Determine test be associated risks. We rarely encounter made available in areas of the country where women who are allergic to penicillin. It must VDRL is unavailable, or where logistics do also be kept in mind that 90% of patients who not allow for the test results to be available say they are ‘allergic’ to penicillin do not have early enough to make a difference, is a valid the true IgE-mediated anaphylactic reaction one. and therefore can still be treated with penicillin (17). We have not encountered Babies born of mothers who do not have patients with Jarisch-Herxheimer reaction. antenatal care are at risk of congenital syphilis, and therefore these mothers should Since the introduction of penicillin (in the have an urgent screening test done at the 1940s), syphilis has declined in incidence in labour or postnatal wards so that they and developed countries. We do not have data their babies can be treated if positive. to demonstrate trends of syphilis since its discovery in PNG in 1960. On the other hand Our standard management protocol allows the incidence may well be getting worse like antenatal attenders only one chance to be many other maternal and child health tested. This policy is inadequate. Some indicators. The rate of positive syphilis women will become infected during serology in the antenatal clinics of the NCD pregnancy and others may seroconvert after in PNG has shown a slow rise from 3.0% in the initial test. There is a need to increase 1996 to 4.4% in 2002. public health awareness to promote early antenatal booking and routine syphilis testing The recent rise in the incidence of syphilis of all pregnant women in PNG. In addition a has been noted to be coexistent with HIV proposal to study syphilis infection during 25 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 pregnancy by repeat testing may reveal Treponematosis along the highlands highway. PNG interesting findings. Med J 1972;15:139-141. 10 Schramm M. Syphilis resurgent. Aust NZ J Obstet Gynaecol 1997;37:337-382. REFERENCES 11 Cunningham FG, MacDonald PC, Gant NF. Sexually transmitted diseases. In: Cunningham FG, 1 Isada NB, Grossman JH III. Perinatal infections. Gant NF, Leveno KJ, Gilstrap LC, Hauth JC, In: Gabbe SG, Niebyl JR, Simpson JL, eds. Wenstrom KD, eds. Williams Obstetrics, 21st edition. Obstetrics: Normal and Problem Pregnancies, 2nd New York: McGraw-Hill, 2001. edition. New York: Churchill Livingstone, 1991:1276. 12 Mgone CS, Passey ME, Anang J, Peter W, Lupiwa 2 Mabey DCW, Lloyd-Evans NE, Conteh S, Forsey T, Russell DM, Babona D, Alpers MP. Human T. Sexually transmitted diseases among randomly immunodeficiency virus and other sexually selected attendees at an antenatal clinic in The transmitted infections among female sex workers in Gambia. Br J Vener Dis 1984;60:331-336. two major cities in Papua New Guinea. Sex Transm 3 Ratnam AV, Din SN, Hira SK, Bhat GJ, Wacha Dis 2002;29:265-270. DS, Rukmini A, Mulenga RC. Syphilis in pregnant 13 Dippel AL. The relationship of congenital syphilis to women in Zambia. Br J Vener Dis 1982;58:355- abortion and miscarriages, and the mechanism of 358. intrauterine protection. Am J Obstet Gynecol 4 Duncan ME. Tropical infections in pregnancy. In: 1944;47:369. Bonnar J, ed. Recent Advances in Obstetrics and 14 Harter C, Benirschke K. Fetal syphilis in the first Gynaecology, 15th edition. London: Churchill trimester. Am J Obstet Gynecol 1976;124:705-711. Livingstone, 1987:33-63. 15 Amoa AB, Klufio CA, Moro M, Kariwiga G, Mola 5 Port Moresby General Hospital. Annual Report of GL. A case-control study of stillbirths at the Port the Division of Obstetrics and Gynaecology for 2002. Moresby General Hospital. PNG Med J Port Moresby General Hospital, Papua New Guinea, 1998;41:126-136. 2003. 16 Williams JW. Obstetrics: A Text-book for the Use 6 National Sex and Reproduction Research Team, of Students and Practitioners, 5th edition. New York: Jenkins C. National Study of Sexual and Appleton, 1923. Reproductive Knowledge and Behaviour in Papua 17 Gravett MG, Sampson JE. Other infectious New Guinea. Papua New Guinea Institute of Medical conditions in pregnancy. In: James DK, Ster PJ, Research Monograph No 10. Goroka: Papua New Weiner CP, Gonik B, eds. High Risk Pregnancy Guinea Institute of Medical Research, 1994. Management Options, 2nd edition. London: WB 7 Bloland P, Slutsker L, Steketee RW, Wirima JJ, Saunders, 1999:559-563. Heymann DL, Breman JG. Rates and risk factors 18 Cotran RS, Kumar V, Robbins SL. Robbins’ for mortality during the first two years of life in rural Pathologic Basis of Disease, 4th edition. Malawi. Am J Trop Med Hyg 1996;55(1 Suppl):82- Philadelphia: Saunders, 1989:368-371. 86. 19 Angue Y, Yauieb A, Mola G, Duke T, Amoa AB. 8 Frank D, Duke T. Congenital syphilis at Goroka Syphilis serology testing: a comparative study of Base Hospital: incidence, clinical features and risk Abbot DetermineR, Rapid Plasma Reagin (RPR) card factors for mortality. PNG Med J 2000;43:121-126. test and Venereal Disease Research Laboratory 9 Garner MF, Hornabrook RW, Backhouse JL. (VDRL) methods. PNG Med J 2005;48:168-173. 26 PNG Med J 2008 Mar-Jun;51(1-2):27-28 Variations of the middle thyroid vein in Papua New Guinean Melanesians DAMIEN JOSEPH HASOLA 1, GEORGE GENDE2 AND OSBORNE LIKO3 Department of General Surgery, Kavieng General Hospital, Papua New Guinea, Head and Neck Unit and Department of Anatomy, Port Moresby General Hospital, Papua New Guinea SUMMARY The middle thyroid vein has been noted to be frequently absent during thyroidectomies in Papua New Guinea (PNG). To verify this and other known inconsistencies in the neck a total of 103 bodies were dissected at the Port Moresby General Hospital in 2002. The middle thyroid vein was absent in 59% of the cases. It would be to the advantage of surgeons in PNG to be aware of this. Introduction excluded. No additional incision was made apart from the usual symphysis menti to The thyroid veins drain the thyroid lobes, symphysis pubis extensile cut for post- the oesophagus and the trachea. The middle mortem studies. For the purpose of the thyroid vein (MTV) is described as short and study, our part of the dissection was confined wide and drains into the internal jugular vein to the neck. Bilateral flaps were raised (IJV) (1). In rare instances it drains into the subplatysmally and the strap muscles were brachiocephalic vein (2). It exhibits a small retracted in the plane of the sternothyroid internal diameter (average of 2.0 mm) (3). muscle to expose the carotid sheath. The The vein is at risk during thyroidectomy and sternothyroid plane was carefully retracted tearing from the IJV can lead to a difficult and the lateral aspect of the thyroid gland haemorrhage. The middle thyroid vein is and the internal jugular vein inspected. The found to be frequently absent in Papua New loose areolar tissue was cleared in search Guineans. This study was undertaken to of the middle thyroid vein. Apart from the establish the prevalence of the middle thyroid MTV other possible variations were also vein in Papua New Guinean Melanesians. considered. Patients and methods Results This study was carried out at the Port The results in Table 1 show that the MTV Moresby General Hospital (PMGH) mortuary was absent in 59% (61/103) of the cases. In from February to August 2002. A males it was absent in 56% and in females consecutive series of 32 females and 71 in 66%. The Chi-squared value with Yates males were dissected. The autopsies were correction equals 0.45 at 1 degree of freedom performed for routine reasons by the resident and the two-tailed p value equals 0.5023. pathologist, such as mandatory coroner’s Thus there was no significant difference cases and others. Consent was obtained between the sexes with respect to MTV from the relatives. One newborn baby was absence. There were no unusual variations 1 Department of General Surgery, Kavieng General Hospital, PO Box 68, Kavieng, New Ireland Province 631, Papua New Guinea 2 Head and Neck Unit, Port Moresby General Hospital, Free Mail Bag, Boroko, National Capital District 111, Papua New Guinea 3 Department of Anatomy, Port Moresby General Hospital, Free Mail Bag, Boroko, National Capital District 111, Papua New Guinea 27 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 1ELBAT OCCURRENCE OF THE MIDDLE THYROID VEIN IN PAPUA NEW GUINEAN MELANESIANS BY SEX Snxe ievdioryhtelddiM Attnesb Plneser atoT M0ela 41317 F1elame 21123 latoT 621 4301 found in this series of cases. any significant variations in anatomy by sex and ethnic origin that may exist will need a Discussion much larger study to demonstrate. In the 103 Melanesians we studied the ACKNOWLEDGEMENTS middle thyroid vein was absent in 59%. There was no significant difference between males We are grateful to Dr P. Golpak, Dr A.K. and females though the vein was slightly Asthana and Professor S. Chung for valuable more often absent in females in this series. comments. Surgeons practising in PNG usually note that the MTV is absent and this study provides REFERENCES the answer for its absence. Although this finding does not absolve the surgeon from 1 McMinn RMH. Last’s Anatomy – Regional and carefully looking for the vein during Applied, 9th edition. Edinburgh: Churchill Livingstone, 1994. thyroidectomy, it does help to put the surgeon 2 Shima H, von Luedinghausen M, Ohno K, Michi at ease when the vein is not found. Our study K. Anatomy of microvascular anastomosis in the illustrates that clinically important information neck. Plast Reconstr Surg 1998;101:33-41. can be gleaned by careful clinical observation 3 Bliss RD, Gauger PG, Delbridge LW. Surgeon’s approach to the thyroid gland: surgical anatomy and and laboratory studies that do not require the importance of technique. World J Surg expensive equipment. This study is small; 2000;24:891-897. 28 PNG Med J 2008 Mar-Jun;51(1-2): 29-42 A focused ethnography about treatment-seeking behaviour and traditional medicine in the Nasioi area of Bougainville JOAN E. MACFARLANE1 Centre for International Health, Curtin University of Technology, Perth, Australia SUMMARY A focused ethnographic study examining knowledge, behaviour and attitudes related to treatment-seeking behaviour and traditional medicine was conducted in the Nasioi area of Bougainville. The study was conducted in two separate locations within the Nasioi area. Some similarities as well as some differences in knowledge and behaviour were exhibited in the two study locations which could be further investigated. People from both areas had a similar perception of common and serious illnesses. People from both areas also attributed illness to both biological factors and supernatural forces. Home management was the most common initial response to illness and the local village clinic or urban health centre were most frequently used as the first treatment resort. The most important factors in determining treatment-seeking responses to illness in both locations were cost and proximity to home or convenience. The study found that there may be considerable potential for an integrated health system in the Nasioi area and provided suggestions as to how this could be progressed. The study also provided information that could be useful in a wider context in informing the implementation of the national policy on traditional medicine in Papua New Guinea. Introduction The North Nasioi Council of Elders (COE) is the local government authority presiding A focused ethnographic study looking at over an area of approximately 240 square knowledge, behaviour and attitudes related kilometres and comprising 8 smaller Village to traditional medicine was conducted in the Council of Chiefs (VCC) areas. These are Nasioi area which lies around the coastal Bava Pirung, Kerei East, Kerei West, Doue, town of Arawa in Central Bougainville. The Tasipo, Konampai, Dangua and Apiatei. purpose of the study was to provide information that could be used to develop a The North Nasioi area includes Panguna, provisional explanatory model of treatment- which was the site of one of the world’s seeking behaviour and design a structured largest open-cut mines, operational in the questionnaire which would be administered 1970s and 1980s and associated with the to a larger cross-section of the Nasioi unrest that forced the closure of the copper population. It was anticipated that mine and led to the 10-year civil conflict information from both the broader population known as the Bougainville Crisis. A blockade sample and the more in-depth information imposed by the government of Papua New from those interviewed for the initial focused Guinea (PNG) during the crisis prevented ethnography would be used to formulate supplies from reaching many parts of the policy recommendations specific to the island and caused services to collapse. Much Nasioi area as well as to inform the infrastructure was destroyed during the implementation of the recently drafted fighting. Consequently people were forced National Policy on Traditional Medicine for to rely on their own local resources including Papua New Guinea (1). traditional medicine (2). This relatively recent 1 Centre for International Health, Curtin University of Technology, GPO Box U1987, Perth, WA 6845, Australia Current address: PO Box 1458, Kokopo, East New Britain Province 613, Papua New Guinea 29 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 resurgence or focus on traditional medicine chief is elected who sits on the VCC, which is predictive of its continued widespread use governs each local area. and makes the Nasioi area of particular interest for a study on traditional medicine. The vast majority of Nasioi people are Catholic. Other religious groups with some Population size and characteristics following are Seventh Day Adventist and United Church. A few other religious groups The National Census data indicate that a have very small followings. total of 5854 persons reside in the North Nasioi area (3). This includes 3078 males 70% of people aged 10 years and over in and 2776 females occupying 1316 houses. the Arawa LLG are literate in one language A breakdown by age and sex is not available (3). The most common languages spoken for the North Nasioi area on its own. are Nasioi, Melanesian Pidgin and English. However, the next largest government area is the Arawa Local Level Government (LLG) The last National Census found that area for which the age and sex population 11,864 people in the Arawa LLG (or 53% of breakdown is shown in Figure 1. The those aged 10 years or more) were structure of the North Nasioi area is expected economically active. The main employment to be similar to that of the Arawa LLG category was agriculture. People were population. involved in various types of agriculture for both income generation and personal Nasioi society is matrilineal. A Nasioi consumption. Cocoa was the main cash person inherits land, if they are female, and crop. Selling produce provided an income clan membership from their mother. The clan for some families and a few people ran their is an important organizational unit in Nasioi own business (3). The number of people society. The other fundamental social unit engaged in various types of income- is the family, which in Nasioi society includes generating activity is shown in Figure 2. the extended family. People typically live in small communities or villages based on clan Arawa is the main urban centre in the and extended family relationships. A village Nasioi area. Some telecommunications, may comprise 3 or 4 clan groups. A village retail outlets and health and education Figure 1. Population by age and sex, Arawa Local Level Government area, 2000. Source: National Statistical Office (3). 30 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Figure 2. Income-generating activities, Arawa Local Level Government area, 2000. Data were obtained from 5863 households in private dwellings headed by a PNG citizen. Source: National Statistical Office (3). services are available in Arawa including a ‘specialists’ are well known for being able to secondary school for students in year 9 and treat 1 or 2 conditions. Some can treat a above. The closest banking and postal range of conditions but have 1 or 2 facilities are located in Buka at a distance of specialties. The different types of traditional 200 kilometres or 4 hours and K50 by public practitioner include herbalists, bonesetters bus. and spiritualists. There are also practitioners who deal with sorcery and illness thought to Health services be caused by evil spirits. Arawa Town Health Centre is the main Methods health facility in the North Nasioi area and is supported by 6 aid posts and 2 smaller Before commencing the study approval clinics. For most people, aid posts are for it to be conducted was obtained from the located at a distance of between 1 and 3 Medical Research Advisory Committee of hours’ walk from home. Arawa Town Health PNG and Curtin University’s Human Centre is a 40-bed facility comprising Research Ethics Committee. Participation postnatal ward, intensive care ward, was voluntary and all respondents were paediatric ward, medical ward and general required to give written or, if not literate, oral ward. Other services available at Arawa consent before interviews were conducted. Town Health Centre include outpatients clinic, maternal and child health and As recommended in rapid assessment antenatal clinic, dental clinic, pathology procedures (4), data were collected using laboratory, delivery room, theatre, X-ray several different methods of assessment and machine (but no technician), dispensary and investigation to obtain information from a an ambulance. Other providers of western variety of sources, allowing for data health care are 1 general practitioner and 2 verification through triangulation. A focused health extension officers who all run private ethnographic approach (5), incorporating in- practices in Arawa. depth and semi-structured interviews, was adopted. Data were collected from key A plethora of traditional healers provide informants, people who had recently health care services to the Nasioi population. experienced an illness episode, community Traditional healers reside in virtually every members and practitioners. Nasioi village although their level of traditional medical knowledge varies. Some For specificity, the research focused on 2 practitioners can treat a range of conditions illness categories: febrile illnesses and skin and are referred to as generalists. Other conditions. Febrile illnesses and skin 31 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 conditions were selected because they groups of research assistants agreed on 8 accounted for more presentations to key informants, including males and females outpatients in Bougainville than any other and comprising traditional healers from conditions between 1997 and 1999. While different villages and providers of western other measures of morbidity and mortality health care in the two locations. All key are available, it was appropriate to base the informants had considerable health selection of conditions for this study on knowledge and were experienced in treating presentation to outpatients as this is the or managing illness. Key informants helped measure most likely to be comparable with to identify people who could provide an illness presentation to traditional healers. narrative, that is, people who had experienced a recent illness episode. The Two teams of 3 research assistants were research assistants identified other engaged to conduct interviews with community respondents. Some health care assistance from the North Nasioi COE. All practitioners were known to research research assistants had grown up and lived assistants and others were identified during in the area where they were to conduct key informant interviews. A sample of health interviews. They were fluent in Nasioi, had care practitioners was chosen for interview a minimum Grade 10 level of education and in each location as the total number identified were adults of some standing in their local included more than it was possible to communities. Each team included a male interview. The samples included the better- and a female. Prior to commencing data known traditional healers as well as some collection, a 3-day training workshop was lesser-known practitioners. All health conducted. workers and nurses providing western health care in each location were included in the The VCCs of Tasipo and Bava Pirung sample of health care practitioners. were purposively selected as the study sites. Two contrasting areas were selected that Almost without exception, the people would together reflect the diversity that exists initially identified as potential key informants within the Nasioi area. Tasipo is in the and community members were interviewed. mountains at a distance of about 7 Among people who were invited to participate kilometres from the town of Arawa. Since in the study only 2 declined, claiming that they there is no public transport in this area getting were too busy. Through the COE and local to Arawa requires more than an hour’s walk. chiefs some information about the study had Bava Pirung lies along the coastline reached residents in each area before the approximately 15 kilometres east of Arawa. actual data collection commenced. A few Access to Arawa may involve walking for people expressed concerns about the nature people from some inland villages but there of the research but once they were reassured are many villages beside the sealed road that no information would be sought about that runs along the coastline. A semi-regular the actual plant or other preparations used public bus service operates along this route in traditional medicine people were receptive and the one-way fare to Arawa is K2. and happy to contribute to the research. The fact that all the research assistants were Selection of respondents within each indigenous to the areas where they location was also purposive to conform to conducted interviews and familiar to the criteria stipulated for the various respondents probably contributed to the respondent categories (6). The research excellent response rate. assistants had an intimate knowledge of their respective locations and so were ideally A series of explicit and detailed question placed to identify suitable key informants and guides and recording sheets were designed other respondents. Before starting to for use with various groups of respondents conduct interviews, each group of research during the qualitative data collection. The assistants and the principal investigator various instruments were designed discussed and collectively agreed upon specifically for this study but were adapted potential key informants, community from instruments used in other focused respondents and health care practitioners ethnographies (6,7). Pre-testing of the whom they would invite to participate in the instruments in the study area was not study. possible due to administrative, logistical and time limitations. Procedures were adjusted Guided by the principal investigator both during the course of data collection as 32 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 deemed necessary; for example, community condition, to which the response was members were asked to list signs and invariably negative. Thus the exercise was symptoms of illnesses that had been not repeated separately for skin conditions. identified by key informants because using Each provider type was given a score based cards proved too cumbersome. Overall the on the number of votes received as a integrity of the original instruments was percentage of the total votes for the pair of maintained. All data collection instruments service providers. Providers were then were in English with a common ranked according to overall popularity using understanding and interpretation across the results obtained in this manner for febrile interviewers attained during the training illnesses. workshop. The number and type of interviews completed in each location are The majority of the data were qualitative shown in Table 1. and analyzed using the qualitative data analysis software package ‘MAXQDA’ The choice of practitioner instrument (registered trademark of VERBI Software). followed a forced-choice methodology (7) MAXQDA was used to facilitate the analysis and was designed to yield information about of the following interview types: Explanatory the respondents’ preferences for different Model, Illness Narratives, Health Care health care providers. In this methodology Practitioner and Illness Descriptions hypothetical scenarios were posed whereby (characteristics and relationships between respondents were asked, if they suffered illnesses) provided by community members. from a febrile illness and only two types of Some of the data collected in Tasipo and service providers were available, which one Bava Pirung were of a quantitative nature. they would approach for treatment. This The statistical package SPSS (registered question was repeated for every possible trademark of SPSS Inc.) Version 10.0 was combination of provider types available in the used to facilitate the analysis of Signs and local community (15 pairs of providers). Symptoms of Common Illnesses, Severity Respondents were then asked if their choices Rating of Common Illnesses and Choice of would be different if they suffered from a skin Practitioners. 