Secondary Logo

Journal Logo

Fever After a Stay in the Tropics: Clinical Spectrum and Outcome in HIV-Infected Travelers and Migrants

Bottieau, Emmanuel MD, PhD*; Florence, Eric MD, PhD*; Clerinx, Jan MD*; Vlieghe, Erika MD*†; Vekemans, Marc MD*; Moerman, Filip MD*; Lynen, Lut MD*; Colebunders, Robert MD, PhD*†; Van Gompel, Alfons MD*; Van den Ende, Jef MD, PhD*†

JAIDS Journal of Acquired Immune Deficiency Syndromes: August 15th, 2008 - Volume 48 - Issue 5 - p 547-552
doi: 10.1097/QAI.0b013e31817bebc5
Clinical Science

Objective: To investigate the epidemiology and clinical spectrum of fever in HIV-infected returning travelers and migrants.

Methods: From April 2000 to December 2006, we explored prospectively, at our referral travel/HIV clinics, the etiology and outcome of febrile illnesses developing within 3 months after a stay in the tropics. For this study, we compared the morbidity profile between HIV-infected individuals and all other cases tested HIV negative.

Results: Of the 1850 adults (15 years and older) evaluated for 1921 fever episodes, 93 (5%) had HIV infection, including 5 presenting with primary infection. HIV prevalence was 2% in western travelers or expatriates, 11% in travelers “visiting friends and relatives,” and 24% in foreign visitors/migrants. Fever episodes (n = 104) occurring in the HIV-infected individuals were mainly due to opportunistic infections (23%, including tuberculosis), respiratory tract infections (20%), sexually transmitted infections (9%), and noninfectious diseases (7%). All these conditions were more frequently diagnosed than in HIV-negative travelers (1035 fever episodes), although tropical infections (mostly malaria) were proportionally less prevalent. Morbidity (rate and duration of hospitalization) was more considerable in HIV-infected patients than in HIV-negative individuals.

Conclusions: HIV infection was frequent in returning travelers and migrants presenting with fever at our setting and affected strongly the diagnostic spectrum and overall morbidity.

From the *Department of Clinical Sciences, Institute of Tropical Medicine, Antwerp, Belgium; and the †Department of Tropical Medicine, University Hospital of Antwerp, Antwerp, Belgium.

Received for publication November 5, 2007; accepted March 11, 2008.

Conflicts of interest: none.

Financial disclosure: none.

Correspondence to: Emmanuel Bottieau, Institute of Tropical Medicine, Nationalestraat 155, 2000 Antwerp, Belgium (e-mail:

Back to Top | Article Outline


The last decades have witnessed an unprecedented increase in international travel and migration and, concomitantly, the emergence of the human immunodeficiency virus (HIV) pandemic and the global reemergence of several other sexually transmitted infections (STIs).1 There is nowadays a considerable overlap between travel medicine and HIV/STI medicine.2 These 2 new disciplines are often closely associated in most institutions historically dedicated to infectious and tropical pathology.

Fever develops in about 10% of international travelers before or upon return. It is a leading reason for consultation in travel clinics and a challenging clinical problem because of the broad differential diagnosis and the risk of severe causative diseases.3,4 In addition, in international travelers and migrants from developing countries, risk for HIV infection and STI is considerably higher.5,6 Therefore, because these conditions may be the etiology of imported fever or may largely influence its clinical presentation, they should be systematically considered during the diagnostic workup. However, few data are available on the burden of HIV infection in travel medicine.7 In the global GeoSentinel survey on illnesses in international travelers conducted in 30 travel clinics worldwide, HIV infection and/or STI were reported in about 1% of the sick travelers, whatever may be the reason for consultation.4 In a 5-year study conducted in our institution and exploring specifically the etiology of fever after a stay in the tropics, prevalence of HIV infection reached 5%.3 We hypothesized that an underlying HIV infection could have a specific influence on the morbidity profile of imported febrile illnesses. The purposes of this study were to describe the epidemiology of HIV infection among travelers and migrants presenting with fever at our referral travel/HIV clinics and to compare the clinical spectrum and outcome of imported fever in these HIV-infected individuals with those of the “classic” traveler population.

