QJM Advance Access originally published online on June 27, 2008
QJM 2008 101(8):649-656; doi:10.1093/qjmed/hcn072
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Falciparum malaria as a cause of fever in adult travellers returning to the United Kingdom: observational study of risk by geographical area
From the 1The Hospital for Tropical Diseases, Mortimer Market Centre, Capper Street and 2London School of Hygiene & Tropical Medicine, Keppel St., London WC1E 6AU, UK
Address correspondence to Prof. Christopher Whitty, The Hospital for Tropical Diseases, Mortimer Market Centre, Capper Street, London WC1E 6AU, UK. email: christopher.whitty{at}lshtm.ac.uk
Received 7 January 2008 and in revised form 16 May 2008
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Background: The probability that a returned traveller with a history of fever has malaria is likely to vary by geographical area, but this has not been quantified in travellers.
Aim: To collect data on prevalence of malaria in outpatients returning with a fever or history of fever from malaria-endemic countries, at the point of presentation for a malaria test.
Design: Observational retrospective study. Consecutive patients presenting to an unselected ‘walk-in’ clinic for returned travellers.
Results: Of 2867 patients meeting inclusion criteria, 337 (11.8%) had malaria, 89.5% originating in sub-Saharan Africa. Of travellers returning from sub-Saharan Africa excluding South Africa with fever/history of fever, 291/1497 had malaria (19.4%, 95% CI 17–21%). A high proportion was visiting friends and relatives. In those from other areas the proportions were: 16/707 (2.3%, 95% CI 1.5–3.8) from Indian subcontinent/Southeast Asia; 2/143 (1.4%) from Southern America; 4/129 (3.1%) from South Africa; 1/44 (2.3%) from North Africa; and 8/41 (19.5%) from Oceania. Compared to other malaria-endemic regions, African travel gave an adjusted odds ratio of 7.8 (95% CI 5.4–11.2, P < 0.0001). Only 45.1% of malaria cases had a fever (
37.5°C) at the time of presentation. Only 3% of all diagnoses of malaria had no history of fever. In 28% of cases parasite count increased in the initial 24 h of antimalarial treatment.
Conclusions: The likelihood that a patient with fever returning from Africa has malaria is high (around 1 in 5), and is significantly lower from other areas. Absence of fever at presentation does not exclude malaria.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Imported malaria is responsible for approximately 11 000 cases per year in the European Union of which around 2000 a year are seen in the UK.1 It is one of the commonest diagnoses in Infectious Diseases units throughout the UK with around 2000 cases a year,2 and is the principal imported tropical infection causing avoidable deaths every year. Most cases are caused by Plasmodium falciparum with a case fatality rate of 1.2% in the UK.3 In many non-tropical settings malaria is the commonest identified cause of fever in returned travellers.4,5 Over the past two decades, travel to malaria-endemic countries has been rising steadily, and health care providers are increasingly faced with returned travellers. Data on the numbers of malaria cases imported into the UK and other European countries help to identify trends for public health planning, but do not provide information directly relevant to the question for clinicians faced with a recently returned traveller presenting as an emergency complaining of fever––what is the chance that this traveller has malaria? This risk is likely to vary by geographical area travelled to, but has not been quantified in the UK or other non-endemic settings.
Diagnosing malaria purely clinically is difficult due to the wide range and non-specific nature of symptoms, although fever is usual, and confirmation with blood film for malaria parasites is required in all suspects. Fever on examination, symptom duration, immigrant travel, travel to sub-Saharan Africa, inadequate prophylaxis, thrombocytopenia, elevated bilirubin and splenomegaly have all been shown to correlate with malaria infection;6–9 however, no single variable or clinical algorithm has been shown to predict malaria reliably.