1ELBAT NUMBER OF COMPLETED INTERVIEWS BY LOCATION Toweivretnifoepy Tgpisa nuriPavaB Fnelirbe Seik Fnlirbe ikS E7698ssenllirofledomyrotanalpx I9evitarranssenll 1313272 M0semanssenlliotsmotpmysdnasngisgnihcta 1019 9 S0gnitarytireve 1019 9 H3ytinummocnisecruosererachtlae 9 C8renoititcarapfoecioh 243 I0senicidemdlohesuohfoyrotnevn 18 foscitsiretcarahcdnanoitisopmocdlohesuoH r0ecnedise 101 I3sredivorpecivreserachtlaehrofediugweivretn 101 33 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Results one of several different agents. Information and knowledge systems Although many respondents attributed febrile illnesses to biological factors such as Key informants in Tasipo and Bava Pirung poor hygiene and poor nutrition it was evident had very similar ideas about which febrile that people in both study locations believe illnesses are the most common and serious. that febrile illnesses are caused by a mixture Most of the illnesses included in the short list of scientific and non-biological agents such of 10 common and serious febrile illnesses as breach of geographic or dietary taboos, (11 in Bava Pirung) were identified in both exposure to the blood of a relative, or spiritual areas. These illnesses included cough (kou), or divine intervention. Taxonomies of illness deep and persistent cough (eenu), showing how most respondents tended to respiratory conditions (domang o), diarrhoea group febrile illnesses are depicted in Figures (kubiri), urinary tract infection (pintuu), 3 and 4 for each respective location. enlarged spleen (maana), fever (malaria), malaria-like fever (pari) and headache (bore As for febrile illnesses there was a high bana). The relationships between and level of concurrence between people in descriptions of each of these illnesses Tasipo and Bava Pirung regarding the most provided by people from both locations were common and/or serious skin conditions. Key also very similar suggesting that people from informants in both locations identified leprosy different parts of the Nasioi area share a (oramu and erepu), sores from head lice common understanding of everyday and (kitei), rashes (kasikasi), ringworm (aaroa), more serious febrile illnesses. whitespot (kokosi), boils (moona) and cellulitis (sisisi) as being among the most Many of the common and serious febrile common types of skin conditions affecting illnesses appeared to be roughly equivalent local communities. Descriptions of each of to particular illnesses known in western these conditions were also very similar medicine. However, it cannot be assumed among respondents from both Tasipo and that any condition described by respondents Bava Pirung although respondents in Bava corresponds precisely to any western Pirung introduced ‘fungal infections’ as a medical condition. Western medical terms separate category, which included 3 skin beside bracketed Nasioi disease names conditions that were classified as ‘sores or denote the closest equivalent condition and skin irritations’ in Tasipo. Thus, the core of should not be interpreted as an exact common knowledge and understanding of translation of the local term. For example, it illnesses among respondents from Tasipo became apparent during the course of the and Bava Pirung extends to skin conditions. study that a diagnosis of malaria under the As was the case for febrile illnesses, Nasioi taxonomy of illness is not the same respondents believed that the causal as the western medical diagnosis of malaria. pathway for skin conditions could be both For Nasioi speakers ‘malaria’ is a syndrome biological and supernatural. Parasites, mites, of signs and symptoms, characterized worms, germs or bacteria and poor personal especially by fever, which can be caused by hygiene were identified by many respondents Figure 3. Taxonomy of febrile illnesses, Tasipo. 34 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Figure 4. Taxonomy of febrile illnesses, Bava Pirung. as causes of skin conditions. Allergies, Pirung treated skin conditions at home they irritant grasses, flies, pigs or other animals more commonly used traditional remedies. also featured in explanations for skin The most common form of home conditions. Traditional beliefs about the management for skin conditions was to bathe cause of skin conditions included ignoring in solutions made from different types of dietary taboos, spitting on relatives, sorcery leaves. Popo, guava, bantu and lime leaves and coming into contact with the blood of a were commonly used. Sometimes juice dead person. Sorcery (nenura), blood extracted from leaves would be mixed with contact and ignoring dietary taboos are lime and rubbed onto the affected area of thought to cause a number of different skin. conditions. Respondents grouped skin conditions according to similarities in signs Most of those who provided an illness and symptoms. Taxonomies of skin narrative reported using treatment services conditions for Tasipo and Bava Pirung are outside the home at some stage during the shown in Figures 5 and 6 respectively. illness episode. 84% of Tasipo and 60% of Bava Pirung respondents sought assistance Treatment-seeking response to illness from a health care provider. The first health care service used outside the home was most The general treatment-seeking response often the local village clinic, followed by to illness was the same for respondents from Arawa Health Centre for people in both Tasipo and Bava Pirung. The initial response locations. The number of respondents who to illness for at least three-quarters of approached various categories of service respondents who provided an illness provider in each location is shown in Table narrative (75% in Tasipo and 84% in Bava 2. The 2 health care services most likely to Pirung) was to try and manage the condition be used by respondents as the first treatment at home. The percentage of respondents resort are both providers of western medical who first used home management was even services. Worsening of the condition, higher (96%) among Bava Pirung residents development of more serious symptoms or who had recently suffered from a skin failure of home treatment to resolve the condition. Either traditional or western condition is what usually prompts medicine or a combination of the two is used respondents to seek treatment from in home management. In Tasipo, half of someone outside their immediate family. those who tried to manage their febrile illness or skin condition at home used some type of Respondents were generally satisfied with traditional medicine. Bava Pirung the service provided by the first service respondents who tried to manage febrile provider they approached outside the home. illnesses at home were more likely to use However, 15 respondents in Tasipo and 8 in western medicine purchased from a store or Bava Pirung found it necessary to seek left over from a previous illness episode than treatment from a second service provider. At a traditional remedy. When people in Bava the second treatment resort there was a more 35 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Figure 5. Taxonomy of skin conditions, Tasipo. Figure 6. Taxonomy of skin conditions, Bava Pirung. even spread across the available service either in concert or sequentially, is widely providers although traditional healers were practised. People were very satisfied with the most popular treatment option of second and had great confidence in the efficacy of resort in both study locations. People pluralistic use of medicine. approached a second service provider because they were not fully satisfied or had Explanatory model for treatment- not fully recovered with the treatment seeking responses received at the health care service provider of first resort. Several factors that are important in treatment-seeking decisions were identified. In Bava Pirung there appeared to be a Respondents’ understanding of illness preference for western health care services pathways includes both biological causative among people in the study population for factors and traditional beliefs. Relationships treating febrile illnesses and when they seek with family and neighbours are believed to treatment outside the home, but it was also have an impact on health as does leading a common for people to use their own spiritual life. Much illness is attributed to traditional treatment at the same time as sorcery. These beliefs about the aetiology western medicine. Sequential use of of disease mean that traditional medicine is traditional and western medicine was more an important part of the local health common in Tasipo. Thus, medical pluralism, paradigm. At the same time people perceive that is, the use of two or more health systems that biological factors contribute to many 36 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 2ELBAT HEALTH CARE SERVICE OF FIRST RESORT S)redivorpecivre %(sesnopserforebmuN Tgopisa nuriPavaB V)cinilcegalli 1)73(0 05(51 A)ertneChtlaeHawar 6)22( 72(8 P)PGetavir 6)22( 0(0 T)relaehlanoitidar 3)11( 02(6 P)OEHetavir 2)7( 3(1 latoT 2))001(7 001(03 renoititcarplareneg=PG reciffonoisnetxehtlaeh=OEH illnesses. With this understanding of disease clinics and non-availability of traditional it is logical that people would resort to both healer. In Tasipo, the most common reason western and traditional health services for given for not using the preferred practitioner treatment. in the illness episode being narrated was cost, nearly always in relation to the cost of In practice, respondents do not necessarily services provided by the private general seek treatment from their preferred provider. practitioner (GP). Despite the cost, the Service provider preferences in each private GP was the treatment option of first location, obtained using the forced choice of resort for 22% of Tasipo respondents. In practitioner instrument, are shown in Table Bava Pirung cost and distance were the 3. Respondents in both Tasipo and Bava factors that most often prevented Pirung indicated that their practitioner choices respondents from seeking treatment from would be the same for febrile illnesses and their preferred provider. skin conditions and so only one column of data is presented for each location. The most important factors in treatment- Comparison of these data with those in Table seeking responses to illness for respondents 2, actual treatment option of first resort, in both Tasipo and Bava Pirung were cost shows some anomalies. Where people say and proximity to home or convenience. they prefer to go does not necessarily Another slightly less important consideration correspond with where they actually go in the was confidence in the effectiveness of the first instance. However, since the data in treatment. The data presented in Tables 4 Table 3 include multiple preferences, strict and 5 relate to the perceived efficacy of comparison between the two tables is not various service providers. In Tasipo, 3 possible. providers of western medical services are seen as providing very effective treatment. Differences between preferences and In particular, treatment provided by the actual behaviour suggest that at the time of private general practitioner is thought to be illness events sometimes conspire to prevent highly effective. The treatment provided by the sick person from getting treatment from traditional practitioners is well regarded their preferred provider. Barriers to getting because it has an immediate effect rather treatment from the preferred provider than because it is better treatment. It would reported by respondents included cost, appear that there is little confidence in distance, non-attendance of staff at village treatment provided by village clinics. In view 37 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 3ELBAT SERVICE PROVIDER PREFERENCES,TASIPO AND BAVA PIRUNG Srredivorpecivre edivorpecivresrofseciohcfoegatnecreP Tgopisa nuriPavaB F3rebmemylima 906 T1relaehlanoitidar 892 V5cinilcegalli 756 A3ertneChtlaeHawar 317 P3PGetavir 244 P1OEHetavir 113 gnuriPavaBni43dnaopisaTni82sawstnednopserforebmuN renoititcarplareneg=PG reciffonoisnetxehtlaeh=OEH 4ELBAT PERCEIVED EFFICACY OF SERVICE PROVIDERS,TASIPO* niPG niOEH awarA egalliV lanoitidarT etavirp etavirp htlaeH cinilc **relaeh cinilc cinilc ertneC segatnavdA I1tceffe/feileretaidemm 13--3 3 S0tnemtaertroirepu 241311 7 segatnavdasiD I-tnemtaertevitceffen 321-0 T lufniapsitnemtaer ----1 0 T-tceffeekatotwolstnemtaer -1-- U sgurdynamootses 1--- - 82=stnednopserforebmunlatoT renoititcarplareneg=PG reciffonoisnetxehtlaeh=OEH * sesnopserforebmunetonedsrebmuN ** tnednopserrepsrelaehlanoitidart4otpurofsesnopsersedulcninmuloc'relaehlanoitidarT' 38 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 of the popularity of village clinics among that incorporates and recognizes both Tasipo respondents perhaps convenience traditional and western medicine. outweighs perceived efficacy in treatment- Practitioners reported that some degree of seeking decisions. Efficacy of treatment may collaboration is already occurring. There is also be of less importance to the Bava Pirung more intra- than inter-medical stream study population than cost and proximity in collaboration and there is only limited support treatment-seeking decisions. The data for more collaboration between traditional shown in Table 5 suggest that the private GP and western practitioners. Reservations and Arawa Health Centre are both thought about collaboration were expressed because to provide a high standard of treatment. It some practitioners prefer to provide would appear that there is less confidence treatment just to their own family members. in treatment provided by the village clinic and Often traditional practitioners are reluctant to yet this was most often the health care share information about their medicine. service of first resort (Table 2). In view of Despite the common view that current levels the usage patterns of the village clinic and of collaboration are adequate, western and the private GP by Bava Pirung respondents, traditional practitioners alike wholeheartedly perhaps convenience and cost outweigh supported the formal recognition of traditional efficacy in treatment-seeking decisions. The medicine as part of the government’s health services that are most used in both locations, care system. village clinics and Arawa Health Centre, both provide western health care services, are In both Tasipo and Bava Pirung, health cheap and are fairly easy to get to. care practitioners identified a range of specific benefits that would accrue from an Potential for integrated health care integrated health care system. Many practitioners could see that services would A sample of western and traditional health be more accessible, affordable, broader and care practitioners in both areas were asked more effective. Some practitioners in Tasipo what they thought about the development of also believed that recognition of traditional an integrated health care system, that is, one medicine could have benefits at the 5ELBAT PERCEIVED EFFICACY OF SERVICE PROVIDERS,BAVA PIRUNG* niPG niOEH awarA egalliV lanoitidarT etavirp etavirp htlaeH cinilc **relaeh ecitcarp ecitcrap ertneC segatnavdA I9tceffe/feileretaidemm 58410 S5tnemtaertroirepu 15518 9 segatnavdasiD I-tnemtaertevitceffen -1-2 T-lufniapsitnemtaer -1-- T-tceffeekatotwolstnemtaer -1-1 43=stnednopserforebmunlatoT renoititcarplareneg=PG reciffonoisnetxehtlaeh=OEH * sesnopserforebmunetonedsrebmuN ** tnednopserrepsrelaehlanoitidart2otpurofsesnopsersedulcninmuloc'relaehlanoitidarT' 39 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 psychological and spiritual levels. In Bava Discussion Pirung a number of practitioners believed that recognition of traditional medicine would Respondents in the two locations shared result in financial benefits for service a core of knowledge and had similar providers. Two Bava Pirung practitioners perceptions about febrile illnesses and skin thought that if traditional medicine was conditions. The usual illness response formally recognized as part of the pathway was also common to both locations. government’s health care system it would At least three-quarters of respondents tried become better organized, which would be to manage their illness at home before advantageous in terms of processing seeking treatment from someone outside the supplies of traditional medicine and family. Village clinics and Arawa Health preserving traditional medical knowledge. Centre were most often the treatment option of first resort once assistance was sought Although support for recognition and from outside the family. Low cost and easy integration of traditional medicine was access were the two most important factors universal and many real benefits were in treatment choices. Efficacy of treatment volunteered, practitioners were cognizant of was also a consideration. Some some obstacles that would need to be discrepancies between treatment overcome if an integrated system is to be preferences, treatment-seeking behaviour achieved. In Tasipo these related to the and the most important determinants of tendency to blame and seek retribution when treatment choices were evident. a patient fails to recover, the imprecision and lack of scientific verification surrounding There is an anomaly between the low use traditional medicine and the anticipated of traditional healers and the importance of reluctance of some practitioners to reveal cost and proximity in treatment-seeking information about their treatments. In Bava decisions. Many traditional healers provide Pirung lack of understanding and trust, services in the villages where respondents disparity in education levels of different types live and do not charge high fees. However, of practitioners, and the sheer number of traditional healers were the treatment option traditional practitioners were identified as of first resort for a minority of respondents in potential barriers to integration. both Tasipo and Bava Pirung. Rank ordering of preferences for the available service Respondents could envisage an providers or treatment options suggested integrated health care system where traditional healers are the preferred provider traditional and western health care services in Tasipo but the least preferred in Bava are co-located. Most practitioners thought Pirung (Table 3). Further investigation may that separate rooms would be required for be warranted to determine why traditional practitioners to be able to work in comfort. practitioners are not popular or very Several respondents also suggested frequently utilized in Bava Pirung and why establishing a traditional medicine section they are popular but infrequently used in within Arawa Health Centre. In Tasipo it was Tasipo despite their services being low in cost suggested that traditional practitioners were and readily accessible to villagers. ideally placed to assume a primary health care role at the village level. Differences between the type of treatment provided by the preferred provider and the Practitioners in both areas were able to treatment option of first resort in Tasipo suggest strategies to progress integration. suggest that, in actual fact, many These included: developing a better respondents do not have a strong preference understanding of the services that each for either traditional or western medicine. stream of medicine can provide; creating a Practical circumstances surrounding an registry of practitioners and conditions that episode of illness may be the factors that each can treat; verifying the curative determine what type of treatment is sought properties of particular traditional treatments because people consider both traditional and and establishing doses and side-effects western medicine to be effective (Tables 4 through scientific testing; registering and and 5). Although respondents perceive the issuing licenses to traditional practitioners. scientific rigour and precision that surround It was suggested that meetings between western medicine as desirable, traditional leaders of each stream of medicine be held medicine is more affordable and readily to work out how integration could proceed. available to the majority of the population who 40 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 live in rural areas. knowledge about febrile illnesses and skin conditions in the two VCC areas although the There appears to be considerable support way people classified or grouped illnesses among health care practitioners in both differed somewhat. The belief that illness Tasipo and Bava Pirung for the concept of a may be caused by both biological factors and health care system where traditional and spiritual or supernatural forces was common western medicine are integrated. in both areas. Home management was the Respondents thought an integrated health most common initial response to illness in system would mean better services for the both areas. Whereas in Tasipo respondents community and preservation of important most often used some form of traditional cultural knowledge on traditional medicine. medicine to treat illness at home, in Bava It was also perceived that traditional healers Pirung western medicine was more likely to would benefit financially under an integrated be used for febrile illnesses. The majority of system. respondents in both areas sought the assistance of a health care practitioner during In order for an integrated health care the course of an illness and in both areas system to be established, open, trusting and the first treatment resort was most often the equal relationships would need to be village clinic followed by Arawa Town Health developed between traditional practitioners Centre. Traditional healers were the most and providers of western health care frequently used second treatment resort for services. Mutual understanding and respect people from both areas. Cost and should be the basis of these relationships and convenience appeared to be the two most would pave the way for integration. Providing important factors in determining treatment basic training courses in each stream of choices. Perceived efficacy of the treatment medicine could help to build this type of appeared to be slightly less important. The understanding and respect. extent to which the two study locations reflect the broader Nasioi population, reasons for Home management is a common initial the apparent anomalies regarding response to illness the world over and the hypothetical preferences and actual Nasioi area is no exception. Some gains in behaviour that were detected in this focused health status in the Nasioi area might be ethnography and further refinement of the achieved by ensuring that basic first aid provisional explanatory model are all issues knowledge, either traditional or western, and that could be elucidated through further the ability to recognize signs and symptoms research. of common ailments and serious illness become more widespread. Health care The research has also provided practitioners said that people often wait too information that could be used by health long and allow their condition to deteriorate authorities who wish to pursue an integrated before seeking treatment. Encouraging early form of health care. Although the sample of presentation to a health care service provider practitioners interviewed expressed limited may also help to improve health status. support for more collaboration than is Traditional healers may be able to play a role currently occurring, there was unanimous in making home management more effective, support for formal recognition of traditional providing health promotion and disease medicine. Practitioners believed an prevention information and encouraging early integrated health service would offer better presentation through referral to an access, affordability and effectiveness than appropriate health care provider. the current formal health system as well as a holistic alternative. Practitioners’ ideas about Conclusion how integration could be progressed contributed to the development of The focused ethnography that has been recommendations that could be useful to reported in this paper produced considerable authorities in the Nasioi area should they data on knowledge of illness, treatment- choose to recognize and incorporate seeking behaviour and attitudes toward an traditional medicine into the formal health integrated health system in two VCCs within system (J.E. Macfarlane and M.P. Alpers, the North Nasioi area. Although some Health care preferences among the Nasioi homogeneity was evident between the two people of Bougainville: steps toward study sites, some differences were also incorporating traditional medicine into the noted. There was a common core of health care system, unpublished 41 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 manuscript). participants, both from the health services and the community, who so willingly The information could also be used to contributed to the study. I am grateful to the inform the implementation of the recently North Nasioi Council of Elders and the Village drafted National Policy on Traditional Council of Chiefs of Tasipo and Bava Pirung Medicine for PNG (1). The National Policy for their approval to conduct the study as well needs to be locally relevant and a better as their ongoing advice and practical support understanding of the knowledge, behaviour which made the study possible. Finally, I am and attitudes of people at the community grateful for the warm reception and hospitality level is important in this context. Of course, I received in both Tasipo and Bava Pirung. the Nasioi are just one amongst PNG’s multitude of cultural groups and REFERENCES implementation of the Policy in other areas would require information about the relevant 1 Ministry of Health. National Policy on Traditional cultural groups. The research method used Medicine for Papua New Guinea. Port Moresby: in this study could be incorporated into Department of Health, 2004:1-14. 