Back to Top | Article Outline


Study Setting and Population

The study was conducted at the Institute of Tropical Medicine (ITM) and at the University Hospital, both located at Antwerp, Belgium. The ITM provides outpatient referral care for travel/tropical and HIV/STI pathology. The University Hospital is a tertiary-level hospital providing emergency and inpatient care under the supervision of physicians affiliated to the ITM for posttravel and/or HIV-related pathology.

From April 2000 to December 2006, we assessed prospectively the etiologic spectrum and outcome of fever (temperature ≥38°C or combination of fever sensation with sweats and chills) occurring in any patient attending our centers within 12 months after a stay in a (sub) tropical area. Tropics and subtropics corresponded to all nonindustrialized countries at least partly situated between the 35° northern and 35° southern latitude. The diagnostic procedure and case management have been extensively described in a previous article reporting on 1842 febrile episodes investigated from April 2000 to March 2005.3 Briefly, clinical data were prospectively registered during consultation. All patients were offered a standard first-line laboratory workup, and additional investigations were ordered according to the clinician's decision. Final diagnosis could be etiologic, if a pathogenic microorganism was demonstrated in a relevant specimen or if serology against an infectious agent was definitive or strongly suggestive, or clinical if fever was undisputedly due to a specific syndrome (like erysipelas or orchitis). In case of concomitant infections, the most severe febrile disease defined the final diagnosis, but all other diagnoses were registered. Febrile episodes not fulfilling any precise etiologic or clinical diagnosis were of “unknown cause”; in these cases, a presumptive diagnosis was reported. A new febrile episode was defined in the same patient if fever and associated symptoms had completely abated during at least 2 weeks. Patients' outcome was assessed within 3 months after the initial contact by follow-up consultations and/or phone calls until complete cure (or death).

Back to Top | Article Outline

Diagnosis and Management of HIV Infection and STI

For the present study, which aimed at describing the morbidity profile in the subset of HIV-infected travelers and migrants, we investigated only the fever episodes occurring in adults (15 years and older) within 3 months upon return/arrival. We had indeed observed in our previous study that a travel-related etiology was very unlikely when fever develops more than 3 months after the tropical exposure. Patients were divided in 3 categories: western travelers or expatriates (natives of western countries having traveled/resided in the tropics); travelers “visiting friends and relatives” (VFR travelers) (natives of the tropics established since more than 1 year in Europe and returning to their home country to visit friends and relatives); and foreign visitors/migrants (natives of the tropics arriving for the first time to Europe).

HIV testing was liberally offered to all febrile patients but performed only according to clinician's judgment and after patient's acceptance. HIV testing was however the rule in hospitalized patients and those diagnosed with STI. HIV infection was diagnosed by 2 positive screening and confirmation testing (identifying both HIV-1 and HIV-2) performed at different moments. Acquired immunodeficiency syndrome (AIDS) was defined according to the Center for Disease Control (CDC) guidelines revised in 1993 by the etiologic diagnosis of any listed opportunistic infection (OI) and/or a CD4 cell count below 200/μL at the initial contact. HIV-infected patients were referred to the HIV clinic for chronic care, including those without legal status in Belgium through the national emergency first aid procedure, which provided to them free follow-up and treatment. Partners of patients newly diagnosed were systematically traced.

Of note, risk of STI related or not to travel was systematically assessed during the first consultation by the attended physician. Workup for STI was realized according to risks and presenting symptoms, and diagnosis and treatment were based on international recommendations. Partners were systematically traced and treated. According to the clinical judgment, STI could be reported as the etiology of fever or as an incidental finding (if another condition was obviously the main cause of fever). Patients with consistent risk behavior during travel were retested for HIV/STI 3 months after their last sexual exposure. Outcome of the febrile episode was assessed within 3 months according to the standard study procedure, and long-term outcome of all HIV-infected patients was reassessed at the end of the study period.