London reports two-thirds10 of the UK's imported malaria. Significant numbers of cases are seen partly due to a large and diverse migrant population11 and also proximity to the world's busiest international airport (Heathrow), making it easier to determine the relative prevalence of malaria in febrile travellers returning from different regions, than almost any other centre. The Hospital for Tropical Diseases (HTD) provides a walk-in (self/GP referral) diagnostic service for any traveller with fever or other symptoms. This study aimed to establish the prevalence of malaria in patients presenting with fever or history of fever by geographical region, to assist physicians faced with a returned febrile traveller in determining the probability that this will be malaria. Additionally, the study aimed to determine whether objective or subjective evidence of fever was a sensitive malaria screening tool.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
This was a retrospective examination of 5 years of patient episodes using data collected prospectively on a standard proforma. All patient episodes were identified by hand-searching of notes and cross-checked by examination of laboratory records. Patients presenting either as self or general practitioner referral to the HTD walk-in clinic between 1 January 2000 and 31 December 2004, with a recorded temperature
37.5°C in clinic, or history of fever (i.e. perceived fever, sweats, rigors, chills) and recent travel outside Europe were potentially eligible for inclusion. Those with a known diagnosis of malaria (made in another laboratory) were not included in the main study. Patients presenting after hours, or via Accident & Emergency, with involvement of the Tropical Medicine service, with a subsequent diagnosis of malaria were included, as risk may vary by time of presentation. Patients under 16-years-old, and those for whom no blood film had been performed because they had a clear alternative diagnosis based on clinical findings (e.g. cellulitis) and epidemiological malaria risk assessment, were excluded. An additional group who did not present with fever, but had other symptoms and epidemiological risk of malaria, and had a malaria film performed, were included for analyses of malaria cases. They were however analysed separately when assessing the risk of malaria in returned travellers with fever or history of fever (Figure 1).
|
In each case a standardised proforma was completed prospectively (Box 1); certain sections by the triage nurse specialist (recorded fever, some demographic details), and the remaining details by the duty tropical medicine doctor. Data on area of travel were collected by the triage nurse and then confirmed with the patient by the duty doctor. Self-reported travel details, ethnicity, place of birth and reason for travel, were recorded. Observations, including temperature, were recorded, and an EDTA blood sample sent for thin film microscopy for malaria parasites. All malaria films were stained with Field stain and readings of both thick and thin films were performed by experienced microscopists in the same on-site parasitology laboratory, which is a national reference laboratory for parasitology. All malaria films were read independently by two microscopists.
Box 1. Information recorded on standardised proforma for patients presenting to the HTD ‘walk-in’ clinic after foreign travel
|
Data were entered into a database using Microsoft Access, and analysed using Stata 9.0 (Statacorp). Area of travel was determined from countries visited, and categorised according to areas of known malaria endemicity, shown in Box 2, based on distinct epidemiological risk, with sub-Saharan Africa being further divided into West Africa, East Africa and Central/Southern Africa (excluding South Africa). Some travellers had complex travel schedules which made it difficult to reliably ascribe a region for acquiring malaria; these were subsequently excluded. Proportion of malaria in returning travellers with a history of fever was determined for each region and the confidence intervals calculated. Rates of malaria were then stratified by recorded reason for travel. Odds ratios for acquiring malaria in sub-Saharan Africa compared to other malaria-endemic regions, and for visiting friends and relatives (VFRs) compared to tourists, were calculated uncorrected, and adjusted for reason for travel and reported chemoprophylaxis using a logistic regression model. Those with missing data for potential confounding variables were classified as ‘missing’ and included in the final model.
Box 2. Region of travel (determined by country visited) categorised according to epidemiological malaria risk
|
All patients with falciparum malaria were admitted for treatment with quinine for at least 24 h and repeat blood films were performed. The change of parasite count, defined as percentage of red blood cells parasitised seen on thin film examination between the first and second day, was calculated.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
A total of 3054 patients potentially fulfilled criteria for inclusion in the study. No blood film was performed in 187, who were excluded. The remaining 2867 patients’ data were analysed. This included the subset of 185 patients who had symptoms and epidemiological risk to warrant performing a malaria film, but who did not have a recorded fever or febrile symptoms. For the main outcomes on all patients, region of travel was available in 2849 (99%), recorded temperature in 2867 (100%) and parasite count in 2867 (100%). Data on the potential confounding factors reason for travel and chemoprophylaxis was available in 2617 (91.3%) and 2638 (92%) respectively. Demographic data and travel details of the population presenting in this clinic are shown in Table 1. Overall, 337 of the 2867 patients (11.8%) had malaria based on blood film result, of whom 10 had no fever or history of fever; parasite count was available on repeat film in 259 (77%) of these.
