Annals of the ACTM V15 N1 May 2014

ANNALS OF THE ACTM AN INTERNATIONAL JOURNAL OF TROPICAL & TRAVEL MEDICINE

IN THIS ISSUE

The physical health impacts of tropical cyclones

Vitamin A, Zinc, Iron and Iodine Interventions in Children Under Five Years of Age

Hearing Loss in Malaria

TB or not TB: The clinical and diagnostic challenge of suspected pericardial tuberculosis

An unusual case of Q fever in an immunised sheep officer ISSN 1448-4706

Official Journal of The Australasian College of Tropical Medicine Volume 15 Number 1 May 2014

Volume 15 Number 1 1 ANNALS OF THE ACTM 1

ANNALS OF THE ACTM AN INTERNATIONAL JOURNAL OF TROPICAL & TRAVEL MEDICINE

CONTENTS MAY 2014 Officers of The Australasian College of Editorial Tropical Medicine Changing of the editorial guard! President Professor John McBride Peter A Leggat �� 1 President Elect

Associate Professor Peter Nasveld Vice President Review articleS Dr Kym Daniell The physical health impacts of tropical cyclones Honorary Secretary Dr Colleen Lau Rob D Mitchell, Peter Aitken and Richard C Franklin �� 2 Honorary Treasurer Professor Peter A. Leggat, AM Council Members Vitamin A, zinc, iron and iodine interventions in children under five years of age Associate Professor Geoff Quail, OAM, Professor Marc Benjamin Wood �� 8 Shaw, Emeritus Professor Richard Speare, AM, Professor Bart Currie, Associate Professor David Porter, Dr John Heydon Hearing loss in malaria Chair, Faculty of Travel Medicine Professor Peter A. Leggat AM (Acting) Eric Fong �� 11 Chair, Faculty of Expedition and Wilderness Medicine Professor Marc Shaw Chairs of Standing Committees CASE REPORT Professor Tim Inglis (Disaster Health) TB or not TB: The clinical and diagnostic challenge of suspected pericardial tuberculosis Dr Richard S. Bradbury (Medical Parasitology & Zoonoses) Associate Professor John Frean (Publications) Victoria G Hall �� 16 Dr Ken D. Winkel (Toxinology)

Secretariat CASE REPORT ACTM Secretariat, PO Box 123, Red Hill QLD 4059 AUSTRALIA An unusual case of Q fever in an immunised sheep officer Tel: +61-7-3872-2246 Frank C H Ng and Peter A Leggat �� 19 Fax: +61-7-3856-4727 Email: [email protected] Website: http://www.tropmed.org

Editorial Board ANNALS OF THE ACTM Editor-in-Chief Associate Professor John Frean Emeritus Editor-in-Chief Professor John M. Goldsmid Professor Derek Smith Executive Editor Professor Peter A. Leggat, AM ACTM Newsletter Editor ACTM Newsletter Wayne Pederick Board Members and Review Panel Dr Irmgard Bauer, Emeritus Professor Roderick SF Campbell, AM, Professor David Durrheim, Dr Michael Humble, Associate Professor Tim Inglis, Cover photo: The Australian Institute of Tropical Medicine in 1916 (photo courtesy of James Cook University) Professor Ahmed Latif, Dr Alan Mills, Professor John H. Pearn, AO, RFD, Dr Ken D. Winkel © Copyright 2014 ACTM

Material published in the Annals of the ACTM is covered by copyright and all rights are reserved, excluding “fair use”, as permitted under copyright law. Permission to use any material published in the Annals of the ACTM should be obtained in writing from the authors and Editorial board. EDITORIAL Changing of the Editorial Guard!

It is with great pleasure to have the opportunity to (Issue 1, 2009) to celebrate 15 years of the Annals (1995-2009). With introduce and formally welcome Associate Professor Associate Professor John Frean taking over the helm as Editor-in-Chief, John Frean (see image at left) as the new Editor-in- the College conferred the title of Editor-in-Chief Emeritus for his dedicated Chief of the Annals of the ACTM. John Frean is in the service to the Annals over a period of 5 years. Derek Smith continues National Institute for Communicable Diseases, and to hold senior appointments on editorial boards of several journals, Associate Professor, University of the Witwatersrand, particularly in the field of occupational and environmental health and safety. Johannesburg, South Africa. John Frean has played a number of key roles in the College, including being a member of the ACTM Professor Derek Smith is the second Editor-in-Chief to be conferred the College Council. In his new role, he also chairs the Publication Committee title of Editor-in-Chief Emeritus. Derek succeeded Emeritus Professor John of the College, which has the Annals as its flagship publication, now in its Goldsmid, who had transitioned the Annals from an irregularly published 14th Volume. monograph series to a peer-reviewed journal. John Goldsmid also ensured that the Annals was published in full colour publication by the end of 2007, Associate Professor John Frean takes over the reins of which encouraged the use of colour images. Following his retirement, the Annals from Professor Derek Smith (see image at Emeritus Professor John Goldsmid was confirmed as the inaugural Editor- right) from the University of Newcastle, Australia. Derek in-Chief Emeritus in 2007, but continued to make contributions to various Smith was appointed Editor-in-Chief of the Annals of College publications. A College award, the John Goldsmid Award, has been the ACTM in 2008. Under Derek’s stewardship, two dedicated to John’s contributions to the journal and is awarded to the best issues of the Annals were published each year in January paper published in the Annals each year. and June/July commencing with Volume 9 (2008) through to Volume 13 (2012). Abstracts from various ACTM symposia Professor Peter A. Leggat, AM and the Townsville Hospital Week were also included in one of the issues, Executive Editor, Annals of the ACTM, and Head, School of Public Health, particularly in more recent years. Derek Smith also produced the first Tropical Medicine and Rehabilitation Sciences, James Cook University, special issue of the Annals and also a consolidated index in Volume 10 Townsville, Queensland, Australia.

Public Health & Tropical Medicine At the Anton Breinl Centre

Public Health and Tropical Medicine at the Anton Breinl Centre seeks to undertake high quality and relevant teaching, research and training in population health, with a special focus on tropical Australia and our neighbours. Postgraduate study programs: • Public Health and Tropical Medicine • Public Health • Disaster and Refugee Health • Aeromedical Retrieval • Communicable Disease Control • Tropical Medicine and Hygiene • Biosecurity and Disaster Preparedness • Health Promotion • Travel Medicine For further information: Phone: +61 7 4781 5836 Email: [email protected] Web: www.jcu.edu.au/phtmrs 33898_JCU1033_MAKADS

Volume 15 Number 1 1 ANNALS OF THE ACTM 1 REVIEW ARTICLE

The physical health impacts of tropical cyclones

Rob D Mitchell,1 Peter Aitken,1,2 and Richard C Franklin2

1 Department of Emergency Medicine, Townsville Hospital, Townsville, Queensland, Australia

2 School of Public Health, Tropical Medicine & Rehabilitation Sciences, James Cook University, Townsville, Queensland, Australia

ABSTRACT

Tropical cyclones, also known as typhoons and hurricanes, are low-pressure weather systems characterised by extremely fast rotational winds. They disproportionately affect developing countries and have the potential to seri- Cyclone Yasi crossing the Queensland coast, February 2011 ously disrupt basic societal functions. Like other natural disasters, cyclones Satellite image originally processed by the Bureau of Meteorology from the geostationary also impact on health. This literature review considers the epidemiology of meteorological satellite MTSAT-2 operated by the Japan Meteorological Agency. these events in terms of trauma, communicable disease and non-communicable disease. The human and environmental impacts of cyclone activity are the result of strong winds, heavy rainfall and storm surge, which often lead to landslides Trauma is a significant cause of morbidity and mortality, although mecha- and flooding.1,5,6 In Australia, severity is rated using a five-category system nisms vary between events. Both drowning and building collapse can lead based on wind velocity.7 Category four and five cyclones are expected to to large numbers of fatalities. Minor injuries represent the primary cause of result in significant structural damage. Different rating systems are used in- morbidity in the post-impact phase but carbon monoxide poisoning from ternationally (such as the Saffir/Simpson Scale in North America) and vary inappropriate generator use is emerging as an important risk in both adult principally according to how wind speed is measured (Figure 1).7,8 and paediatric patients. While communicable disease outbreaks are uncom- Cyclones have the potential to cause significant morbidity and mortality. mon in developed settings, gastroenteritis, vector-borne disease and acute They are estimated to have contributed to 1.9 million deaths in the last two respiratory illness pose particular problems in developing communities. centuries, and approximately 250,000 in the period 1980-2000.2,9 Their ef- Non-communicable disease constitutes a considerable proportion of health- fects on health can be both direct (as a result of severe weather forces) and care attendances following cyclones, and exacerbations as well as treatment indirect (as a result of infrastructure damage, environmental devastation, disruptions of chronic diseases are increasingly being recognised as a cause population displacement and economic hardship). Historically, developing of cyclone-related morbidity and mortality. nations of the Asia-Pacific region have experienced the greatest absolute and proportionate mortality from tropical cyclones and other natural disasters.2 Although Australia has well-developed public health infrastructure, major cyclones still have the potential to cause significant human suffering. This In Australia, cyclones in the last 150 years are estimated to have caused review will assist tropical communities in planning for these severe weather approximately 2,000 deaths, with a financial cost in excess of 6 trillion dol- 10 events. Implicit in its findings is the importance of injury prevention strate- lars. Over half of these deaths were caused by just five events occurring in gies, communicable disease surveillance and provision for managing chron- a 50-year period from the late 1800s to the early 1900s (Figure 2). Cyclone ic illnesses. Mahina (Cooktown, 1899) led to over 400 fatalities, representing the sixth highest death toll from any disaster in Australia since 1850.10 In more recent memory, Cyclone Tracy (Darwin, 1974) resulted in 71 deaths and approxi- Keywords: cyclones, hurricanes, typhoons, trauma, disease, communicable, 11 non-communicable mately 650 injuries. Although cyclone-related morbidity and mortality in the last two decades has Ann Australas Coll Trop Med 2014;15(1):2-8. been relatively small,10 recent events have illustrated the ongoing potential for harm. Cyclone Yasi, a category five system, brought winds in excess of 285 km/h when it crossed the Queensland coast near Mission Beach in Background February 2011.12 Although Queensland escaped without a major impact on health, the cyclone still led to at least one death13 and a record number of at- Tropical cyclones, also known as typhoons and hurricanes, are low-pressure tendances to Townsville Hospital Emergency Department (personal commu- weather systems that develop over warm ocean waters. They are character- nication, P Aitken). The event also prompted the evacuation of both of Cairns’ ised by extremely fast rotational winds that are generally in excess of 100 hospitals (public and private), necessitating the transportation of over 350 km/h.1 Approximately 119 million people are affected by an average of 80-90 patients to Brisbane (approximately 1700 km south).14,15 individual cyclone systems each year,2,3 making them one of the most common natural disasters worldwide.4 The threat posed by tropical cyclones is unlikely to abate. As a consequence of global warming and rising sea temperatures, the frequency and intensity As cyclogenesis requires warm ocean waters, cyclones typically occur of extreme weather events is expected to increase.16-18 Consistent with this, between the latitudes of 30 degrees north and 30 degrees south. Regions upward trends have been observed in the estimated lifetime-maximum wind within 500 kilometres of the equator are spared because cyclonic winds ro- speeds of the very strongest tropical cyclones (above the 99th percentile) tate in different directions in the two hemispheres. As a consequence of this over each ocean basin.19 distribution, cyclones predominantly affect developing communities. Of 84 nations with a history of recurrent exposure to cyclonic events, only a small Given the ongoing threat of tropical cyclones in northern Australia and the minority (including Australia, New Zealand, Japan and the United States of Asia-Pacific,1,2,7 it is appropriate to consider their potential impact on hu- America) are high-income countries.2 man health. While severe cyclone activity is known to have psychological

2 ANNALS OF THE ACTM May 2014 Tropical Cyclone Severity Categories

10-min 30 40 50 60 70 80 90 100 110 120 130 Mean Wind Knots 1-min 40 50 60 70 80 90 100 110 120 130 140 150

Australian Categories ONE TWO THREE FOUR FIVE US Saffir Simpson

40 60 80 100 120 140 160 180 Knots Max Gust 80 100 120 140 160 180 200 220 240 260 280 300 320 340 Km/h

Source: Australian Government Bureau of Meteorology (http://www.bom.gov.au/cyclone/faq/index.shtml#definitions)

Figure 1 Tropical cyclone severity categories

sequelae,1 the consequences for mental health are beyond the scope of this paper. This literature review builds on a comprehensive analysis by Shultz et al published in 20051 and includes a specific focus on epidemiological data from recent events. It has been designed to capture the significant amount of literature produced in the wake of Hurricane Katrina (United States of America, 2005). Methods PubMed was keyword and MeSH term searched for ‘health’ with each of ‘cyclone’, ‘hurricane’ and ‘typhoon’. Results were limited to human data, English language and the last seven years (April 2005 – March 2012), as the 2005 review by Shultz et al1 was assumed to have included relevant peer- reviewed literature up until its publication date. This broad search strategy identified 813 articles, most of which (62%) did not directly address the review question: 180 were only concerned with mental health, 246 were unrelated to health impacts or cyclones and 82 were editorials or opinion pieces. The abstracts of the remaining papers were re- Figure 2 Cyclone-related deaths in Australia 1871-2010 viewed and a majority were found to be qualitative in nature. Full text was Source: Emergency Management Australia obtained for the relevant articles and bibliographies of key references were scanned for applicable articles, book chapters and grey literature (Figure 3).

The 105 papers identified using this strategy were considered alongside the

813 identified primary references quoted by Shultz et al (where accessible). They identified using search 151 relevant studies and reports, including those related to mental health.1 strategy The results of this search confirm that the volume of literature related to trop-

180 excluded for 246 excluded for ical cyclones has grown significantly since 2005. In keeping with historical relating only to being unrelated to experience, most of the recent papers are retrospective descriptive studies. mental health health or tropical cyclones In the discussion below, prominence has been given to more recent data. Where studies referred to the same data set, only one has been referenced. 82 excluded for

being editorials or Findings are presented in the categories of trauma, communicable disease commentary and non-communicable disease.

305 reviewed for

relevance and new Discussion data Trauma

200 excluded for Trauma constitutes the majority of cyclone-related morbidity, and may occur being qualitative, 1,6 commentary or pre-impact, at impact or post-impact. Pre-impact injury tends to be the re- primarily mental sult of preparation efforts, including reinforcing buildings, preparing genera- health related tors and cleaning outdoor areas. Impact-related trauma relates to the direct impacts of destructive winds, heavy rain and storm surge, as well as the 105 included flooding and landslides that may result. Potentially harmful consequences of

these forces include structural collapse, wind-blown debris, falling trees and Figure 3. Search strategy downed power-lines. Post-impact trauma is largely a consequence of clean- Figure 3 Search strategy up related activities as well as the loss of normal public services. Injury