2 Boge V. Conflict potential and violent conflicts in studies designed to collect the required the South Pacific – options for a civil peace service. information in other parts of PNG where Working Paper No 1. Research Unit of Wars, conditions are conducive to the development Armament and Development, University of of an integrated form of health care. Hamburg, 2001:1-94. 3 National Statistical Office. Census Data, Papua New Guinea. Port Moresby: National Statistical ACKNOWLEDGEMENTS Office, 2000. 4 Manderson L, Aaby P. Can rapid anthropological The material presented in this paper was procedures be applied to tropical diseases? Health Policy Plan 1992;7:46-55. collected as part of a broader study 5 Pelto PJ, Pelto GH. Studying knowledge, culture, completed in partial fulfilment of the and behaviour in applied medical anthropology. Med requirements for a Professional Doctorate of Anthropol Q 1997;11:147-163. International Health through the Centre for 6 World Health Organization. Focused International Health at Curtin University of Ethnographic Study of Acute Respiratory Infections. Technology. I thank both my supervisors for Geneva: World Health Organization,1993:1-203. 7 Scrimshaw S, Hurtado E. Rapid Assessment their wise advice and encouragement over Procedures for Nutrition and Primary Health Care. the duration of my studies. I also thank my Los Angeles: Regents of the University of California, dedicated research assistants and all 1987. 42 PNG Med J 2008 Mar-Jun;51(1-2):43-46 Primary repair of colonic injuries at the Kundiawa and Madang General Hospitals, Papua New Guinea JERZY KUZMA1,2 AND JAN JAWORSKI1 Kundiawa General Hospital, Papua New Guinea, Modilon General Hospital and Divine Word University, Madang, Papua New Guinea SUMMARY In this study, we evaluated the safety of primary repair of colon injury in a low-volume tropical hospital setting. Between 1998 and 2005, 18 consecutive patients who underwent emergency operation for civilian traumatic colon injury were studied. The main outcome measures were the mortality and morbidity rates and the total length of the hospital stay. The mean hospital stay for one-stage repair was 12 days versus 29 days for the two-stage procedure, which was a significant difference (p = 0.009). There was no death reported from this study. There was no significant difference in postoperative septic complications between the one-stage and two-stage procedures. One-stage repair of colonic injury is a safe and cost-effective option for selected patients in the tropical hospital setting. Introduction Patients and methods Over the last two decades there has been Between January 1998 and April 2005 in a shift in the management of colon injuries Kundiawa General Hospital and Modilon toward one-stage repair. Before that, most General Hospital an open case series study colonic injuries were managed by a two-stage was conducted on 18 consecutive patients procedure including colostomy because of who underwent operation for colonic injury. the fear of anastomotic and wound breakdowns. The criteria for inclusion were full- thickness colonic injury and informed However, at present in specialized centres consent obtained from the participant. All the results of one-stage repair are so superior patients qualified for laparotomy had that the two-stage procedure has become penetrating abdominal injury and were obsolete in the management of these injuries diagnosed with local or diffuse peritonitis. (1-7). Although there are a few reports from There was no exclusion from the study high-volume tropical hospitals (1,5) where except lack of consent. The study was colonic injuries were safely managed with approved by the local ethics committee. one-stage repair, there is a paucity of data (2,8) on one-stage primary repair in a low- To classify patients in terms of general volume tropical hospital. physical fitness we employed the American Society of Anesthesiologists classification The primary purpose of this study was to (ASA). The patients in a good general evaluate the safety of primary repair of colon condition (ASA grade 1 and 2) were allocated injury in a low-volume tropical hospital to one-stage repair whereas those critically setting. Additionally, we aimed to determine ill (ASA grade 4 and 5) were managed by whether one-stage primary repair in selected the two-stage procedure. The selection for patients is safe in the presence of early one-stage repair or colostomy for the diffuse peritonitis. patients in the third ASA grade was at the 1 Kundiawa General Hospital, PO Box 346, Kundiawa, Simbu Province 461, Papua New Guinea 2 Modilon General Hospital and Divine Word University, PO Box 483, Madang, Madang Province 511, Papua New Guinea 43 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 surgeon’s discretion. The longer the time administered to all patients at the time of from the injury to the operation and the more induction of general anaesthesia and severe the associated injury, the more likely continued for at least 48 hours. the patient was to be selected for the two- Postoperative management included stage procedure. monitoring of the patient’s vital signs, continuation of intravenous fluid until the The severity of peritonitis was classified patient resumed oral diet, early according to the modified Hughes’ postoperative oral feeding (12 hours after classification (9, as quoted in 10): grade II, operation) and early mobilization. localized peritonitis or paracolic abscess; and grade III, diffuse faecal or purulent Comparison between means was peritonitis. performed using the 2-sided t test for independent samples, while comparison of We compared the one-stage versus the frequencies was performed by χ2 using two-stage group with regard to the main Yates’ correction. We considered a outcome measures: morbidity rate and the difference with p <0.05 to be significant. total length of the hospital stay. Additionally, we compared the main outcome measures Results between the subgroups with local peritonitis and diffuse peritonitis. A total of 18 patients (13 men and 5 women) with penetrating colon injury were All data, including mechanism of injury, assessed. The median age was 25 years time from the injury to the operation, ASA (range 6-50). The distribution of the injured grade, intraoperative findings, severity of segment of colon was as follows: 7 in the peritonitis, early postoperative septic transverse colon, 5 in the descending colon, complications, mortality rate and overall 5 in the sigmoid colon, and 1 in the caecum. hospital stay, were recorded prospectively. Overall hospital stay was expressed in days Regarding associated injuries, 6 patients in hospital after operation including the had small bowel injury, 3 mesentery injury, 2 hospitalization period of patients readmitted spleen injury, 2 vagina injury and 1 each of for second-stage repair or because of the following injuries: retroperitoneal complications related to colon injury. haematoma, diaphragm, stomach, liver, lung, pancreas and urinary bladder. The patients were resuscitated and optimized for surgery according to Advanced The relationship between the surgical Trauma Life Support guidelines with procedure and ASA grade is presented in crystalloids and blood transfusion if Table 1. The average duration of time necessary. Antibiotic prophylaxis was between the injury and operation for the 1ELBAT AMERICAN SOCIETY OF ANESTHESIOLOGISTS )ASA( GRADE AND SURGICAL PROCEDURE FOR PENETRATING COLON INJURY Tenoitarepofoepy dargASA IIIIIVI IV P23600riaperyramir R0ymotcelocimehthgi 0100 C0riaperdnaymotsolo 0220 H0erudecorps'nnamtra 0101 latoT 23121 0 44 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 whole group was 17.4 hours; for the one- after primary repair as compared to stage repair it was 12.9 hours while for the colostomy. Curran and Borzotta’s review two-stage procedure it was 81.6 hours. of 2964 cases of primary repair of colonic injury (7) revealed that the leak rate after The mean overall hospital stay for the one- simple suture of perforation is very low stage procedure was significantly shorter (1.4%). In addition, the advantage of one- than for the two-stage procedure: 11.9 stage repair is not only reduction in morbidity (±5.64) versus 28.8 (±18.63) days (p = but also decrease in the cost associated with 0.009). colostomy (11,12). The difference between the one-stage and Another question is whether one-stage the two-stage procedure with respect to early repair can be carried out safely in the postoperative septic complications was not presence of peritonitis. Our findings suggest significant (Table 2). There was no death reported in the series and there was one that diffuse peritonitis is not a significant risk relaparotomy due to intraperitoneal abscess factor despite a tendency for an increased following colostomy. rate of postoperative septic complications and longer hospital stay. The classical Our findings also showed a tendency for approach in the presence of diffuse increased mean overall hospital stay and peritonitis is a two-stage procedure. postoperative complication rate in the diffuse Recently, many reports recommend primary versus the localized peritonitis subgroup but the difference was not considered significant. anastomosis as a safe management in the presence of diffuse purulent peritonitis Discussion (3,11,13). At present, 60% to 84% of patients with In our study the mean overall hospital stay colonic injuries are treated by primary repair was significantly longer for the two-stage (1-3). than for the one-stage procedure. Similarly, Berne at al. (12) have shown that the two- Our main finding is that there was no stage colostomy group had increased significant difference in the rate of early overall hospital stay compared to patients postoperative septic complication between in the one-stage primary repair group. the one-stage and two-stage procedures for perforating colon injuries. These results are Although in this series the choice for the in agreement with others who reported that two-stage procedure was guided by the the one-stage compared to the two-stage severity of the patient’s general condition procedure has not increased postoperative and the presence of such risk factors as the septic complications and mortality (3-6). time from the injury to the operation longer Furthermore, several prospective than 12 hours, associated injury, severe randomized trials found either lower (4) or anaemia or presence of septic shock, there similar (3-6,11) rates of septic complications is a lot of discussion whether primary repair 2ELBAT EARLY POSTOPERATIVE INFECTIOUS COMPLICATIONS AFTER ONE-STAGE AND TWO-STAGE PROCEDURES FOR PENETRATING COLON INJURY )stneitap21(erudecorpegats-enO )stneitap6(erudecorpegats-owT W))stneitap2(noitcefnidnuo stneitap3(noitcefnidnuoW )tneitap1(ssecsbalaenotirepartnI )tneitap1(ssecsbalaenotirepartnI A))tneitap1(sitimydidipeetuc tneitap1(ainomuenP latoT )%33(4: latoT )%38(5: 45 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 can be safely performed in the presence of REFERENCES these risk factors. A prospective multicentre study (3) has shown that severe faecal 1 Bowley DM, Boffard KD, Gossen J, Bebington contamination, transfusion of more than 4 BD, Plani F. Evolving concept in the management of colonic injury. Injury 2001;32:435-439. units of blood within the first 24 hours and 2 Baako BN. Colostomy: its place in the management single-agent antibiotic prophylaxis are of colorectal injuries in civilian practice. West Afr J independent risk factors for abdominal Med 1998;17:109-112. complications. Though mortality and rate of 3 Demetriades D, Murray JA, Chan L, Ordoñez C, septic complications are generally higher in Bowley D, Nagy KK, Cornwell EE 3rd, Velmahos GC, Muñoz N, Hatzitheofilou C, Schwab CW, patients with a greater number of associated Rodriguez A, Cornejo C, Davis KA, Namias N, organ injuries, it was demonstrated in Wisner DH, Ivatury RR, Moore EE, Acosta JA, randomized trials (4,10) that there was no Maull KI, Thomason MH, Spain DA, Committee difference in the postoperative complication on Multicenter Clinical Trials, American Association for the Surgery of Trauma. rate between the one-stage and the two- Penetrating colon injuries requiring resection: stage groups. Furthermore, some have diversion or primary anastomosis? An AAST recommended colostomy for patients with prospective multicenter study. J Trauma colon injuries presenting in shock or with 2001;50:765-775. 4 Kamwendo NY, Modiba MCM, Matlala NS, Becker associated multiple injuries (14). More PJ. Randomized clinical trial to determine if delay recently, Kamwendo et al. (4) in a from time of penetrating colonic injury precludes randomized trial have demonstrated that primary repair. Br J Surg 2002;89:993-998. primary repair of civilian penetrating colonic 5 Thomson SR, Baker A, Baker LW. Prospective audit of multiple penetrating injuries to the colon: injury is safe irrespective of the presence of further support for primary closure. J R Coll Surg high-risk factors such as shock, massive Edinb 1996;41:20-24. blood transfusion, delay in presentation, 6 Jacobson LE, Gomez GA, Broadie TA. Primary intraabdominal faecal soiling and associated repair of 58 consecutive penetrating injuries of the injuries. colon: should colostomy be abandoned? Am Surg 1997;63:170-177. 7 Curran TJ, Borzotta AP. Complications of primary It is acknowledged that the sample size repair of colon injury: literature review of 2,964 cases. of our study is small, and the results need to Am J Surg 1999;177:42-47. be validated in a larger multicentre study with 8 Naraynsingh V, Ariyanayagam D, Pooran S. Primary repair of colon injuries in a developing possible random allocation to the treatment country. Br J Surg 1991;78:319-320. groups. 9 Hughes ESR, Cuthbertson AM, Carden AB. The surgical management of acute diverticulitis. Med J In summary, our findings suggest that in Aust 1963;1:780-782. 10 Gooszen AW, Tollenaar RAEM, Geelkerken RH, a low-volume tropical hospital setting, one- Smeets HJ, Bemelman WA, Van Schaardenburgh stage repair of colonic injury is a safe and P, Gooszen HG. Prospective study of primary cost-effective option. However, having in anastomosis following sigmoid resection for mind the current trend towards one-stage suspected acute complicated diverticular disease. Br J Surg 2001;88:693-697. repair of colonic injury, and on the other hand 11 Gonzalez RP, Falimirski ME, Holevar MR. Further insufficient evidence from powered evaluation of colostomy in penetrating colon injury. randomized trials on the association Am Surg 2000;66:342-346. 12 Berne JD, Velmahos GC, Chan LS, Ansensio JA, between the risk factors and results in both Demetriades D. The high morbidity of colostomy treatment groups, we suggest caution and closure after trauma: further support for primary opt rather for two-stage repair in critically ill repair of colon injuries. Surgery 1998;123:157-164. patients with late diffuse peritonitis. 13 Biondo S, Jaurrieta E, Marti Ragué J, Ramos E, Deiros M, Moreno P, Farran L. Role of resection and primary anastomosis of the left colon in the ACKNOWLEDGEMENT presence of peritonitis. Br J Surg 2000;87:1580- 1584. We are indebted to Professor Francis 14 Chappius CW, Frey DJ, Dietzen CD, Panetta TP, Buechter KJ, Cohn I Jr. Management of Hombhanje for his advice in writing and penetrating colon injuries. A prospective randomized revision of the paper. trial. Ann Surg 1991;213:492-497. 46 PNG Med J 2008 Mar-Jun;51(1-2): 47-55 Barriers to the delivery of the hepatitis B birth dose: a study of five Papua New Guinean hospitals in 2007 S.G. DOWNING 1,2, W. LAGANI 3, R. GUY 1,4 AND M. HELLARD 1,4 Centre for Epidemiology and Population Health Research, Burnet Institute, Melbourne, Australia, National Centre for Epidemiology and Population Health, Australian National University, Canberra, National Department of Health, Port Moresby, Papua New Guinea and Department of Epidemiology and Preventative Medicine, Monash University, Melbourne, Australia SUMMARY Hepatitis B is highly endemic in Papua New Guinea (PNG). Vaccination at birth is a key mother-to-child transmission prevention strategy. Despite recommendations for newborns to be vaccinated within 24 hours of delivery, a 2005 survey showed 23% coverage among children born in health facilities. Our study examined hepatitis B birth- dose coverage and knowledge, practices and barriers to vaccine delivery in five major PNG hospitals. Data on births and vaccines administered were sourced from the National Department of Health (NDoH) and directly from the five hospitals. A maternity unit audit and staff survey were undertaken. Across the five hospitals, the hospital-level data of hepatitis B birth-dose coverage was 79% (range: 40-96%) compared to 19% from national data (range: 0-106%). Despite hospitals having adequate vaccine supply, access to appropriately stored vaccine in maternity units was compromised with only one unit having a vaccine-specific temperature-monitored refrigerator. In interviews of 25 staff, incorrect reasons given for delaying vaccination were prematurity (60%), low birthweight (48%) and difficult birth (36%). This study found encouraging birth-dose coverage rates in five major hospitals but 20% of babies still missed receiving the recommended vaccine. The NDoH Immunization Unit will use the results of this study to inform strategies to improve hepatitis B birth-dose coverage in hospitals. Introduction approximately 500,000 to 750,000 deaths worldwide annually (1). The development Hepatitis B virus (HBV) is an important of chronic HBV infection is inversely related cause of morbidity and mortality worldwide. to age: it occurs in approximately 90% of The World Health Organization estimates persons infected perinatally, in 30% infected that two billion people have serological in early childhood and in 6% infected after evidence of past or present HBV infection, five years of age (4-7). There is a highly with 360 million chronically infected (1). An effective vaccine for HBV infection (1) and estimated 15-25% of people with chronic the provision of infant hepatitis B HBV infections will die prematurely from immunization is a proven public health cirrhosis or hepatocellular carcinoma (2,3). strategy in preventing HBV transmission at These two conditions account for population level (8,9). 1 Centre for Epidemiology and Population Health Research, Macfarlane Burnet Institute for Medical Research and Public Health, GPO Box 2284, Melbourne, Victoria 3001, Australia 2 Master of Applied Epidemiology Program, National Centre for Epidemiology and Population Health, Australian National University, Canberra, ACT 0200, Australia 3 Health Improvement Branch, National Department of Health, PO Box 807, Waigani, National Capital District 131, Papua New Guinea 4 Department of Epidemiology and Preventative Medicine, Monash University, Melbourne, Victoria 3800, Australia 47 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 In highly endemic countries (defined as grouped into four geographical regions: >8% of the population having chronic HBV Highlands, Southern, Islands and Momase. carriage (10)) HBV is most commonly spread The selection of the five hospitals included from mother to child at birth (perinatal in the study was based on representation of transmission) or from person to person in the four PNG regions and the national capital early childhood (10). Infants born to mothers hospital, the provision of specialist obstetric with hepatitis B surface antigen have a 5% services and high numbers of reported births. to 20% risk of perinatal transmission which Hospitals selected were Port Moresby increases to 70% to 90% if the mother is General (National Capital District), Goroka hepatitis B e antigen positive (11). There are (Highlands Region), Madang (Momase multiple hepatitis B infant vaccination Region), Nonga (Islands Region) and Alotau schedules recommended throughout the (Southern Region). world, ranging from two to four doses. In highly endemic countries the World Health There were three main sources of Organization recommends that the initial information: (i) a comparison of nationally hepatitis B vaccine dose be given within 24 reported and hospital-level immunization hours of birth (12). This initial dose acts as a data; (ii) a maternity unit audit and post-exposure prophylaxis following possible observation of vaccination practices; and (iii) exposure to HBV during the birthing process. staff interviews on knowledge and attitudes, Birth-dose administration will prevent practices and barriers. perinatal transmission in 80% to 95% of cases; however, the efficacy declines with Comparison of nationally reported and increasing intervals between birth and hospital-level immunization data administration of the vaccine (12). Hepatitis B immunization coverage rates HBV seroprevalance studies have shown were calculated with data obtained from two that the disease is highly endemic in Papua sources: National Department of Health New Guinea (PNG) with carrier rates of 12% (NDoH) and directly from the five hospitals. to 37% (13-15). The PNG national immunization guidelines recommend that National level hepatitis B vaccine be administered at birth and at one and three months of age (16) but The NDoH collates data provided by all despite this the 2005 PNG vaccination cluster provincial health offices (PHOs) including the coverage survey showed that the coverage number of hospital births and vaccines rate for hepatitis B vaccine at birth was low. administered at the facility. The timing of the Only 16% of all children surveyed were administration of the birth-dose vaccine in reported to have received the hepatitis B birth relation to time of delivery is not captured dose within 24 hours of birth, irrespective of through this system. The number of births place of delivery, and of the children born in and hepatitis B birth-dose vaccinations health facilities the coverage was only slightly provided during 2006 at each of the selected higher (23%) (17). These data suggest that hospitals was obtained from this database. within PNG there are many missed opportunities for a hepatitis B birth dose Hospital level within health facilities. The five hospitals in this study maintain a We report the findings of a study paper-based birth register of all births that examining HBV coverage, the attitudes and take place in that facility. This register practices of staff, and the barriers to delivery records standard treatments provided to the of a hepatitis B birth dose in five major PNG baby at birth, including administration of hospitals. vaccines, although the recording of this information is not routine for all maternity Methods units. A separate immunization recording book is also maintained by the maternity units The study was undertaken during October to document all vaccines administered to and November 2007 in PNG. babies in the labour ward. The number of births and the number of hepatitis B birth Site selection doses administered over a minimum two- month period were obtained from the five PNG has 20 provinces which can be hospitals using the birth register and the 48 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 immunization recording books. The two most Results recent complete months of data were collected with the months varying between Comparison of nationally reported and hospitals according to the timing of the visit. hospital-level immunization data If time permitted further months of hospital data were collected. The 2006 NDoH data for the five selected hospitals recorded that 17,395 children were For both national and hospital-level data born and 3265 received a hepatitis B birth- the proportion of newborns receiving a dose vaccination, giving an overall coverage hepatitis B birth dose (coverage rate) in each rate of 19%. This rate varied widely across facility was calculated by dividing the number hospitals; three facilities reported coverage of documented vaccines administered by the rates of less than 3% while the other two number of births reported for the same time hospitals had coverage rates greater than period. 