|
Of the 333 malaria cases where region of travel was clear, 89.5% were excluded from sub-Saharan Africa (90.9% due to P. falciparum) and 5.7% from Asia (84.2% due to P. vivax). Contributions of other regions are shown in Table 2. The risk of malaria in patients with fever/history of fever returning from each region is shown in Table 3. The prevalence was highest in patients returning from Oceania although numbers were small (19.5%, total n = 41, CI = 9.5–33), followed by sub-Saharan Africa (19.4%, total n = 1497, CI = 17–21). For all other regions the percentage of patients, with a history of fever, who had malaria parasites, was
3%. The odds ratio (OR) of acquiring malaria in sub-Saharan Africa vs. all other malaria-endemic areas was 7.82 (95% CI 5.5–11.2, P < 0.0001). Adjusting for reason for travel, country of birth and prophylaxis OR was 7.83 (95% CI 5.4–11.2).
|
|
Purpose of travel and region of travel were recorded in 313 of 337 malaria cases (92%) and for 2300 of 2530 non-malaria cases (90.9%). Of the malaria cases 47% occurred in those VFR, 20.1% in tourists, 13.4% in business travellers, 13.1% in new entrants to UK, 2.9% in expatriates and 2.6% in foreign students visiting the UK. Two cases occurred in volunteers, and one in the military. The majority of malaria, 43.5%, occurred in VFR travellers to sub-Saharan Africa, followed by 15.8% in tourists visiting Africa, 12.9% in new entrants to the UK from Africa and 11.4% in business travellers to Africa; this is broadly similar to UK data as a whole.3 The risk of malaria, when stratified by purpose of travel for each region, is shown in Table 4; history of travel to Africa was a stronger predictor of malaria than any reason for travel, and was the strongest predictor of malaria in all the travel groups.
|
Ethnic origin (self-defined) was known for 296 malaria cases, of which 62.8% were Black African (all had travelled to sub-Saharan Africa), 30.7% were Caucasian and 3.4% were Asian.
Temperature was recorded in all study participants. In the 337 malaria cases, 45.1% had a recorded fever (37.5°C), mean temperature was 37.6°C and 25% had a temperature >38.5°C. In non-malaria cases, 16.4% had a recorded fever, mean temperature was 36.9°C and only 5% had a temperature >38.3°C. Of 2299 afebrile patients at presentation, 8.1% had malaria, compared to 26.8% of 568 febrile patients. Ten of the 337 malaria cases (3%) reported no fever or history of fever. Of these, four were born in Africa, and therefore likely to have had previous exposure; but five were born in non-endemic countries and in one case, region of birth was not recorded. Seven of the ten were current UK residents.
Restricting the data to where chemoprophylaxis data were available, 37% of 308 malaria cases had taken prophylaxis, in contrast to 51.1% of 2330 non-malaria cases. For 2616 travellers to endemic areas with chemoprophylaxis data, 14.4% of those not taking prophylaxis developed malaria, in contrast to 8.8% of those who did.
Time interval between return to UK and presentation to HTD was recorded in 321 malaria cases (95.3% of the total). The median was 11 days (range 0–541), in comparison to 10 days (range 0–1597) for non-cases. Around 35.2% of malaria cases were presented within 1 week, 75.4% within 3 weeks, 93.5% within 3 months, 96.6% within 6 months and 98.2% within 1 year. Within 3 weeks of return, 84.4% of 269 P. falciparum cases were presented. About 16.6% of 30 P. vivax cases were presented after 1 year.
In 52/190 (28%) of P. falciparum cases, there was an increase in parasite count between Day 1 and 2 of treatment. Of these, 2.5% increased substantially to >2%, including observed increases from 0.4–10.7%, 3.9–12% and 0.01–3.8% in three cases.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
There is very little data on the likelihood that a patient returning from a tropical area presenting to a doctor with a history of fever has malaria. This study of unselected consecutive patients presenting with a fever or history of fever, after recent travel to malaria-endemic countries, demonstrated that in those presenting from sub-Saharan Africa or Oceania, irrespective of reason for travel, the probability of this being due to malaria is substantial (around 1 in 5). Prevalence in those returned from all other malaria-endemic areas is around 10 times lower, but still significant for a potentially fatal disease. Although a recorded fever increased the likelihood of malaria, over half of the malaria cases were afebrile at presentation and 3% had neither fever nor history of fever. Malaria parasitaemia was increased after 24 h of treatment with quinine in a third of those with falciparum malaria. This study demonstrates malaria is substantially more likely in those returning with history of fever from the tropics than is usually realised, and that a normal temperature does not rule it out.