Volume 15 Number 1 3 ANNALS OF THE ACTM 3

36 can be associated with utilities damage and the subsequent requirements Major injuries for temporary measures (including stored water and power generators).1,6 Life-threatening injuries may also result from tropical cyclones and tend to Morbidity and mortality related to each of these stages varies according to occur during the impact phase. For instance, following Cyclone Ike in the US, the degree of development. Deaths tend to be highest in the impact phase in 47 (64%) of all deaths resulted from injuries, compared with 23 (31%) from developed settings, whereas developing communities tend to be affected to illnesses. Nine percent of overall deaths were caused by falling trees.44 Fol- a greater extent in the wake of the storm front.1,6,20-22 lowing Katrina, 25% of deaths were attributed to injury and trauma, and older people were disproportionately affected.29 Motor vehicle accidents, chain- Data from Hurricane Katrina found that certain groups are at higher risk of saw injuries, electrocutions and blunt trauma in the post-impact phase have trauma (and other health impacts) following cyclones. These include elderly, also been reported.1,5,29,44 homeless, poor, unemployed and non-English speaking citizens as well as certain racial and ethnic groups.23-29 A recent study has identified that many In a developing context, collapsed buildings were the most frequent cause of of these groups are less likely to receive emergency broadcasts and evacua- death following Typhoon Rananim in China (2004). Flying debris, collapsing tion orders, placing them at higher risk of trauma.30 buildings and motor vehicle collisions together caused 46% of all significant injuries. In that event, staying outdoors and not receiving (or ignoring) the Drowning typhoon alert were found to be statistically significant risk factors for serious Drowning has traditionally been the major cause of cyclone-related deaths, injury.36 and typically results from storm surge and flooding in the immediate wake Carbon monoxide poisoning of the event.31 In recent times, the risk of drowning has been mitigated by the introduction of advanced warning, evacuation and shelter systems.1,5,32 Cyclone-related carbon monoxide poisoning was first described more than twenty years ago,45 but is increasingly reported as a major cause of morbid- Despite these developments, drowning still constitutes a significant pro- ity.33,37,45-51 A majority of cases relate to the use of generators following power portion of cyclone-related mortality. In the United States of America (US), outages. 40% of the 971 deaths directly attributed to Hurricane Katrina (2005) were thought to result from drowning.29 An empirical relationship between the Following Cyclone Ike, thirteen deaths were attributed to CO poisoning and at depth of flooding water and mortality has been described.26 Following Hur- least fifteen individuals required treatment with hyperbaric therapy.33,46 It was ricane Ike (2008), drowning constituted only 7% of overall deaths but was estimated that 82-87% of exposures were the result of improper generator the most frequent cause of those directly related to the storm.33 Similar rates use, and children were disproportionately affected.46,47 At one hospital, 54% have been seen following other US events.34 of 37 individuals presenting with CO poisoning were under the age of 18, and 75% of these resulted from generator-powered television or video-game Although limited data is available for developing settings, there are reports use.47 In Australia, the one death attributed to Cyclone Yasi (2011) was re- of drowning causing large numbers of fatalities following cyclones in Ban- portedly the result of asphyxiation from inappropriate generator use.13 gladesh, Mexico and the Philippines.1,35 On the contrary, however, Typhoon Ranamin in China (2004) resulted in no deaths due to drowning but a signifi- Communicable diseases cant number as a result of building collapse.36 Infectious disease outbreaks are not a major cause of morbidity following Minor injuries tropical cyclones.1,52-54 As with other disasters, the risk of outbreaks is associated with the size, health status and living conditions of the affected popu- Minor injuries have been observed to constitute over 70% of presentations lation. Crowding, inadequate water and sanitation, as well as poor access to following tropical cyclones in developed countries.1,6,21,34 Most emergency health services, increase the risk of disease transmission.52,53,55,56 For these department visits occur when weather conditions have stabilised.22,37 reasons, outbreaks tend to occur in developing countries far more frequently A 1993 review of cyclone-related injuries found significant variation in the than they do in developed environments and almost always manifest in the patterns between events, as well inconsistency in the classification and cod- post-impact phase.1,55,56 ing of injuries.6 In recent events, a wide spectrum of injuries has again been Diarrhoeal disease observed, including puncture wounds, lacerations, abrasions, contusions, sprains and fractures.34 Following Hurricane Katrina, the commonest pre- Diarrhoeal disease is the commonest infectious consequence of cyclone sentations were for lacerations, blunt trauma and puncture wounds29 and the activity and is frequently the only reported communicable challenge in de- most frequently reported mechanisms were falls and cuts.38 In other settings veloped countries.1,52 A variety of causative pathogens have been reported, (eg, Cyclone Gonu in Oman, 2007) orthopaedic injuries and lacerations have which differ between settings based on local endemicity.56 predominated.39 Outbreaks of gastroenteritis have been documented following some cyclones Although lower limb injuries were common among Hurricane Katrina evacu- in the US.57-62 Most recently, surveillance systems detected a norovirus out- ees in the Houston Astrodome/Reliant Park Complex site,40 this has not been break in a large evacuation centre in the aftermath of Hurricane Katrina.57-59 At the experience with other US events.34 Following Hurricane Andrew (1992), a temporary medical facility erected at the Reliant Park Complex site, 1,173 the most common body parts affected in non-fatal injuries were the upper patients were treated for gastroenteritis, representing 17% of all clinic visits extremities, including the fingers, hands, and arms.34,41 A recent review of and 4% of all patients who resided there for nine days.58 On the peak days, all cyclone-related injuries presenting to emergency departments in Hong 40% of all paediatric presentations were for diarrhoea.57 While not all disease Kong over a six-year period supports this observation. In that study, the head shelters experienced outbreaks, 61 sporadic cases of Salmonella spp. and (33.5%) and upper limbs (32.5%) were most often involved, with falls the toxigenic and non-toxigenic Vibrio cholerae were also reported.62 commonest mechanism of injury (42.6%).21.40 Together, these data confirm Diarrhoeal disease outbreaks have been reported with greater frequency that injury patterns following tropical cyclones are highly variable. in developing countries.52,53,63-67 Recent examples include Cyclone Aila, in A recent study from the US has found that older citizens who were displaced West Bengal, India (2008),65,66 hurricanes Gustav and Ike in Cuba (2008)68 from their homes had an increased fracture risk extending to 12 months and Cyclone Nargis in Burma (2008).67 In the latter, recently published data following Hurricane Katrina.42 This persisted after controlling for other risk describes a four-fold increase in severe watery diarrhoea (suspected chol- factors. Rates of accelerant-related burn injuries have also been noted to era) in the immediate wake of the event. The overall incidence of diarrhoea increase following hurricane seasons in the US, which is possibly related to increased from 571 per 100,000 persons per year in 2007 to 799 in the the burning of cyclone-generated debris.43 post-Nargis months of 2008. Rates of dysentery also peaked sharply but had no impact on mortality.67 Following Aila, increases in diarrhoeal rates of

4 ANNALS OF THE ACTM May 2014 between 30 and 60% were seen in affected districts. Vibrio cholerae (54%) diseases by changing vector distribution and breeding sites. As with other was the commonest agent, followed by Shigella spp. (5%).66 In Cuba, at least communicable diseases, the subsequent risk of outbreaks is compounded a 10% increase in gastroenteritis was seen following cyclones Gustav and by crowding, loss of shelter, weakened public health systems and the in- Ike.68 More recently, reports have emerged of increased diarrhoeal disease terruptions of ongoing control programs.83 Outbreaks can only occur if the following typhoon events in the Philippines.69 disease is endemic in the region prior to the event.84 Acute respiratory infections While increases in vector population following cyclones have been documented (examples include Belize post-Alma in 2008 and Mexico post-Isidore As with other natural disasters and situations involving displaced persons in 2002),85,86 the extent to which this has translated to disease is variable. and overcrowded shelters, cyclones carry an increased risk of acute respira- While cases of malaria have peaked sharply in some settings,87 they have tory infection.52,53,70 Like diarrhoeal illness, developing settings are dispro- been seen to decline in others.85 Recently-published data describes an in- portionately affected. crease in malaria cases following Cyclone Nargis, but it is uncertain if this is The most recently published data describes a peak in the incidence of acute a reflection of normal seasonal variation or a post-disaster phenomenon.67 respiratory infections (ARI) following Cyclone Nargis. A rate of 4,042 per Some have hypothesised that, in certain settings, the risk of arboviral infec- 100,000 persons per year among children under 5 increased to 7,280 in tions might surpass the risk from other vector-borne diseases.85 This is sup- 2008 following the event; it then fell back to 4,662 in 2009.67 More profound ported by evidence of significant spikes in dengue cases following Cyclone increases were seen in Nicaragua in the 30 days following Hurricane Mitch Mitch.84,87 (1998), where a four-fold rise in incidence was reported.63 Evidence of vector-borne disease in developed settings is slim. While there There are few reports of epidemic ARI following cyclones in developed set- were concerns about a potential outbreak of West Nile virus following Hur- tings. While some data suggested an increased number of adult presenta- ricane Katrina, this did not occur to a significant extent.49,88 One group has tions with acute respiratory infection in the wake of Hurricane Katrina, the reported that no mosquitoes collected from the Gulf Coast in the six weeks trend was attributed to multiple cases among a single National Guard bat- following that event had evidence of arbovirus infection.89 There is, however, talion.71 There was, however, a significant spike in the number of cases of data to support an increase in the incidence of animal bites and stings among ARI among children and adolescents, which was thought to be associated exposed persons.1,48,54 with increased environmental exposures.72 Firefighters exposed to floodwater reported increased rates of upper respiratory tract infection symptoms.73 Leptospirosis Wound infections Leptospirosis is a zoonotic disease that has potential to cause significant morbidity, primarily as a result of acute renal failure. It has been commonly Despite the theoretical risk of wound infections following cyclonic activity, described following cyclones in developing settings.55,69,90 For example, one there is little published data to support this. This may reflect reporting bias. study reports that 19.2% of tested individuals in four flooded villages had Following Hurricane Katrina, primary care providers noted a large number of serological evidence of active leptospirosis following a cyclone in Orissa, wound infections on the lower limbs, but the severity has not been quanti- India (1999).91 A significant outbreak was seen in Taiwan following Typhoon fied.74 After Hurricane Ivan hit Grenada in 2004, there was an increase in Nali (2001)90 and, more recently, speculative evidence has emerged of an the number of patients admitted to surgical facilities as a result of wound epidemic following consecutive events in the Philippines in 2009.69 There are complications and diabetic foot infections.75 no reports of leptospirosis outbreaks in developed settings. A single Centers for Disease Control and Prevention report describes 18 cas- Other infections es of Vibrio spp.-infected wounds following Hurricane Katrina. Most patients The review by Shultz et al revealed that a variety of other communicable had associated co-morbidities, and many had been wading in flood-waters. disease outbreaks have been documented, including balantidiasis and ty- Five of these individuals subsequently died.76 In the wake of the same event, phoid.1,55 These have not been reported in more recent literature. one hospital receiving evacuees identified a link between the presence of a wound and colonisation with multi-drug resistant organisms, but the nature Of note, some infections associated with other natural disasters have not of the association remains unclear.77 been demonstrated to follow tropical cyclones. These include tetanus, hepatitis A and hepatitis E, which have been reported in the wake of tsunamis Sexually transmitted infections and inland flooding.92 Their absence in the cyclone literature may represent A small number of reports from the US suggest that the incidence of sexu- publication bias. There are no recent reports on the impact on tuberculosis, ally transmitted infections may increase following cyclones.78,79 This has not except from one post-Katrina study which demonstrated no adverse conse- been reported in developing regions affected by cyclones,67 but may reflect quences despite a short period of treatment program closure.93 the integrity of surveillance systems. Although there is a theoretical risk of mould-associated infections following Following Katrina, the prevalence of gonorrhoea in high school students in- water inundation, there is little data to suggest that this risk has ever been creased from 2.3% (8/346) to 5.1% (17/333). Although of weak statistical realised to a significant degree. Presentations of chromoblastomycosis were significance, there was an independent positive trend toward testing positive noted following Hurricane Ike94 and mould was included amongst a list of for gonorrhoea after the hurricane. The exact reason for this is not yet under- environmental exposures thought to be associated with increased ARI post stood, and the results may represent sampling error.78 Katrina.72 This link has been questioned, however.95 There is also emerging data from the US about adverse impacts on human In another study following Hurricane Katrina, homes with greater flood dam- immunodeficiency virus (HIV).79-82 One study has reported a link between age demonstrated higher levels of mould growth compared with homes with Katrina-associated post-traumatic stress disorder (PTSD) and viral replica- little or no flooding; however, no increase in the occurrence of adverse health tion. Those with PTSD were more likely than those without PTSD to have outcomes was observed.96 In one paediatric sample, there was an overall detectable plasma viral loads and CD4 cell counts <200/mm3 two years post- decrease in mould exposure and respiratory symptoms following the event.97 disaster. In addition, they were more likely to have had anti-retroviral inter- Following Typhoon Morakot in southern Taiwan (2009), Hsu et al found in- ruptions as a consequence of the event.79 Robinson et al have also recently creased Aspergillus spp. in both indoor and outdoor environments in the suggested that residential displacement might impact on CD4 counts.80 immediate wake of the cyclone.98 These effects are unlikely to be cyclone specific. Non-communicable disease Vector-borne diseases More is being published about the impact of tropical cyclones on non-com- Cyclones have the potential to impact on the transmission of vector-borne municable disease (NCD) in terms of chronic illness, acute exacerbations

Volume 15 Number 1 5 ANNALS OF THE ACTM 5 and new diagnoses. This is in keeping with a trend towards increased em- interrupted by the event.118 One small study has identified an increase in con- phasis on NCD in the broader disaster and public health literature. genital cleft palate pathology. Goenjian et al report a statistically significant increase in the cases beginning 9 months after the cyclone.119 An environ- The significant, overall impact of tropical cyclones on NCD has recently been mental cause has been hypothesised. illustrated by a retrospective study that sought to quantify the number of direct and indirect deaths resulting from hurricanes Charley, Frances, Ivan Hurricane Katrina may have impacted on birth rates. The total number of and Jeanne in Florida, 2004. There was an elevated mortality for up to two births in selected hurricane-effected counties decreased 19 percent in the months following each hurricane, resulting in a total of 624 direct and in- 12 months after Hurricane Katrina compared with the 12 months before.120 direct deaths. In contrast, the official count was 31 direct and 113 indirect The decrease was not consistent across all areas however, which may reflect deaths resulting from the four hurricanes combined. Indirect mortality ac- population shifts.115 counted for most deaths, and was primarily due to heart disease (34%) and Ischaemic heart disease cancer (19%). Diabetes (5%) and accident-related deaths (9%) accounted for a smaller but not insignificant percentage of the elevated mortality.99 Although increased rates of ischaemic heart disease have been associated The elevated mortality were deaths that would not have otherwise occurred with other natural disasters,121 only recently has data emerged specifically within that hurricane season. linking acute myocardial infarction (AMI) and tropical cyclones. A single-centre retrospective cohort study following Hurricane Katrina found a three-fold Pre-existing illness increased incidence in AMI after the event. In the two years post-cyclone, In developed settings, pre-existing illness accounts for a significant propor- there were 246 admissions for AMI (2.18% of all presentations) compared tion of health care visits in the aftermath of events; following hurricanes with 150 admissions (0.71%) prior to the event.122 The authors suggest that Andrew and Katrina, greater than 33% of presentations were for chronic dis- changes in caseload, patient demographics and health service availability ease.20,100,101 In the case of the latter, the majority of visits were for endocrine, only partly account for this significant discrepancy. It has been hypothesised cardiovascular, and psychiatric disorders.102 Evacuees were noted to have a that AMI may be a manifestation of chronic stress121-123 but increased to- high burden of disease and most medications dispensed to displaced per- bacco use and medical non-compliance may also contribute.123 At a different sons were for chronic conditions.103-105 institution, there was no impact on cardiothoracic operative mortality despite a significant increase in demand.124 Cyclones have the potential to directly precipitate exacerbations of pre-existing illness as well as disrupt treatment regimes. A preliminary report of Cancer mortality associated with Hurricane Charley suggested that at least six of Recent analysis has confirmed that a significant number of patients recently the atraumatic deaths were the direct result of exacerbated cardiac or respi- diagnosed with cancer were potentially displaced following Hurricane Ka- ratory conditions.106 Surveys conducted following Hurricane Katrina found trina.125 Decreased access to care led to delayed presentations and treatment that 20.6% of survivors with chronic disease cut back or terminated their of patients with head and neck malignancies, but the clinical significance of treatment because of the event. Disruptions in treatment were significantly this finding is unknown.126 more common among the non-elderly, uninsured, socially isolated, those with housing needs and for conditions remaining relatively asymptomatic Other conditions but still dangerous if untreated. Disruptions were frequently the result of A significant increase in alcohol use and binge drinking was observed fol- problems accessing doctors and medications.107 This data is congruent with lowing Hurricane Katrina. This correlated with the number of exposures to the profound disruption to primary and secondary health services that fol- hurricane-related traumatic events.127 In a cohort of evacuees, females and lowed Katrina,108,109 which included the closure of 94 dialysis facilities along younger persons were found to be at increased risk of excess alcohol and to- the Gulf Coast110,111 and the interruption of specialist HIV/AIDS clinics.81 bacco use.128 In the wake of Hurricane Rita, alcohol use was seen to increase There is emerging evidence regarding the impact of treatment disruptions among a small group of adolescents displaying symptoms of post-traumatic 129 following cyclones. For instance, Katrina led to an increase in HbAIC for sub- stress. sections of the cyclone-affected community with diabetes112 and negatively Preliminary data from New Orleans also suggests that hurricanes Katrina and impacted on the provision of chronic pain management.113 Among patients Rita may have altered the redistribution of certain environmental contami- requiring haemodialysis, 44% of patients reported missing at least one ses- nants (including lead), potentially leading to an exposure risk for children in sion and almost 17% reported missing three or more. The adjusted odds highly-contaminated areas.130-132 ratio of hospitalisation among the latter group compared to those who did not miss any dialysis sessions was 2.16.114 Conclusion Until recently, there has been limited data in support of a direct link between Cyclones are severe weather events that occur relatively frequently in tropical mortality and cyclone-related disruptions to public health services. While regions. They disproportionately affect developing countries and have the deaths due to exacerbations of chronic disease have previously been attrib- potential to cause serious disruption to basic societal functions. Given the uted to cyclone events, there has been a lack of clarity about the causative ongoing threat of tropical cyclones in northern Australia and the Asia-Pacific factors.29,99,102,106 The significant number of deaths among hospitalised and region, there is value in identifying their potential impacts on health. chronically ill patients following Hurricane Katrina has helped quantify the This review was designed to expand on previous analyses by focussing on impact of health system disruption.29 more recent literature, most of which has been generated in the US following Peri-natal issues Hurricane Katrina. The expanded volume of literature has provided a much more detailed picture of the health impact of tropical cyclones, particularly A 2009 literature review found that, following Katrina, obstetric, prenatal, in relation to developed countries. It has also confirmed that there is a large and neonatal care were compromised, but increases in adverse birth out- degree of variability in the causes of morbidity and mortality between events comes (such as preterm delivery, low birth-weight and maternal complica- and settings. tions) were mostly limited to women seriously impacted by the cyclone. They concluded that, with a few specific exceptions, post-disaster concerns and Trauma remains a significant cause of both, although mechanisms vary health outcomes for pregnant and postpartum women were similar to those between events. Both drowning and building collapse can lead to large of other people exposed to the event.115 numbers of fatalities. Minor injuries represent the primary cause of morbidity in the post-impact phase but carbon monoxide poisoning from inap- Other studies have also failed to identify a persisting negative impact on propriate generator use is emerging as an important risk in both adult and infant and child mortality from Hurricane Katrina,116 despite speculation to paediatric patients. While communicable disease outbreaks are uncommon the contrary.117 Routine screening of newborns was, however, temporarily in developed settings, gastroenteritis, vector-borne disease and ARI pose