Back to Top | Article Outline

Statistical Analysis

Statistical analyses were performed with SPSS software, version 15 (SPSS Inc., Chicago, IL). The χ2 test was used to compare categorical variables; parametric and nonparametric tests were used when appropriate for continuous variables. All tests were 2 tailed and P values <0.05 indicated statistical significance.

Back to Top | Article Outline

Ethical Issue

All patients were informed about the objectives of this observational study. Patient data were rendered anonymous for further analyses, according to the Belgian legislation. The study was designed, conducted, and analyzed independently of any sponsoring. The protocol was approved by the ethical committee of the ITM.

Back to Top | Article Outline


Characteristics of the Study Population

During the study period, we investigated in total 1921 febrile episodes occurring in 1850 adults within 3 months after a tropical stay (1.04 fever episode in average per patient). Of them, 104 febrile episodes occurred in 93 patients with HIV infection (1.12 episode per HIV-infected patient), giving an overall HIV prevalence of 5%. HIV testing had been performed in 1139 (59%) of all febrile episodes.

As shown in Table 1, the rate of HIV testing differed between western travelers or expatriates, VFR travelers, and foreign visitors/migrants (respectively 54%, 75%, and 78%) as well as did the prevalence of HIV infection (respectively 2%, 11%, and 24%) (P < 0.001 between traveler categories for both comparisons).



At first inclusion in the study, mean age of patients with HIV infection was 37 years (range: 15-65 years). Each of the 3 categories of travelers contributed for about one third of the study participants. Male patients predominated in western travelers or expatriates (79%), but sex distribution was nearly equal in the other 2 categories. The vast majority of VFR travelers and foreign visitors/migrants were native from sub-Saharan Africa (respectively 74% and 77%). Most western travelers (55%) and all expatriates were returning from sub-Saharan Africa. One third of all patients were referred by other physicians, and this proportion was similar for the 3 traveler categories.

Back to Top | Article Outline

Clinical Spectrum and Outcome of Febrile Illnesses According to HIV Status

Diagnosis of HIV infection was made during the fever workup in about 50% of the study participants (Table 1), either by the first-line referring physician (n = 22) or directly at the Institute (n = 26). HIV infection was already known in the remaining cases at inclusion in the study.

Table 2 compares the 104 fever episodes in HIV-infected patients with the 1035 episodes evaluated in travelers tested negative for HIV during the fever workup. Demographic data were similar between these 2 groups (except for the category of traveler). The median delay before attending our center was slightly longer in HIV-infected patients [5 days (interquartile range: 3-11) versus 4 days (interquartile range: 3-7) in HIV-negative patients, P = 0.01]. At presentation, HIV-positive travelers had more often respiratory symptoms (61%), enlarged lymph nodes (25%), and abnormal lung auscultation (20%) than the rest of the traveler population. In contrast, fever with no focal symptom was very uncommon (<5%) in the HIV-infected population.



OIs (including tuberculosis) and respiratory tract infections were the leading conditions observed in the HIV-infected patients, accounting each for about 20% of the final diagnoses (Table 2). OIs, respiratory tract infections, STI (other than HIV infection), and noninfectious diseases were more prevalent in HIV-infected individuals than in the other travelers; in contrast, tropical conditions (mostly malaria) and fever of unknown etiology were less frequently found (Table 2). Strikingly, primary HIV infection was diagnosed during the fever workup in 5 patients (4 males, 1 female), all belonging to the group of western travelers or expatriates (median duration of tropical stay = 45 days). Route of infection had been heterosexual for 3 of them and homosexual for the other 2.

An STI was diagnosed as the definite etiology of fever in 9 episodes occurring in HIV-infected patients (Table 2) and found concomitantly with another febrile condition in an additional 4 episodes (Mycoplasma genitalium cervicitis, n = 2; Chlamydia trachomatis urethritis, n = 1; Norwegian scabies, n = 1). There was no difference in prevalence of specific STI according to HIV status, but the sample size was small.