The 11.8% overall risk of malaria compares with studies from various countries which quote between 13 and 47.6% of cases of fever in the returning traveller to be due to malaria.8,9,12–17 Our slightly lower prevalence may be due to the fact that we examined outpatient presentations only (the normal initial presentation to medical services) in unselected patients who had not previously been tested, albeit in the setting of a ‘walk-in’ tropical diseases service in central London. Most studies are analyses of hospital admissions which deal with more severe clinical presentations, eliminating cases where malaria has been excluded by blood film. Malaria is inevitably a much higher proportion of the inpatient population. The easy accessibility of our outpatient clinic, which allows self-referral, is likely to lead to an increased proportion of patients presenting with less severe illness, more typical of a general practice or A&E setting than a referral service. General practitioners and A&E officers seeing patients returned from the tropics need to test for malaria irrespective of geographical area, and in those from sub-Saharan Africa the probability of malaria is high.
Malaria cases at HTD show similar patterns to data collected by the Malaria Reference Laboratory (MRL) and European surveillance with respect to region acquired, causative species, reasons for travel and ethnic group, suggesting the case-mix is fairly representative.3,18 Our data demonstrated a slightly higher percentage of cases from sub-Saharan Africa, reflecting the large migrant population from the African diaspora in London returning to visit friends and relatives, as well as increasing travel to this region for tourism.19 The high proportion of malaria cases at HTD seen in VFR travellers also supports existing literature,12,18,20,21 with almost a threefold increased risk of acquiring malaria in VFR travellers to Africa compared to tourists (29.6 vs. 11%). VFR travellers are more likely to visit poorly developed and remote regions, stay for longer periods and appear less likely to seek health advice prior to travel or take chemoprophylaxis,21,22 therefore we continue to emphasise the increased risk of malaria in this population, and the need for a low threshold for further investigation with blood film in this subgroup.
Less than half of our malaria cases had recorded fever (37.5°C) on the day of presentation making this an unreliable predictor in the diagnosis of malaria. These findings support results obtained in a similar study of clinical presentation and examination of 555 confirmed malaria cases in Canada.7 Indeed, in ten malaria cases (3%) there was no recorded fever or history of febrile symptoms.
In those who reported taking prophylaxis malaria was proved in 8.8%. Reported use and actual use may vary, but history of prophylaxis in a patient with fever returning from an endemic area should not discourage clinicians from testing for malaria. The increase in malaria parasitaemia over the first 24 h is in line with previous data,23 and reflects the complex life-cycle of the malaria parasite. This rise does not demonstrate that a drug is failing and should only prompt treatment change if the patient is deteriorating clinically. It is likely that this observation will be less true for those treated with artemisinin-based combination treatment.
Retrospective studies have known limitations, and recording bias in particular can limit the reliability of data. This is however unlikely to be a significant issue for the three main outcomes in this study. Malaria parasites are recorded in both notes and laboratory records and are objective data, which will not differ between prospective and retrospective studies. In this specialist hospital, history of travel is central information and is recorded on proformas by both nurses and doctors and data were complete. Temperature was measured in all patients and is again objective data. Data on reason for travel, country of birth and chemoprophylaxis are more liable to reporting errors. These are explanatory rather than primary outcomes in this study, so this is unlikely to affect the central points of the study and completion rates were high for all. The fact that the HTD outpatient clinic uses a proforma for each attendant meant that, although the study was retrospective, documentation enabled a high yield of results. The other limitation of any study is generalisability. We do not consider the population representative of febrile patients in the UK, but it is likely to be representative of the population of patients with febrile illness after return from malaria-endemic countries, presenting in the major cities in Europe with significant migrant populations.
When assessing patients in general practice, accident and emergency departments, or outpatient clinics, clinicians must maintain a high index of suspicion for malaria in any patient with a fever or history of fever who has recently returned from an endemic area, particularly VFRs from sub-Saharan Africa. Absence of recorded fever should not be interpreted as low risk of malaria, and patients with a history of fever should have a malaria blood test performed promptly, by whichever means available.
|
Acknowledgements |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
We would like to thank the staff of the HTD outpatients and parasitology laboratory for their assistance with this study. M.A. is supported by the HTD Special Trustees.
Conflict of interest: None declared.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
1. Smith AD, Bradley DJ, Smith V, Blaze M, Behrens RH, Chiodini PL, et al. Imported malaria and high risk groups: an observational study using UK surveillance data 1987-2006. BMJ (2008) 337:a120.