6 ANNALS OF THE ACTM May 2014 particular problems in developing communities. Non-communicable disease 41� McNabb SJ, Kelso KY, Wilson SA, McFarland L, Farley TA. Hurricane Andrew-related injuries and illnesses, Louisiana, 1992. South Med J 1995;88:615-8. constitutes a considerable proportion of healthcare attendances following 42� Uscher- Pines L, Vernick JS, Curriero F, Lieberman R, Burke TA. Disaster-related injuries in the period of cyclones, and exacerbations as well as treatment disruptions of chronic dis- recovery: the effect of prolonged displacement on risk of injury in older adults. J Trauma 2009;67:834-40. 43� r ainey S, Cruse CW, Smith JS, Smith KR, Jones D, Cobb S. The occurrence and seasonal variation of accel- eases are increasingly being recognised as a cause of cyclone-related mor- erant-related burn injuries in central Florida. J Burn Care Res 2007;28:675-80. bidity and mortality. 44� ragan P, Schulte J, Nelson SJ, Jones KT. Mortality surveillance: 2004 to 2005 Florida hurricane-related deaths. Am J Forensic Med Pathol 2008;29:148-53. Although Australia has well-developed public health infrastructure, any cy- 45� Hampson NB, Stock AL. Storm-related carbon monoxide poisoning: lessons learned from recent epidemics. Undersea Hyperb Med 2006;33:257-63. clone event that features destructive winds and rain, population displace- 46� Centers for Disease Control and Prevention. Carbon monoxide exposures after hurricane Ike - Texas, Septem- ment and interruption of basic societal functions has the potential to im- ber 2008. MMWR Morb Mortal Wkly Rep 2009;58:845-9. 47� Fife CE, Smith LA, Maus EA, et al. Dying to play video games: carbon monoxide poisoning from electrical pact on human health. Data from the US following Hurricane Katrina has generators used after Hurricane Ike. Pediatrics 2009;123:e1035-8. demonstrated that developed settings are not immune. The results of this 48� Forrester MB. Impact of Hurricane Ike on Texas poison center calls. Disaster Med Public Health Prep 2009;3:151-7. review will assist tropical communities in preparing for the threats to health 49� institute of Medicine (US). Roundtable on environmental health sciences and medicine. Environmental public posed by tropical cyclones. Implicit in its findings is the importance of injury health impacts of disasters: Hurricane Katrina: workshop summary. Washington; 2007. 50� Cukor J, Restuccia M. Carbon monoxide poisoning during natural disasters: the Hurricane Rita experience. J prevention strategies, communicable disease surveillance and provision for Emerg Med 2007;33:261-4. managing chronic illnesses. 51� Audin C. Carbon monoxide poisoning following a natural disaster: a report on Hurricane Rita. J Emerg Nurs 2006;32:409-11. 52. Ivers LC, Ryan ET. Infectious diseases of severe weather-related and flood-related natural disasters. Curr Opin References Infect Dis 2006;19:408-14. 53� Watson JT, Gayer M, Connolly MA. Epidemics after natural disasters. Emerg Infect Dis 2007;13:1-5. 1� Shultz JM, Russell J, Espinel Z. Epidemiology of tropical cyclones: the dynamics of disaster, disease, and 54� perez E, Thompson P. Natural hazards: causes and effects. Lesson 5 -tropical cyclones (hurricanes, typhoons, development. Epidemiologic Reviews 2005;27:21-35. baguios, cordonazos, tainos). Prehosp Disaster Med 1995;10:202-15; quiz 16-7. 2� United Nations Development Programme. Reducing disaster risk: a challenge for development. New York: 55� Toole MJ. Communicable disease and disease control. In: Noji EK, ed. The Public Health Consequences of United Nations Development Programme; 2004. Disasters. New York: Oxford University Press; 1997. 3� Henderson-Sellers A, Zhang H, Berz G, et al. Tropical cyclones and global climate change: a post-IPCC assess- 56� Watkins RR. Gastrointestinal infections in the setting of natural disasters. Curr Infect Dis Rep 2012;14:47-52. ment. Bulletin of the American Meteorological Society 1998;79:19-38. 57� Centers for Disease Control and Prevention. Norovirus outbreak among evacuees from hurricane Katrina - 4� international Federation of the Red Cross and Red Crescent Societies. World Disasters Report. Geneva; 2001. Houston, Texas, September 2005. MMWR Morb Mortal Wkly Rep 2005;54:1016-8. 5. Keim M. Cyclones, Tsunamis, and human health: the key role of preparedness. Oceanography 2006;19:40-9. 58� yee EL, Palacio H, Atmar RL, et al. Widespread outbreak of norovirus gastroenteritis among evacuees of Hur- 6� Noji EK. Analysis of medical needs during disasters caused by tropical cyclones: anticipated injury patterns. J ricane Katrina residing in a large ‘megashelter’ in Houston, Texas: lessons learned for prevention. Clin Infect Trop Med Hyg 1993;96:370-6. Dis 2007;44:1032-9. 7� Tropical cyclone intensity and impacts. (Accessed 5 September 2011, at http://www.bom.gov.au/cyclone/ 59� Murray KO, Kilborn C, DesVignes-Kendrick M, et al. Emerging disease syndromic surveillance for Hurricane about/intensity.shtml.) Katrina evacuees seeking shelter in Houston’s Astrodome and Reliant Park Complex. Public Health Rep 8� Hurricane Research Division. 2010. (Accessed May 19, 2012, at http://www.aoml.noaa.gov/hrd/tcfaq/ 2009;124:364-71. D1.html.) 60� Waring SC, Reynolds KM, D’Souza G, Arafat RR. Rapid assessment of household needs in the Houston area 9� Nicholls R, Mimura N, Topping J. Climate change in south and south-east Asia: Some implications for coastal after Tropical Storm Allison. Disaster Manag Response 2002:3-9. areas. Journal of Global Environmental Engineering 1995;1:137-54. 61� Cavey AM, Spector JM, Ehrhardt D, et al. Mississippi’s infectious disease hotline: a surveillance and education 10. EMA Disasters Database Report. (Accessed 22 April 2012, at http://www.ema.gov.au ) model for future disasters. Prehosp Disaster Med 2009;24:11-7. 11. Arthur C, Schofield A, Cechet B. Assessing the impacts of tropical cyclones. Australian Journal of Emergency 62� Centers for Disease Control and Prevention. Infectious disease and dermatologic conditions in evacuees and Management 2008;23:14-20. rescue workers after Hurricane Katrina - multiple states, August-September, 2005. MMWR Morb Mortal Wkly 12� Severe Tropical Cyclone Yasi. (Accessed 13 September 2011, at http://www.bom.gov.au/cyclone/history/yasi. Rep 2005;54:961-4. shtml.) 63� Campanella N. Infectious diseases and natural disasters: the effects of Hurricane Mitch over Villanueva munici- 13. AAP. Man in cyclone zone suffocates. Sydney Morning Herald 2011 February 4 2011. pal area, Nicaragua. Public Health Rev 1999;27:311-9. 14. Woods CE, Goodman D, Mills J, Usher K, McBride WJ. Weather to evacuate? Med J Aust 2011;195:712-3. 64. McCurry J. Burma: after the cyclone. Lancet 2008;371:1737-8. 15� Little M, Stone T, Stone R, et al. The evacuation of Cairns Hospitals due to Severe Tropical Cyclone Yasi. 65� Bhunia R, Ghosh S. Waterborne cholera outbreak following Cyclone Aila in Sundarban area of West Bengal, Academic Emergency Medicine 2012 (in press). India, 2009. Trans R Soc Trop Med Hyg 2011;105:214-9. 16� Webster PJ, Holland GJ, Curry JA, Chang HR. Changes in tropical cyclone number, duration, and intensity in 66� panda S, Pati KK, Bhattacharya MK, Koley H, Pahari S, Nair GB. Rapid situation & response assessment of a warming environment. Science 2005;309:1844-6. diarrhoea outbreak in a coastal district following tropical cyclone Aila in India. Indian J Med Res 2011;133:395- 400. 17� Emanuel K. Increasing destructiveness of tropical cyclones over the past 30 years. Nature 2005;436:686-8. 67� Myint NW, Kaewkungwal J, Singhasivanon P, et al. Are there any changes in burden and management of 18� i ntergovernmental Panel on Climate Change. IPCC Fourth Assessment Report: Climate Change 2007. Geneva: communicable diseases in areas affected by Cyclone Nargis? Confl Health 2011;5:9. IPCC; 2007. 68. Miranda DS, Choonara I. Hurricanes and child health: lessons from Cuba. Arch Dis Child 2011;96:328-9. 19� Elsner JB, Kossin JP, Jagger TH. The increasing intensity of the strongest tropical cyclones. Nature 2008;455:92-5. 69. McCurry J. Philippines struggles to recover from typhoons. Lancet 2009;374:1489. 20. Alson R, Alexander D, Leonard RB, Stringer LW. Analysis of medical treatment at a field hospital following 70. Bohn D, Kanter RK, Burns J, Barfield WD, Kissoon N. Supplies and equipment for pediatric emergency mass Hurricane Andrew, 1992. Ann Emerg Med 1993;22:1721-8. critical care. Pediatr Crit Care Med 2011;12:S120-7. 21� r otheray KR, Aitken P, Goggins WB, Rainer TH, Graham CA. Epidemiology of injuries due to tropical cyclones 71� Centers for Disease Control and Prevention. Surveillance for illness and injury after hurricane Katrina--New in Hong Kong: A retrospective observational study. Injury 2011. Orleans, Louisiana, September 8-25, 2005. MMWR Morb Mortal Wkly Rep 2005;54:1018-21. 22� Smith CM, Graffeo CS. Regional impact of Hurricane Isabel on emergency departments in coastal southeast- 72� r ath B, Young EA, Harris A, et al. Adverse respiratory symptoms and environmental exposures among chil- ern Virginia. Acad Emerg Med 2005;12:1201-5. dren and adolescents following Hurricane Katrina. Public Health Rep 2011;126:853-60. 23� Curtis A, Li B, Marx BD, Mills JW, Pine J. A multiple additive regression tree analysis of three exposure mea- 73� Tak S, Bernard BP, Driscoll RJ, Dowell CH. Floodwater exposure and the related health symptoms among sures during Hurricane Katrina. Disasters 2011;35:19-35. firefighters in New Orleans, Louisiana 2005. Am J Ind Med 2007;50:377-82. 24� Zoraster RM. Vulnerable populations: Hurricane Katrina as a case study. Prehosp Disaster Med 2010;25:74-8. 74� Edwards TD, Young RA, Lowe AF. Caring for a surge of Hurricane Katrina evacuees in primary care clinics. Ann Fam Med 2007;5:170-4. 25� Mortensen K, Wilson RK, Ho V. Physical and mental health status of Hurricane Katrina evacuees in Houston in 2005 and 2006. J Health Care Poor Underserved 2009;20:524-38. 75� Sjoberg L, Yearwood R. Impact of a category-3 hurricane on the need for surgical hospital care. Prehosp Disaster Med 2007;22:194-8. 26. Jonkman SN, Maaskant B, Boyd E, Levitan ML. Loss of life caused by the flooding of New Orleans after Hur- ricane Katrina: analysis of the relationship between flood characteristics and mortality. Risk Anal 2009;29:676- 76� Centers for Disease Control and Prevention. Vibrio illnesses after Hurricane Katrina--multiple states, August- 98. September 2005. MMWR Morb Mortal Wkly Rep 2005;54:928-31. 27� Burton LC, Skinner EA, Uscher-Pines L, et al. Health of Medicare Advantage plan enrollees at 1 year after Hur- 77� Seybold UMD, White NRN, Wang Yun FP, Halvosa JSM, Blumberg Henry MMD. Colonization with multi- ricane Katrina. Am J Manag Care 2009;15:13-22. drug resistant organisms in evacuees after Hurricane Katrina� infection Control and Hospital Epidemiology 2007;28:726-9. 28� do MP, Hutchinson PL, Mai KV, Vanlandingham MJ. Disparities in health care among Vietnamese New Orlea- nians and the impacts of Hurricane Katrina. Res Sociol Health Care 2009;27:301-19. 78� Nsuami MJ, Taylor SN, Smith BS, Martin DH. Increases in gonorrhea among high school students following hurricane Katrina. Sex Transm Infect 2009;85:194-8. 29� Brunkard J, Namulanda G, Ratard R. Hurricane Katrina deaths, Louisiana, 2005. Disaster Med Public Health Prep 2008;2:215-23. 79� reilly KH, Clark RA, Schmidt N, Benight CC, Kissinger P. The effect of post-traumatic stress disorder on HIV disease progression following hurricane Katrina. AIDS Care 2009;21:1298-305. 30� Taylor-Clark KA, Viswanath K, Blendon RJ. Communication inequalities during public health disasters: Ka- trina’s wake. Health Commun 2010;25:221-9. 80� r obinson WT, Wendell D, Gruber D. Changes in CD4 count among persons living with HIV/AIDS following Hurricane Katrina. AIDS Care 2011;23:803-6. 31. Noji EK. The public health consequences of disasters: Oxford University Press; 1997. 81� Clark RA, Besch L, Murphy M, et al. Six months later: The effect of Hurricane Katrina on health care for persons 32� Haque U, Hashizume M, Kolivras KN, Overgaard HJ, Das B, Yamamoto T. Reduced death rates from cyclones living with HIV/AIDS in New Orleans. AIDS Care 2006;18 Suppl 1:S59-61. in Bangladesh: what more needs to be done? Bull World Health Organ 2012;90:150-6. 82� Kissinger P, Liddon N, Schmidt N, Curtin E, Salinas O, Narvaez A. HIV/STI Risk behaviors among Latino 33� Zane DF, Bayleyegn TM, Hellsten J, et al. Tracking deaths related to Hurricane Ike, Texas, 2008. Disaster Med migrant workers in New Orleans post-Hurricane Katrina disaster. Sex Transm Dis 2008;35:924-9. Public Health Prep 2011;5:23-8. 83� Lifson A. Mosquitoes, models, and dengue. The Lancet 1996;347:1201-2. 34� Centres for Disease Control and Prevention. Injuries and illnesses related to Hurricane Andrew--Louisiana, 1992. MMWR Morb Mortal Wkly Rep 1993;42:242-4, 9-51. 84� Shoaf KI, Rottman SJ. Public health impacts of disasters. Australian Journal of Emergency Management 2000;15:60-3. 35� UN ICEF Cyclone Evaluation Team. Health Effects of the 1991 Bangladesh Cyclone: Report of a UNICEF Evalu- ation Team. Disasters 1993;17:153-65. 85� Morrow MG, Johnson RN, Polanco J, Claborn DM. Mosquito vector abundance immediately before and after tropical storms Alma and Arthur, northern Belize, 2008. Rev Panam Salud Publica 2010;28:19-24. 36� gong Z, Chai C, Tu C, Lin J, Gao Y, Qiu Y. Injuries after a typhoon in China. New England Journal of Medicine 2007;356:196-7. 86� g uzman-Tapia Y, Ramirez-Sierra MJ, Escobedo-Ortegon J, Dumonteil E. Effect of Hurricane Isidore on Tri- atoma dimidiata distribution and Chagas’ disease transmission risk in the Yucatan Peninsula of Mexico. Am J 37� p latz E, Cooper HP, Silvestri S, Siebert CF. The impact of a series of hurricanes on the visits to two central Trop Med Hyg 2005;73:1019-25. Florida emergency departments. J Emerg Med 2007;33:39-46. 87� p an-American Health Organization. Impact of Hurricane Mitch on Central America. Epidemiological Bulletin 38� Sullivent EE, 3rd, West CA, Noe RS, Thomas KE, Wallace LJ, Leeb RT. Nonfatal injuries following Hurricane 1998;19:1-14. Katrina - New Orleans, Louisiana, 2005. J Safety Res 2006;37:213-7. 88� Caillouet KA, Michaels SR, Xiong X, Foppa I, Wesson DM. Increase in West Nile neuroinvasive disease after 39� Alhinai MY. Tropical Cyclone Gonu: number of patients and pattern of illnesses in the primary health centers Hurricane Katrina. Emerg Infect Dis 2008;14:804-7. in A’Seeb area, Muscat, Sultanate of Oman. Oman Med J 2011;26:223-8. 89� Foppa IM, Evans CL, Wozniak A, Wills W. Mosquito fauna and arbovirus surveillance in a coastal Mississippi 40� Faul M, Weller NF, Jones JA. Injuries after Hurricane Katrina among Gulf Coast evacuees sheltered in Houston, community after Hurricane Katrina. J Am Mosq Control Assoc 2007;23:229-32. Texas. J Emerg Nurs 2011;37:460-8.