Hospitalization rate was higher (55%), and median hospital stay was longer (8 days) in HIV-positive than in HIV-negative individuals (Table 2). In contrast, mortality rate did not differ noticeably according to the HIV status; only one HIV-related fatality occurred during the course of the febrile episode (1 patient with multicentric Castleman disease). Of note, none of the HIV-infected patients died of Plasmodium falciparum malaria in contrast with 5 among the HIV-uninfected travelers.

Two additional HIV-related deaths occurred in the long-term follow-up (2 AIDS patients with septic shock). Of the 93 enrolled HIV-infected patients, 82 (88%) were still actively followed at the end of the study period (mean duration of follow-up = 39 months); 66 (80%) of them were receiving an antiretroviral therapy, of whom 58 (88%) had a full suppression of HIV replication (HIV-RNA <50 copies/mL).

Back to Top | Article Outline

Morbidity Profile of HIV-Positive Patients According to the Category of Traveler

At inclusion in this study, HIV infection had been previously diagnosed in about 60% of the western travelers or expatriates and of the VFR travelers. Most of them were actively followed in our HIV clinic, and half of them were taking a highly active antiretroviral therapy (HAART) while traveling. Only 2 HIV-infected patients did travel with cotrimoxazole prevention because of low CD4 count. Median exposure to HAART was 30 months (range: 1-47 months) and was similar for both traveler categories. There was no significant difference between the western travelers or expatriates and the VFR travelers regarding the median CD4 cell count at first inclusion in the study [414/μL (interquartile range: 232-694) versus 255/μL (52-553)] and the rate of AIDS (29% versus 38%). Disease pattern and hospitalization rate were also rather similar between these 2 categories (but still differed from those of the HIV-uninfected travelers). In sharp contrast, about 70% of the foreign visitors/migrants were diagnosed with HIV infection during the fever workup. Immunity was often severely compromised at inclusion [median CD4 cell count = 88/μL (interquartile range: 45-221)], with a high rate of AIDS (83%). Tuberculosis was the main diagnosis in 28% of the febrile episodes in foreign visitors/migrants (versus 5% in the other 2 categories) and other OIs in another 28% (versus 8%), (P = 0.003 and 0.006, respectively). Also, the rate of hospitalization was 75% for the HIV-infected foreign visitors/migrants compared with 44% for the other 2 subgroups (P = 0.003).

Back to Top | Article Outline


In our referral HIV/travel center, about 5% of the patients presenting with fever after a tropical stay have an underlying HIV infection.3 The present study describes more in detail this subset of travelers with febrile disease. Not surprisingly, HIV prevalence depended largely on the traveler category, ranging from 2% in western travelers or expatriates to 11% in VFR travelers and even reaching 24% in foreign visitors/migrants. Also, the pattern of associated diseases differed substantially in HIV-infected patients compared with the classic traveler population.

This observational study has several limitations. First, it only quantifies the HIV burden in travelers with fever (entry criterion) and therefore does not reflect the overall prevalence of HIV infection in international travelers. Second, it describes the epidemiology as observed in a European referral center, which might not be transposable as such to other nonreferral travel clinics or HIV clinics. Third, although the postexposure period of inclusion was restricted to 3 months, some of the febrile illnesses may have not been strictly travel related. Finally, all travelers and migrants have not been systematically tested for HIV infection. There was no “opt-out” testing strategy during the study period.8 However, most physicians involved in this study were used to ask routinely for sexual behavior and to offer HIV testing, including a systematic control after 3 months in case of evidence of exposure. Most patients not tested for HIV were probably at extremely low risk of HIV infection, and therefore, few HIV-infected cases might have been missed.