2. Harling R, Crook P, Lewthwaite P, Evans M, Schmid ML, Beeching NJ. Burden and cost of imported infections admitted to infectious disease units in England and Wales in 1998 and 1999. J Infect (2004) 48:139–44.[CrossRef][Web of Science][Medline]
3. Foreign Travel-associated Illness: England, Wales and Northern Ireland – Annual Report 2005 (2005) London: Health Protection Agency.
4. Ansart S, Perez L, Vergely O, Danis M, Bricaire F, Caumes E. Illnesses in travellers returning from the tropics: a prospective study of 622 patients. J Travel Med (2005) 12:312–8.[Web of Science][Medline]
5. O’Brien DP, Leder K, Matchett E, Brown GV, Torresi J. Illness in returned travellers and immigrants/refugees: the 6-year experience of two Australian infectious diseases units. J Travel Med (2006) 13:145–52.[CrossRef][Web of Science][Medline]
6. Svenson JE, Gyorkos T, MacLean J. Diagnosis of malaria in the febrile traveller. Am J Trop Med Hyg (1995) 53:518–21.
7. Svenson JE, MacLean JD, Gyorkos TW, Keystone J. Imported malaria: clinical presentation and examination of symptomatic travelers. Arch Intern Med (1995) 155:861–8.
8. Casalino E, Le Bras J, Chaussin F, Fichelle A, Bouvet E. Predictive factors of malaria in travellers to areas where malaria is endemic. Arch Intern Med (2002) 162:1625–30.
9. D’Acremont V, Landry P, Mueller I, Pecoud A, Genton B. Clinical and laboratory predictors of imported malaria in an outpatient setting: an aid to medical decision making in returning travelers with fever. Amer J Trop M Hyg (2002) 66:481–6.
10. Health Protection Agency. Illness in England, Wales and Northern Ireland Associated with Foreign Travel – A Baseline Report to 2002 (2004) London: Health Protection Agency.
11. Health Protection Agency. Migrant Health: Infectious Diseases in non-UK Born Populations in England, Wales and Northern Ireland. A Baseline Report - 2006 (2006) London: Health Protection Agency.
12. Antinori S, Galimberti L, Gianelli E, Calattini S, Piazza M, Morelli P, et al. Prospective observational study of fever in hospitalised returning travellers and migrants from tropical areas. J Travel Med (2004) 11:135–42.[Web of Science][Medline]
13. Klein JL, Milman G. Prospective, hospital based study of fever in children in the United Kingdom who had recently spent time in the tropics. Br Med J (1998) 316:1425–6.
14. West NS, Riordan F. Fever in returned travellers: a prospective review of hospital admissions for a 2
year period. Arch Dis Child (2003) 88:432–4.
15. Doherty JF, Grant AD, Bryceson ADM. Fever as the presenting complaint of travellers returning from the tropics. QJM (1995) 88:277–81.
16. MacLean JD, Lalonde RG, Ward B. Fever from the tropics. Travel Med Advisor (1994) 5. 27.1–27.14.
17. O’Brien D, Tobin S, Brown GV, Torresi J. Fever in returned travelers: review of hospital admissions for a 3 year period. Clin Infect Dis (2001) 33:603–9.[CrossRef][Web of Science][Medline]
18. Jelinek T, Schulte C, Behrens R, Grobusch MP, Couland JP, Bisoffi Z, et al. Imported falciparum malaria in Europe: sentinel surveillance data from the European network on surveillance of imported infectious diseases. Clin Infect Dis (2002) 34:572–6.[CrossRef][Web of Science][Medline]
19. Travel Trends 2003: A Report on the International Passenger Survey (2004) London: Office for National Statisics, HMSO.
20. Leder K, Black J, O’Brien D, Greenwood Z, Kain KC, Schwartz E, et al. Malaria in travellers: a review of the geosentinel surveillance network. Clin Infect Dis (2004) 39:1104–12.[CrossRef][Web of Science][Medline]
21. Phillips-Howard PA, Radalowicz A, Mitchell J, Bradley DJ. Risk of malaria in British residents returning from malarious areas. Br Med J (1990) 300:499–503.
22. Angell SY, Behrens RH. Risk assessment and disease prevention in visiting friends and relatives. Infect Dis Clin North Amer (2005) 19:49–65.[CrossRef]
23. Bunn A, Escombe R, Armstrong M, Whitty CJM, Doherty JF. Falciparum malaria in malaria-naïve travellers and African visitors. QJM (2004) 97:645–9.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||