Volume 15 Number 1 7 ANNALS OF THE ACTM 7 90� yang HY, Hsu PY, Pan MJ, et al. Clinical distinction and evaluation of leptospirosis in Taiwan - a case-control 114. Anderson AH, Cohen AJ, Kutner NG, Kopp JB, Kimmel PL, Muntner P. Missed dialysis sessions and hospital- study. J Nephrol 2005;18:45-53. ization in hemodialysis patients after Hurricane Katrina. Kidney Int 2009;75:1202-8. 91� Sehgal S, Sugunan A, Vijayachari P. Outbreak of leptospirosis after the cyclone in Orissa. The National Medical 115. Harville EW, Xiong X, Buekens P. Hurricane Katrina and perinatal health. Birth 2009;36:325-31. Journal of India 2002;15:22. 116. Kanter RK. Child mortality after Hurricane Katrina. Disaster Med Public Health Prep 2010;4:62-5. 92� World Health Organization. Epidemic-prone disease surveillance and response after the tsunami in Aceh Prov- 117. Callaghan WM, Rasmussen SA, Jamieson DJ, et al. Health concerns of women and infants in times of natural ince, Indonesia. Wkly Epidemiol Rec 2005;80:160-4. disasters: lessons learned from Hurricane Katrina. Matern Child Health J 2007;11:307-11. 93� Centers for Disease Control and Prevention. Tuberculosis control activities after Hurricane Katrina--New Or- 118. Lobato MN, Yanni E, Hagar A, et al. Impact of Hurricane Katrina on newborn screening in Louisiana. Pediatrics leans, Louisiana, 2005. MMWR Morb Mortal Wkly Rep 2006;55:332-5. 2007;120:e749-55. 94� riddel CE, Surovik JG, Chon SY, et al. Fungal foes: presentations of chromoblastomycosis post-Hurricane Ike. 119. Goenjian HA, Chiu ES, Alexander ME, St Hilaire H, Moses M. Incidence of cleft pathology in greater New Cutis 2011;87:269-72. Orleans before and after Hurricane Katrina. Cleft Palate Craniofac J 2011;48:757-61. 95� r abito FA, Perry S, Davis WE, Yau CL, Levetin E. The relationship between mold exposure and allergic re- 120. Hamilton BE, Sutton PD, Mathews TJ, Martin JA, Ventura SJ. The effect of Hurricane Katrina: births in the U.S. sponse in post-Katrina New Orleans. J Allergy (Cairo) 2010;2010. Gulf Coast region, before and after the storm. Natl Vital Stat Rep 2009;58:1-28, 32. 96� Barbeau DN, Grimsley LF, White LE, El-Dahr JM, Lichtveld M. Mold exposure and health effects following 121. Bhattacharyya MR, Steptoe A. Emotional triggers of acute coronary syndromes: strength of evidence, biologi- hurricanes Katrina and Rita. Annu Rev Public Health 2010;31:165-78 1 p following 78. cal processes, and clinical implications. Progress in Cardiovascular Diseases;49:353-65. 97� rabito FA, Iqbal S, Kiernan MP, Holt E, Chew GL. Children’s respiratory health and mold levels in New Orleans 122. Gautam S, Menachem J, Srivastav SK, Delafontaine P, Irimpen A. Effect of Hurricane Katrina on the incidence after Katrina: a preliminary look. J Allergy Clin Immunol 2008;121:622-5. of acute coronary syndrome at a primary angioplasty center in New Orleans. Disaster Med Public Health Prep 98. Hsu NY, Chen PY, Chang HW, Su HJ. Changes in profiles of airborne fungi in flooded homes in southern 2009;3:144-50. Taiwan after Typhoon Morakot. Sci Total Environ 2011;409:1677-82. 123. Jiao Z, Kakoulides SV, Moscona J, et al. Effect of Hurricane Katrina on incidence of acute myocardial infarction 99� McKinney N, Houser C, Meyer-Arendt K. Direct and indirect mortality in Florida during the 2004 hurricane in New Orleans three years after the storm. Am J Cardiol 2012;109:502-5. season. Int J Biometeorol 2011;55:533-46. 124. Bakaeen FG, Huh J, Chu D, et al. Hurricane Katrina: impact on cardiac surgery case volume and outcomes. Tex 100. Miller AC, Arquilla B. Chronic diseases and natural hazards: impact of disasters on diabetic, renal, and cardiac Heart Inst J 2008;35:273-8. patients. Prehosp Disaster Med 2008;23:185-94. 125. Joseph DA, Wingo PA, King JB, et al. Use of state cancer surveillance data to estimate the cancer burden in 101. Centers for Disease Control and Prevention. Morbidity surveillance after Hurricane Katrina - Arkansas, Louisi- disaster-affected areas - Hurricane Katrina, 2005. Prehosp Disaster Med 2007;22:282-90. ana, Mississippi, and Texas, September 2005. MMWR Morb Mortal Wkly Rep 2006;55:727-31. 126. Loehn B, Pou AM, Nuss DW, et al. Factors affecting access to head and neck cancer care after a natural 102. Mokdad AH, Mensah GA, Posner SF, Reed E, Simoes EJ, Engelgau MM. When chronic conditions become disaster: a post-Hurricane Katrina survey. Head Neck 2011;33:37-44. acute: prevention and control of chronic diseases and adverse health outcomes during natural disasters. Prev 127. Cerda M, Tracy M, Galea S. A prospective population based study of changes in alcohol use and binge drinking Chronic Dis 2005;2 Spec no:A04. after a mass traumatic event. Drug Alcohol Depend 2011;115:1-8. 103. Jhung MA, Shehab N, Rohr-Allegrini C, et al. Chronic disease and disasters medication demands of Hurricane 128. Cepeda A, Valdez A, Kaplan C, Hill LE. Patterns of substance use among hurricane Katrina evacuees in Hous- Katrina evacuees. Am J Prev Med 2007;33:207-10. ton, Texas. Disasters 2010;34:426-46. 104. Greenough PG, Lappi MD, Hsu EB, et al. Burden of disease and health status among Hurricane Katrina-dis- 129. Rohrbach LA, Grana R, Vernberg E, Sussman S, Sun P. Impact of hurricane Rita on adolescent substance use. placed persons in shelters: a population-based cluster sample. Ann Emerg Med 2008;51:426-32. Psychiatry 2009;72:222-37. 105. Vest JR, Valadez AM. Health conditions and risk factors of sheltered persons displaced by Hurricane Katrina. 130. Abel MT, Cobb GP, Presley SM, et al. Lead distributions and risks in New Orleans following Hurricanes Katrina Prehosp Disaster Med 2006;21:55-8. and Rita. Environ Toxicol Chem 2010;29:1429-37. 106. Centers for Disease Control and Prevention. Preliminary medical examiner reports of mortality associated with 131. Abel MT, Suedel B, Presley SM, et al. Spatial distribution of lead concentrations in urban surface soils of New Hurricane Charley--Florida, 2004. MMWR Morb Mortal Wkly Rep 2004;53:835-7. Orleans, Louisiana USA. Environ Geochem Health 2010;32:379-89. 107. Kessler RC. Hurricane Katrina’s impact on the care of survivors with chronic medical conditions. J Gen Intern 132. Shi H, Witt EC, 3rd, Shu S, Su T, Wang J, Adams C. Toxic trace element assessment for soils/sediments Med 2007;22:1225-30. deposited during Hurricanes Katrina and Rita from southern Louisiana, USA: a sequential extraction analysis. 108. Zwillich T. Health care remains basic in New Orleans. The Lancet 2006;367:637-8. Environ Toxicol Chem 2010;29:1419-28. 109. Rudowitz R, Rowland D, Shartzer A. Health care in New Orleans before and after Hurricane Katrina. Health Aff (Millwood) 2006;25:w393-406. 110. Kopp JB, Ball LK, Cohen A, et al. Kidney patient care in disasters: lessons from the hurricanes and earthquake Corresponding Author of 2005. Clin J Am Soc Nephrol 2007;2:814-24. 111. Kleinpeter MA, Norman LD, Krane NK. Dialysis services in the hurricane-affected areas in 2005: lessons Dr RD Mitchell learned. Am J Med Sci 2006;332:259-63. 112. Fonseca VA, Smith H, Kuhadiya N, et al. Impact of a natural disaster on diabetes: exacerbation of disparities Department of Emergency Medicine, Townsville Hospital, and long-term consequences. Diabetes Care 2009;32:1632-8. Townsville, Queensland, Australia. 113. Potash MN, West JA, Corrigan S, Keyes MD. Pain management after Hurricane Katrina: outcomes of veterans enrolled in a New Orleans VA pain management program. Pain Med 2009;10:440-6. Email: [email protected]

keting and sound science can dramatically improve the health of a targeted REVIEW ARTICLE population. There are still areas for improvement though with regards to micronutrient interventions, and it is important that new strategies are utilised appropriately. Further research is needed with regards to iron supple- Vitamin A, Zinc, Iron and Iodine mentation in malaria hyper-endemic regions.

Interventions in Children Under Five Keywords: micronutrients, vitamin A, iron, zinc, iodine, deficiencies, chil- Years of Age dren, supplementation, fortification

Benjamin Wood Ann Australas Coll Trop Med 2014;15(1):8-11.

Sir Charles Gairdner Hospital, Perth, Western Australia; School of Public Health, Tropical Medicine and Rehabilitation Sciences, James Cook University, Townsville, Queensland Background Malnutrition is estimated to account for 11% of the global burden of dis- 1 ABSTRACT ease. Malnutrition can be divided into two major categories: protein-energy malnutrition and micronutrient deficiencies.2 These two areas of nutrition Background: micronutrient deficiency accounts for approximately 10% of deficiencies are interconnected and very often simultaneously exist. They all deaths and disability adjusted life years in children under the age of five. are both complexly intertwined with geographical, socio-economic and po- Vitamin A, iron, zinc and iodine are considered the most important of the litical influences. The scope of this literature review will focus on four major micronutrients and have massive global health implications for this demo- micronutrient deficiencies with major global health implications for children graphic. – vitamin A, zinc, iodine and iron. It is estimated that micronutrient deficiencies affect at least 2 billion people Methods: major strategies for vitamin A, zinc, iron and iodine interventions worldwide. In children less than five years of age, micronutrient deficiencies including education and dietary diversification, supplementation, fortification account for approximately 10% of deaths and disability-adjusted life years.3 and integrated approaches, were reviewed. Much of this burden stems from deficiencies in vitamin A and zinc.3 Deficien- cies in vitamin A, zinc, iodine and iron are undoubtedly massive global health Conclusion: vitamin A and zinc supplementation programs remain the most concerns that predominately affect developing countries. Adequate provi- cost effective of all health interventions as judged by the Copenhagen con- sion of these important micronutrients is vital to ensure appropriate physical, sensus panel. Salt iodisation is the best example of micronutrient fortifica- mental and cognitive development, and long-term health.4 tion to date, and is an excellent example of how political will, effective mar-

8 ANNALS OF THE ACTM May 2014 Roughly 250 million people, mainly pregnant women and young children, education and information alone.13 An advantage of educational programs are affected by vitamin A deficiency (VAD).5 Vitamin A is important for the is that they are sustainable, and if delivered effectively, promote long-lasting immune system and it is vital for the physical health of children exposed to behaviour change in a population.13 There is evidence that these programs illnesses.2 VAD dramatically increases the risk and mortality from typical that specifically target education can work in relatively advantaged popula- childhood infections such as diarrhoeal disease.5 Vitamin A is also integral to tions.14 However, there are several drawbacks to education programs alone. the functioning of a healthy eye, being vital to the retina’s ability to adjust to It is important to realise that changes in behaviour and practices will be con- dark lighting, and to the protection of the cornea from scarring.4 VAD results fined to the socio-economic and geographical capabilities of the targeted in a condition called xerophthalmia, a disorder which is the leading cause of population.13 The take-home message of many nutritional reviews is that preventable blindness in children. An estimated 250 000 to 500 000 children education and dietary diversification is important in all nutritional interven- who are vitamin A deficient become blind annually. Half of these children die tions. However, it should not be used as a stand-alone technique, but in within a year of becoming blind.5 combination with other proven techniques that will be mentioned. About 2 billion people worldwide are zinc deficient.2 Zinc is essential for Supplementation promoting immunity, resisting infections, and for promoting the production Micronutrient supplementation of targeted populations of young children of antibodies against gut pathogens. Zinc supplementation, alongside oral has been proven to be very cost-effective, particularly for vitamin A and zinc. rehydration therapy, reduces diarrhoeal rates in children by 27%. Consider- Supplementation is usually delivered through primary healthcare systems, ing 18% of deaths under the age of five are attributed to diarrhoea, this is a and commonly coupled with immunisation programs.13 Supplementation major finding. Zinc supplementation can also lower the rates of acute lower programs are only effective if the supplements reach a high percentage of respiratory infections by 15%,6 and reduces the rates of malaria.7 Zinc is also people in the vulnerable population, and if the delivery system has built-in important for the growth and development of the nervous system.4 quality control.13 Iron is vital for human development and functioning.4 Iron deficiency is the Vitamin A supplementation most common nutritional deficiency globally and affects both developing and developed nations.2 It is the major cause of anaemia worldwide, and Vitamin A (VA) supplementation is undoubtedly one of the most effective increases the mortality and morbidity in children. It is estimated that 40% child survival interventions.4 Systematic review of RCTs showed that one of preschool children are anaemic.8 Communicable diseases such as worm dose of VA reduced the mortality risk of children aged between 6 months infections, malaria, HIV and TB aggravate iron deficiency anaemia, making it and 5 years by 12%, with two doses reducing the risk by 22%.3 Studies in a major issue in many developing nations.7 Iron deficiency is also a common Asia showed that neonatal supplementation reduced mortality of children cause of psychomotor disorders and cognitive impairment, and has been between the age of 2 days and 6 months by 21%.3 VA was also shown to shown to decrease the level of physical activity undertaken.4 reduce the burden of multiple diseases, including persistent diarrhoea with a rate ratio of 0.45.3 It was estimated that nearly two-thirds of children in Iodine is critical element required for nervous system development.4 Iodine developing countries under the age of five received two high-dose VA sup- deficiency is the most common cause of brain damage worldwide, yet it is plements.15 What is even more impressive is that 82% of children under the easily preventable.9 Iodine deficiency is a significant public health issue in age of five in the least developed countries of the world received coverage, 130 countries, with about 750 million people affected.10 In those communi- highlighting what happens when political will and international resources are ties where iodine intake is sufficient, the average IQ, on average, has been available.7 VA supplementation has also been shown in studies to improve shown to be about 13 points higher than those where intake is insufficient.11 the efficacy of iron supplementation by increasing haemoglobin levels. This Objectives believed to be done stimulating erythroid precursors and improving the availability of iron to the bone marrow.16 The objectives of this review are to highlight the global health significance of the four major micronutrient deficiencies focusing on their impact on the Zinc supplementation morbidity and mortality of children, review public health interventions that In 2004, a joint statement was made by UNICEF and WHO recommending aim to address these deficiencies, and to review what needs to be done with zinc supplementation (ZS) in the management of diarrhoea.7 Despite these future efforts. recommendations, as of 2010, only 46 countries utilised a zinc policy in their Methods child health delivery.7 Four systematic reviews have shown that preventive ZS in children greater than 6 months was shown to reduce the overall mor- PubMed, Google Scholar and the WHO database were used as search en- tality risk by 9%, and reduce the odds of stunting by 15%.3 A meta-analysis gines for this review using the following keywords: micronutrients, iron, zinc, revealed that ZS reduces the duration of diarrhoea by 15-24%,3,17,18 reduced vitamin A, iodine, deficiencies, children, public health interventions. the episodes of diarrhoea and dysentery3 and the frequency of lower respira- Discussion tory infections by 15-24%.17 Many programs addressing micronutrient deficiencies have yielded proven Iron supplementation success in improving physical and cognitive capacity. The excellent cost- Studies have shown that iron supplementation (IrS) resulted in a haemo- effectiveness of micronutrient interventions have received many accolades, globin concentration increase of 7.4 g/L.19 IrS alone reduced the occurrence and in 2008, vitamin A and zinc supplementation for children was stated by of anaemia in regions not endemic for malaria from 6-32%.16 However, it the Copenhagen consensus panel, out of all global development efforts, to is important to note that potentially there is an increased risk of childhood provide the very best return on investment. Iron and iodine food fortification mortality in hyper-endemic malarial regions. This recommendation was was also ranked third on the list.11 brought about by a trial in Tanzania which showed there was an increase in Education and dietary diversification mortality in children receiving IrS, and hence the trial was stopped prema- turely.20 Several other studies have also showed a correlation between IrS Education directed towards populations at micronutrient deficiency risk is and communicable diseases, although further research is needed to make clearly important in any intervention program. These programs mainly look this conclusive.21 at the importance of dietary diversification of foods containing important micronutrients. One study has found that education and dietary diversifica- Iodine supplementation tion slightly improved haemoglobin levels from 102 to 107g/L,3 with another Iodine supplementation is generally reserved for pregnant women. Iodine showing it slightly improved vitamin A.3 Recent literature supports improve- fortification is the leading intervention for targeting children populations ment of dietary zinc through the production and consumption of animal which are iodine deficient. source foods.12 However, there is limited documentation of the impact of