95%. Maternity unit audit Based on data sourced directly from the five hospitals, the overall proportion of An audit was undertaken of the maternity children receiving a hepatitis B birth dose unit at each hospital using a pre-tested was 79% with less variation by hospital. Four standardized structured data collection tool of the five selected hospitals showed covering the following main areas: vaccine coverage levels of greater than 80% for the and cold chain management, documentation, birth dose in the months examined (Table 1, policies/guidelines and clinical practices. The Figure 1). most senior nursing staff member on duty at the time of the audit provided the information and observations of clinical practice were Maternity unit audit also documented during this process. Vaccine and cold-chain management Staff interviews The vaccine management systems and Confidential interviews with individual staff cold-chain practices varied considerably were conducted using a pre-tested across hospitals. Three hospitals maintained standardized semi-structured interview tool. their vaccine stock in temperature-monitored, All willing maternity unit health care workers vaccine-only, ice-lined refrigerators. One who were involved with vaccination and were refrigerator was located in the labour ward on duty at the time of the visit were eligible and the remaining two were located in other to participate. The questionnaire collected wards and could be accessed by labour ward sociodemographic details and contained a staff twice a day at designated times. The series of statements relating to HBV remaining two hospitals used standard knowledge that required a true or false refrigerators without temperature monitors to response from the interviewee. If the maintain a one-week supply of vaccine in the interviewee appeared to have difficulty labour ward. Both of these refrigerators were understanding the question it was explained used for storage of other medications and/or in different terms or translated into Tok Pisin. staff lunches. The questionnaire also contained open- ended questions related to common Four maternity units used vaccine carriers vaccination practices and barriers/enabling to store a small quantity of vaccines required factors to delivery of hepatitis B vaccination. for each shift. This practice is recommended as a means to minimize the opening and Interview findings were entered into an closing of vaccine refrigerators. The vaccine Access database designed for this survey. carriers in one unit were an older style with The sociodemographic characteristics of the ill-fitting ice packs. The remaining maternity interviewed staff were analysed descriptively. unit used vaccines directly from the Responses to the 14 knowledge statements refrigerator. were tallied and individuals grouped as either having ≥10 or ≤9 correct answers. A Documentation, policies and guidelines univariate analysis using STATA v9 with a cut off of p <0.05 was undertaken to identify All maternity units had good predictors of staff knowledge. documentation of information in their birth 49 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 1ELBAT NATIONAL HEALTH INFORMATION SYSTEM )SIHN( AND HOSPITAL-LEVEL HEPATITIS B BIRTH-DOSE COVERAGE RATES Hhlatipso tlaeHlanoitaN sdrocerlevel-latipsoH ecnereffiD metsySnoitamrofnI *7002doirephtnom4-2 neewteb 6002ceD-naJ dnaSIHN atadlatipsoh Bhshtri triB Csegarevo Bhhtri triB egarevoC sesod sesod NN % N N % % H01latipso 200537 999 865 868 H52latipso 13260 2211 1317 1554 838 H43latipso 13330 05. 851 302 404 H74latipso 1074 1665 170 405 443 921 H95latipso 1925 1784 96400 369 91 latoT 155937 3962 144783 3954 706 * )tcO-guA(5latipsoH,)tcO-luJ(4latipsoH,)tcO-luJ(3latipsoH,)peS-guA(2latipsoH,)peS-luJ(1latipsoH Figure 1. Hospital-level hepatitis B birth-dose coverage rates, 2007. registers and vaccination record books, the labour ward within two hours; therefore, evidenced by complete (no obvious missing it would be reasonable to assume that entries) and legible information. The vaccines administered in the labour ward maternity unit birth register was completed were given within 24 hours of birth. by the person who delivered the baby and the vaccination record book was completed The senior staff from four of the five by the person administering the vaccine. maternity units reported the total number of Neither register records the timing of the births and vaccinations to the hospital records hepatitis B birth-dose administration in unit. The hospital records unit is responsible relation to birth. It was usual practice for for forwarding these data to the PHO, who in mothers and babies to be transferred out of turn report to the NDoH. The remaining 50 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 maternity unit reported vaccination data revealed that none of the interviewed staff directly to the PHO and only provided data were aware of the rationale for providing the on the vaccinations provided in the post-natal birth dose within 24 hours of delivery. ward and not those administered in the labour ward. Vaccine contraindications The staff at the maternity units said that The only contraindication to administration all mothers received a record of immunization of hepatitis B vaccine in newborns is a fever for their baby before discharge. Baby health of >38.5oC; however, 52% of staff did not books were provided free of charge at one correctly identify this as a contraindication. hospital and cost one to three kina at the On the other hand, a number of other other hospitals. During the hospital visit it conditions were incorrectly reported as was recorded that only one maternity unit had reasons for delaying vaccination: prematurity a copy of the PNG immunization guidelines (60%), low birthweight (48%) and a difficult (16) and that it was kept in a locked office birth such as forceps or vacuum delivery occupied by the sister in charge of the unit. (36%). The majority of staff (56%) were unsure or unable to identify inflammation at Clinical practices the injection site and fever as potential side- effects of hepatitis B vaccination. Hepatitis B vaccine was routinely provided in labour wards in all five hospitals; newborns Vaccination practices and cold-chain transferred to the Special Care Nursery, maintenance however, had vaccination delayed until their condition was stable, unlike other routine The majority (88%) of staff correctly stated interventions such as administration of that hepatitis B vaccine is not administered vitamin K and eye ointment, which were subcutaneously but 52% of staff were unsure provided irrespective of the newborn’s whether opened multidose vaccine vials condition. A range of incorrect practices were could be reused or not. Discussions revealed observed, including: changing needles after a generally poor understanding of cold-chain drawing up the vaccine (all units), large-bore management amongst staff. The majority of needles being left inserted in the vial for staff were unaware of the recommended drawing up purposes (two units) and hard practice of monitoring refrigerator plastic bottles being used instead of sharps temperatures and of the vaccine vial monitors disposal boxes (one unit). All units had used to monitor the heat exposure of multiple injecting equipment options and individual vials. reported inconsistent supply of auto-disable syringes. HBV transmission Staff interviews Over 80% of staff correctly answered that people with hepatitis B can look and feel well Interviews were conducted with 25 staff but still spread the infection and that infection representing 28% of all staff on the maternity as a child can lead to chronic infection and unit rosters at the five selected hospitals. The liver disease. Although over 84% of staff proportion of those interviewed varied from knew that hepatitis B can be spread from 23% to 33% across individual maternity units. person to person via infected body fluids, The majority of staff interviewed were 52% incorrectly reported that it could be midwives (64%), were aged 36 years and spread via contaminated food and water. above (72%) and had worked in maternity units for more than ten years (52%) (Table Univariate analysis showed no statistically 2). significant association between sociodemographic characteristics (hospital, Immunization cadre, year of graduation, years working in maternity units, age group) and HBV Of the 25 staff interviewed, nearly three- knowledge. quarters (68%) were aware that HBV can be spread from mother to child at birth and 80% Barriers to the delivery of a timely hepatitis agreed that it was important to give hepatitis B birth-dose vaccination B vaccine within 24 hours of delivery (see Table 3). Further informal questioning Discussions with staff revealed that a lack 51 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 2ELBAT SOCIODEMOGRAPHIC CHARACTERISTICS OF MATERNITY UNIT STAFF INTERVIEWED Cdscitsiretcarah eweivretnirebmuN N% Ceerda M6fiwdi 146 R4esrunderetsige 16 C5rekrowhtlaehytinummo 20 Y0noitaudargforae B0991erofe 104 115991-199 4 161002-699 24 287002-200 32 A5)sraey(eg <002 24603-5 1 33253-1 1 38204-6 3 >004 104 nignikrowsraeY 0782- 2 stinuytinretam 3325- 1 62801- >301 125 of hepatitis B vaccine stock in the labour ward Enabling factors in the delivery of a was the main and often the only reason staff timely hepatitis B birth-dose did not routinely offer vaccination to vaccination newborns. Daily collection of vaccine from another part of the hospital at designated Informal discussions with staff revealed times was also reported to be a barrier to that a designated vaccine refrigerator in the maintaining adequate stock at the labour labour ward to maintain vaccine stock levels ward level. Staff confirmed that a poor and having hepatitis B vaccination as part of understanding of contraindications to routine care of the newborn were good vaccination resulted in babies who were strategies to ensure that newborns received transferred to the Special Care Nursery a timely hepatitis B birth-dose vaccination. having their hepatitis B vaccination delayed unnecessarily. Attendance at the 2007 Family Health 52 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 3ELBAT HEPATITIS B KNOWLEDGE OF 52 MATERNITY UNIT STAFF IN FIVE GNP HOSPITALS,7002 TrstnemetatsegdelwonkeslaFroeur otcerrocnI srewsnaerusnu N% daerpsllitsnactubllewleefdnakoolnacnoitcefniVBHhtiwelpoeP Bsititapeh 321 dnanoitcefnicinorhcotdaelnacdlihcasaBsititapehhtiwnoitcefnI esaesidrevil 461 ebtondluohsC°5.83nahtretaergfoerutarepmetahtiwseibaB noitaniccavBsititapehehtnevig 123 5 ylsuoenatucbusderetsinimdasieniccavBsititapeH 321 nihtiwesodhtribBsititapehehtnevigeraseibabtahttnatropmisitI htribfosruoh42 502 nezorfebrevendluohseniccavBsititapeH 100 4 ro/dnaniapedulcnieniccavBsititapehfostceffe-edislaitnetoP revefdnaetisnoitcejniehttanoitammalfni 164 5 ylerutamerpnrobsiybabehtfideyaledebdluohsnoitaniccaV 105 6 tcatnochguorhtnosrepotnosrepmorfdaerpsebnacBsititapeH )cteavilas,egrahcsiddnuow,doolb(sdiulfydobdetcefnihtiw 461 nairaseac,noitcartxemuucav(htribtluciffidadahevahohwseibaB deyalednoitaniccavBsititapehevahdluohs)ctenoitces 963 retawdnadoofdetanimatnocaivdaerpsebnacBsititapeH 123 5 deyalednoitaniccavBsititapehevahdluohsseibabthgiewhtribwoL 182 4 htribtadlihcotrehtommorfdaerpsebnacBsititapeH 823 ehtotdenruterebnaceniccavBsititapehfoslaivesoditlumdenepO sesodtneuqesbusrofdesudnarotaregirfer 123 5 surivBsititapeh=VBH Services coordinators’ meeting by senior at discharge from the post-natal ward, maternity and paediatric nursing staff from ensuring that it is given within 24 hours of one hospital had resulted in the appointment birth. of a designated immunization nurse and increased awareness within the labour ward Univariate analysis revealed no of hepatitis B vaccination. Since this statistically significant sociodemographic intervention, recorded hepatitis B birth-dose predictors of knowledge or any difference administration had risen from less than 40% between the knowledge of staff from the five to over 80% coverage. Service improvement hospitals. planning undertaken at another hospital had resulted in a recent change in practice with Discussion hepatitis B vaccination now being administered in the labour ward rather than The findings of the most recent (2005) 53 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 national vaccination coverage survey that many staff were aware that the hepatitis B only 23% of children born in health facilities birth dose should be given within 24 hours received a hepatitis B birth dose was of delivery, none had a clear understanding concerning. Similarly the combined of the rationale. coverage across the five hospitals at the national level through the National Health This study has some limitations: firstly, it Information System (NHIS) was only 19%. was restricted to five major government-run The results of our review were far more hospitals and the results cannot be encouraging with more than 80% of babies generalized to all hospitals in PNG. It is not born in four of the five selected hospitals unreasonable to expect that the main themes receiving a hepatitis B birth-dose vaccination identified will be similar to those in other major before discharge. In the fifth, the proportion hospitals; however, the situation in rural was 40%; however, during one of the months hospitals and health centres may be different when data were collected structural work as they are more likely to have issues with being conducted in the maternity unit resulted vaccine supply given the increased distance in increased logistical challenges accessing from the central supply stores. Secondly, vaccine. Based on our review, the data only about a quarter of all staff on maternity available at the primary source, ie, maternity unit rosters were included in the staff surveys; units, appears to be accurate and the wide nevertheless, participation was high for the discrepancy between the sources suggests shift. It is unlikely that findings would vary there are limitations in reporting systems considerably between shifts as all hospitals where information at the national level has have rosters requiring staff to rotate through passed through both hospital and PHO all shifts. reporting processes. This study has found higher than While an 80% coverage rate in four of the anticipated hepatitis B birth-dose coverage five hospitals is encouraging, the missed levels in five major PNG hospitals and vaccination opportunities for the other 20% identified barriers to provision of the birth or more of children born in these health dose in maternity units. The large difference facilities must be addressed. Two key identified between coverage levels recorded barriers to hepatitis B birth-dose vaccination at hospital level and that reported nationally emerged during this study: i) lack of access suggests that there could be value in a more to hepatitis B vaccine in the maternity units; comprehensive audit and investigation of and ii) limited staff knowledge and reporting practices. Accurate reporting, awareness. including time of vaccination in relation to birth, would greatly assist decision-making Access to appropriately stored vaccine by and planning at health facility, provincial and maternity unit staff was compromised in national level. The two key barriers to various ways despite all five hospitals having increasing vaccination coverage could be adequate supply of hepatitis B vaccine. Two rectified by i) locating vaccine-specific maternity units did not have their own refrigerators in maternity units and ii) training refrigerators and were reliant on twice-daily sessions or other strategies to raise collection of vaccine from other parts of the awareness and improve knowledge about hospital. Of the three maternity units with vaccine practices and the importance of the refrigerators, only one had appropriate hepatitis B birth dose, among both clinical temperature-monitoring systems in place. staff and management. Further studies in smaller and rural facilities would be useful to Limited staff knowledge and awareness identify their specific issues and develop resulted in vaccination being delayed for targeted strategies ensuring that all babies newborns transferred to the Special Care born in health facilities receive a timely Nursery. The optimal timing for hepatitis B birth dose. administration of the hepatitis B birth dose is within 24 hours of delivery and delays may ACKNOWLEDGEMENTS reduce the effectiveness of the vaccine in preventing mother-to-child transmission of We acknowledge the assistance of the hepatitis B (12). It was concerning that most management and unit staff of the staff incorrectly believed that prematurity, low participating hospitals, participating provincial birthweight and a difficult birth were reasons health offices, the NDoH Monitoring, to delay vaccination. Furthermore, although Evaluation and Research Branch, the NDoH 54 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Expanded Program on Immunization (EPI) 1985;151:599-603. Unit, the Macfarlane Burnet Institute for 8 Huang K, Lin S. Nationwide vaccination: a success story in Taiwan. Vaccine 2000;18(Suppl 1):S35-S38. Medical Research and Public Health, the 9 Viviani S, Jack A, Hall AJ, Maine N, Mendy M, World Health Organization PNG office and Montesano R, Whittle HC. Hepatitis B vaccination the Australian Government Department of in infancy in The Gambia: protection against carriage Health and Ageing for funding support for the at 9 years of age. Vaccine 1999;17:2946-2950. 10 Vryheid RE, Kane MA, Muller N, Schatz GC, Master of Applied Epidemiology Program. Bezabeh S. Infant and adolescent hepatitis B immunization up to 1999: a global overview. Vaccine Funding for this project was provided by 2000;19:1026-1037. the World Health Organization. 11 Beasley RP, Trepo C, Stevens CE, Szmuness W. The e antigen and vertical transmission of hepatitis B surface antigen. Am J Epidemiol 1977;105:94- REFERENCES 98. 12 World Health Organization. Preventing Mother to 1 World Health Organization. Hepatitis B vaccines. Child Transmission of Hepatitis B: Operational Field Wkly Epidemiol Rec 2004;28:255-263. Guidelines for Delivery of the Birth Dose of Hepatitis 2 Beasley RP, Hwang LY, Lin CC, Chien CS. B Vaccine. Manila: Western Pacific Regional Office, Hepatocellular carcinoma and hepatitis B virus. A 2006. prospective study of 22 707 men in Taiwan. Lancet 13 Sanders RC, Lewis D, Dyke T, Alpers MP. Markers 1981;2:1129-1133. of hepatitis B infection in Tari District, Southern 3 McMahon BJ, Alberts SR, Wainwright RB, Highlands Province, Papua New Guinea. PNG Med Bulkow L, Lanier AP. Hepatitis B-related sequelae. J 1992;35:197-201. Prospective study in 1400 hepatitis B surface 14 Spooner V, Richens J, Sanders R. Hepatitis B antigen-positive Alaska native carriers. Arch Intern surface antigen, e antigen and HBV DNA in healthy Med 1990;150:1051-1054. antenatal patients attending Goroka Hospital and 4 Edmunds WJ, Medley GF, Nokes DJ, Hall AJ, their relationship to tattooing practices. PNG Med J Whittle HC. The influence of age on the 1990;33:11-15. development of the hepatitis B carrier state. Proc 15 Yamaguchi K, Inaoka T, Ohtsuka R, Akimichi T, Biol Sci 1993;253:197-201. Hongo T, Kawabe T, Nakazawa M, Futatsuka M, 5 Hsieh CC, Tzonou A, Zavitsanos X, Kaklamani Takatsuki K. HTLV-I, HIV-I, and hepatitis B and C E, Lan SJ, Trichopoulos D. Age at first viruses in Western Province, Papua New Guinea: a establishment of chronic hepatitis B virus infection serological survey. Jpn J Cancer Res 1993;84:715- and hepatocellular carcinoma risk. A birth order study. Am J Epidemiol 1992;136:1115-1121. 719. 6 Hyams KC. Risks of chronicity following acute 16 Expanded Program on Immunization Unit. hepatitis B virus infection: a review. Clin Infect Dis Immunization Manual: A Handbook for Staff in Health 1995;20:992-1000. Centres. Port Moresby: Papua New Guinea National 7 McMahon BJ, Alward WL, Hall DB, Heyward WL, Department of Health, 2002. Bender TR, Francis DP, Maynard JE. Acute 17 Papua New Guinea National Department of hepatitis B virus infection: relation of age to the Health. National immunization coverage survey. clinical expression of disease and subsequent Papua New Guinea National Department of Health, development of the carrier state. J Infect Dis Port Moresby, 2005. 55 PNG Med J 2008 Mar-Jun;51(1-2): 56-59 The prevalence of HIV infection in women attending antenatal clinics in Fiji CHARLES H. WASHINGTON1,2, LAUREN M. SINGER1,3, TAUTA MCCAIG4, LISI TIKODUADUA4, SOPHAGANINE T. A LI1, JAMES FONG5, JIKO LUVENI6, THANE O. KYAW-MYINT1, STUART WATSON7 AND FIONA RUSSELL4 United Nations Children’s Fund, Colonial War Memorial Hospital, Ministry of Health and United Nations AIDS, Suva, Fiji SUMMARY HIV (human immunodeficiency virus) is an increasing concern in the South Pacific. We estimate, based on reported figures, that the prevalence of HIV infection in women attending antenatal clinics in Fiji in 2003 was 0.04%. The number of children born to HIV-positive mothers is small, though perinatal transmission appears to be high. Fiji’s preliminary strategies for prevention of perinatal transmission have been significant, but require ongoing support and implementation. Introduction Approximately one-third of all deliveries in Fiji occur at CWMH (4). Other hospitals There is a significant push to implement commenced antenatal HIV testing in later HIV (human immunodeficiency virus) and years. In 1999, Fiji adopted a prevention of AIDS (acquired immune deficiency mother-to-child transmission (pMTCT) syndrome) prevention programs in Pacific protocol at CWMH based on the ‘Thai short- Island countries, including Fiji, to prevent a long’ regimen (5). full-scale epidemic as is occurring in nearby Papua New Guinea (1, as cited in 2). Fiji’s The Fiji protocol recommends maternal first official case of HIV infection was zidovudine (AZT) 300 mg twice daily from diagnosed in 1989. As of September 2003, 34 weeks gestation to delivery by elective 129 cases of HIV infection have been caesarian at 38 weeks, and 6 weeks of AZT serologically confirmed (3). More than half 2 mg/kg four times daily for the infant. No of these cases have been diagnosed in the AZT is administered during caesarian section last three years, marking an increase in as there is currently no access to intravenous detection and possibly transmission. AZT. Artificial feeding of infants is recommended for infants born to HIV- In 1995 antenatal testing began at Fiji’s positive mothers. only tertiary hospital, the Colonial War Memorial Hospital (CWMH). During a routine The goals of this retrospective study were antenatal visit pregnant women were to estimate the prevalence of HIV infection provided pre-HIV test counselling and in pregnant women, and to document cases subsequently offered an HIV-antibody test. of perinatal transmission and the outcome in 1 United Nations Children’s Fund, Private Mail Bag, Suva, Fiji 2 Present address: School of Medicine, University of Washington, A-300 Health Sciences Building, Box 356340, Seattle, WA 98195-6340, United States of America 3 Present address: 7209 Sleep Soft Circle, Columbia, MD 21045, United States of America 4 Department of Paediatrics, Colonial War Memorial Hospital, GPO Box 115, Toorak, Suva, Fiji 5 Department of Obstetrics and Gynaecology, Colonial War Memorial Hospital, GPO Box 115, Toorak, Suva, Fiji 6 Department of Statistics, Ministry of Health, Dinem House, Toorak, Suva, Fiji 7 United Nations AIDS, Private Mail Bag, Suva, Fiji 56 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 affected infants. The study was approved by the Fiji National Research Ethics Review Methods Committee. We retrospectively studied clinic records Results of all 23 hospitals providing antenatal care in Fiji. These hospitals had various HIV 52,783 pregnant women were tested for testing implementation dates ranging from HIV in Fiji from 1995 to 2003. The yearly 1995 to 2003. HIV-positive cases were maternal prevalence of HIV infection ranged compiled from the Ministry of Health, the from 0 to 0.05% (Table 1). In 2003, Department of Statistics and laboratory approximately 64% (10,837) of pregnant records. All hospitals providing antenatal women received antenatal testing and 4 were care were visited or contacted by telephone found to be HIV positive (0.037%). Of the to verify that no cases had been missed. Only 12 HIV-positive women whose ethnicity was births that occurred when HIV testing was known, 11 (92%) were indigenous Fijian available at each specific hospital were women. included in estimates of prevalence. 