This study is the first of its kind to specifically investigate the epidemiology and morbidity profile of imported fever in HIV-infected travelers. Our observations might be useful in various clinical settings because such patients may attend either their usual HIV physician or any primary care or travel physician. In our study, one group of patients was composed mostly, but not exclusively, of migrants with HIV infection diagnosed during the fever workup, underlining the crucial role that primary care or travel physicians may play in early HIV recognition. Our rate of HIV testing was high in migrants and VFR travelers (about 80%) because HIV infection is highly endemic in the developing countries of origin. The lower rate of testing (55%) in western travelers or expatriates probably reflects somehow a lower perceived risk of infection. However, one should note that the few but highly contagious primary HIV infections were exclusively diagnosed in this category. Casual sexual experiences during travel have been reported at rates up to 50% for classic travelers,6 and rates were even higher for expatriates, VFR travelers, military personnel, seamen, or men who have sex with men.9 About half of the travelers engaging in sex do not use condoms consistently.7,9 It would be therefore worth investigating the yield in HIV diagnosis, which could be obtained by implementing an opt-out strategy in travel clinics, but several legitimate concerns should first be addressed before generalizing this option,10 in particular the appropriate follow-up in this highly mobile population. Also, the substantial rate of acute HIV infection and other STIs found in the study population underlines once again the importance of systematic pretravel counseling about STI prevention. This applies to HIV-negative individuals and to HIV-infected patients alike because most STIs may have a deleterious impact on HIV disease progression and HIV transmission.11

The second group of study participants was composed of individuals aware of their HIV infection, most of the time stabilized in our HIV clinic before traveling. HIV-infected patients traveled already a lot before the HAART era,12 and small studies have reported that 30%-40% of them became ill abroad or upon return.13,14 The proportion of HIV-infected patients who travel has risen with the clinical benefits of HAART.5,15 In a more recent and larger survey, about 20% of such patients sought medical care,15 a proportion which is quite similar to that observed in HIV-uninfected travelers.16 This trend underlines the increasing need for improving specific pre- and posttravel care for HIV-infected travelers. Adequate recommendations for immunization, malaria management, and also prevention of STI acquisition (as mentioned above) are required,17,18 and some gaps in the current knowledge of potential interactions between antimalarial and antiretroviral drugs need to be urgently addressed.19

The differential diagnosis of imported fever in HIV-infected travelers seemed to be different and more complex than in the classic travelers. Malaria was notably less prevalent (and less severe) than in other travelers. This may be explained by the higher proportion of semi-immune migrants in the study population and by the better pretravel counseling and the more consistent use of malaria prophylaxis in the HIV-infected western and VFR travelers followed at our clinics. Only very few HIV-positive patients took cotrimoxazole prevention during travel. Therefore, our study does not allow any conclusion about its protective effect against malaria.20 Besides a high proportion of lower respiratory tract infections, tuberculosis, and STI, the diagnostic spectrum of imported fever in HIV-infected travelers included also a large variety of OIs and of noninfectious conditions. The management of such a population requires, therefore, a wide knowledge of HIV and travel-related pathology. Not surprisingly, tuberculosis was highly prevalent in migrants found to be HIV positive, confirming the close and reciprocal interactions between both diseases in hyperendemic areas. However, some HIV-infected travelers might also be at high risk of acquiring tuberculosis, with subsequently a substantial impact on HIV disease progression and even mortality.21 Such patients might therefore benefit of specific preventive measures (including isoniazid prophylaxis or even early HAART) or at least of close posttravel follow-up (with repeated tuberculin skin testing or interferon-γ release assays).

Finally, our study provides a new illustration that travel clinics may be considered as a window on the global disease epidemiology.22 Its findings expose indeed current gaps in the global village, with individuals residing in western countries benefiting mostly of proper disease management and migrants from developing world still lacking elementary care.23 Although short-term morbidity was often severe, prompt recognition of HIV infection might even benefit immunodepressed patients mostly because long-term prognosis was very favorable with adequate management.24

Back to Top | Article Outline


HIV infection should be systematically considered in febrile patients attending travel clinics and particularly in VFR travelers and migrants. Imported fever in HIV-infected travelers is extremely challenging because of the very wide differential diagnosis and the severe associated morbidity. They should be considered as complex travelers requiring specific preventive and curative care. Our work is also a call for travel and HIV physicians to develop an adequate knowledge in both disciplines to improve health of populations on the move.