Volume 15 Number 1 9 ANNALS OF THE ACTM 9 Fortification monitoring.4 Let us take a look at Zambia, which in 2000 with the assistance of UNICEF, has been implementing half-annual CHWs. Here, alongside Food fortification with micronutrients is a constantly evolving public health routine vaccinations, deworming, malaria prevention strategies and growth intervention. Fortification can be broken down into two general approaches monitoring, vitamin A supplements are given. Evaluation performed in 2003 – central and peripheral fortification.13 Central fortification involves adding showed that through these CHWs, 77% of children received VA. This was micronutrients to foods in a centralised manner prior to distribution or mar- shown to halve the level of VA deficiency, with anaemia being reduced by keting.13 Peripheral fortification involves adding micronutrients to food in 12%.4 the household.13 The great advantage of central fortification done in a safe manner is that minimal behavioural change or motivation is necessary in the The Essential Nutrition Actions (ENA) approach population to instigate health benefits. This approach does require sound The ENA approach evolved from the Lancet Child Survival series which re- scientific evidence, political will and leadership and effective implementation. ported a set of nutritional interventions that were shown to reduce childhood Also, ensuring equal coverage is paramount, with rural populations often mortality.29 It was developed by WHO, UNICEF and the US Agency for Inter- neglected in such interventions. Peripheral fortification can be utilised to national Development and includes a focus on the first two years of child- target certain vulnerable sub-groups such as young children. This approach hood.13 Vitamin A, zinc supplementation and exclusive breastfeeding in the requires active participation form household leaders.13 first 6 months of life are strategies in the ENA framework.13 This integrated Vitamin A fortification approach has been shown to be effective in numerous developing countries, with an example being Madagascar.13 There have been interventions of VA fortification to foods such as cooking oils, monosodium glutamate (MSG) and sugar.6 Evidence for the effective- Conditional cash transfers (CCTs) ness of fortification of these foods with VA is difficult to find, however a study CCTs have become widespread in Latin America, and are based on financial showed a 30% mortality reduction in children aged 6 months – 5 years in a payments to poor and vulnerable populations in exchange for completing population in Indonesia given commercially fortified MSG.22 certain conditions. These conditions include a certain school attendance rate Zinc fortification for children, use of primary healthcare services and nutrition programs.13 WHO recommends zinc as a curative measure only for childhood illnesses Perhaps the most successful of the CCT programs has been in Mexico, and diarrhoea,23 and there is minimal data to support central zinc fortifica- which includes food fortification and micronutrient supplementation of chil- tion.24 As a result, there is limited data on zinc fortification strategies. A dren aged between 6 months and 2 years. The effect on anaemia prevalence study in Senegal showed food containing zinc-fortification did not increase and infant growth has been markedly positive since the commencement of plasma zinc concentration, unlike ZS for 15 days which did.25 A study in this program.30 northern India where children were given fortified milk with multiple nutri- ents, including zinc and iron, had an 18% lower incidence of diarrhoea and a What needs to be done 26% lower incidence of pneumonia.26 National governments and development partners need to increase long-term investment to yield greater and further returns.11 More affordable, feasi- Iron fortification ble and evidence-based programs are emerging, and should be utilised to There has been much debate about the efficacy of iron fortification. Similar expand the potential benefits of addressing these major micronutrient defi- to IrS, there is unclear data about whether this intervention is safe for chil- ciencies.11 Monitoring and surveillance of nutritional programs is paramount dren in malaria hyper-endemic regions. Iron fortification has been shown for the success of current and future interventions. This includes popula- to reduce the rate of anaemia by more than two-thirds in school children in tion monitoring assessing change in dietary behaviours which will guide Chile.27,28 It has been estimated that iron fortification of staple foods could alterations of programs, allowing them to be more efficient and effective. potentially prevent 8% of DALYs of children – a massive 123 000 DALYs.3 Resources need to be mobilised in national budgets so that each country can A study in India where food was fortified with iron and riboflavin showed a sustain appropriate nutritional interventions on a long term basis. The me- reduction in the prevalence of anaemia by more than half.13 More research, dia, including social media, should be utilised for awareness, motivation and however, needs to be done in hyper-endemic malarial regions before wide- mobilisation of the masses, and should be incorporated into national health spread interventions can be made with sound scientific backing. systems and schools.4 The 2009 Global Report on micronutrient deficiencies sets the coverage target of those receiving twice-yearly VA supplementation Iodine fortification on a repeated basis as greater than 80%.4 The other ‘difficult-to-reach’ 20% Undoubtedly the greatest success in food fortification has been with the who are missed in regular programs will need to be targeted with special, iodisation of salt. This is a fantastic global health example of how good context-specific programs such as integrated outreach programs.4 Manda- governance and market opportunity can improve the health of a population.4 tory legislation of salt iodisation, with appropriate enforcement capabilities, Salt iodisation is estimated to reduce DALYs from iodine deficiency by 41% should be made. Incentives should be made to salt processors in developing in children.3 Less than one in five developing world households were using countries to ensure they iodise their salt, with a transition from a donor- iodised salt in 1990. In 2010, this number increased to 70%, with more than dependent to a market-supported supply important for sustainability.4 Zinc 30 developing countries reaching the universal iodised salt goal.4 Having said supplementation needs to be implemented into each nation’s policy for the this, 24 countries in 2010 showed no growth and even decline in iodised management of diarrhoea in children.4 This will involve ensuring an adequate salt coverage. In half of these countries, one in five is consuming adequate zinc supply, with appropriate delivery strategies. Safe strategies to improve iodised salt levels.4 Iodised water has been shown to reduce infant mortality iron intake need to be made clear for policy makers and program designers. by more than a half in studied areas,16 and neonatal mortality by nearly two Further directed research needs to be done with regards to iron supplemen- thirds.16 tation in hyper-endemic malaria regions.4 Integrated approaches Conclusion Integrated approaches include child nutrition programs, the Essential Nutri- Micronutrient deficiencies, especially that of vitamin A, zinc, iron and iodine, tion Actions approach and conditional cash transfers. result in an enormous health burden on children under the age of five. Over the last few decades, there have been a large number of programs designed Child nutrition through child health programs to target this issue. Some of these programs have been touted as being Child health weeks (CHWs) are an excellent example of delivering a range of some of the most cost-effective health programs that exist today. Sup- services for child health through an existing health system. This requires a plementation programs with zinc and vitamin A have produced staggering sound and effective public health structure with appropriate evaluation and results in terms of the reduction of childhood mortality and morbidity. How-

10 ANNALS OF THE ACTM May 2014 ever, these programs need to be scaled up to ensure coverage is expanded 17� Sazawal S, Black RE, Ramsan M, et al. Effect of zinc supplementation on mortality in children aged 1–48 months: a community-based randomised placebo-controlled trial. Lancet 2007;369:927-34. and performed on a recurrent basis. Coverage needs to include targeting the 18� Tielsch J, Khatry S, Stoltzfus R, et al. Effect of routine prophylactic supplementation with iron and folic acid on neglected groups of a population that have been missed with other routine preschool child mortality in southern Nepal: community-based, cluster-randomised, placebo-controlled trial. programs. Nutritional intervention is an evolving field, with food fortifica- Lancet 2006;367: 144–52. 19� gera T, Sachdev HP, Nestel P, et al. Effect of iron supplementation on haemoglobin response in children: tion strategies emerging as an area of great potential. Further research and systematic review of randomised controlled trials. J Pediatr Gastroenterol Nutr 2007;44:468–86. data is needed however with food fortification strategies, and governments 20� Sazawal S, Black R, Ramsan M, et al. Effects of routine prophylactic supplementation with iron and folic need to ensure they maintain the political will to ensure up-to-date legislative acid on admission to hospital and mortality in preschool children in a high malaria transmission setting: community-based, randomised, placebo-controlled trial. Lancet 2006;367:133-43. changes that promote better nutritional health. 21. Iannotti L, Tielsch J, Black M, et al. Iron supplementation in early childhood: health benefits and risks. Am J Clin Nutr 2006;84:1261-76. References 22. Muhilal, Permeisih D, Idjradinata YR, et al. Vitamin A-fortified monosodium glutamate and health, growth, and survival of children: a controlled field trial. Am J Clin Nutr 1988;48:1271-76. 1� World Health Assembly. Maternal, infant and young child nutrition. Sixty-Fifth World Health Assembly. WHO. 2012;13(3):1-2. 23� World Health Organization, UNICEF. Joint statement on the management of acute diarrhoea. Geneva: WHO, 2004. p13. 2. Muller O, Krawinkel M. Malnutrition and health in developing countries. CMAJ. 2005;173(3):279-285. 24. Salgueiro M, Zubigalla M, Lysioek A, et al. Fortification strategies to combat zinc and iron deficiency. Nutr Rev 3� Bhutta Z, Ahmad T, Black RE, et al. What works? Interventions for maternal and child undernutrition and 2002;60:52-8. survival. Lancet 2008;10(1):41-46. 25� Ba Lo N, Aaron GJ, Hess SY, et al. Plasma zinc concentration responds to short-term zinc supplementation, 4. The Micronutrient Initiative. Investing in the Future: A United Call to Action on Vitamin and Mineral Deficien- but not zinc fortification, in young children in Senegal. Am J Clin Nutr 2011;93(6):1348-55. cies. 2009. Available from: http://www.unitedcalltoaction.org/flashreport.asp (accessed May 3 2013). 26. Sazawal S, Dhingra U, Dhingra P, et al. Effects of fortified milk on morbidity in young children in north India: 5. World Health Organization. Programmes: Nutrition. Vitamin A deficiency. Geneva: WHO, 2013. Available from: community based, randomised, double masked placebo controlled trial. Br Med J 2007;334:140. http://www.who.int/nutrition/topics/vad/en/index.html (accessed May 5 2013.) 27. Nadiger H, Krishnamachari K, Naidu A, et al. The use of common salt (sodium chloride) fortified with iron to 6� World Health Organization. Comprehensive implementation plan on maternal, infant and young child nutrition. control anaemia: results of a preliminary study. Br J Nutr 1980;43:45-51. 2012;A65(11):1-14. 28. Walter T, Hertrampf E, Pizarro F, et al. Effect of bovine-hemoglobin-fortified cookies on iron status of school- 7. Khan Y, Bhutta ZA. Nutritional deficiencies in the developing world: current status and opportunities for inter- children: a nationwide program in Chile. Am J Clin Nutr 1993;57:190-4. vention. Pediatr Clin North Am 2010;57:1409-1441. 29� Jones G, Steketee R, Black R, Buttha Z, et al. How many child deaths can we prevent this year? Lancet 8. World Health Organization. Programmes: Nutrition. Iron deficiency anaemia . Geneva: WHO, 2013. Available 2003;362:65-71. from: http://www.who.int/nutrition/topics/ida/en/index.html (acessed May 6 2013.) 30� rivera J, Sotres-Alvarez D, Habicht J, et al. Impact of the Mexican program for education, health and nutrition 9. World Health Organization. Programmes: Nutrition. Iodine deficiency disorders. Geneva: WHO, 2013. Avail- (Progresa) on rates of growth and anemia in infants and young children: a randomized effectiveness study. able from: http://www.who.int/nutrition/topics/idd/en/index.html (accessed May 6 2013). JAMA 2004;291:2563-70. 10. Prinzo Z, De Benoist B. Meeting the challenges of micronutrient deficiencies in emergency-affected populations. Department of Nutrition for Health and Development, WHO. 2002;61:251-257. 11� Food and Agriculture Organization of the United Nations. Undernourishment around the world. The state of food insecurity in the world. Rome: FAO, 2004. Corresponding Author 12. Gibson R, Anderson V. A review of interventions based on dietary diversification or modification strategies with the potential to enhance intakes of total and absorbable zinc. Food Nutr Bull 2009;30 (1):108-43. Dr Benjamin Wood 13. Harrison G. Public Health Interventions to Combat Micronutrient Deficiencies. Publ Hlth Rev 2010 Sir Charles Gairdner Hospital, Perth, Western Australia; Jul;32(1):256-266. School of Public Health, Tropical Medicine and Rehabilitation Sci- 14� Contento I, Balch GI, Bronner YL, et al. The effectiveness of nutrition education and implications for nutrition education policy, programs and research: a review of research. J Nutr Educ 1995;27:279-83. ences, James Cook University, Townsville, Queensland 15� Bhutta Z. Micronutrient needs of malnourished children. Curr Opin Clin Nutr Metab Care 2008;11:309-14. Email: [email protected] 16. Khan Y, Bhutta ZA. Nutritional deficiencies in the developing world: current status and opportunities for intervention. Pediatr Clin North Am 2010;57:1409-1441.

cause hearing loss through biological processes not well understood. No REVIEW ARTICLE prospective studies on this issue have been done to date. Collectively, reports have shown minimal evidence that antimalarial medications cause ototoxicity, except for quinine, which is well recognised as an ototoxic agent. Hearing Loss in Malaria Conclusion: This study highlights evidence that malaria may cause hearing Eric Fong impairment. Although there is limited definitive evidence, the impact malaria may have on altering hearing thresholds is increasingly recognised. Most School of Public Health and Tropical Medicine, James Cook University, Townsville antimalarial medications are safe for the ear. Quinine produces a predictable, but reversible, hearing loss. More research should aim to identify whether ABSTRACT a direct link exists between malaria and hearing loss because malaria is a global burden, the impact of hearing loss can be substantial, and studies Background: Malaria is a preventable and treatable infectious disease; how- evaluating drug toxicity may be confounded by deafness due to malaria itself. ever, malaria is extremely common in many parts of the world. Hearing loss can be devastating and adversely affect individuals and populations. Ma- Keywords: malaria, hearing loss, ototoxicity, antimalarials, chemotherapy laria and antimalarial regimens may affect hearing, thus causing significant functional, social, and developmental issues. This article explores the link Ann Australas Coll Trop Med 2014;15(1):11-15. between hearing loss and malaria, and ototoxicity from antimalarial chemotherapy. Introduction Aim: The aim of this study was to perform a systematic literature review on Impact of malaria hearing loss in malaria. Malaria is a disease of poverty that affects hundreds of millions of people 1 Design: A systematic search of the literature and a discussion of the results. worldwide, with many more at risk of exposure. Malaria can cause neurocognitive disorders, and it is well recognised that Plasmodium falciparum causes language disorders in children, although the mechanism is unclear.2 Method: A search was performed in MEDLINE via OvidSP, Scopus, and Hearing is a component of language and is known to be affected by some PubMed databases. antimalarial agents.3 Although ototoxicity studies have been done to evaluate these drugs, most studies assume that the affected individual has normal Results: Of 13 papers analysed, one was a systematic review on the direct hearing, since audiometric testing is usually done during and after treatment effect of malaria on hearing loss, which suggested that malaria may directly

Volume 15 Number 1 11 ANNALS OF THE ACTM 11 for acute malaria.4,5 Therefore, malaria and antimalarial agents may alter Methods hearing thresholds causing deafness and ultimately impairing language. The Two systematic literature searches were undertaken. A systemic literature impact of deafness on the individual and the population includes functional, search was performed on MEDLINE via OvidSP, Scopus, and PubMed data- social, and economic aspects, both in the child and adult, so exploring this bases. The keyword ‘malaria’ was used with ‘hearing loss or ototoxicity’. A proposed association may help identify new avenues for preventing hearing second systematic search was completed using the keywords ‘neurologi- loss.6 This review of the current literature was undertaken to investigate cal or neurocognitive deficits’ and ‘malaria’ and these articles were selected whether malaria and antimalarial treatment regimens cause hearing prob- based on available internal data on hearing loss. Articles were limited to lems, as the use of these drugs for febrile illness is widespread.3 English. Evidence for safety profiles of chemotherapeutic agents was limited Malaria disease to clinical studies. Malaria is a parasitic infection that affected 216 million people in 2010 with Results 3.3 billion people at risk.1 Young children are at increased risk of malaria Thirteen articles were selected after analysis as meeting the inclusion criteria because immunity to malaria has not yet developed, and as a result young for this review. There were three studies looking at the direct effect of ma- children living in endemic areas of malaria are a particular population at laria on hearing, which include a systematic review, a clinical trial, and one risk.1 Given that childhood development is sensitive to environmental in- laboratory report. Results for studies on safety of antimalarials included four sults, malaria can adversely affect normal growth.1 Deficiencies in areas clinical trials, two case control trials, one retrospective and two prospective such as language can occur due to restricted brain development.2 This can studies, and a case report. The findings are summarised in Tables 1 and 2 restrict learning, social interaction, and functional capacity to work, which and will be discussed next. has serious effects not only for the affected individual but at the social and economic level.6 These impairments are associated with a significant bur- Discussion den, especially in those affected in sub-Saharan Africa, which has high levels Direct effect of malaria on hearing of malaria transmission.7 There are a limited number of studies that directly investigate the association Chemotherapy agents used for malaria treatment may also be involved in between malaria and hearing loss (Table 1.) The most recent review pub- damaging the brain and, in particular, hearing function. These agents can lished by Zhao and Mackenzie in 2011 analysed 14 studies and suggested an 8 be toxic enough to be harmful to sensitive organs such as the inner ear. association of deafness due to the neurological deficits secondary to cerebral For example, artemether-lumefantrine is an effective antimalarial for uncom- malaria, but could not show a strong causal relationship between uncom- 9 plicated Plasmodium falciparum. Dose-response relationships have been plicated malaria and deafness.20 Their study found a small but significant established for artemisinin derivatives and toxicity on brain areas involved in proportion of hearing loss was associated with a non-specific fever or febrile 3,10,11 hearing in animal studies. In stable malaria transmission situations, qui- illness in countries where the most common cause of fever is malaria.20 Re- nine is commonly used as a first-line agent in children with a febrile illness call and measurement bias in these studies, however, makes it unknown as for presumed malaria, and can exert neurotoxic effects, as demonstrated in to whether causation exists. They conclude that clinically-significant hearing 3 the laboratory. Mefloquine is mainly a chemoprophylactic agent, but its impairment occurs only in severe disease, and that less-noticeable hearing side effect profile includes neuropsychiatric disorders that have been linked deficits, which may not be recognised, may be present in uncomplicated to toxicity in the brain. The evidence that these agents may be involved in malaria.20 ototoxicity and significant hearing loss in humans will be reviewed. Zhao and Mackenzie suggest that during a malarial infection prior to treat- Hearing loss ment, the level of hearing is reduced, and thus hearing may appear to be Hearing loss is a well-recognised global issue that affects around 278 mil- affected during the treatment period. Since baseline hearing function has lion people in the world.12 In developing countries, 50% of hearing loss been assumed to be normal in other studies, the actual amount of hearing has been estimated to be preventable. Many cases in the developing world loss is unknown and there could be potential loss due to the infection prior to may have no explanation as to the cause of deafness.12 Although there are treatment.20 These authors suggest that if this occurs, treatment may further numerous aetiologies for hearing loss, malaria or its treatment regimens damage hearing, and although hearing improves at follow-up compared to may help explain cases of unknown origin, and clarifying its role in hearing testing at pre-treatment, this might not represent the true hearing threshold impairment may allow public health strategies to be put into practice to avoid as no baseline of hearing was recorded prior to parasitic infection in the stud- a potentially preventable cause of deafness.12 Cerebral malaria, which is a ies they reviewed.20 These theories are supported by work in Ghana which complication of severe malaria, may cause persisting neurological issues, showed that hearing thresholds differed significantly (p<0.001) between including deafness.13 However, many individuals do not experience cerebral children with an acute illness of uncomplicated P. falciparum and healthy malaria and yet unexplained deafness is common in malaria-endemic areas.4 controls.21 After 9 months follow-up the study measured hearing thresholds Malaria may explain this large proportion of preventable cases of deafness. again, and found thresholds were not significantly different among the two groups.21 This suggests if P. falciparum affects hearing, the deafness may The effect of deafness on paediatric populations has been studied and evi- be transient and reversible. This study has implications for research in drug dence suggests that deafness in children may result in poor attention span, therapy for ototoxicity, since evaluating ototoxicity during the period of acute leading to limited class participation and hindering the learning experience, illness may be confounded by the presumed ototoxic effect of malaria.21 restricting their potential.6,14 An association between malaria parasite den- sity and level of concentration of schoolchildren, showing a dose-response Although clinical studies have not been performed sufficiently to establish an relationship, has been suggested in a number of studies.15-18 The cause of association, a laboratory study by Schmutzhard et al demonstrated signifi- this cognitive decline is unclear and is likely due to anaemia and brain hy- cant hearing loss in mice following cerebral malaria.22 They concluded that poxia,15-18 but hearing loss has been suggested to be a plausible explana- malaria significantly causes hearing impairment in mice by studying auditory tion.4 This can have serious consequences, as the effects of hearing loss evoked brainstem responses (ABRs), which measures hearing thresholds, can result in functional, social, and developmental issues in the developing before and after infection. In addition, they demonstrated that higher hearing child and these problems may continue into adulthood.6 The deaf adult may thresholds occurred in mice with cerebral malaria compared to uncompli- have poor language abilities, and consequently, limited social interaction and cated malaria, which suggests the severity of malaria has a more significant occupational capacity.14 This translates into public health and economic is- effect on hearing.22 There are no human studies demonstrating similar find- sues to allocate resources towards screening, rehabilitation, and hearing aid ings, probably because the mouse model of cerebral malaria is not directly programs, for an issue which may be preventable.19 comparable to human cerebral malaria.