13 HIV-positive women have given birth The national HIV testing protocol states since the implementation of the pMTCT that children born to HIV-positive mothers are protocol. Of these, 5 women received to be tested in the first week of life, with antenatal AZT and delivered by elective additional testing performed at follow-up caesarian, 3 women received antenatal AZT visits and at 18 months of age. Initial testing but delivered vaginally, and 5 women did not for both pregnant women and infants was receive antenatal AZT and delivered done by Serodia HIV-1/2 (Fujirebio Inc, vaginally. Of the 8 women receiving Tokyo, Japan) antibody detection tests. antenatal AZT, all initiated treatment between Positive or indeterminate tests were retested 35 and 37 weeks gestation. Of the 5 women on-site and at CWMH (the national not receiving any interventions, 4 women laboratory). Confirmatory HIV testing by were diagnosed postpartum (3 of these Western blot was performed at the Australian women were determined to be HIV positive National Reference Laboratory (NRL). after their children developed symptoms of 1ELBAT ANNUAL PREVALENCE OF VIH INFECTION IN WOMEN WITH ANTENATAL CLINIC BOOKINGS BEFORE DELIVERY Yerae NdvitisopVIHrebmu N)etsetrebmu %(ecnelaverP 10599 3086 100.0 12699 3228 950.0 10799 3089 400.0 10899 5071 200.0 12999 5734 130.0 21000 5807 510.0 21100 6641 710.0 24200 7099 750.0 24300 17380 730.0 57 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 HIV infection including failure to thrive and positive women of unknown immune status severe malnutrition) and 1 woman was has been shown to increase morbidity and previously known to be HIV positive but did mortality, particularly in women from not seek antenatal care before delivery. Of developing country settings (6,7). In Fiji, the 11 women with follow-up information, 4 women have access to sound perinatal died, 2 were sick with possible AIDS-related services; therefore judgment on whether to conditions and the other 5 were healthy. perform a caesarian section or not will Causes of death included oesophageal depend on the clinical status of the pregnant candidiasis, cytomegalovirus infection and woman, as determined by the treating other AIDS-related conditions. physician, in the absence of being able to determine immune status. Medical records were identified for 12 children from 12 HIV-positive women. 10 of Interviews conducted during the course of these deliveries occurred after 1999. Of the study revealed multiple challenges to those 10, all 8 children for whom their further interruption of vertical transmission of mother’s status was known before delivery HIV. Structural challenges include: lack of received AZT. The first HIV test occurred at infrastructure for monitoring adherence; birth for 5 children, at 1 month of age for 1 stigma and discrimination within the health child, and by 4 months of age for the other 2 care system; insufficient medical, social and children. The last available test results for mental health support programs; and limited these 8 children were: 2 were positive (1 by counselling and confidentiality. Cultural NRL, 1 by Serodia), 3 were negative (2 by traditions also need to be addressed. For NRL, 1 by Serodia), and 3 were example, one Fijian custom encourages the indeterminate (all NRL). Overall, 4 of these father’s parents to determine the age at which children are known to be dead (1 positive by the child should be weaned from NRL, 1 negative by NRL, 2 indeterminate) breastfeeding. This is one of the barriers to and 2 are known to be alive; the remaining 6 the uptake of artificial feeding, along with the have not had recent follow-ups and their stigma and suspicion associated with women status is unknown. Co-morbidities in all of who do not breastfeed their infants. the children included scabies, malnutrition, failure to thrive, persistent pneumonia and The current World Health Organization anaemia, but the proximate cause of death guidelines recommend exclusive formula was not able to be determined. feeding, if it is “acceptable, feasible, affordable, sustainable and safe”; otherwise Discussion exclusive breastfeeding is recommended (8). These guidelines stem from the large Our study documents a relatively low reduction in diarrhoeal and respiratory prevalence of HIV infection among pregnant disease in breastfed infants in developing women and infants. Although there has been countries (9). The breastmilk immune factors an increase in HIV-positive mothers in Fiji, it of HIV-infected women are similar to those is difficult to determine if this increase is a of HIV-uninfected women; this argues for trend. There has been a rapid uptake of HIV exclusive breastfeeding for the first 4-6 testing in antenatal clinics in Fiji, progressing months of life followed by early and sudden from the testing of 3681 women in 1995 to weaning, which provides the best protection 10,837 women in 2003, although testing is from early infant death for babies of HIV- still not universal. Given that approximately positive mothers in developing countries or 89% of women in Fiji receive antenatal care settings where mixed methods of feeding are and 97% of births occur in a hospital, used (10-13). In the context of Fiji, a low- antenatal testing provides a systematic middle-income country, with relatively low method of providing HIV testing to a infant mortality indicating a sound health care population in which prevention of disease system and reasonable access to transmission is possible (3). environmental hygiene practices, it is not clear if exclusive breastfeeding with rapid Given the small number of cases and high weaning should be the universal loss to follow-up, our study is not able to recommendation for all. However, given the assess the effectiveness of the current social stigma associated with not pMTCT policy, including elective caesarean breastfeeding, the same recommendation as sections (CS), in Fiji; however, transmission for higher-mortality countries may well be the appears to be high. Performing CS in HIV- preferred option. 58 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Counselling (pre- and post-test) is not only Statistics. Fiji Ministry of Health, Suva, Fiji Islands, necessary, but it is also an opportunity for Sep 2003. 4 Fiji Ministry of Health. Report of the Division of HIV education of the general population Statistics. Fiji Ministry of Health, Suva, Fiji Islands, rather than just the positive women. The 2002. counselling should be continued throughout 5 Shaffer N, Chuachoowong R, Mock PA, the pregnancy and include infant feeding, Bhadrakom C, Siriwasin W, Young NL, Chotpitayasunondh T, Chearskul S, family planning, nutrition, overall care, and Roongpisuthipong A, Chinayon P, Karon J, support. Additionally, national testing Mastro TD, Simonds RJ, Bangkok Collaborative guidelines, alteration and implementation of Perinatal HIV Transmission Study Group. Short- the pMTCT protocol, increased laboratory course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. capacity for early diagnosis and improved Lancet 1999;353:773-780. monitoring of HIV-positive women and 6 Muula AS. Ethical and practical consideration of children are required and are currently being women choosing cesarean section deliveries without addressed by national policy. ‘medical indication’ in developing countries. Croat Med J 2007;48:94-102. 7 Read JS, Newell MS. Efficacy and safety of This study has several limitations, cesarean delivery for prevention of mother-to-child including small numbers, lack of a country- transmission of HIV-1. Cochrane Database Syst Rev wide antenatal clinic HIV screening program, 2005;4:CD005479. high loss to follow-up, lack of definitive 8 World Health Organization. HIV and Infant Feeding Technical Consultation held on behalf of the diagnosis of children in Fiji, and inability to Inter-agency Task Team on Prevention of HIV assess adherence to the antiretroviral Infections in Pregnant Women, Mothers and Their therapy and artificial feeding protocols. Our Infants. Geneva: World Health Organization, 2006. data are limited, but they document the 9 World Health Organization Collaborative Study Team on the Role of Breastfeeding on the annual prevalence of HIV infection in women Prevention of Infant Mortality. Effect of visiting antenatal clinics and give insight into breastfeeding on infant and child mortality due to the climate and challenges surrounding HIV infectious diseases in less developed countries: a in Fiji. pooled analysis. Lancet 2000;355:451-455. 10 Shapiro RL, Lockman S, Kim S, Smeaton L, Rahkola JT, Thior I, Wester C, Moffat C, Arimi P, ACKNOWLEDGEMENTS Ndase P, Asmelash A, Stevens L, Montano M, Makhema J, Essex M, Janoff EN. Infant morbidity, Assistance was received from the Fiji mortality, and breast milk immunologic profiles Ministry of Health, the Colonial War Memorial among breast-feeding HIV-infected and HIV- uninfected women in Botswana. J Infect Dis Hospital, UNICEF (United Nations Children’s 2007;196:563-569. Fund) (especially Judith Leveillee) and the 11 Becquet R, Ekouevi DK, Menan H, Amani-Bosse many patients who every day deal with the C, Bequet L, Viho I, Dabis F, Timite-Konan M, morbidity, mortality and stigma of HIV Leroy V: ANRS 1201/1202 Ditrame Plus Study Group. Early mixed feeding and breastfeeding infection. UNICEF, Suva, Fiji gave financial beyond 6 months increase the risk of postnatal HIV support. transmission: ANRS 1201/1202 Ditrame Plus, Abidjan, Côte d’Ivoire. Prev Med 2008;47:27-33. REFERENCES 12 Fowler MG. Further evidence that exclusive breast- feeding reduces mother-to-child HIV transmission 1 Papua New Guinea Department of Health. Papua compared with mixed feeding. PLoS Med 2008 Mar New Guinea National Health Plan 1996-2000. 11;5(3):e63. Volume 2. Port Moresby: Department of Health, 13 Fox MP, Brooks D, Kuhn L, Aldrovandi G, Sinkala 1996:251. M, Kankasa C, Mwiya M, Horsburgh R, Thea DM. 2 Caldwell JC, Isaac-Toua J. AIDS in Papua New Reduced mortality associated with breast-feeding- Guinea. Situation in the Pacific. J Health Popul Nutr acquired HIV infection and breast-feeding among 2002;20:104-111. HIV-infected children in Zambia. J Acquir Immune 3 Fiji Ministry of Health. Report of the Division of Defic Syndr 2008;48:90-96. 59 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 MEDLARS BIBLIOGRAPHY PUBLICATIONS OF RELEVANCE TO PAPUA NEW GUINEA AND MELANESIA Bibliographic Citation List generated from MEDLARS 1 Appleyard B, Tuni M, Cheng Q, Chen N, Bryan J, the most populated country in the Pacific. The McCarthy JS. investigations carried out in sentinel sites showed a Malaria in pregnancy in the Solomon Islands: barriers public health impact, by the fall in the number of to prevention and control. microfilaria carriers, often down to a rate <1%. Am J Trop Med Hyg 2008 Mar;78(3):449-454. However, the rate of circulating antigen prevalence A study of malaria in pregnancy (MIP) was often remained above the required threshold of 1%. undertaken in Marovo Lagoon, Solomon Islands, to Prevalence surveys carried out in 2007, in different evaluate pregnancy-specific control strategies for endemic countries, revealed the necessity to malaria. Peripheral parasitemia was present in 18% intensify efforts and to refine the strategy for the (19/106) of women: 15 Plasmodium falciparum and elimination of LF from the Pacific. A lot of progress 4 P. vivax. Primigravidae were twice as likely to be was made, but a few problems were identified. parasitemic as multigravidae (31% versus 14%; Reevaluations are urgently required and are in relative risk: 2.24; 95% confidence interval: 1.01- progress. They deal with maintenance of the MDA 4.96; p = 0.05). Although ante-natal clinic attendance coverage rates in the face of a certain lassitude in was high, women booked late (mean, 19.7 weeks) the populations and among health staff, the methods and attended irregularly. Free insecticide-treated to evaluate the effectiveness of MDA, the reliability nets (ITN) were not distributed despite government of the diagnostic tools to decide when to stop MDA policy. Primigravidae were less likely to have an and to certify the absence of transmission, the ITN in their homes than multigravidae (relative risk: relevance of universal biological criteria for the whole 2.13; 95% confidence interval: 1.03-4.40). Coverage Pacific area, and the need for active surveillance for with chloroquine prophylaxis was low. This study several years after stopping MDA, particularly in the revealed barriers to control of MIP at both the health countries affected by the very efficient vector Aedes service and client level. To develop an evidence- polynesiensis. Seven years after its launching, based malaria control policy in pregnancy for this despite undeniable success, the PacELF program region, further study of the epidemiology of malaria has not achieved its very ambitious goal of stopping and its effects, including social and behavioral transmission. Three years before its term, strong aspects, is needed. efforts have to be made and additional strategies implemented. However, it is reasonable to expect 2 Bulletin of the World Health Organization. the prolongation of the program in order to achieve Primary health care: Fiji’s broken dream. the final objective. Beyond that, in some countries, Bull World Health Organ 2008 Mar;86(3):166-167. it will still be necessary to ensure a sustained global drug pressure and an active surveillance program 3 Chanteau S, Roux JF. to prevent the re-emergence of the disease. Bancroftian lymphatic filariasis: toward its elimination from the Pacific? [Fr] 4 Chao YC, Huang CS, Lee CN, Chang SY, King Bull Soc Pathol Exot 2008 Jun;101(3):254-260. CC, Kao CL. The region of the Pacific is historically affected Higher infection of dengue virus serotype 2 in human by lymphatic filariasis (LF). Following the World monocytes of patients with G6PD deficiency. Health Assembly resolution in 1997, the Global PLoS ONE 2008 Feb 13;3(2):e1557. Program to Eliminate Lymphatic Filariasis (GPELF) The prevalence of glucose-6-phosphate was launched. In the Pacific, the World Health dehydrogenase (G6PD) deficiency is high in Asia. Organization (WHO) has implemented from 1999 An ex vivo study was conducted to elucidate the the Pacific Program to Eliminate Lymphatic Filariasis association of G6PD deficiency and dengue virus (PacELF) bringing together the 22 countries and (DENV) infection when many Asian countries are territories in a common effort to eliminate the hyper-endemic. Human monocytes from peripheral disease. The strategy is based on mass drug mononuclear cells collected from 12 G6PD-deficient administration (MDA); an annual single dose during patients and 24 age-matched controls were infected 5 years of a diethylcarbamazine/albendazole with one of two DENV serotype 2 (DENV-2) strains combination is distributed to all the populations at – the New Guinea C strain (from a case of dengue risk. Among the 22 countries and territories of the fever) or the 16681 strain (from a case of dengue Pacific, 16 are endemic and 6 are non-endemic. The hemorrhagic fever) with a multiplicity of infection of classification is based according to the filarial antigen 0.1. The infectivity of DENV-2 in human monocytes prevalence being above or below 1%. MDA is was analyzed by flow cytometry. Experimental indicated when the rate of the filarial antigen results indicated that the monocytes of G6PD- prevalence is >1%. The objective of PacELF is to deficient patients exhibited a greater level of infection reduce this rate down to <1%, the threshold under with DENV-2 New Guinea C strain than did those which transmission is supposed to be stopped. From from healthy controls [mean±SD:33.6%±3.5 vs 1999 to 2007, 14 of the 16 endemic countries 20.3%±6.2, p<0.01]. Similar observations were organized MDA. Eleven of them completed the cycle made of infection with the DENV-2 16681 strain of 5 treatments and even beyond. But these MDA [40.9%±3.9 vs 27.4%±7.1, p<0.01]. To our reached only 19% of the at-risk population, mainly knowledge, this study demonstrates for the first time because of logistic difficulties in Papua New Guinea, higher infection of human monocytes in G6PD 60 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 patients with the dengue virus, which may be randomly selected from specific sites on the islands important in increasing epidemiological transmission of Efate (urban setting) and Ambae (rural setting). and perhaps with the potential to develop more Pap smears were collected, screened and reported. severe cases pathologically. SETTING: The first collection site was the Women’s Health/Antenatal Care Clinic at Vila Central Hospital 5 Chaves LF, Kaneko A, Taleo G, Pascual M, Wilson in Port Vila, the capital city located on the island of ML. Efate, and the second collection site was a rural Malaria transmission pattern resilience to climatic village on a sparsely populated inhabited northern variability is mediated by insecticide-treated nets. island, Ambae. PARTICIPANTS: A total of 905 Ni- Malar J 2008 Jun 2;7:100. Vanuatu women participants: Efate (n = 562) 62%, BACKGROUND: Malaria is an important public- and Ambae (n = 343) 38%. The mean age was 35.8 health problem in the archipelago of Vanuatu and years: Efate 32.6 years, and Ambae 40.8 years. climate has been hypothesized as an important MAIN OUTCOME MEASURE: The presence or influence on transmission risk. Beginning in 1988, absence of T. vaginalis in these Pap smears was a major intervention using insecticide-treated bed documented during the study’s cervical screening nets (ITNs) was implemented in the country in an process. RESULTS: The overall prevalence of T. attempt to reduce Plasmodium transmission. To vaginalis within the study participants was 25.3%. date, no study has addressed the impact of ITN Almost half of the infected sample group were in the intervention in Vanuatu, how it may have modified age group of 30-39 years (43.8%). The prevalence the burden of disease, and whether there were any of T. vaginalis in Efate was 14.7%, compared with changes in malaria incidence that might be related 43.4% in Ambae. CONCLUSION: The prevalence to climatic drivers. METHODS AND FINDINGS: of T. vaginalis in Vanuatu women is significantly Monthly time series (January 1983 through higher than in developed countries. Women in rural December 1999) of confirmed Plasmodium settings are less likely to have access to sexually falciparum and Plasmodium vivax infections in the transmitted disease prevention and treatment archipelago were analysed. During this 17-year programs, thus contributing to high infection rates period, malaria dynamics underwent a major regime compared with women in urban settings. Cultural shift around May 1991, following the introduction of and educational differences in the rural setting might bed nets as a control strategy in the country. By also contribute to higher sexually transmitted disease February of 1994 disease incidence from both rates among these women. parasites was reduced by at least 50%, when at most 20% of the population at risk was covered by ITNs. 8 Fowkes FJ, Allen SJ, Allen A, Alpers MP, Seasonal cycles, as expected, were strongly Weatherall DJ, Day KP. correlated with temperature patterns, while inter- Increased microerythrocyte count in homozygous annual cycles were associated with changes in alpha(+)-thalassaemia contributes to protection precipitation. Following the bed net intervention, the against severe malarial anaemia. influence of environmental drivers of malaria PLoS Med 2008 Mar 18;5(3):e56. dynamics was reduced by 30-80% for climatic forces, BACKGROUND: The heritable and 33-54% for other factors. A time lag of about haemoglobinopathy alpha(+)-thalassaemia is five months was observed for the qualitative change caused by the reduced synthesis of alpha-globin (“regime shift”) between the two parasites, the chains that form part of normal adult haemoglobin change occurring first for P. falciparum. The latter (Hb). Individuals homozygous for alpha(+)- might be explained by interspecific interactions thalassaemia have microcytosis and an increased between the two parasites within the human hosts erythrocyte count. Alpha(+)-thalassaemia and their distinct biology, since P. vivax can relapse homozygosity confers considerable protection after a primary infection. CONCLUSION: The against severe malaria, including severe malarial Vanuatu ITN programme represents an excellent anaemia (SMA) (Hb concentration <50 g/l), but does example of implementing an infectious disease not influence parasite count. We tested the control programme. The distribution was undertaken hypothesis that the erythrocyte indices associated to cover a large, local proportion (approximately with alpha(+)-thalassaemia homozygosity provide a 80%) of people in villages where malaria was haematological benefit during acute malaria. present. The successful coverage was possible METHODS AND FINDINGS: Data from children because of the strategy for distribution of ITNs by living on the north coast of Papua New Guinea who prioritizing the free distribution to groups with had participated in a case-control study of the restricted means for their acquisition, making the protection afforded by alpha(+)-thalassaemia against access to this resource equitable across the severe malaria were reanalysed to assess the population. These results emphasize the need to genotype-specific reduction in erythrocyte count and implement infectious disease control programmes Hb levels associated with acute malarial disease. focusing on the most vulnerable populations. We observed a reduction in median erythrocyte count of approximately 1.5 x 1012/l in all children with 6 Chung S, Watters D. acute falciparum malaria relative to values in Academic surgery in Papua New Guinea. community children (p <0.001). We developed a ANZ J Surg 2008 May;78(5):347-349. simple mathematical model of the linear relationship between Hb concentration and erythrocyte count. 