Back to Top | Article Outline


1. Gerbase AC, Rowley JT, Heymann DH, et al. Global prevalence and incidence estimates of selected curable STDs. Sex Transm Infect. 1998;74(Suppl 1):S12-S16.
2. Hill DR, Ericsson CD, Pearson RD, et al. The practice of travel medicine: guidelines by the Infectious Diseases Society of America. Clin Infect Dis. 2006;43:1499-539.
3. Bottieau E, Clerinx J, Schrooten W, et al. Etiology and outcome of fever after a stay in the tropics. Arch Intern Med. 2006;166:1642-1648.
4. Freedman DO, Weld LH, Kozarsky PE, et al. Spectrum of disease and relation to place of exposure among ill returned travelers. N Engl J Med. 2006;354:119-130.
5. Castelli F, Patroni A. The human immunodeficiency virus-infected traveler. Clin Infect Dis. 2000;31:1403-1408.
6. Matteelli A, Carosi G. Sexually transmitted diseases in travelers. Clin Infect Dis. 2001;32:1063-1067.
7. Steffen R, deBernardis C, Banos A. Travel epidemiology-a global perspective. Int J Antimicrob Agents. 2003;21:89-95.
8. Branson BM, Handsfield HH, Lampe MA, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep. 2006;55:1-17.
9. Ward BJ, Plourde P. Travel and sexually transmitted infections. J Travel Med. 2006;13:300-317.
10. Lifson AR, Rybicki SL. Routine opt-out HIV testing. Lancet. 2007;369:539-540.
11. Palacios R, Jimenez-Onate F, Aguilar M, et al. Impact of syphilis infection on HIV viral load and CD4 cell counts in HIV-infected patients. J Acquir Immune Defic Syndr. 2007;44:356-359.
12. Kemper CA, Linett A, Kane C, et al. Frequency of travel of adults infected with HIV. J Travel Med. 1995;2:85-88.
13. Kemper CA, Linett A, Kane C, et al. Travels with HIV: the compliance and health of HIV-infected adults who travel. Int J STD AIDS. 1997;8:44-49.
14. Simons FM, Cobelens FG, Danner SA. Common health problems in HIV-infected travelers to the (sub)tropics. J Travel Med. 1999;6:71-75.
15. Salit IE, Sano M, Boggild AK, et al. Travel patterns and risk behaviour of HIV-positive people travelling internationally. CMAJ. 2005;172:884-888.
16. Hill DR. Health problems in a large cohort of Americans traveling to developing countries. J Travel Med. 2000;7:259-266.
17. Chadwick DR, Geretti AM. Immunization of the HIV infected traveller. AIDS. 2007;21:787-794.
18. Spacek LA, Quinn TC. International travel: recommendations for the HIV-infected patient. Curr Infect Dis Rep. 2004;6:399-403.
19. Khoo S, Back D, Winstanley P. The potential for interactions between antimalarial and antiretroviral drugs. AIDS. 2005;19:995-1005.
20. Thera MA, Sehdev PS, Coulibaly D, et al. Impact of trimethoprim-sulfamethoxazole prophylaxis on falciparum malaria infection and disease. J Infect Dis. 2005;192:1823-1829.
21. Badri M, Ehrlich R, Wood R, et al. Association between tuberculosis and HIV disease progression in a high tuberculosis prevalence area. Int J Tuberc Lung Dis. 2001;5:225-232.
22. Jelinek T, Muhlberger N. Surveillance of imported diseases as a window to travel health risks. Infect Dis Clin North Am. 2005;19:1-13.
23. Montaner JS, Hogg RS. Hungering for HAART. Ann Intern Med. 2007;146:609-610.
24. Staehelin C, Rickenbach M, Low N, et al. Migrants from Sub-Saharan Africa in the Swiss HIV Cohort Study: access to antiretroviral therapy, disease progression and survival. AIDS. 2003;17:2237-2244.

fever; tropics; HIV; travelers; migrants

© 2008 Lippincott Williams & Wilkins, Inc.