12 ANNALS OF THE ACTM May 2014 Table 1 Summary of evidence suggesting the direct effects of malaria on hearing

Method of Study type, sample size FINDINGS Author & reference measurement

Falciparum infection may potentially Systematic literature review Review of 14 studies Zhao and Mackenzie20 lead to hearing loss

Clinical trial artesunate- amodiaquine (n=37), Audiometric thresholds compared In the acute setting malaria may artemether- Adjei et al21 to controls cause elevated hearing thresholds lumefantrine (n=35), or amodiaquine (n=8) in children

Mice murine cerebral malaria Auditory brainstem response testing Significant hearing impairment Schmutzhard et al22 (n=20)

Proposed mechanisms for hearing loss in malaria trial demonstrated that artemether-lumefantrine does not appear to cause hearing impairment. The one case of a patient on artemether-lumefantrine Zhao and Mackenzie suggest that multiple mechanisms ultimately lead to with significant deafness on day 90 was attributed to malaria reinfection.28 damage of three core areas involved in hearing. These include the coch- lea, the cochlear nerve, and the brainstem and its associated structures that In the same study, it was found that quinine was associated with a hearing participate in hearing.20 Indirectly, malaria may increase risk of acquiring loss of 30 dB, although this was transient.28 A significant number of patients other auditory damaging infections which can lead to hearing loss.20 Lo- (9 out of 30) on day 7 of quinine therapy complained of hearing issues, but cal microvascular changes of cochlear end arteries has been suggested to all but one complaint about deafness disappeared by the day 28 visit, and no be the prime mechanism of contributing directly towards hearing loss and hearing problems were reported by day 90.28 Quinine is well recognised to this pathological alteration may be implicated in lowering resistance to other cause toxicity to the outer hair cells of the ear, but its effect is reversible and infections.23 the patients in this study did not complain of hearing issues by day 90.28 His- torically, quinine derivatives have a well-recognised ototoxic profile, as cin- The lack of causation between uncomplicated malaria and hearing loss may chonism, a syndrome that occurs from quinine toxicity, results in headache, be because any deafness suffered from uncomplicated malaria is not rec- deafness, and anaphylactic shock.3 In Tanzania, a large number of children ognised or is underreported.20 This might be compounded by the fact that were suspected of having profound sensorineural deafness.29 A subsequent the issue has not been studied in detail, according to the lack of research in pilot study found 36% of deaf children had received intramuscular quinine the literature. The reviewed clinical studies support the need for further re- and/or gentamicin for ‘fever’ on admission.29 This highlights the potential search in the area due to the considerable absence of evidence, especially in for ototoxicity due to the improper use and monitoring of these two drugs, humans. Prospective studies should be performed to help establish whether as they both have ototoxic properties. A study by Flanagan et al evaluat- causation exists. ing the role of measuring routine quinine levels found that hearing loss due Antimalarial agents and ototoxicity to quinine occurs regardless of oral or parenteral route but is likely to be completely reversible.30 Quinine plasma levels do not appear to correlate Artemisinins have been thoroughly studied due to concerns regarding well with hearing loss, and individual risk may be better assessed by hearing neurotoxicity based on animal studies.10,11 Amodiaquine-artesunate and tests rather than measuring quinine levels.30 This is supported by an older artemether-lumefantrine were randomised in an efficacy and safety trial in study that produced similar results, with reversal of hearing loss within one Ghana, powered by 227 children aged six months to 14 years with uncom- week, according to audiograms.31 Evidence suggests that although quinine plicated malaria.24 No significant differences in hearing were reported during may produce deafness, it is a transient event that may completely return to the one year study, which included a monthly follow-up period. A matched normal, and that audiometric testing rather than blood tests may be done for case-control that involved 68 adults on artemether-lumefantrine compared those patients on quinine to reliably monitor its ototoxic effects. with 68 matched-controls similarly did not show evidence that artemether- lumefantrine causes hearing impairment.25 Hearing loss was common in Mefloquine might potentially affect hearing, since the side effect profile of this study in both groups, but statistical analysis showed that only age was mefloquine includes neuropsychiatric disturbances, likely due to its neuro- significantly associated, and was unrelated to artemether-lumefantrine expo- toxic properties.32,33 However, an open-label phase IV study of African chil- sure.25 These findings are supported by a randomised prospective study by dren with uncomplicated malaria treated with three-day artesunate and me- Carrasquilla et al that found no significant difference in auditory brainstem floquine failed to demonstrate convincing evidence of hearing loss (95% CI: response (ABR) in 265 patients randomised to artemether-lumefantrine or 0.0 to 1.73 % of patients), although this was not the direct end-point of their artesunate-mefloquine.26 One early study evaluating audiometric changes study.34 A prospective study of 93 patients with uncomplicated malaria on during use of artemether-lumefantrine for uncomplicated malaria was a three-day mefloquine-artesunate in Thailand also could not suggest clinical matched case-control trial study of 300 patients.27 This study found a sig- evidence via audiometric and ABR testing.35 This study tested ABR on day nificant (p<0.05), albeit small, hearing impairment at frequencies of 1-8 kHz 0 and day 7, and did not find evidence to suggest artesunate or mefloquine but not in the 500 to 250 Hz range. However, the samples collected were are toxic to the auditory system.35 One Canadian case report concerned a from a construction site, and there was no information provided about sam- 67-year-old woman who had mefloquine chemoprophylaxis and developed pling and differences of noise exposure between the two groups, making unilateral sudden profound sensorineural hearing loss and tinnitus, which conclusions drawn from this study less reliable.27 A randomised trial as- resolved spontaneously at day 93 of her illness, but no causal interpretation sessing artemether-lumefantrine, quinine, and atovaquone-proguanil evalu- could be made.36 ated hearing function with tuning fork tests, pure tone audiometry, transient Artemether-lumefantrine does not appear to cause hearing loss and this evoked and distortion product otoacoustic emission, and ABR before, dur- evidence supports its safety in this context. These clinical studies confirm ing, and after treatment on days 0, 7, 28, and 90.28 This comprehensive

Volume 15 Number 1 13 ANNALS OF THE ACTM 13 Table 2 Summary of studies on antimalarials and relationship with hearing function

Study Type and Total Author & Study Factor, Sample Size, and Methodology Findings Size reference

Random treatment of uncomplicated malaria in children 6 months Rare instances of side effects Randomised efficacy and to 14 years to amodiaquone-artesunate (n=116) or artemether- and no reports of hearing Adjei et al24 safety trial (n=227) lumefantrine (n=111), follow-up in 28 days for reports of adverse impairment nor abnormal events neurological signs

Case-control Group treated with artemether-lumefantrine within 5 years (n=68) No audiometric difference Hutagalung et al25 (n=136) vs matched controls (n=68) compared for audiometric differences between cases and controls

Randomised, prospective Random assignment to artemether-lumefantrine (n=159), No difference in ABR; 2-4 open-label three-arm atovaquone-proguanil (n=53), or artesunate-mefloquine (n=53) and dB pure-tone threshold Carrasquilla et al26 study (n=265) auditory brain response (ABR) comparison improvement

Artemether-lumefantrine (n=150) to treat uncomplicated malaria Subjects in the artemether- Matched case-control trial vs age-, gender-, weight- and race-matched controls (n=150) who lumefantrine group had Toovey and (n=300) did not suffer malaria and did not have antimalarial therapy with significant hearing loss Jamieson27 comparison of audiometric findings compared to controls

No impact of standard oral dosing of artemether- Randomised double- Random assignment to receive artemether/lumefantrine (n=23) or lumefantrine on either blinded controlled trial quinine (n=26) or atovaquone/proguanil (n=28) and comparison of peripheral or brainstem Gurkov et al28 (n=77) peripheral and brainstem auditory pathway assessment auditory pathways; transient hearing loss on day 7 with quinine therapy

Hearing loss occurs almost Retrospective audit invariably but may not Quinine plasma level results and correlation with toxicity effects Flanagan et al30 (n=82) occur in a dose-response relationship

Experimental study Audiometric testing before and after doses of quinine administered High frequency loss reversible Tange et al31 (n=22) in 12 healthy subjects and 10 patients with falciparum malaria within one week

Treatment of African children with uncomplicated P. falciparum Experimental phase IV treated over three days with artesunate-mefloquine combination No reports of hearing loss Frey et al34 open label study (n=213) with follow-up to report adverse events

3-day oral artesunate-mefloquine in patients with uncomplicated No threshold changes of Prospective study (n=93) falciparum and auditory testing before the first dose and seven days significance (changes did not Carrara et al35 after the first dose was started exceed 10 dB)

Sudden right ear severe sensorineural hearing loss Case report Case analysis of a woman on chemoprophylactic weekly mefloquine and tinnitus with spontaneous Wise and Toovey36 resolution after day 93 of hearing loss

14 ANNALS OF THE ACTM May 2014 existing knowledge of quinine as an ototoxic drug that is used in malaria 16� Nankabirwa J, Wandera B, Kiwanuka N, Staedke SG, Kamya MR, Brooker SJ. Asymptomatic plasmodium infection and cognition among primary schoolchildren in a high malaria transmission setting in Uganda. Am J treatment. Limited available evidence does not currently support the hy- Trop Med Hyg. 2013 Apr 15. pothesis that mefloquine is ototoxic. There were no other clinical reports 17� Thuilliez J, Sissoko MS, Toure OB, Kamate P, Berthelemy JC, Doumbo OK. Malaria and primary education in of ototoxicity of other antimalarial drugs, including atovaquone-proguanil, Mali: a longitudinal study in the village of Doneguebougou. Soc Sci Med. 2010;71(2):324-34. sulfadoxine-pyrimethamine, or halofantrine. 18� Fernando D, de Silva D, Carter R, Mendis KN, Wickremasinghe R. A randomized, double-blind, placebo-controlled, clinical trial of the impact of malaria prevention on the educational attainment of school children. Am J Conclusions Trop Med Hyg. 2006 Mar;74(3):386-93. 19. Kotby MN, Tawfik S, Aziz A, Taha H. Public health impact of hearing impairment and disability. Folia Phoniatr This review of the literature found a small number of studies that support an Logop. 2008;60(2):58-63. association between malaria and hearing loss. This suggests that further re- 20� Zhao SZ, Mackenzie IJ. Deafness: malaria as a forgotten cause. Ann Trop Paediatr. 2011;31(1):1-10. search should be done to better clarify this association and its impact on the 21� Adjei GO, Goka BQ, Kitcher E, Rodrigues OP, Badoe E, Kurtzhals JA. Reversible audiometric threshold changes individual. In particular, studies should concentrate on the long-term conse- in children with uncomplicated malaria. J Trop Med. 2013;2013:360540. quences, if any, of this phenomenon. Most antimalarial agents appear not to 22� Schmutzhard J, Kositz CH, Lackner P, Dietmann A, Fischer M, Glueckert R, et al. Murine malaria is associated cause ototoxic reactions, with the exception of quinine, which has a predict- with significant hearing impairment. Malar J. 2010;9:159. able response. The evidence for deafness due to mefloquine is poor. The lack 23. Dunmade AD, Segun-Busari S, Olajide TG, Ologe FE. Profound bilateral sensorineural hearing loss in Nigerian of strong evidence of a direct correlation between malaria and hearing loss children: any shift in etiology? J Deaf Stud Deaf Educ. 2007 Winter;12(1):112-8. may be because any deafness suffered from malaria is not appreciated or is 24. Adjei G, Kurtzhals J, Rodrigues O, Alifrangis M, Hoegberg L, Kitcher E, et al. Amodiaquine-artesunate vs artemether-lumefantrine for uncomplicated malaria in Ghanaian children: a randomized efficacy and safety underreported. This might be compounded by the fact that the issue has not trial with one year follow-up. Malaria Journal. 2008;7(1):127. been well recognised, suggested by the limited studies within the literature. 25. Hutagalung R, Htoo H, Nwee P, Arunkamomkiri J, Zwang J, Carrara VI, et al. A case-control auditory evaluation In addition, there are few studies which record a baseline auditory threshold of patients treated with artemether-lumefantrine. Am J Trop Med Hyg. 2006 Feb;74(2):211-4. before becoming infected with malaria. Both clinical and non-clinical studies 26. Carrasquilla G, Barón C, Monsell EM, Cousin M, Walter V, Lefèvre G, et al. Randomized, prospective, three- support further clinical research to identify an association and clarify mecha- arm study to confirm the auditory safety and efficacy of artemether-lumefantrine in Colombian patients with uncomplicated Plasmodium falciparum malaria. The American Journal of Tropical Medicine and Hygiene. nisms that cause deafness during a malarial episode, as this may provide a 2012 January 1, 2012;86(1):75-83. strategy to prevent potentially avoidable cases of hearing loss. 27. Toovey S, Jamieson A. Audiometric changes associated with the treatment of uncomplicated falciparum malaria with co-artemether. Trans R Soc Trop Med Hyg. 2004 May;98(5):261-7; discussion 268-9.

References 28. Gurkov R, Eshetu T, Miranda I, Berens-Riha N, Mamo Y, Girma T, et al. Ototoxicity of artemether/lumefantrine in the treatment of falciparum malaria: a randomized trial. Malaria Journal. 2008;7(1):179. 1� WHO. World Malaria Report, 2011. Geneva: World Health Organization. Available from: http://www.who.int/ malaria/world_malaria_report_2011/en/index.html (accessed 23/04/2013) 29. Freeland A, Jones J, Mohammed NK. Sensorineural deafness in Tanzanian children--is ototoxicity a significant cause? A pilot study. Int J Pediatr Otorhinolaryngol. 2010 May;74(5):516-9. 2. Carter JA, Mung’ala-Odera V, Neville BGR, Murira G, Mturi N, Musumba C, et al. Persistent neurocognitive impairments associated with severe falciparum malaria in Kenyan children. Journal of Neurology, Neurosurgery 30. Flanagan KL, Buckley-Sharp M, Doherty T, Whitty CJ. Quinine levels revisited: the value of routine drug level & Psychiatry. 2005;76(4):476-481. monitoring for those on parenteral therapy. Acta Trop. 2006 Feb;97(2):233-7. 3. AlKadi HO. Antimalarial drug toxicity: a review. Chemotherapy. 2007;53(6):385-391. 31. Tange RA, Dreschler WA, Claessen FA, Perenboom RM. Ototoxic reactions of quinine in healthy persons and patients with Plasmodium falciparum infection. Auris Nasus Larynx. 1997 Apr;24(2):131-6. 4. Chukuezi A. Hearing loss: a possible consequence of malaria. Afr Health. 1995;17(6):18-9. 32. Schneider C, Adamcova M, Jick SS, Schlagenhauf P, Miller MK, Rhein HG, et al. Antimalarial chemoprophy- 5. Smith AW. The World Health Organization and the prevention of deafness and hearing impairment caused by laxis and the risk of neuropsychiatric disorders. Travel Med Infect Dis. 2013 Mar-Apr;11(2):71-80. noise. Noise Health. 1998;1(1):6-12. 33. Toovey S. Mefloquine neurotoxicity: A literature review. Travel Medicine and Infectious Disease. 2009;7(1):2- 6. WHO . Deafness and hearing loss. Geneva: World Health Organization, 2013. Available from: http://www.who. 6. int/mediacentre/factsheets/fs300/en/ (accessed 11/05/2013) 34. Frey SG, Chelo D, Kinkela MN, Djoukoue F, Tietche F, Hatz C, et al. Artesunate-mefloquine combination therapy 7. Mung’Ala-Odera V, Snow RW, Newton CR. The burden of the neurocognitive impairment associated with in acute Plasmodium falciparum malaria in young children: a field study regarding neurological and neuropsy- Plasmodium falciparum malaria in sub-saharan Africa. Am J Trop Med Hyg. 2004;71(2 Suppl):64-70. chiatric safety. Malar J. 2010;9:291. 8. Taylor WR, White NJ. Antimalarial drug toxicity: a review. Drug Saf. 2004;27(1):25-61. 35. Carrara V, Phyo A, Nwee P, Soe M, Htoo H, Arunkamomkiri J, et al. Auditory assessment of patients with acute uncomplicated Plasmodium falciparum malaria treated with three-day mefloquine-artesunate on the north- 9. Omari Aika AA, Gamble Carrol L, Garner P. Artemether-lumefantrine (six-dose regimen) for treating uncompli- western border of Thailand. Malaria Journal. 2008;7(1):233. cated falciparum malaria. Cochrane Database of Systematic Reviews [serial on Internet]. 2005; (4): Available from: http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD005564/abstract. 36. Wise M, Toovey S. Reversible hearing loss in temporal association with chemoprophylactic mefloquine use. Travel Med Infect Dis. 2007 Nov;5(6):385-8. 10� genovese RF, Newman DB. Understanding artemisinin-induced brainstem neurotoxicity. Arch Toxicol. 2008;82(6):379-85.