7 Fotinatos N, Warmington A, Walker T, Pilbeam This model predicted that children homozygous for M. alpha(+)-thalassaemia lose less Hb than children of Trichomonas vaginalis in Vanuatu. normal genotype for a reduction in erythrocyte count Aust J Rural Health 2008 Feb;16(1):23-27. of >1.1 x 1012/l as a result of the reduced mean cell OBJECTIVE: To assess the prevalence of Hb in homozygous alpha(+)-thalassaemia. In Trichomonas vaginalis in two island populations of addition, children homozygous for alpha(+)- Vanuatu using the Pap smear as the screening thalassaemia require a 10% greater reduction in technique. STUDY DESIGN: Women were erythrocyte count than children of normal genotype 61 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 (p = 0.02) for Hb concentration to fall to 50 g/l, the regions. It also shows how textured human cutoff for SMA. We estimated that the population variation can be in particular haematological profile in children homozygous for circumstances. Genetic diversity within individual alpha(+)-thalassaemia reduces the risk of SMA Pacific populations is shown to be very low, while during acute malaria compared to children of normal differentiation among Melanesian groups is high. genotype (relative risk 0.52; 95% confidence interval Melanesian differentiation varies not only between [CI] 0.24-1.12, p = 0.09). CONCLUSIONS: The islands, but also by island size and topographical increased erythrocyte count and microcytosis in complexity. The greatest distinctions are among the children homozygous for alpha(+)-thalassaemia may isolated groups in large island interiors, which are contribute substantially to their protection against also the most internally homogeneous. The pattern SMA. A lower concentration of Hb per erythrocyte loosely tracks language distinctions. Papuan- and a larger population of erythrocytes may be a speaking groups are the most differentiated, and biologically advantageous strategy against the Austronesian or Oceanic-speaking groups, which significant reduction in erythrocyte count that occurs tend to live along the coastlines, are more intermixed. during acute infection with the malaria parasite A small ‘Austronesian’ genetic signature (always Plasmodium falciparum. This haematological profile <20%) was detected in less than half the Melanesian may reduce the risk of anaemia by other Plasmodium groups that speak Austronesian languages, and is species, as well as other causes of anaemia. Other entirely lacking in Papuan-speaking groups. host polymorphisms that induce an increased Although the Polynesians are also distinctive, they erythrocyte count and microcytosis may confer a tend to cluster with Micronesians, Taiwan Aborigines, similar advantage. and East Asians, and not Melanesians. These findings contribute to a resolution to the debates over 9 Fowkes FJ, Michon P, Pilling L, Ripley RM, Tavul Polynesian origins and their past interactions with L, Imrie HJ, Woods CM, Mgone CS, Luty AJ, Day Melanesians. With regard to genetics, the earlier KP. studies had heavily relied on the evidence from single Host erythrocyte polymorphisms and exposure to locus mitochondrial DNA or Y chromosome variation. Plasmodium falciparum in Papua New Guinea. Neither of these provided an unequivocal signal of Malar J 2008 Jan 3;7:1. phylogenetic relations or population intermixture BACKGROUND: The protection afforded by proportions in the Pacific. Our analysis indicates human erythrocyte polymorphisms against the the ancestors of Polynesians moved through malaria parasite, Plasmodium falciparum, has been Melanesia relatively rapidly and only intermixed to a proposed to be due to reduced ability of the parasite very modest degree with the indigenous populations to invade or develop in erythrocytes. If this were the there. case, variable levels of parasitaemia and rates of seroconversion to infected-erythrocyte variant 11 Genton B, D’Acremont V, Rare L, Baea K, Reeder surface antigens (VSA) should be seen in different JC, Alpers MP, Müller I. host genotypes. METHODS: To test this hypothesis, Plasmodium vivax and mixed infections are P. falciparum parasitaemia and anti-VSA antibody associated with severe malaria in children: a levels were measured in a cohort of 555 prospective cohort study from Papua New Guinea. asymptomatic children from an area of intense PLoS Med 2008 Jun 17;5(6):e127. malaria transmission in Papua New Guinea. Linear BACKGROUND: Severe malaria (SM) is mixed models were used to investigate the effect of classically associated with Plasmodium falciparum alpha+-thalassaemia, complement receptor-1 and infection. Little information is available on the South-east Asian ovalocytosis, as well as glucose- contribution of P. vivax to severe disease. There are 6-phosphate dehydrogenase deficiency and ABO some epidemiological indications that P. vivax or blood group on parasitaemia and age-specific mixed infections protect against complications and seroconversion to VSA. RESULTS: No host deaths. A large morbidity surveillance conducted in polymorphism showed a significant association with an area where the four species coexist allowed us both parasite prevalence/density and age-specific to estimate rates of SM among patients infected with seroconversion to VSA. CONCLUSION: Host one or several species. METHODS AND FINDINGS: erythrocyte polymorphisms commonly found in This was a prospective cohort study conducted within Papua New Guinea do not effect exposure to blood the framework of the Malaria Vaccine Epidemiology stage P. falciparum infection. This contrasts with and Evaluation Project. All presumptive malaria data for sickle cell trait and highlights that the above- cases presenting at two rural health facilities over mentioned polymorphisms may confer protection an 8-y period were investigated with history taking, against malaria via distinct mechanisms. clinical examination, and laboratory assessment. Case definition of SM was based on the World Health 10 Friedlaender JS, Friedlaender FR, Reed FA, Kidd Organization (WHO) criteria adapted for the setting KK, Kidd JR, Chambers GK, Lea RA, Loo JH, (i.e., clinical diagnosis of malaria associated with Koki G, Hodgson JA, Merriwether DA, Weber JL. asexual blood stage parasitaemia and recent history The genetic structure of Pacific Islanders. of fits, or coma, or respiratory distress, or anaemia PLoS Genet 2008 Jan;4(1):e19. [haemoglobin <5 g/dl]). Out of 17,201 presumptive Human genetic diversity in the Pacific has not malaria cases, 9,537 (55%) had a confirmed been adequately sampled, particularly in Melanesia. Plasmodium parasitaemia. Among those, 6.2% As a result, population relationships there have been (95% confidence interval [CI] 5.7%-6.8%) fulfilled the open to debate. A genome scan of autosomal case definition of SM, most of them in children <5 markers (687 microsatellites and 203 insertions/ years. In this age group, the proportion of SM was deletions) on 952 individuals from 41 Pacific 11.7% (10.4%-13.2%) for P. falciparum, 8.8% (7.1%- populations now provides the basis for understanding 10.7%) for P. vivax, and 17.3% (11.7%-24.2%) for the remarkable nature of Melanesian variation, and mixed P. falciparum and P. vivax infections. P. vivax for a more accurate comparison of these Pacific SM presented more often with respiratory distress populations with previously studied groups from other than did P. falciparum (60% versus 41%, p = 0.002), 62 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 but less often with anaemia (19% versus 41%, p = by extensive 1D and 2D NMR and MS data analyses 0.0001). CONCLUSION: P. vivax monoinfections and by comparison with the spectroscopic data of as well as mixed Plasmodium infections are apratoxins A and C. Apratoxin D (1) showed potent associated with SM. There is no indication that in vitro cytotoxicity against H-460 human lung cancer mixed infections protected against SM. Interventions cells with an IC 50 value of 2.6 nM. targeted toward P. falciparum only might be insufficient to eliminate the overall malaria burden, 14 Habgood PJ, Franklin NR. and especially severe disease, in areas where P. The revolution that didn’t arrive: a review of falciparum and P. vivax coexist. Pleistocene Sahul. J Hum Evol 2008 Aug;55(2):187-222. Epub 2008 12 Gilpin CM, Simpson G, Vincent S, O’Brien TP, May 15. Knight TA, Globan M, Coulter C, Konstantinos There is a ‘package’ of cultural innovations that A. are claimed to reflect modern human behaviour. The Evidence of primary transmission of multidrug- introduction of the ‘package’ has been associated resistant tuberculosis in the Western Province of with the Middle-to-Upper Palaeolithic transition and Papua New Guinea. the appearance in Europe of modern humans. It Med J Aust 2008 Feb 4;188(3):148-152. has been proposed that modern humans spread OBJECTIVE: To review patient outcomes and from Africa with the ‘package’ and colonised not only the molecular epidemiology of multidrug-resistant Europe but also southern Asia and Australia tuberculosis (MDR-TB) strains isolated from patients (McBrearty and Brooks, 2000; Mellars, 2006a). In living in the Western Province of Papua New Guinea order to evaluate this proposal, we explore the late (PNG) seeking treatment in Australia. DESIGN, Pleistocene archaeological record of Sahul, the SETTING AND PARTICIPANTS: Review of all cases combined landmass of Australia and Papua New of MDR-TB among people living in the open border Guinea, for indications of these cultural innovations region between the Western Province of PNG and at the earliest sites. It was found that following initial the Torres Strait Islands of Australia who presented occupation of the continent by anatomically and to health clinics in the region between 2000 and behaviourally modern humans, the components 2006. All cases of suspected TB were were gradually assembled over a 30,000-year period. bacteriologically confirmed at the time of We discount the idea that the ‘package’ was lost en presentation by the Mycobacterium Reference route to Sahul and assess the possibility that the Laboratory in Brisbane. MAIN OUTCOME ‘package’ was not integrated within the material MEASURES: Drug resistance patterns; drug use and culture of the initial colonising groups because they duration; molecular typing of TB strains; patient may not have been part of a rapid colonisation outcomes. RESULTS: Between 2000 and 2006, 60 process from Africa. As the cultural innovations patients from the Western Province of PNG were appear at different times and locations within Sahul, diagnosed with TB, of which 15 had MDR-TB. the proposed ‘package’ of archaeologically visible Mortality was high, although no patient who was able traits cannot be used to establish modern human to maintain access to supervised therapy died. All behaviour. Whilst the potential causal role of 15 MDR-TB isolates were Beijing-family strains increasing population densities/pressure in the showing the same unique mycobacterial appearance of the ‘package’ of modern human interspersed repetitive unit (MIRU) profile, with the behaviour in the archaeological record is exception of a single strain that differed by a single acknowledged, it is not seen as the sole explanation repeat at one locus. Restriction fragment length because the individual components of the ‘package’ polymorphism (RFLP) typing on 10 of these strains appear at sites that are widely separated in space further differentiated them into two distinct clusters. and time. CONCLUSION: Transmission of MDR-TB is occurring in the Western Province of PNG. Additional 15 Hasan AU, Suguri S, Fujimoto C, Itaki RL, Harada resources are urgently needed to interrupt the M, Kawabata M, Bugoro H, Albino B, Tsukahara ongoing transmission of MDR-TB from the Western T, Hombhanje F, Masta A. Province of PNG to the Torres Strait Islands. Good Phylogeography and dispersion pattern of Anopheles supervision and management of patient treatment, farauti sensu stricto mosquitoes in Melanesia. which includes ensuring a regular supply of second- Mol Phylogenet Evol 2008 Feb;46(2):792-800. line anti-TB drugs, are essential elements of TB control. 16 Hawkins VN, Auliff A, Prajapati SK, Rungsihirunrat K, Hapuarachchi HC, Maestre A, 13 Gutiérrez M, Suyama TL, Engene N, Wingerd JS, O’Neil MT, Cheng Q, Joshi H, Na-Bangchang K, Matainaho T, Gerwick WH. Sibley CH. Apratoxin D, a potent cytotoxic cyclodepsipeptide Multiple origins of resistance-conferring mutations from Papua New Guinea collections of the marine in Plasmodium vivax dihydrofolate reductase. cyanobacteria Lyngbya majuscula and Lyngbya Malar J 2008 Apr 28;7:72. sordida. BACKGROUND: In order to maximize the useful J Nat Prod 2008 Jun;71(6):1099-1103. Epub 2008 therapeutic life of antimalarial drugs, it is crucial to Apr 30. understand the mechanisms by which parasites Cancer cell toxicity-guided fractionation of resistant to antimalarial drugs are selected and extracts of the Papua New Guinea marine spread in natural populations. Recent work has cyanobacteria Lyngbya majuscula and Lyngbya demonstrated that pyrimethamine-resistance sordida led to the isolation of apratoxin D (1). conferring mutations in Plasmodium falciparum Compound 1 contains the same macrocycle as dihydrofolate reductase (dhfr) have arisen rarely de apratoxins A and C but possesses the novel 3,7- novo, but spread widely in Asia and Africa. The origin dihydroxy-2,5,8,10,10-pentamethylundecanoic acid and spread of mutations in Plasmodium vivax dhfr as the polyketide moiety. The planar structures and were assessed by constructing haplotypes based stereostructures of compound 1 were determined on sequencing dhfr and its flanking regions. 63 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 METHODS: The P. vivax dhfr coding region, 792 bp Malar J 2008 Apr 25;7:66. upstream and 683 bp downstream were amplified BACKGROUND: Aggregated haemozoin and sequenced from 137 contemporary patient crystals within malaria-infected erythrocytes confer isolates from Colombia, India, Indonesia, Papua New susceptibility of parasitized cells to a magnetic field. Guinea, Sri Lanka, Thailand, and Vanuatu. A repeat Here the utility of this method for diagnosis of human motif located 2.6 kb upstream of dhfr was also malaria is evaluated in a malaria-endemic region of sequenced from 75 of 137 patient isolates, and Papua New Guinea (PNG). METHODS AND mutational relationships among the haplotypes were FINDINGS: Individuals with Plasmodium falciparum visualized using the programme Network. malaria symptoms (n = 55) provided samples for RESULTS: Synonymous and non-synonymous conventional blood smear (CBS) and magnetic single nucleotide polymorphisms (SNPs) within the deposition microscopy (MDM) diagnosis. Standard dhfr coding region were identified, as was the well- Giemsa staining and light microscopy was performed documented in-frame insertion/deletion (indel). to evaluate all preparations. Plasmodium falciparum SNPs were also identified upstream and downstream parasitaemia observed on MDM slides was of dhfr, with an indel and a highly polymorphic repeat consistently higher than parasitaemia observed by region identified upstream of dhfr. The regions CBS for ring (CBS = 2.6 vs. MDM = 3.4%; t-test p- flanking dhfr were highly variable. The double mutant value = 0.13), trophozoite (CBS = 0.5 vs. MDM = (58R/117N) dhfr allele has evolved from several 1.6%; t-test p-value = 0.01), schizont (CBS = 0.003 origins, because the 58R is encoded by at least 3 vs. MDM = 0.1%; t-test p-value = 0.08) and different codons. The triple (58R/61M/117T) and gametocyte (CBS = 0.001 vs. MDM = 0.4%; t-test quadruple (57L/61M/117T/173F, 57I/58R/61M/117T p-value = 0.0002) parasitaemias. Gametocyte and 57L/58R/61M/117T) mutant alleles had at least prevalence determined by CBS compared to MDM three independent origins in Thailand, Indonesia, and increased from 7.3% to 45%, respectively. Papua New Guinea/Vanuatu. CONCLUSION: It was CONCLUSION: MDM increased detection sensitivity found that the P. vivax dhfr coding region and its of P. falciparum-infected, haemozoin-containing flanking intergenic regions are highly polymorphic erythrocytes from infected humans while maintaining and that mutations in P. vivax dhfr that confer detection of ring-stage parasites. Gametocyte antifolate resistance have arisen several times in prevalence five-fold higher than observed by CBS the Asian region. This contrasts sharply with the suggests higher malaria transmission potential in selective sweep of rare antifolate resistant alleles PNG endemic sites compared to previous estimates. observed in the P. falciparum populations in Asia and Africa. The finding of multiple origins of 20 Kayser M, Choi Y, van Oven M, Mona S, Brauer resistance-conferring mutations has important S, Trent RJ, Suarkia D, Schiefenhövel W, implications for drug policy. Stoneking M. The impact of the Austronesian expansion: evidence 17 Holden C. from mtDNA and Y chromosome diversity in the Evolutionary genetics. Polynesians took the express Admiralty Islands of Melanesia. train through Melanesia to the Pacific. Mol Biol Evol 2008 Jul;25(7):1362-1374. Epub 2008 Science 2008 Jan 18;319(5861):270. Apr 3. The genetic ancestry of Polynesians can be 18 Ichiyama T, Matsushige T, Siba P, Suarkia D, traced to both Asia and Melanesia, which presumably Takasu T, Miki K, Furukawa S. reflects admixture occurring between incoming Cerebrospinal fluid levels of matrix Austronesians and resident non-Austronesians in metalloproteinase-9 and tissue inhibitor of Melanesia before the subsequent occupation of the metalloproteinase-1 in subacute sclerosing greater Pacific; however, the genetic impact of the panencephalitis. Austronesian expansion to Melanesia remains J Infect 2008 May;56(5):376-380. Epub 2008 Apr largely unknown. We therefore studied the diversity 18. of nonrecombining Y chromosomal (NRY) and OBJECTIVES: To investigate the brain mitochondrial (mt) DNA in the Admiralty Islands, inflammation and damage in subacute sclerosing located north of mainland Papua New Guinea, and panencephalitis (SSPE), the cerebrospinal fluid updated our previous data from Asia, Melanesia, and (CSF) concentrations of matrix metalloproteinase-9 Polynesia with new NRY markers. The Admiralties (MMP-9) and tissue inhibitor of metalloproteinase-1 are occupied today solely by Austronesian-speaking (TIMP-1) were determined in SSPE patients. groups, but their human settlement history goes back METHODS: CSF MMP-9 and TIMP-1 levels were 20,000 years prior to the arrival of Austronesians measured in 23 patients with SSPE in Papua New about 3,400 years ago. On the Admiralties, we found Guinea by ELISA. RESULTS: CSF MMP-9 levels substantial mtDNA and NRY variation of both and MMP-9/TIMP-1 ratios of SSPE patients were Austronesian and non-Austronesian origins, with significantly higher than controls (p<0.001 and higher frequencies of Asian mtDNA and Melanesian p=0.005, respectively). There were no significant NRY haplogroups, similar to previous findings in differences in CSF TIMP-1 levels between SSPE Polynesia and perhaps as a consequence of patients and controls. CONCLUSIONS: Previous Austronesian matrilocality. Thus, the Austronesian studies suggested that CSF MMP-9 levels reflect language replacement on the Admiralties (and inflammatory damage to the brain. Our findings elsewhere in Island Melanesia and coastal New suggest that the MMP-9 level in CSF is an indicator Guinea) was accompanied by an incomplete genetic of inflammatory damage to the brain in SSPE. replacement that is more associated with mtDNA than with NRY diversity. These results provide 19 Karl S, David M, Moore L, Grimberg BT, Michon further support for the ‘Slow Boat’ model of P, Mueller I, Zborowski M, Zimmerman PA. Polynesian origins, according to which Polynesian Enhanced detection of gametocytes by magnetic ancestors originated from East Asia but genetically deposition microscopy predicts higher potential for mixed with Melanesians before colonizing the Plasmodium falciparum transmission. Pacific. We also observed that non-Austronesian 64 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 groups of coastal New Guinea and Island Melanesia indigenous Melanesians are genetically closer to had significantly higher frequencies of Asian mtDNA Asians than to Africans and European Americans. haplogroups than of Asian NRY haplogroups, Population structure analyses revealed that the suggesting sex-biased admixture perhaps as a Tongan population is genetically originated from consequence of non-Austronesian patrilocality. We Asians at 70% and indigenous Melanesians at 30%, additionally found that the predominant NRY which thus supports the so-called Slow Train model. haplogroup of Asian origin in the Admiralties (O- We also applied the SNP data to genome-wide scans M110) likely originated in Taiwan, thus providing the for positive selection by examining haplotypic first direct Y chromosome evidence for a Taiwanese variation and identified many candidates of locally origin of the Austronesian expansion. Furthermore, selected genes. Providing a clue to understand we identified a NRY haplogroup (K-P79, also found human adaptation to environments, our approach on the Admiralties) in Polynesians that most likely based on evolutionary genetics must contribute to arose in the Bismarck Archipelago, providing the first revealing unknown gene functions as well as direct link between northern Island Melanesia and functional differences between alleles. Conversely, Polynesia. These results significantly advance our this approach can also shed some light onto the understanding of the impact of the Austronesian invisible phenotypic differences between expansion and human history in the Pacific region. populations. 21 Kayser M, Lao O, Saar K, Brauer S, Wang X, 23 King CL, Michon P, Shakri AR, Marcotty A, Nürnberg P, Trent RJ, Stoneking M. Stanisic D, Zimmerman PA, Cole-Tobian JL, Genome-wide analysis indicates more Asian than Mueller I, Chitnis CE. Melanesian ancestry of Polynesians. Naturally acquired Duffy-binding protein-specific Am J Hum Genet 2008 Jan;82(1):194-198. binding inhibitory antibodies confer protection from Analyses of mitochondrial DNA (mtDNA) and blood-stage Plasmodium vivax infection. nonrecombining Y chromosome (NRY) variation in Proc Natl Acad Sci USA 2008 Jun 17;105(24):8363- the same populations are sometimes concordant but 8368. Epub 2008 Jun 3. sometimes discordant. Perhaps the most dramatic Individuals residing in malaria-endemic regions example known of the latter concerns Polynesians, acquire protective immunity after repeated infection in which about 94% of Polynesian mtDNAs are of with malaria parasites; however, mechanisms of East Asian origin, while about 66% of Polynesian Y protective immunity and their immune correlates are chromosomes are of Melanesian origin. Here we poorly understood. Blood-stage infection with analyze on a genome-wide scale, to our knowledge Plasmodium vivax depends completely on for the first time, the origins of the autosomal gene interaction of P. vivax Duffy-binding protein (PvDBP) pool of Polynesians by screening 377 autosomal with the Duffy antigen on host erythrocytes. Here, short tandem repeat (STR) loci in 47 Pacific Islanders we performed a prospective cohort treatment/ and compare the results with those obtained from reinfection study of children (5-14 years) residing in 44 Chinese and 24 individuals from Papua New a P. vivax-endemic region of Papua New Guinea Guinea. Our data indicate that on average about (PNG) in which children were cleared of blood-stage 79% of the Polynesian autosomal gene pool is of infection and then examined biweekly for reinfection East Asian origin and 21% is of Melanesian origin. for 25 weeks. To test the hypothesis that naturally The genetic data thus suggest a dual origin of acquired binding inhibitory antibodies (BIAbs) Polynesians with a high East Asian but also targeting PvDBP region II (PvDBPII) provide considerable Melanesian component, reflecting sex- protection against P. vivax infection, we used a biased admixture in Polynesian history in agreement quantitative receptor-binding assay to distinguish with the Slow Boat model. More generally, these between antibodies that merely recognize PvDBP results also demonstrate that conclusions based and those that inhibit binding to Duffy. The presence solely on uniparental markers, which are frequently of high-level BIAbs (>90% inhibition of PvDBPII- used in population history studies, may not Duffy binding, n = 18) before treatment was accurately reflect the history of the autosomal gene associated with delayed time to P. vivax reinfection pool of a population. diagnosed by light microscopy (p = 0.02), 55% reduced risk of P. vivax reinfection (hazard ratio = 22 Kimura R, Ohashi J, Matsumura Y, Nakazawa M, 0.45, p = 0.04), and 48% reduction in geometric Inaoka T, Ohtsuka R, Osawa M, Tokunaga K. mean P. vivax parasitemia (p < 0.001) when Gene flow and natural selection in oceanic human compared with children with low-level BIAbs (n = populations inferred from genome-wide SNP typing. 148). Further, we found that stable, high-level BIAbs Mol Biol Evol 2008 Aug;25(8):1750-1761. Epub 2008 displayed strain-transcending inhibition by reducing Jun 3. reinfection with similar efficiency of PNG P. vivax It is suggested that the major prehistoric human strains characterized by six diverse PvDBPII colonizations of Oceania occurred twice, namely, haplotypes. These observations demonstrate a about 50,000 and 4,000 years ago. The first settlers functional correlate of protective immunity in vivo are considered as ancestors of indigenous people and provide support for developing a vaccine against in New Guinea and Australia. The second settlers P. vivax malaria based on PvDBPII. are Austronesian-speaking people who dispersed by voyaging in the Pacific Ocean. In this study, we 24 Kuruppu S, Smith AI, Isbister GK, Hodgson WC. performed genome-wide single-nucleotide Neurotoxins from Australo-Papuan elapids: a polymorphism (SNP) typing on an indigenous biochemical and pharmacological perspective. Melanesian (Papuan) population, Gidra, and a Crit Rev Toxicol 2008;38(1):73-86. Polynesian population, Tongans, by using the Most of the medically important snakes in Papua Affymetrix 500K assay. The SNP data were analyzed New Guinea and Australia belong to the family together with the data of the HapMap samples Elapidae and are referred to as ‘Australo-Papuan’ provided by Affymetrix. In agreement with previous elapids. Neurotoxicity is often a life-threatening studies, our phylogenetic analysis indicated that symptom of envenoming by these snakes; therefore, 65 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 much attention has been paid to the isolation and for CQ and 54 microg/l (22, 106) for DECQ. detailed pharmacological and biochemical Estimated absolute and relative infant doses were characterization of the presynaptic (beta) and 34 microg/kg/day (7, 50) and 15 microg/kg/day (4, postsynaptic (alpha) neurotoxins from these elapid 26), and 2.3% (0.5, 3.6) and 1.0% (0.4, 2.0) for CQ venoms. These studies have highlighted the and DECQ (as CQ equivalents), respectively. potential for these toxins to be used as highly potent CONCLUSION: Infant exposure to CQ and DECQ and selective probes for biomedical research and, during pregnancy will be similar to that in the perhaps, the potential for their use as lead maternal circulation, and dependent on maternal compounds for the development of pharmaceutical dose and frequency. The median CQ + DECQ agents. Historically, the potency of neurotoxins/crude relative infant dose of 3.2% (as CQ equivalents) was venoms has been determined using murine LD50 low, confirming that use of CQ during lactation is (lethal dose) assays. However, a different rank order compatible with breastfeeding. of potency often results when crude venoms/toxins are ranked based on their in vitro pharmacological 26 Lusida MI, Nugrahaputra VE, Soetjipto, parameters (e.g., t90 values). The lack of Handajani R, Nagano-Fujii M, Sasayama M, neurotoxicity following envenoming by brown snakes, Utsumi T, Hotta H. despite the presence of a potent neurotoxin in their Novel subgenotypes of hepatitis B virus genotypes venom, has puzzled clinical toxinologists for years. C and D in Papua, Indonesia. This paradox also appears to include envenoming J Clin Microbiol 2008 Jul;46(7):2160-2166. Epub by the Stephen’s banded snake. Lastly, the in vitro 2008 May 7. studies examining the effectiveness of antivenoms Eight genotypes (A to H) and nine subtypes as well as the potential for alternative compounds (adw2, adw4, ayw1, ayw2, ayw3, ayw4, adrq+, adrq- to reverse/prevent neurotoxicity are discussed. This and ayr) of hepatitis B virus (HBV) have been review presents for the first time a detailed identified worldwide. They appear to be associated comparative analysis of the pharmacology and with geographical distribution, virological biochemistry of neurotoxins isolated from the characteristics, and possibly clinical outcomes. We Australo-Papuan elapids, placing emphasis on the performed sequence analysis of part of the S gene time taken for onset of action, receptor binding and the entire precore/core gene of HBV isolates parameters, reversibility, and the methods for obtained from HBsAg-positive blood donors in Papua determining potency. Province, Indonesia. Phylogenetic analysis of the S gene sequences revealed that 23 (85.2%) of the 25 Law I, Ilett KF, Hackett LP, Page-Sharp M, Baiwog 27 HBV isolates tested belonged to genotype C F, Gomorrai S, Mueller I, Karunajeewa HA, Davis (HBV/C) and 2 (7.4%) each to HBV/B and HBV/D. TM. Interestingly, 19 (82.6%) of the 23 isolates of HBV/ Transfer of chloroquine and desethylchloroquine C clustered in a branch that was distinct from the across the placenta and into milk in Melanesian previously reported subgenotypes C1 to C5 (HBV/ mothers. C1 to HBV/C5). Similarly, two isolates of HBV/D Br J Clin Pharmacol 2008 May;65(5):674-679. Epub clustered in a branch distinct from the reported 2008 Feb 15. subgenotypes HBV/D1 to HBV/D5. Phylogenetic WHAT IS ALREADY KNOWN ABOUT THIS analysis of the entire precore/core gene confirmed SUBJECT: The literature on placental and milk the consistent presence of the distinct branches in transfer of chloroquine and its major bioactive HBV/C and HBV/D. We therefore propose novel metabolite desethylchloroquine is sparse and subgenotypes designated HBV/C6 and HBV/D6. incomplete. WHAT THIS STUDY ADDS: We have The majority of HBV/C6 isolates in Papua had provided data on the transplacental transfer of alanine at positions 159 and 177 (A159/A177) in the chloroquine and desethylchloroquine in Melanesian HBsAg. A159/A177 is different from the women (n = 19), measured transfer of these drugs determinants for adrq+ (A159/V177), found into breast milk (n = 16) and estimated absolute and throughout Asia, and adrq- (V159/A177), found in relative infant doses for the breastfed infant. The New Caledonia and Polynesia, possibly representing data for desethylchloroquine are novel. In all three a unique antigenic group (provisionally referred to areas we have significantly increased both quantity as adrq indeterminate). In conclusion, we have and quality of the available database. AIMS: To identified two novel HBV subgenotypes, HBV/C6 and investigate the transfer of chloroquine and its major HBV/D6, the first of which is the most prevalent bioactive metabolite desethylchloroquine across the subgenotype of HBV in Papua, Indonesia. placenta and into breastmilk. METHODS: In Papua New Guinea, chloroquine (CQ; 25 mg base/kg) is 27 Marfurt J, Müller I, Sie A, Oa O, Reeder JC, Smith recommended for prophylaxis of malaria during TA, Beck HP, Genton B. pregnancy, and at the Alexishafen Health Centre The usefulness of twenty-four molecular markers in women are routinely prescribed CQ at the time of predicting treatment outcome with combination delivery. Fetal-cord and maternal serum samples therapy of amodiaquine plus sulphadoxine- were collected at delivery (n = 19) and milk samples pyrimethamine against falciparum malaria in Papua were collected from day 3 to day 17-21 after delivery New Guinea. (n = 16). CQ and its primary active metabolite Malar J 2008 Apr 19;7:61. desethylchloroquine (DECQ) were quantified by BACKGROUND: In Papua New Guinea (PNG), high-performance liquid chromatography. For both combination therapy with amodiaquine (AQ) or CQ and DECQ cord/maternal ratios (C/M) were chloroquine (CQ) plus sulphadoxine-pyrimethamine calculated to characterize placental transfer, and (SP) was introduced as first-line treatment against infant exposure via milk was estimated by standard uncomplicated malaria in 2000. METHODS: We methods. RESULTS: The median (interquartile assessed in vivo treatment failure rates with AQ+SP range) C/M was 1.1 (0.9, 1.6) for CQ and 1.2 (0.5, in two different areas in PNG and twenty-four 1.8) for DECQ. The average concentration in milk molecular drug resistance markers of Plasmodium over the time of sampling was 167 microg/l (27, 340) falciparum were characterized in pre-treatment 66 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 samples. The aim of the study was to investigate gradient antithetic to that of malaria incidence. We the association between infecting genotype and therefore suggest that Plasmodium falciparum may treatment response in order to identify useful have exerted a negative selection on this gene. This predictors of treatment failure with AQ+SP. might be due to a higher susceptibility to severe RESULTS: In 2004, Day-28 treatment failure rates forms of malaria, associated with HLA-B27 or other for AQ+SP were 29% in the Karimui and 19% in the close gene(s). In addition, we suggest here that the South Wosera area, respectively. The strongest same selective pressure that has contributed to independent predictors for treatment failure with reduce the HLA-B27 frequency in some regions has AQ+SP were pfmdr1 N86Y (OR = 7.87, p <0.01) favoured the fixing of newly generated B27 subtypes and pfdhps A437G (OR = 3.44, p <0.01). Mutations included in more advantageous HLA haplotypes. In found in CQ/AQ related markers pfcrt K76T, A220S, some cases, as for B*2709 in Sardinia and B*2706 N326D, and I356L did not help to increase the in Southeast Asia, these haplotypes may harbour predictive value, the most likely reason being that factors that protect from ankylosing spondylitis, an these mutations reached almost fixed levels. Though autoimmune disease strongly associated with HLA- mutations in SP related markers pfdhfr S108N and B27, thus offering a novel, powerful tool to dissect C59R were not associated with treatment failure, disease pathogenesis, and to identify additional they increased the predictive value of pfdhps A437G. genetic factors of susceptibility. The difference in treatment failure rate in the two sites was reflected in the corresponding genetic 30 McBride WJ, Hannah RC, Le Cornec GM, Bletchly profile of the parasite populations, with significant C. differences seen in the allele frequencies of mutant Cutaneous chancroid in a visitor from Vanuatu. pfmdr1 N86Y, pfmdr1 Y184F, pfcrt A220S, and Australas J Dermatol 2008 May;49(2):98-99. pfdhps A437G. CONCLUSION: The study provides A 23-year-old woman from Vanuatu presented evidence for high levels of resistance to the to an Australian hospital with a 3-week history of a combination regimen of AQ+SP in PNG and non-healing ulcer on the lower leg. A swab was indicates which of the many molecular markers submitted for a multiplex polymerase chain reaction analysed are useful for the monitoring of parasite designed to investigate genital ulcerative conditions. resistance to combinations with AQ+SP. Haemophilus ducreyi was detected and the gene product was subsequently sequenced, confirming 28 Matai S, Peel D, Wandi F, Jonathan M, Subhi R, the diagnosis of cutaneous chancroid. The lesion Duke T. responded to intramuscular benzathine penicillin. Implementing an oxygen programme in hospitals in This report adds further evidence that cutaneous Papua New Guinea. chancroid should be considered in the evaluation of Ann Trop Paediatr 2008 Mar;28(1):71-78. skin ulcers in the south Pacific. In Papua New Guinea (PNG), the most common cause of death among children under 5 years of age 31 Olsson DJ, Grant WD, Glick JM. is pneumonia. Children with severe pneumonia need Conjunctivitis outbreak among divers. antibiotics and oxygen but oxygen shortages are Undersea Hyperb Med 2008 May-Jun;35(3):169- common owing to the cost and complex logistics of 174. transporting it in cylinders. Detection of hypoxaemia In March 2006, an outbreak of conjunctivitis that using clinical signs can be difficult, especially in occurred over a six day period among twenty-nine highly pigmented children in whom cyanosis is individuals who partook in recreational scuba diving difficult to recognise. Pulse oximetry is the most trips on two boats off Vitu Levu Island, Fiji. We reliable, non-invasive way of detecting hypoxaemia. investigated the likelihood that a communal container However, most hospitals in PNG do not have pulse used to store diving masks facilitated the spread of oximetry. We proposed that the installation of a conjunctivitis among individuals. The diagnosis of reliable, sufficient and cheap supply of oxygen in conjunctivitis was based on clinical assessment by hospitals coupled with the use of pulse oximetry a physician. Transmission of conjunctivitis from would make a significant difference to child survival person to person was documented with eventual rates in PNG. Oxygen concentrators, which extract identification of the index case, the dive master, a oxygen from ambient air, were installed in the Fijian resident. Topical antibiotics were dispensed children’s wards of five hospitals during 2005. Pulse accordingly and detergent and bleach were used as oximeters were also introduced to enable better mask cleaning agents in an effort to control the detection of hypoxaemia. This paper describes the outbreak. Follow-up surveys were mailed to all technical aspects of this programme: the equipment twenty-nine participants. Ultimately, fourteen cases used and the rationale behind choosing it, the of conjunctivitis were documented (46.7%). Eleven installation, commissioning and testing processes. cases were verified during the six days in Fiji, two The ongoing training of clinical and engineering staff upon arrival back in the US, and one case of familial as well as two follow-up evaluations are described. transmission in the US. All but two cases resolved within one week. Unknown to these divers was a 29 Mathieu A, Cauli A, Fiorillo MT, Sorrentino R. coincidental, generalized outbreak of acute HLA-B27 and ankylosing spondylitis geographic haemorrhagic conjunctivitis among the Fijian distribution as the result of a genetic selection residents. The communal container used to store induced by malaria endemic? A review supporting diving masks was the likely vector for the spread of the hypothesis. infectious conjunctivitis, the first such documented Autoimmun Rev 2008 May;7(5):398-403. Epub 2008 outbreak involving communal diving equipment. Apr 9. The geographic distribution of HLA-B27 shows 32 Pauli J, Gundelach R, Vanelli-Rees A, Rees G, a latitude-related gradient inverse to that of malaria Campbell C, Dubey S, Perry C. endemic. An apparent exception occurs in New Juvenile nasopharyngeal angiofibroma: an Guinea, a region where malaria is present, but where immunohistochemical characterisation of the stromal HLA-B27 frequency shows, however, an orographic cell. 67 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Pathology 2008 Jun;40(4):396-400. Angers University Hospital (France). Ten patients AIMS: Juvenile nasopharyngeal angiofibroma showed very evocative lesions with a positive direct (JNA) is a rare tumour occurring almost exclusively examination, but T. concentricum was only isolated in young adult males. Although histologically benign, from three patients. Identification of the strains was it can be locally aggressive with a significant confirmed by sequencing of the internal transcribed recurrence rate. The finding of activating beta- spacer (ITS) regions. With the increase in catenin gene mutations in the stromal cells indicates international travel, one cannot disregard that this these are the neoplastic cells and supports the very rare species may be isolated by mycologists in association of JNA and familial adenomatous temperate areas from patients coming from endemic polyposis (FAP). Previous immunohistochemical foci. studies have demonstrated a null or focal myoepithelial immunophenotype in the stromal cells. 34 Poespoprodjo JR, Fobia W, Kenangalem E, Recently, expression of several growth factors and Lampah DA, Warikar N, Seal A, McGready R, oncoproteins including CD117 (c-kit) in the stromal Sugiarto P, Tjitra E, Anstey NM, Price RN. cells has been demonstrated. Our objective is to Adverse pregnancy outcomes in an area where evaluate the immunohistochemical phenotype of the multidrug-resistant Plasmodium vivax and stromal cell of JNA, particularly within the proliferative Plasmodium falciparum infections are endemic. zone of the tumour, by application of antibodies Clin Infect Dis 2008 May 1;46(9):1374-1381. against MNF116, CAM5.2, S-100, CD31, CD34, BACKGROUND: Plasmodium falciparum CD99, CD68, vimentin, EMA, SMA, desmin, infection exerts a considerable burden on pregnant calponin, Bcl-2 and (CD117) c-kit in a series of 54 women, but less is known about the adverse cases. METHODS: A routine immunohistochemical consequences of Plasmodium vivax infection. protocol was applied to representative paraffin METHODS: In Papua, Indonesia, where multiple sections of 54 JNAs collected from the Port Moresby drug resistance to both species has emerged, we General Hospital, Papua New Guinea, and Princess conducted a cross-sectional hospital-based study to Alexandra and Royal Brisbane Hospitals, quantify the risks and consequences of maternal Queensland, Australia. Immunoexpression of each malaria. RESULTS: From April 2004 through antigen was assessed in the stromal cells and the December 2006, 3046 pregnant women were vessels. RESULTS: The majority of stromal cells in enrolled in the study. The prevalence of parasitemia more than half of the cases demonstrated no staining at delivery was 16.8% (432 of 2570 women had with any of the 14 antibodies other than vimentin. infections), with 152 (35.2%) of these 432 infections Of 54 cases, 22 contained a microvascular being associated with fever. The majority of component (usually peripherally located and infections were attributable to P. falciparum (250 indicating the active growth front of the tumour) in [57.9%]); 146 (33.8%) of the infections were which the stromal cells demonstrated a hybrid attributable to P. vivax, and 36 (8.3%) were immunophenotype with both smooth muscle and coinfections with both species. At delivery, P. endothelial differentiation; c-kit was negative in all falciparum infection was associated with severe cases. CONCLUSIONS: The majority of stromal anemia (hemoglobin concentration, <7 g/dL; odds cells have an undifferentiated immunophenotype ratio [OR], 2.8; 95% confidence interval [95% CI], with no evidence of epithelial, myoid, endothelial or 2.0-4.0) and a 192 g (95% CI, 119-265) reduction in other lineage specific differentiation. In the mean birth weight (p<0.001). P. vivax infection was microvascular component the stromal cells appear associated with an increased risk of moderate able to show smooth muscle or endothelial anemia (hemoglobin concentration, 7-11 g/dL; OR, differentiation. No c-kit expression was identified. 1.8; 95% CI, 1.2-2.9; p=0.01) and a 108 g (95% CI, 17.5-199) reduction in mean birth weight (p<0.019). 33 Pihet M, Bourgeois H, Mazière JY, Berlioz- Parasitemia was associated with preterm delivery Arthaud A, Bouchara JP, Chabasse D. (OR, 1.5; 95% CI, 1.1-2.0; p=0.02) and stillbirth (OR, Isolation of Trichophyton concentricum from chronic 2.3; 95% CI, 1.3-4.1; p=0.007) but was not cutaneous lesions in patients from the Solomon associated with these outcomes after controlling for Islands. the presence of fever and severe anemia, suggesting Trans R Soc Trop Med Hyg 2008 Apr;102(4):389- that malaria increases the risk of preterm delivery 393. Epub 2008 Mar 4. and stillbirth through fever and contribution to severe Tinea imbricata, also known as ’tokelau’, is an anemia rather than through parasitemia per se. uncommon superficial mycosis caused by the CONCLUSIONS: These observations highlight the anthropophilic dermatophyte Trichophyton need for novel, safe, and effective treatment and concentricum. Cutaneous lesions appear prevention strategies against both multidrug- characteristically as scaly and concentric rings that resistant P. falciparum and multidrug-resistant P. may cover all parts of the body. Often acquired in vivax infections in pregnant women in areas of mixed childhood, tinea imbricata is a chronic disease and endemicity. lichenification is extremely common due to pruritus. The dermatophytosis mainly occurs in the South 35 Ricaut FX, Thomas T, Arganini C, Staughton J, Pacific, but also in some regions of Southeast Asia Leavesley M, Bellatti M, Foley R, Mirazon Lahr and Central or South America. Tinea imbricata M. usually affects people living in primitive and isolated Mitochondrial DNA variation in Karkar islanders. conditions. Mycological analysis is required for the Ann Hum Genet 2008 May;72(Pt 3):349-367. Epub diagnosis. The epidemiological and mycological 2008 Feb 28. study reported here took place in the Solomon We analyzed 375 base pairs (bp) of the first Islands from June to September 2006. Skin hypervariable region (HVS-I) of the mitochondrial scrapings were collected from 29 Melanesian DNA (mtDNA) control region and intergenic COII/ patients (aged 8 months to 58 years) with chronic tRNALys 9-bp deletion from 47 Karkar islanders cutaneous lesions and were analysed mycologically (north coast of Papua New Guinea) belonging to the in the Laboratory of Parasitology and Mycology of Waskia Papuan language group. To address 68 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 questions concerning the origin and evolution of this Plasmodium vivax. population we compared the Karkar mtDNA Antimicrob Agents Chemother 2008 Mar;52(3):1040- haplotypes and haplogroups to those of neighbouring 1045. Epub 2008 Jan 7. East Asians and Oceanic populations. The results In Papua, Indonesia, the antimalarial of the phylogeographic analysis show grouping in susceptibility of Plasmodium vivax (n = 216) and P. three different clusters of the Karkar islander mtDNA falciparum (n = 277) was assessed using a modified lineages: one group of lineages derives from the first schizont maturation assay for chloroquine, Pleistocene settlers of New Guinea-Island amodiaquine, artesunate, lumefantrine, mefloquine, Melanesia, a second set derives from more recent and piperaquine. The most effective antimalarial arrivals of Austronesian speaking populations, and against P. vivax and P. falciparum was artesunate, the third contains lineages specific to the Karkar with geometric mean 50% inhibitory concentrations islanders, but still rooted to Austronesian and New (IC50s) (95% confidence intervals [CI]) of 1.31 nM Guinea-Island Melanesia populations. Our results (1.07 to 1.59) and 0.64 nM (0.53 to 0.79), (i) suggest the absence of a strong association respectively. In contrast, the geometric mean between language and mtDNA variation and (ii) chloroquine IC50 for P. vivax was 295 nM (227 to reveal that the mtDNA haplogroups F1a1, M7b1 and 384) compared to only 47.4 nM (42.2 to 53.3) for P. E1a, which probably originated in Island Southeast falciparum. Two factors were found to significantly Asia and may be considered signatures of influence the in vitro drug response of P. vivax: the Austronesian population movements, are preserved initial stage of the parasite and the duration of the in the Karkar islanders but absent in other New assay. Isolates of P. vivax initially at the trophozoite Guinea-Island Melanesian populations. These stage had significantly higher chloroquine IC50s (478 findings indicate that the Karkar Papuan speakers nM [95% CI, 316 to 722]) than those initially at the retained a certain degree of their own genetic ring stage (84.7 nM [95% CI, 45.7 to 157]; p <0.001). uniqueness and a high genetic diversity. We present Synchronous isolates of P. vivax and P. falciparum a hypothesis based on archaeological, linguistic and which reached the target of 40% schizonts in the environmental datasets to argue for a succession of control wells within 30 h had significantly higher (partial) depopulation and repopulation and geometric mean chloroquine IC50s (435 nM [95% expansion events, under conditions of structured CI, 169 to 1118] and 55.9 nM [95% CI, 48 to 64.9], interaction, which may explain the variability respectively) than isolates that took more than 30 h expressed in the Karkar mtDNA. (39.9 nM [14.6 to 110.4] and 36.9 nM [31.2 to 43.7]; p <0.005). The results demonstrate the marked 36 Rogerson SJ, Carter R. stage-specific activity of chloroquine with P. vivax Severe vivax malaria: newly recognised or and suggest that susceptibility to chloroquine may rediscovered. be associated with variable growth rates. These PLoS Med 2008 Jun 17;5(6):e136. findings have important implications for the Comment on: PLoS Med 2008 Jun 17;5(6):e127 phenotypic and downstream genetic characterization (Genton et al.) and PLoS Med 2008 Jun 17;5(6):e128 of P. vivax. (Tjitra et al.). 