11� Erickson RI, Defensor EB, Fairchild DG, Mirsalis JC, Steinmetz KL. Neurological assessments after treatment with the antimalarial beta-arteether in neonatal and adult rats. Neurotoxicology. 2011;32(4):432-40.

12� Tucci D, Merson MH, Wilson BS. A summary of the literature on global hearing impairment: current status and priorities for action. Otol Neurotol. 2010;31(1):31-41. Corresponding Author

13� i dro R, Marsh K, John CC, Newton CRJ. Cerebral malaria: mechanisms of brain injury and strategies for Dr Eric Fong improved neurocognitive outcome. Pediatr Res. 2010;68(4):267-274. School of Public Health and Tropical Medicine, 14� ramkalawan TW, Davis AC. The effects of hearing loss and age of intervention on some language metrics in James Cook University, Townsville young hearing-impaired children. British Journal of Audiology. 1992;26(2):97-107.

15� Al Serouri AW, Grantham-McGregor SM, Greenwood B, Costello A. Impact of asymptomatic malaria parasi- Email: [email protected] taemia on cognitive function and school achievement of schoolchildren in the Yemen Republic. Parasitology. 2000;121 ( Pt 4):337-45.

Volume 15 Number 1 15 ANNALS OF THE ACTM 15 CASE REPORT

TB or not TB: The clinical and diagnostic challenge of suspected pericardial tuberculosis

Victoria G Hall

Royal Melbourne Hospital, Grattan St, Parkville VIC

ABSTRACT

A 61-year-old Vietnamese-born male presented to an inner-city tertiary hospital after urgent medivac retrieval from Vietnam. He had a presumed diagnosis of severe cardiogenic failure, and on admission rapidly deteriorated with a large pericardial effusion and cardiac tamponade seen on echocardiogram. He was admitted to the intensive care unit, and underwent investigation and treatment for suspected tuberculous (TB) pericarditis. All diagnostic mea- Figure 1 Initial chest x-ray sures available, however, were negative for TB. This case highlights the clinical and diagnostic challenge of pericardial TB, and that further research is needed to aid clinicians. A few hours after admission the patient required emergency attention for hypotension, with systolic blood pressure recorded as 75 mmHg. An urgent Keywords: tuberculosis, pericarditis, diagnosis, interferon-gamma, PCR bedside transthoracic echocardiogram (TTE) demonstrated a large pericardial effusion (4 cm), preserved left ventricular function with mild left ventric- Ann Australas Coll Trop Med 2014;15(1):16-19. ular hypertrophy, and evidence of early signs of cardiac tamponade. There was no evidence of hypertrophic cardiomyopathy. Pericardiocentesis was performed, with 1.2 litres of blood-stained pericardial fluid drained. Results Introduction of the pericardial fluid analysis, showing an exudate, are detailed in Table 1. Tuberculous pericarditis is an uncommon complication of TB, and rarely The patient was reviewed by the Infectious Disease team, who thought that seen in the developed world due to low overall prevalence.1 However, world- a diagnosis of pericardial TB was unlikely, due to the absence of prodomal wide, TB is still a major cause of morbidity and mortality, with progress sub- symptoms and no known contacts with TB, with the only risk factor being stantially undermined by the HIV epidemic and emerging drug resistance.2 country of birth. They suggested investigation for TB, a liver cirrhosis work- It remains a disease of poverty, linked to overcrowding and undernutrition.2 up including abdominal ultrasound, and to continue meropenem to cover for The increasing prevalence of co-infection with HIV in developing countries ongoing possible sepsis. has led to a marked increase in rates of TB pericarditis, and in endemic areas, On day three of admission, the patient was again hypotensive and febrile. aetiology of pericarditis is most commonly attributed to TB.1 Vancomycin was added to meropenem to cover for nosocomially-acquired Case report infection. Repeat TTE showed small pericardial effusion and normal pericardial thickness. The patient was transferred to the ICU for haemodynamic A 61-year-old Vietnamese-born male, having lived in Australia for the past six monitoring and support. years, presented to a large inner-city tertiary hospital after urgent medivac retrieval from Vietnam. He had a background history notable for atrial fibrilla- All cultures as part of the septic screen were negative, as were sputum tion (on warfarin), with pacemaker for tachy-brady syndrome, hypertension, samples for acid-fast bacilli and a QuantiFERON-TB Gold test. Abdominal dyslipidemia and non-alcoholic steatohepatitis. He was previously independ- ultrasound revealed small volume ascites and the liver cirrhosis screen was ent, a practising Buddhist, and lived alone with a supportive family close by. normal. Further investigations were done: rheumatological screen, serology He often holidayed in Vietnam to visit friends and family. for schistosomiasis and stronglyloides, cytomegalovirus, Epstein-Barr virus and HIV serology; all were negative. Whilst in Vietnam he was admitted to hospital two months prior to his transfer with severe biventricular heart failure. His admission was complicated by A computed tomography of the chest, abdomen and pelvis was ordered for an extended-spectrum beta-lactamase E. coli catheter-related sepsis, requir- further assessment and to investigate the possibility of malignancy. It re- ing transfer to the intensive care unit (ICU). Despite a prolonged course of vealed a re-accumulated large pericardial effusion and large bilateral pleural meropenem and amikacin, he had persistent hypotension and anasarca un- effusions. There was no evidence of malignancy or lymphadenopathy. responsive to aggressive diuresis. Echocardiography performed in Vietnam The impression from the Infectious Disease team was that there was not was reported as asymmetric hypertrophic cardiomyopathy with no evidence enough evidence to diagnose and treat pericardial TB. However, as it re- of flow obstruction and moderate pulmonary hypertension. mained the primary differential diagnosis, pericardial and pleural biopsy was On admission to hospital in Australia, the patient appeared to be in cardio- recommended. This occurred uneventfully (Table 2) and the patient was genic shock, with a systolic blood pressure of 85 mmHg. He was jaundiced, commenced on standard anti-TB therapy and high-dose prednisolone on with tender hepatomegaly and severe peripheral oedema. Chest x-ray (Fig. 1) day eight of admission. showed a large globular heart and bilateral pleural effusions. The patient was The patient remained in the ICU for a further three weeks with minimal clini- afebrile, with normal white cell count and C-reactive protein of 20 mg/L. Liver cal response. Ongoing issues of hypotension requiring significant inotropic function tests revealed mixed picture deranged liver enzymes, hyperbiliru- support, severe right heart failure, sepsis unresponsive to broad antimicro- binemia and hypoalbuminemia. The working diagnosis was one of ongoing bial treatment, and multi-organ failure ensued. The patient remained on anti- cardiogenic failure, complicated by altered conscious state and congestive TB therapy and high-dose prednisolone. All results, including the pericardial hepatopathy due to severe right heart failure.

16 ANNALS OF THE ACTM May 2014 Table 1 Results of pericardial fluid analysis

Investigation Findings

Sodium 140 mmol/L Potassium 3.6 mmol/L Chloride 95 mmol/L Creatinine 89 umol/L Urea 10.9 mmol/L Glucose 4.3 mmol/L Albumin 23 g/L Lactate dehydrogenase 957 IU/L

Leukocytes + Microscopy & culture No organisms seen on Gram stain.

Microscopy for TB Auramine-rhodamine stain: no acid-fast bacilli detected

The smears contain scant mesothelial cells with foamy macrophages and inflammatory cells including eosinophils. No Cytology malignant cells are identified. Consistent with inflammation.

M. tuberculosis Negative complex DNA PCR

Mycobacterial culture Negative (after four weeks)

Table 2 Results of pericardial biopsy analysis

Investigation Findings

Sections show fibroadipose connective tissue. The stroma contains a sparse lymphoplasmacytic chronic inflammatory cell Histopathology infiltrate. No active or granulomatous inflammation is seen. No refractile foreign body material is identified. There is no evidence of malignancy. Conclusion: mild non-specific chronic inflammation.

Microscopy for TB Auramine-rhodamine stain: no acid-fast bacilli detected

M. tuberculosis Negative complex DNA PCR

Mycobacterial culture Negative

biopsy, were negative for TB. Repeat CT chest however revealed new right sweats and weight loss4; cough, dyspnoea and pleuritic chest pain are also upper lobe pulmonary nodules and sub-centimetre mediastinal lymph nodes common1. A minority of patients, however, present with features of constric- not noted on previous imaging. tive pericarditis;5,6 of these, most present with symptoms of cardiac failure.6 In light of this, and due to ongoing haemodynamic instability, the patient Pericardial involvement by M. tuberculosis commonly develops through underwent a full heart study. Haemodynamic findings were consistent with lymphatic spread from adjacent lung, tracheal, bronchial or mediastinal constriction or restriction. Definitive diagnosis could not be made as the lymph nodes, or via haematogeneous spread from primary infection.1,4 It is patient was in atrial fibrillation and respiratory variation was not able to be often due to reactivation and the primary source of infection is not found. It assessed due to the patient’s conscious state. A pericardectomy, from these is the immune response to the viable acid-fast bacilli in the pericardium that results, was not recommended. is responsible for morbidity, as a hypersensitivity reaction mediated by TH1 lymphocytes causes the resultant pericardial effusion.4 Despite maximal ICU support, the patient’s clinical state continued to dete- riorate. A decision was made, in accordance with the family, to cease active Constrictive pericarditis tends to manifest in one of three syndromes: cardiac management. The patient was palliated and died one day later. Unfortunately, tamponade, constrictive pericarditis, or effusive-constrictive pericarditis.6 due to religious beliefs, an autopsy was not performed. Effusive-constrictive pericarditis has evidence of both constriction and pericardial effusion, complicated by early signs of cardiac tamponade.6 It often Discussion becomes apparent after pericardiocentesis because constriction remains TB pericarditis typically presents with non-specific symptoms of fever, night even after removal of pericardial fluid. Chronic granulomatous inflammation

Volume 15 Number 1 17 ANNALS OF THE ACTM 17 leads to thickened, fibrosed pericardium with evidence of calcification.6,7 This patients from non-endemic areas, as it may add further morbidity without underlying pathology inhibits adequate diastolic filling, reducing venous re- diagnostic value.1,4,12 turn and lowering cardiac output.6 PCR for M. tuberculosis is available, as was performed in this case. Cegielski Prompt diagnosis and treatment of constrictive pericarditis due to TB is et al compared the use of PCR, culture and histopathology for diagnosis of paramount to reduce mortality.7,8,9 Diagnosis relies on isolation of tubercle TB on either pericardial fluid or tissue biopsy.15 The sensitivity of PCR was bacilli in pericardial fluid, which is often difficult.9 A definitive diagnosis is higher with tissue specimens (12 of 15; 80%) than with fluid specimens (2 only made with one of positive culture, acid-fast bacilli stain in pericardial of 13; 15%; P = 0.002).15 In their study, including both specimen types, TB fluid or detection of tubercle bacilli, and/or typical caseating granuloma on infection was correctly identified by culture in 30 out of 43 patients (70%) pericardial biopsy.2,8 It is considered likely in the presence of pericardial ef- compared with 14 out of 28 patients (50%) by PCR.15 There was one false fusion and TB isolated from sputum or other sites, with a positive clinical positive PCR result of a patient with Staphyloccocus aureus pericarditis.15 response to anti-TB therapy.2,9 The standard four drug anti-tuberculous therapy remains the cornerstone Negative initial investigations do not rule out TB as the cause of pericardi- of treatment of TB pericarditis, with reduction in mortality from 80-90% to tis. Smears with Ziehl-Neelsen (ZN) staining of pericardial fluid have poor 8-17% in HIV-negative patients, and the development of constrictive pericar- sensitivity, with only 40-60% being positive.8 Culture is often slow, taking ditis reduced from 88% to 10-20%.1 up to 6-8 weeks, and is also of low sensitivity.2,8 Empiric therapy is often The exact role of adjuvant corticosteroids remains unclear.1,4,7,16,17 Only small commenced with clinicians relying heavily on clinical grounds only for diag- studies have been conducted, with two of them before the HIV era. Limited nosis.8 However, this subjects the patient to unnecessary adverse effects and data suggests that the use of corticosteroids can shorten the time of clinical long duration of anti-TB chemotherapy.8 symptoms and prevent fluid reaccumulation.17 Their use does not appear Initial evaluation of TB pericarditis is as demonstrated by the case. Echo- to affect the likelihood of progression to constriction.1 Guidelines from the cardiography is useful and accurate for the detection of pericardial effusion American Thoracic Society, Centers for Disease Control, and Infectious Dis- and tamponade.2,8 The suggestion of tamponade is an absolute indication for eases Society of America issued in 2003 support the use of prednisolone pericardiocentesis. Resultant pericardial fluid analysis due to TB is classically 60 mg/day for 4 weeks with a tapering dose over the next 7 weeks.1 Further an exudate, with high protein level and lymphocytic counts.2 research is needed, with much anticipated results from the Investigation of the Management of Pericarditis (IMPI) trial, a large multicentre randomised Unfortunately, due to resource limitation, pericardial fluid in our case was not controlled trial currently underway.5 analysed for adenosine deaminase concentration (ADA). Its measurement has been found to be useful, with levels ranging from 30-60 U/L indicative Pericardectomy is indicated in patients with persistent constrictive peri- of the presence of TB pericarditis.2 An observational study by Reuter et al8 carditis. The timing and appropriate candidacy for surgical intervention is on 233 patients in South Africa found that an ADA level of >40 U/L had an controversial. Some recommend a trial of medical therapy, with pericardiec- 87% sensitivity and 89% specificity for TB. Another study in Korea used both tomy for patients who do not clinically respond after 4 to 8 weeks.1, 4 Others ADA and carcinoembryonic antigen to investigate the aetiology of pericardial suggest all patients with constrictive pericarditis should undergo pericardec- fluid. They found no significant difference in ADA level between the group tomy once medical therapy has commenced.1 of patients with definite TB and those with suspected TB. Their study found a sensitivity and specificity of 93% and 97% respectively for a cut-off level Conclusion of >40 U/L.10 Higher levels of ADA have also been found to be a prognostic There is still much to know about tuberculous pericarditis. There is scope for indicator for the development of constrictive pericarditis and subsequent further research in diagnosis and clinical management, including the use of pericardectomy.2,11 corticosteroids. TB pericarditis remains an uncommon but important clinical manifestation of extra-pulmonary TB, requiring a high index of suspicion on The advent of testing for pericardial interferon-gamma, using the T.SPOT TB the clinician’s behalf. Worldwide, ongoing measures to reduce the incidence assay, and its diagnostic potential for TB pericarditis has been described in of infection with M. tuberculosis must be continued with vigilance. numerous studies.8,9,12,13 The T.SPOT TB assay is an enzyme-linked immuno- spot (ELISPOT) assay that measures the gamma interferon response of lymphocytes to the tuberculosis antigens ESAT-6 and CFP-10.13,14 The result is reported as number of IFN-gamma-producing T cells (spot-forming cells).14 If the spot counts in the TB antigen wells exceed a specific threshold relative References to the control wells, an individual is considered positive for M. tuberculo- 1� Stout J. Tuberculous pericarditis. UptoDate, 2013. Available from: http://www.uptodate.com/contents/tuber- 14 culous-pericarditis?source=search_result&search=tuberculous+pericarditis&selectedTitle=1~18 (accessed sis infection. The test should also be performed concurrently on peripheral 22 May 2013) blood; a significantly positive ratio between pericardial and peripheral blood 2� World Health Organization. Tuberculosis Fact Sheet. Available from: http://www.who.int/mediacentre/fact- suggests compartmentalisation and adds further evidence to the presence of sheets/fs104/en/ (accessed 22 May 2013) tuberculous pericarditis.9,12 3� Hoa NB, Sy DN, Nhung NV, Tiemersma EW, Borgdorff MW, Cobelens FGJ. National survey of tuberculosis The TB.SPOT TB assay differs from the QuantiFERON-TB Gold In-Tube (QFT- prevalence in Viet Nam. Bulletin of the World Health Organization 2010:88:273-280. Available from: http:// www.who.int/bulletin/volumes/88/4/09-067801/en/ GIT) assay used in Australia, an ELISA-based test that uses three TB antigens (ESAT-6, CFP-10, TB7.7).14 Both have FDA approval. QFT-GIT is a quantifica- 4. Bongani M, Burgess LJ, Anton FD. Tuberculous pericarditis. Circulation 2005; 112:3608-3616.

tion of the IFN-gamma response to TB antigens, and is considered positive 5� Mayosi BM, Ntsekhe M, Bosch J, Pogue J, Gumedze F, Badri M et al. Rationale and design of the Investigation for M. tuberculosis infection if the response is above a test cut-off.14 The sen- of the Management of Pericarditis (IMPI) trial: a 2 × 2 factorial randomized double-blind multicenter trial of adjunctive prednisolone and Mycobacterium w immunotherapy in tuberculous pericarditis. American Heart sitivity of T-SPOT.TB has been reported as 90%, superior to that of QFT-GIT, Journal 2013;165(2):109-115. with a sensitivity of 80%. Both have specificity greater than 95%.14 Neither 6� Hoit BD. Constrictive pericarditis. UptoDate, 2013. Available from: http://www.uptodate.com/contents/con- is accurate in the differentiation between latent and active tuberculosis infec- strictive-pericarditis?source=search_result&search=tuberculous+pericarditis&selectedTitle=4~18 (accessed tion.14 There is no available literature on the use of QFT-GIT as a diagnostic 22 May 2013)

aid for pericardial TB. 7� y en-Wen L, Huey-Ru T, Wen-Huang L, Li-Jen L, Jyh-Hong C. Tuberculous constrictive pericarditis with con- current active pulmonary tuberculosis infection: a case report. Cases Journal 2009;2:7010. Pericardial tissue can be sent for microscopy and culture, and examined his- 1 8� r euter H, Burgess L, van Vuuren W, Doubell A. Diagnosing tuberculous pericarditis. European Heart Journal tologically for granulomatous inflammation and caseous necrosis. Biopsy of 2006;99(12):827. the pericardium is usually reserved for when clinical suspicion is still high, duration of illness more than 3 weeks, and the aforementioned investigations 9� Biglino A, Crivelli P, Concialdi E, Bolla C, Montrucchio G. Clinical usefulness of ELISPOT assay on pericardial fluid in a case of suspected tuberculous pericarditis. Infection 2008;36:601-604. have not yielded a positive result.1,4 However, it is usually only performed on

18 ANNALS OF THE ACTM May 2014 10� Koh KK, Kim EJ, Cho CH, Choi MJ, Cho SK, Kim SS et al. Adenosine deaminase and carcinoembryonic antigen agement of pericardial diseases executive summary; the task force on the diagnosis and management of in pericardial effusion diagnosis, especially in suspected tuberculous pericarditis. Circulation 2004;89:2728- pericardial diseases of the European Society of Cardiology. European Heart Journal 2004;25(7):587. 2735. 17� Al-Shehri F, Chan K, Dennie C, Veinot J. Tuberculous pericarditis. Current Cardiology Reviews 2005;1(2):165- 11. Komsuoğlu B, GöldelïO, Kulan K, Komsuoğlu SS. The diagnostic and prognostic value of adenosine deaminase 169. in tuberculous pericarditis. European Heart Journal 1995;16(8):1126. 18� Critchley JA, Young F, Orton L, Garner P. Corticosteroids for prevention of mortality in people with tuberculo- 12� Mehta PK, Raj A, Singh N, Khuller GK. Diagnosis of extrapulmonary tuberculosis by PCR. Immunology and sis: a systematic review and meta-analysis. Lancet Infectious Diseases 2013;13(3):16-188. Medical Microbiology 2012;66:20-36.