39 Schoepflin S, Marfurt J, Goroti M, Baisor M, 37 Rothwell SP, Rosengren DJ. Mueller I, Felger I. Severe exercise-associated hyponatremia on the Heterogeneous distribution of Plasmodium Kokoda Trail, Papua New Guinea. falciparum drug resistance haplotypes in subsets of Wilderness Environ Med 2008 Spring;19(1):42-44. the host population. Exercise-associated hyponatremia is the most Malar J 2008 May 6;7:78. common medical complication of ultradistance BACKGROUND: The emergence of drug exercise and is usually caused by excessive resistance is a major problem in malaria control. For hypotonic fluid intake. We report a case of severe mathematical modelling of the transmission and hyponatremia in a healthy male trekking the Kokoda spread of drug resistance the determinant Trail in the remote Southern Highlands of Papua New parameters need to be identified and measured. The Guinea. A 43-year-old male collapsed and had a underlying hypothesis is that mutations associated generalized seizure in the afternoon of the third day with drug resistance incur fitness costs to the parasite of a guided trek. He was evacuated the following in absence of drug pressure. The distribution of drug morning and was found to have a serum sodium resistance haplotypes in different subsets of the host level of 107 mmol/L on arrival to hospital. The case population was investigated. In particular newly highlights that a high index of suspicion is required acquired haplotypes after radical cure were to identify patients with exercise-associated characterized and compared to haplotypes from hyponatremia. Early diagnosis and appropriate persistent infections. METHODS: Mutations management is critical to avoid the potentially fatal associated with antimalarial drug resistance were consequences of severe hyponatremia. The analysed in parasites from children, adults, and new diagnosis and treatment of exercise-associated infections occurring after treatment. Twenty-five hyponatremia is particularly challenging in the known single nucleotide polymorphisms from four remote Papua New Guinea jungle. Education of trek Plasmodium falciparum genes associated with drug leaders, medics, and trekkers in appropriate resistance were genotyped by DNA chip technology. preventative measures and the rapid treatment of RESULTS: Haplotypes were found to differ between exercise-associated hyponatremia is essential to subsets of the host population. A seven-fold mutated avoid recurrences of this life-threatening condition. haplotype was significantly reduced in adults compared to children and new infections, whereas 38 Russell B, Chalfein F, Prasetyorini B, parasites harbouring fewer mutations were more Kenangalem E, Piera K, Suwanarusk R, frequent in adults. CONCLUSION: The reduced Brockman A, Prayoga P, Sugiarto P, Cheng Q, frequency of highly mutated parasites in chronic Tjitra E, Anstey NM, Price RN. infections in adults is likely a result of fitness costs Determinants of in vitro drug susceptibility testing of of drug resistance that increases with number of 69 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 mutations and is responsible for reduced survival of of injury among girls. An intentional injury was mutant parasites. reported more often by those who had been bullied (OR 1.40-1.66, p <0.05), by regular smokers in Tonga 40 Singh PI, Carapetis JR, Buadromo EM, and Vanuatu (OR 1.52-2.21, p <0.05), and illicit drug Samberkar PN, Steer AC. users in Pohnpei and Vanuatu (OR 1.87-1.92, p The high burden of rheumatic heart disease found <0.05). CONCLUSION: Intentional injury was on autopsy in Fiji. reported extensively in these three populations. Cardiol Young 2008 Feb;18(1):62-69. Epub 2007 Interventions directed towards the school Dec 20. environment and which take into account the role of Rheumatic heart disease causes more than bullying and drug use need to be considered. 200,000 deaths worldwide annually, with the vast majority of these deaths occurring in developing 42 Steer AC, Jenney AJ, Oppedisano F, Batzloff MR, countries, yet there are few autopsy studies of Hartas J, Passmore J, Russell FM, Kado JH, rheumatic heart disease in these countries. We Carapetis JR. performed a retrospective review of 6218 autopsies High burden of invasive beta-haemolytic performed during the period from 1990 through 2006, streptococcal infections in Fiji. searching for cases of rheumatic heart disease Epidemiol Infect 2008 May;136(5):621-627. Epub based upon the macroscopic pathologic examination 2007 Jul 16. of the heart. We found 147 cases (2.4%) of We undertook a 5-year retrospective study of rheumatic heart disease. There was an apparent group A streptococcal (GAS) bacteraemia in Fiji, increase in the number of cases in the past 5 years. supplemented by a 9-month detailed retrospective There were 95 deaths that were directly attributable study of beta-haemolytic streptococcal (BHS) to rheumatic heart disease, with congestive cardiac infections. The all-age incidence of GAS failure being the most common cause of death in 75 bacteraemia over 5 years was 11.6/100,000. cases. The mean age at death due to rheumatic Indigenous Fijians were 4.7 times more likely to heart disease was 38 years. There were more cases present with invasive BHS disease than people of of rheumatic heart disease in Indigenous Fijians than other ethnicities, and 6.4 times more likely than Indo- Indo-Fijians, with an adjusted relative risk of 1.26 Fijians. The case-fatality rate for invasive BHS (95% confidence intervals from 0.87 to 1.86). Our infections was 28%. On 23 isolates emm-typing was findings reflect the high burden and early age of performed: 17 different emm-types were found, and death due to rheumatic heart disease in Fiji and the the emm-type profile was different from that found Pacific region generally, and underline the need for in industrialized nations. These data support the early detection and adequate secondary penicillin contentions that elevated rates of invasive BHS and prophylaxis in this region. GAS infections are widespread in developing countries, and that the profile of invasive organisms 41 Smith BJ, Phongsavan P, Bampton D, Peacocke in these settings reflects a wide diversity of emm- G, Gilmete M, Havea D, Chey T, Bauman AE. types and a paucity of types typically found in Intentional injury reported by young people in the industrialized countries. Federated States of Micronesia, Kingdom of Tonga and Vanuatu. 43 Tisch DJ, Bockarie MJ, Dimber Z, Kiniboro B, BMC Public Health 2008 Apr 30;8:145. Tarongka N, Hazlett FE, Kastens W, Alpers MP, BACKGROUND: Intentional injury presents a Kazura JW. threat to the physical and psychological well being Mass drug administration trial to eliminate lymphatic of young people, especially in developing countries, filariasis in Papua New Guinea: changes in which carry the greatest part of the global injury microfilaremia, filarial antigen, and Bm14 antibody burden. While the importance of this problem is after cessation. recognized, there are limited population data in low Am J Trop Med Hyg 2008 Feb;78(2):289-293. and middle income countries that can guide public Laboratory tools to monitor infection burden are health action. The present study investigates the important to evaluate progress and determine prevalence and distribution of intentional injury endpoints in programs to eliminate lymphatic among young people in three Pacific Island societies, filariasis. We evaluated changes in Wuchereria and examines behavioural and psychosocial factors bancrofti microfilaria, filarial antigen and Bm14 related to risk of intentional injury. METHODS: antibody in individuals who participated in a five-year Population surveys were conducted with students mass drug administration trial in Papua New Guinea. aged 11-17 years in Pohnpei State in the Federated Comparing values before treatment and one year States of Micronesia (n = 1495), the Kingdom of after four annual treatments, the proportion of Tonga (n = 2808) and Vanuatu (n = 4474). Surveys microfilaria positive individuals declined to the measured self-reported injury and intentional injury, greatest degree, with less marked change in sources of intentional injury, and the range of antibody and antigen rates. Considering children behavioural, psychological, educational and social as sentinel groups who reflect recent transmission variables that may be related to injury risk. intensity, children surveyed before the trial were more RESULTS: Among boys and girls aged 14-17 years frequently microfilaria and antibody positive than the respective period prevalence of intentional injury those examined one year after the trial stopped. In was 62% and 56% in Pohnpei, 58% and 41% in contrast, antigen positive rates were similar in the Tonga, and 33% and 24% in Vanuatu. The two groups. All infection indicators continued to prevalence of intentional injury declined with age in decline five years after cessation of mass drug Tonga and Vanuatu, but there was little evidence of administration; Bm14 antibody persisted in the an age-trend in Pohnpei. Across the three societies, greatest proportion of individuals. These data the major sources of intentional injury among boys suggest that Bm14 antibody may be a sensitive test were ‘other persons’ followed by boyfriends/ to monitor continuing transmission during and after girlfriends and fathers. Mothers, boyfriends/ mass drug administration aimed at eliminating girlfriends and other persons were primary sources transmission of lymphatic filariasis. 70 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 44 Tjitra E, Anstey NM, Sugiarto P, Warikar N, Same-zone villages spoke different languages: one Kenangalem E, Karyana M, Lampah DA, Price Austronesian and three Papuan (Arapesh, Abelam, RN. and Boiken). Our analysis examined whether Multidrug-resistant Plasmodium vivax associated language or geography better predicted gene flow. with severe and fatal malaria: a prospective study in In general, language better predicted genetic Papua, Indonesia. affinities. Boiken villages across all four zones PLoS Med 2008 Jun 17;5(6):e128. showed no significant genetic difference (F(ST) p BACKGROUND: Multidrug-resistant value >0.05). In contrast, the Austronesian village Plasmodium vivax (Pv) is widespread in eastern was significantly different to most other villages (p Indonesia, and emerging elsewhere in Asia-Pacific <0.05). Only the mountains and coast showed zonal and South America, but is generally regarded as a gene flow (p >0.05). We interpret the data to reflect benign disease. The aim of the study was to review limited gene flow inland by Austronesians the spectrum of disease associated with malaria due overshadowed by a regional displacement by inland to Pv and P. falciparum (Pf) in patients presenting to Boiken speakers migrating seaward. These results a hospital in Timika, southern Papua, Indonesia. are consistent with oral histories and ethnographic METHODS AND FINDINGS: Data were accounts. prospectively collected from all patients attending the outpatient and inpatient departments of the only 46 Wadsworth JD, Joiner S, Linehan JM, hospital in the region using systematic data forms Desbruslais M, Fox K, Cooper S, Cronier S, and hospital computerised records. Between Asante EA, Mead S, Brandner S, Hill AF, Collinge January 2004 and December 2007, clinical malaria J. was present in 16% (60,226/373,450) of hospital Kuru prions and sporadic Creutzfeldt-Jakob disease outpatients and 32% (12,171/37,800) of inpatients. prions have equivalent transmission properties in Among patients admitted with slide-confirmed transgenic and wild-type mice. malaria, 64% of patients had Pf, 24% Pv, and 10.5% Proc Natl Acad Sci USA 2008 Mar 11;105(10):3885- mixed infections. The proportion of malarial 3890. Epub 2008 Mar 3. admissions attributable to Pv rose to 47% (415/887) Kuru provides our principal experience of an in children under 1 year of age. Severe disease epidemic human prion disease and primarily affected was present in 2634 (22%) inpatients with malaria, the Fore linguistic group of the Eastern Highlands with the risk greater among Pv (23% [675/2937]) of Papua New Guinea. Kuru was transmitted by the infections compared to Pf (20% [1570/7817]; odds practice of consuming dead relatives as a mark of ratio [OR] = 1.19 [95% confidence interval (CI) 1.08- respect and mourning (transumption). To date, 1.32], p = 0.001), and greatest in patients with mixed detailed information of the prion strain type infections (31% [389/1273]); overall p <0.0001. propagated in kuru has been lacking. Here, we Severe anaemia (haemoglobin < 5 g/dl) was the directly compare the transmission properties of kuru major complication associated with Pv, accounting prions with sporadic, iatrogenic, and variant for 87% (589/675) of severe disease compared to Creutzfeldt-Jakob disease (CJD) prions in Prnp-null 73% (1144/1570) of severe manifestations with Pf transgenic mice expressing human prion protein and (p <0.001). Pure Pv infection was also present in in wild-type mice. Molecular and neuropathological 78 patients with respiratory distress and 42 patients data from these transmissions show that kuru prions with coma. In total 242 (2.0%) patients with malaria are distinct from variant CJD and have transmission died during admission: 2.2% (167/7722) with Pf, properties equivalent to those of classical (sporadic) 1.6% (46/2916) with Pv, and 2.3% (29/1260) with CJD prions. These findings are consistent with the mixed infections (p = 0.126). CONCLUSIONS: In hypothesis that kuru originated from chance this region with established high-grade chloroquine consumption of an individual with sporadic CJD. resistance to both Pv and Pf, Pv is associated with severe and fatal malaria particularly in young 47 Wall JD, Cox MP, Mendez FL, Woerner A, children. The epidemiology of P. vivax needs to be Severson T, Hammer MF. re-examined elsewhere where chloroquine A novel DNA sequence database for analyzing resistance is increasing. human demographic history. Genome Res 2008 Aug;18(8):1354-1361. Epub 2008 45 Vilar MG, Kaneko A, Hombhanje FW, Tsukahara May 20. T, Hwaihwanje I, Lum JK. While there are now extensive databases of Reconstructing the origin of the Lapita Cultural human genomic sequences from both private and Complex: mtDNA analyses of East Sepik Province, public efforts to catalog human nucleotide variation, PNG. there are very few large-scale surveys designed for J Hum Genet 2008;53(8):698-708. Epub 2008 May the purpose of analyzing human population history. 23. Demographic inference from patterns of SNP The colonization of Oceania occurred in two variation in current large public databases is waves. By 32,000 BP, humans had reached New complicated by ascertainment biases associated Guinea and settled all intervisible islands east to the with SNP discovery and the ways that populations Solomon Islands. Around 3,500 BP, a distinct and regions of the genome are sampled. Here, we intrusive group from Southeast Asia reached coastal present results from a resequencing survey of 40 New Guinea, integrated their components with independent intergenic regions on the autosomes indigenous resources, and gave rise to the Lapita and X chromosome comprising approximately 210 Cultural Complex. Within 2,500 years, Lapita and kb from each of 90 humans from six geographically its descendant cultures colonized the Pacific. To diverse populations (i.e., a total of approximately 18.9 uncover the origin of the Lapita Cultural Complex, Mb). Unlike other public DNA sequence databases, we analyzed the hypervariable region I of the we include multiple indigenous populations that serve mitochondrial deoxyribonucleic acid (mtDNA) in 219 as important reservoirs of human genetic diversity, individuals from eight East Sepik Province villages: such as the San of Namibia, the Biaka of the Central two villages in each of four environmental zones. African Republic, and Melanesians from Papua New 71 Papua New Guinea Medical Journal Volume 51, No 1-2, Mar-Jun 2008 Guinea. In fact, only 20% of the SNPs that we find strengthened cataract services, although its effect are contained in the HapMap database. We identify on any cataract backlog is unknown; developed a several key differences in patterns of variability in diabetic eye disease diagnosis and treatment our database compared with other large public service, but its reach and effectiveness are unknown; databases, including higher levels of nucleotide provided accessible comprehensive eye care, but diversity within populations, greater levels of its effect on the prevalence of vision impairment is differentiation between populations, and significant unknown; and established medical records and data differences in the frequency spectrum. Because collection systems, but these need more attention. variants at loci included in this database are less DISCUSSION: This programme achieved much. likely to be subject to ascertainment biases or linked However, the evaluation highlighted the limitations to sites under selection, these data will be more of inadequate project design and that, without useful for accurately reconstructing past changes in addressing further human resource development size and structure of human populations. and the Ministry of Health’s wavering financial commitment, there are potential risks to ongoing 48 Warrell DA. services. That revenue generating capacity was not Unscrupulous marketing of snake bite antivenoms incorporated into this programme may prove to be a in Africa and Papua New Guinea: choosing the right flaw that will limit ongoing access to eye care, product – ‘what’s in a name?’. especially in rural areas. Trans R Soc Trop Med Hyg 2008 May;102(5):397- 399. Epub 2008 Mar 21. 50 Zhang X, Perugini MA, Yao S, Adda CG, Murphy Snake bite envenoming, mainly caused by the VJ, Low A, Anders RF, Norton RS. saw-scaled or carpet viper (Echis ocellatus), is a Solution conformation, backbone dynamics and lipid neglected disease of West Africa. Specific interactions of the intrinsically unstructured malaria antivenoms can save life and limb but, for various surface protein MSP2. reasons, supply of these essential drugs to Africa J Mol Biol 2008 May 23;379(1):105-121. Epub 2008 has dwindled to less than 2% of estimated Mar 28. requirements. Other problems include Merozoite surface protein 2 (MSP2), one of the maldistribution, inadequate conservation and most abundant proteins on the surface of the inappropriate clinical use of antivenoms. In the face merozoite stage of Plasmodium falciparum, is a of this crisis, several promising new antivenoms have potential component of a malaria vaccine, having been developed. However, some dangerously shown some efficacy in a clinical trial in Papua New inappropriate products of Indian origin are being Guinea. MSP2 is a GPI-anchored protein consisting marketed by unscrupulous manufacturers or of conserved N- and C-terminal domains and a distributors in Africa and Papua New Guinea, with variable central region. Previous studies have shown disastrous results. A major source of confusion is that it is an intrinsically unstructured protein with a labelling antivenom with ambiguous snake names high propensity for fibril formation, in which the that fail to distinguish the Asian species whose conserved N-terminal domain has a key role. venoms are used in their production from the local Secondary structure predictions suggest that MSP2 snakes whose venoms are antigenically dissimilar. contains long stretches of random coil with very little alpha-helix or beta-strand. Circular dichroism 49 Williams C, Szetu JL, Ramke J, Palagyi A, du Toit spectroscopy confirms this prediction under R, Brian G. physiological conditions (pH 7.4) and in more acidic Evaluation of the first 5 years of a national eye health solutions (pH 6.2 and 3.4). Pulsed field gradient programme in Vanuatu. NMR diffusion measurements showed that MSP2 Clin Experiment Ophthalmol 2008 Mar;36(2):162- under physiological conditions has a large effective 167. hydrodynamic radius consistent with an intrinsic pre- PURPOSE: To evaluate against its objectives molten globule state, as defined by Uversky. This the achievements of the first 5 years of a national was supported by sedimentation velocity studies in eye health programme in Vanuatu. METHODS: the analytical ultracentrifuge. NMR resonance Programme clinical activity data were collated from assignments have been obtained for FC27 MSP2, surgical logs, clinic and outreach reports, and patient allowing the residual secondary structure and register books. Cataract surgical outcomes were backbone dynamics to be defined. There is some retrieved from monitoring software. Programme motional restriction in the conserved C-terminal annual reports provided information about region in the vicinity of an intramolecular disulfide management, infrastructure improvements, bond. Two other regions show motional restrictions, equipment supplied, repaired or replaced, the supply both of which display helical structure propensities. and use of consumables, and human resource One of these helical regions is within the conserved development and deployment. Costs were determined from project budgets and acquittals. N-terminal domain, which adopts essentially the RESULTS: The programme promoted eye health, same conformation in full-length MSP2 as in including through the integration of eye care into corresponding peptide fragments. We see no existing health services; established adequate evidence of long-range interactions in the full-length facilities, at referral hospitals, provincial hospitals and protein. MSP2 associates with lipid micelles, but rural health clinics, with equipment and manpower predominantly through the N-terminal region rather to provide eye care appropriate to the location; than the C terminus, which is GPI-anchored to the established a primary eye care programme; membrane in the parasite. 72 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$100 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. Figures to be sent as separate referees’ comments as part of the peer review jpeg or tiff images. Do not paste the images into process. Word. Photographs should be glossy prints, either 7 cm or 14.5 cm in width. Original Articles: Reports of original and new Photomicrographs should have internal scale investigations or contributions. markers. Each table should have a heading and footnotes which make it understandable without Brief Communications and Case Reports: reference to the text. Each figure should have a Contents similar to that of original articles but legend; figure legends should be typed together text should be no more than a total of 4 Journal on a separate sheet. 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 double spacing. Manuscripts should not be right SUBSCRIPTIONS AND ADVERTISEMENTS justified. Each paper should include an informative Summary, Introduction, Patients/ Communications relating to advertisements or Materials and Methods, Results, Discussion and subscriptions should be addressed to the Journal References. The title page should include the as above. Matters related to the Society should title, full names of all authors, names and be addressed to the Medical Society of Papua addresses of institutions where the work has New Guinea, PO Box 6665, Boroko, NCD 111, been done and full present address of the first Papua New Guinea. or corresponding author. Subscriptions: Members of the Medical Society References should be kept to a minimum, and of Papua New Guinea receive the Journal as part must be in the Vancouver style. Authors should of their annual subscription. Others may check all references against the original source. subscribe and should contact the subscription Sample references are shown below. secretary for a price. Papua New Guinea Medical Journal Volume 51, Number 1-2, March-June 2008 CONTENTS EDITORIAL Treatment of uncomplicated malaria in Papua New Guinea: what should be done? T.M.E. Davis 1 ORIGINAL ARTICLES Knowledge and attitudes about infant feeding among nulliparous and parous women in Port Moresby: a comparative study D. Frank, P. Ripa, J.D. Vince and N. Tefuarani 5 A review of the current state of malaria among pregnant women in Papua New Guinea I. Mueller, S. Rogerson, G.D.L. Mola and J.C. Reeder 12 A case-control study of VDRL-positive antenatal clinic attenders at the Port Moresby General Hospital Antenatal Clinic and Labour Ward to determine outcomes, sociodemographic features and associated risk factors G.D.L. Mola, A. Golpak and A.B. Amoa 17 Variations of the middle thyroid vein in Papua New Guinean Melanesians D.J. Hasola, G. Gende and O. Liko 27 A focused ethnography about treatment-seeking behaviour and traditional medicine in the Nasioi area of Bougainville J.E. Macfarlane 29 Primary repair of colonic injuries at the Kundiawa and Madang General Hospitals, Papua New Guinea J. Kuzma and J. Jaworski 43 Barriers to the delivery of the hepatitis B birth dose: a study of five Papua New Guinean hospitals in 2007 S.G Downing, W. Lagani, R. Guy and M. Hellard 47 The prevalence of HIV infection in women attending antenatal clinics in Fiji C.H. Washington, L.M. Singer, T. McCaig, L. Tikoduadua, S.T. Ali, J. Fong, J. Luveni, T.O. Kyaw-Myint, S. Watson and F. Russell 56 MEDLARS BIBLIOGRAPHY 60