13� Bathoorn E, Limburg A, Bouwman JJ, Bossink AW, Thijsen SF. Diagnostic potential of an enzyme-linked im- munospot assay in tuberculous pericarditis. Clinical Vaccine Immunology 2011;18:874-878.

14� Madhukar P, Menzies D. Interferon-gamma release assays for diagnosis of latent tuberculosis infection. Corresponding Author UptoDate, 2013. Available from: http://www.uptodate.com/contents/interferon-gamma-release-assays-for- diagnosis-of-latent-tuberculosis-infection?source=search_result&search=IGRA&selectedTitle=1%7E60 (ac- Dr Victoria G Hall cessed 22 May 2013) Royal Melbourne Hospital, Grattan St, Parkville VIC 15� Cegielski JP, Devlin BH, Morris AJ, Kitinya JN, Pulipaka UP, Lema LE et al. Comparison of PCR, culture, and histopathology for diagnosis of tuberculous pericarditis. Journal of Clinical Microbiology 1997;35:3254. Email: [email protected]

16. Maisch B, SeferovićPM, RistićAD, Erbel R, Rienmüller R, Adler Y et al. Guidelines on the diagnosis and man-

Case report CASE REPORT A 24-year-old Victorian sheep officer presented to her general practitioner in June 2009 with a 3-month history of tiredness, intermittent headaches, dry cough, vomiting and diarrhoea. On presentation, all her vital signs and An unusual case of Q fever in an physical examination were unremarkable, except that she was obese with a immunised sheep officer body mass index of 35. Investigations included full blood examination, elec- trolytes, thyroid function test, iron, vitamin B12, folate and coeliac serology; Frank C H Ng1,2 and Peter A Leggat2,3 all were within normal limits. Infectious diseases screening for toxoplasmo- sis and Epstein-Barr virus was negative. Cytomegalovirus IgG was positive, 1. Diamond Creek Medical Centre, Diamond Creek, Victoria, Australia but IgM was negative. Subsequent Q fever serology (Panbio ELISA, Alere) 2. School of Public Health, Tropical Medicine and Rehabilitation Sciences, James Cook revealed positive phase 2 IgG and negative phase 2 IgM, whereas the phase University, Townsville, Queensland, Australia 1 complement fixation titre (CFT) was 16 and phase 2 CFT was 8 (Table 1). The pathologist’s interpretation was ‘Q fever serology suggestive of chronic 3. School of Public Health, University of the Witwatersrand, Johannesburg, South Africa Q fever infection. The seropositivity unlikely to be due to vaccination’.2 The patient returned for follow up a few months later in October 2009. She ABSTRACT continued to experience intermittent headaches, fatigue, dry cough and vomiting. On further questioning, she had received Q fever vaccination on 19th It is often thought that Q fever vaccination is 100% protective against the March 2009, with a negative pre-vaccination test performed on 12th March disease. An unusual case of Q fever and its debilitating consequences is de- 2009, one week prior to the immunisation. She was tested by her company scribed here in a sheep industry development officer, who had been immun- as she took part in a Sheep Centres for Research Excellence project looking ised. This report highlights the need for further research into improving Q at DNA markers in specific recorded flocks from around the state. Being a fever testing and taking appropriate preventive measures against the disease. sheep officer, her other duties involved attending abattoirs for carcase com- An urgent review of the current occupational health and safety policy is rec- petitions, as well as going to sheep yards, shearing sheds, and sale yards to ommended, with implementation of universal mandatory Q fever immunisa- investigate sheep mortalities. She recalled that she had commenced work tion for all new abattoir, cattle, goat, sheep and other workers, who may be three weeks prior to the testing. The patient’s employer did not enquire about exposed to Q fever, prior to the commencement of any field work. her Q fever vaccination status prior to her employment and the employer was not concerned that she started work prior to being tested and immunised. Keywords: Q fever, zoonosis, vaccination, agriculture, prevention Furthermore, the patient was unaware that she needed to avoid exposure to Q fever for two weeks following the vaccination, as it takes approximately Ann Australas Coll Trop Med 2014;15(1):19-20. two weeks for protective immunity to develop.3 Further investigations to exclude Q fever complications were normal, including transthoracic echocardiogram, chest X-ray, and brain computed tomog- Background raphy scan. Repeat full blood examination revealed mild leukocytosis (11.8 Q fever is a common zoonosis, with an almost worldwide distribution, caused x 109/L; reference range (RR) 4-11 x 109/L) and lymphocytosis (4.1 x 109/L; by Coxiella burnetii.1 Farm animals and pets are the principal reservoirs of in- RR 1-4 x 109/L). Erythrocyte sedimentary rate was raised (23 mm/hr; RR, fection and transmission to humans is usually via inhalation of contaminated <12 mm/hr). Brucellosis and leptospirosis serology was negative. Repeat Q aerosols. Occupational groups with close association with farm or wild ani- fever serology was identical to that performed earlier. mals are most at risk. Immunisation is usually protective. A case of Q fever The patient was referred to an infectious disease physician. A 4-week course in a previously immunised Victorian sheep industry development officer (a of doxycycline 100 mg twice daily was prescribed. In the following month, ‘sheep officer’) is presented. In general, the qualification requirements of the patient reported moderate improvement in symptoms, with less cough, a sheep officer include an undergraduate degree in the areas of science or fewer headaches and resolution of diarrhoea. In December 2009, the patient agriculture, with optional industrial training or postgraduate education. At the still complained of ongoing fatigue, dizziness, cough, headache and myalgia. time of the project, there was universal no Q fever vaccination program nor a In February 2010, her symptoms remained the same. Q fever serology was requirement for vaccination in place for new staff such as this sheep officer. again repeated which revealed a reduction in the level of phase 1 CFT (from Although presumably she would have been aware of the disease, there was 16 to 8). Other serological parameters remained the same (see Table 1). no formal training in prevention of Q fever required for the role.

Volume 15 Number 1 19 ANNALS OF THE ACTM 19 Table 1 Q fever CFT serology titres With respect to treatment, doxycycline 100 mg twice daily was recommended to treat acute Q fever.8 Although a prolonged course of treatment is sug- JUNE OCTOBER FEBRUARY 9 Investigation gested to treat chronic Q fever endocarditis, there is no universal consensus 2009 2009 2010** as to the best treatment for chronic infection. In this instance, as the initial Phase 1 CFT serology was low positive and without any systemic complications, only a 16 16 8 titre one month course of doxycycline was prescribed, with good response. Despite treatment, the patient continued to have symptoms ascribed to post- Phase 2 CFT 8* 8 8 Q fever fatigue syndrome. Patients often present with prolonged lassitude, titre myalgia, arthralgia, sweats and painful lymphadenopathy.1,10 The dominant Phase 2 IgM Negative Negative Negative symptom is an incapacitating fatigue out of proportion to the degree of prior activity, as exemplified in this case. This often lasts beyond a year, up to Phase 2 IgG Positive Positive Positive more than 10 years.11 A United Kingdom study reported that chronic fatigue was experienced in 65% of Q fever sufferers.12 In Australia, post-Q fever fa- *A single CFT titre of >8 for antibodies to phase II antigen indicates past infection.2 tigue syndrome is the most common chronic sequel to acute Q fever, affect- ** A trial of doxycycline was given in Nov 2009. ing 10-15% of patients.13 It is believed that cytokines and other immune me- diators generated by the dysregulated cell-mediated immunity in response to The patient continued to experience fatigue with minimal exertion physically a persisting low level of C. burnetii antigens, result in the symptoms.14 and mentally. Work performance had been greatly affected and she could only work up to 5 hours/day. Her headaches and dizziness persisted over This case highlights several important issues in relation to microbiology and the next 12 months. Worker’s compensation was involved and the claim was occupational health. Firstly, more research is needed to ascertain the current accepted and substantiated. On occasions when she wanted to work longer vaccine’s claimed protectiveness. Secondly, as the current serology and skin hours such as going to field trips, she often had to spend the next day in bed, test may not detect recent subclinical infection, there is a need to develop recovering from her exertion. In December 2010, she was diagnosed with and employ more sensitive diagnostic tests to detect Q fever infection earlier. post-Q fever fatigue syndrome by an infectious diseases expert. No further The use of PCR for earlier diagnosis might be a cost-effective consideration. courses of antibiotics were prescribed. She was referred to a psychologist Most importantly, with respect to occupational safety, there should be an for ongoing support and counselling. Due to fatigue and impaired working urgent call to review the current occupational safety policy, with implementa- performance, she constantly experienced frustration and low mood. In Au- tion of universal mandatory Q fever immunisation of all new abattoir, cattle, gust 2011, she was diagnosed with depression and the antidepressant ser- goat and sheep workers, that may be exposed to Q fever, prior to the com- traline was trialled without much improvement. She was unable to work and mencement of any field work. They should be advised that it takes around had to resign from her job in June 2012. Her current treatment comprises two weeks before the vaccination becomes protective and that they should

regular vitamin D, vitamin C, zinc, and vitamin B12 supplements. To date, the avoid exposure in the meantime. patient has remained debilitated from fatigue and not capable of resuming Acknowledgements any work. This has affected her family relationships significantly. We would like to acknowledge A/Prof Ross Philpot, infectious disease physi- Discussion cian, for his assistance in the preparation of the manuscript. Q fever is a zoonosis caused by the highly infectious C. burnetii, an obligate Gram-negative intracellular organism. Cattle, sheep and goats are the main References reservoirs for infection in humans. Humans are infected by contact with in- 1. Maurin M, Raoult D. Q fever. Clinical Microbiology Reviews 1999;12:518-53. fected animals or their products. In Australia, since the introduction of the Q 2� The Royal College of Pathologists of Australia. Q fever antibodies – serum. RCPA Manual. Updated May 28, 2006. [accessed 2014 Feb 7]. Available from: http://www.rcpamanual.edu.au/index.php?option=com_ fever vaccination (first in the abattoirs in 1991, and later more widely in the pttests&task=show_test&id=26&Itemid=27 rural community in 2001-2006 through the National Q Fever Management 3� Worswick D, Marmion BP. Antibody responses in acute and chronic Q fever and in subjects vaccinated against Program), the incidence rate has dropped substantially, from 800 cases an- Q fever. Medical Microbiology 1985;19:281-96. 4 4. Australian Government Department of Health and Ageing. National Notifiable Diseases Surveillance System. nually prior to the program to 303 notified cases in 2009. Number of notifications by year, Australia, 2005 to 2009. Available from: http://www.aihw.gov.au/australias- health-2010-data-tables/?id=6442475642 (accessed Dec 17 2013). In this case, chronic Q fever infection was diagnosed on the basis of both 5. Fournier PE, Marrie TJ, Raoult D. Diagnosis of Q fever. Journal of Clinical Microbiology 1998;36: 1823. 2 positive phase 1 and phase 2 CFT (titres of 16, 8 respectively). C. burnetii 6� CSL Biotherapies. Q-VAX®. Approved Product Information. Updated Nov, 2013. Available from: http://www. csl.com.au/s1/cs/auhq/1196562765747/Web_Product_C/1196562634452/ProductDetail.htm (accessed Feb has phase 1 and phase 2 lipopolysaccharides against which the human body 7 2014). mounts an antibody immune response. Detecting these antibodies forms the 7. Ackland JR, Worswick DA, Marmion BP. Vaccine prophylaxis of Q fever. A follow-up study of the efficacy of basis of laboratory diagnosis. Phase 2 antibodies are positive in acute Q fe- Q-Vax (CSL) 1985-1990. Medical Journal of Australia 1994;160:704-8. ver, whereas phase 1 antibodies remain elevated in chronic disease.5 During 8. Q fever [revised 2010 Aug]. In: eTG complete [CD-ROM]. Melbourne: Therapeutic Guidelines Limited, 2012. 9� raoult D, Houpikian P, Tissot DH, et al. Treatment of Q fever endocarditis: comparison of 2 regimens contain- acute Q fever, IgM antibodies develop against phase 1 and phase 2 forms, ing doxycycline and ofloxacin or hydroxychloroquine. Archives of Internal Medicine 1999;159:167-73. whereas IgG antibodies develop only against the phase 2 forms. The current 10� Terheggen U, Leggat PA. Clinical manifestations of Q fever in adults and children. Travel Medicine and Infec- Q fever vaccine (Q-VAX, CSL, Australia)6 composed of a killed suspension tious Disease 2007;5:159-64. 11. Ayres JG, Smith EG, Flint N. Protracted debility and fatigue after acute Q fever. Lancet 1996; 347:978-9. of virulent phase 1 organisms, stimulates immune response predominantly 12� Wildman MJ, Smith EG, Groves J, et al. Chronic fatigue following infection by Coxiella burnetii (Q fever): ten- against phase 1 antigen. year follow-up of the 1989 UK outbreak cohort. Quarterly Journal of Medicine 2002;95:527-538. 13� Marmion BP, Shannon M, Maddocks I, et al. Protracted debility and fatigue after acute Q fever. Lancet In Australia, Q-VAX can only be given to people who are tested negative on 1996;347:977-8. both serology and skin tests,6 and this patient was no exception. However, 14� penttila IA, Harris RJ, Storm P, et al. Cytokine dysregulation in the post-Q-fever fatigue syndrome. Quarterly she developed dry cough, headaches and diarrhoea shortly after vaccination. Journal of Medicine 1998;91:549-60. There are a few possible scenarios. Firstly, she could have contracted Q fever just prior to the pre-vaccination tests. As the antibodies often only develop 2 to 3 weeks after inoculation,3 it was possible that her pre-vaccination tests Corresponding Author were performed prior to the development of detectable antibodies. Secondly, Dr Frank CH Ng she could have contracted Q fever after being immunised but before the de- Diamond Creek Medical Centre, 82 Main Hurstbridge Rd, Diamond velopment of protective immunity, as vaccination during the incubation of Creek, VIC, 3089, Australia Q fever does not prevent the onset of the disease.6 Thirdly, the vaccination might have failed to confer immunity. This would challenge the 100% protec- Email: [email protected] tion claimed for Q fever vaccination.7

20 ANNALS OF THE ACTM May 2014

ANNALS OF THE ACTM AN INTERNATIONAL JOURNAL OF TROPICAL & TRAVEL MEDICINE

INSTRUCTIONS FOR AUTHORS

The format of the Annals of the ACTM will, in general, follow guidelines of the “Uniform requirements for manuscripts submitted to biomedical journals” and published by the International Committee of Medical Journal Editors (http://www.icmje.org/index.html). The Annals will appear twice a year and will consider for publication, papers on a wide range of topics relating to tropical and travel medicine. All papers will be refereed prior to acceptance for publication. Papers will be included in one of the following categories: a) Review Articles (5,000-10,000 words) b) Research Articles (up to 5,000 words) c) Case Reports (1,000-2,000 words) d) Research Reports (1,000-2,000 words) e) letters (200-500 words) Figures to be included: 1/4 page size = 250 words; 1/2 page size = 500 words etc. One page with images is approximately 900 words, two pages with image is approximately 1,800 words. Manuscripts should be double spaced and a short summary should be included at the beginning of the paper after the title and author details. Title page with contributor names and addresses should be on a separate page. Each table and figure should be on a separate page together with an appropriate caption, explanatory notes etc. Any acknowledgements should be included at the end of the paper before the references. Where appropriate, authors must confirm in the paper that experimental procedures on humans and animals conformed to accepted international ethical guidelines. References should be numbered consecutively in order of first appearance in the text. For details of references, consult the “Uniform requirements for manuscripts submitted to biomedical journals” available at http://www.icmje.org/index.html. In the first instance, papers submitted for consideration should be sent to: The Editorial Board Annals of The Australasian College of Tropical Medicine ACTM Secretariat PO Box 123, Red Hill Queensland 4059 Australia Tel: + 61-7-3872-2246 Fax: +61-7-3856-4727 Email: [email protected] Statements or opinions in papers published in the Annals of the ACTM are solely those of the authors and not necessarily those of the Editorial Board of The Australasian College of Tropical Medicine. The inclusion of commercial advertising material in the Annals or the College. The College disclaims any responsibility for any injury to persons or property resulting from publishing material or products referred to in articles or advertisements. On acceptance of an article for publication in the Annals, copyright of the article is automatically transferred to the ACTM.

Volume 15 Number 1 21 ANNALS OF THE ACTM 21i