FEVER FOLLOWING TRAVEL ABROAD
Approach to the Febrile Traveler
About 10% of international travelers will experience a febrile illness either during their travels or within 2 weeks of returning from abroad. Travelers who present with febrile illnesses to their primary care physicians offer difficult diagnostic challenges. The geographic distribution of exotic diseases is complex, and the signs and symptoms of imported infections are rarely unique. Discounting the importance of international travel, especially if the trip was brief, patients may neglect to volunteer salient aspects of the epidemiologic history. In addition, travelers in whom a febrile or gastrointestinal disorder develops may be given empiric antimicrobial therapy that can suppress the clinical manifestations of the illness or impair the microbiologic evaluation; for example, antimicrobials can eliminate the bacteremia associated with typhoid fever but not eradicate the infection.
Changes in the natural history of disorders that were once prevalent in the United States can contribute to confusion in the diagnosis of the foreign traveler’s illness. For example, most persons born before 1957 are likely to have had measles and therefore are usually protected from that infection; persons born after that date who did not have natural infection or who received an ineffective vaccine are susceptible to measles, which remains common in many underdeveloped countries of the world. Measles, therefore, must be considered in the differential diagnosis of the traveler with fever, cough, and a rash on the palms and soles.
The possibility of noninfectious disorders also exists in the febrile traveler. For example, the older person with cardiovascular disease is at risk to experience thrombophlebitis with pulmonary embolism as the result of sitting in a cramped position during a prolonged flight (so-called “economy class syndrome”). Similarly, the pyrexic patient taking trimethoprim-sulfamethoxazole for a diarrheal illness may have drug fever.
Detailed descriptions of the geographic distribution, modes of transmission, and clinical manifestations of exotic diseases can be found in a number of textbooks of tropical medicine (see annotated references); the liberal use of reference material is essential in evaluating the febrile traveler. The clinician must also realize that a number of uncommon bacterial (typhoid fever, ehrlichiosis, plague), rickettsial (Q fever, scrub typhus), fungal (histoplasmosis, coccidioidomycosis), and parasitic (babesiosis) diseases can be acquired by travel within the United States. Conversely, although Rocky Mountain spotted fever is often considered endemic only to the United States, the disease occurs in regions of Canada, Mexico, Colombia, and Brazil. Thus, the question, Unde venis? (“Where do you come from?”) remains an essential component of the history of patients with febrile illnesses of obscure origin.
About 25% of travelers to the tropics who return with a febrile illness will remain without a diagnosis despite extensive evaluations and will experience a resolution of their fever within 1 to 2 weeks. The fact that some pyrexial illnesses are benign and self-limited and the realization that the international traveler is also at risk to acquire common diseases should temper diagnostic zeal and serve to focus the evaluation; thus, the differential diagnosis of the person who presents in January with fever, headache, malaise, and myalgias following a trip to Africa must include influenza as well as malaria. Similarly, the possibility of infectious mononucleosis exists in the young adult who has vacationed abroad and is found to have fever, lymphadenopathy, and a splenomegaly. The sexually promiscuous person who has been abroad and presents with fever is at risk to have a number of exotic diseases as a consequence of international travel as well as underlying infection with HIV. Indeed, high-risk behavior abroad, such as sexual intercourse with prostitutes, can result in the acquisition of HIV, and the fever experienced on returning home may represent a manifestation of primary HIV infection.
Approach to the Febrile Traveler
The initial encounter with a traveler who has fever will dictate the speed and direction of the subsequent medical evaluation; therefore, a detailed history and thorough physical examination are essential to elicit clues that can provide insight into the cause of the patient’s illness. The interval between the trip abroad and the onset of symptoms should be determined. Many infections become clinically apparent within 21 days of exposure. Relatively short-incubation disorders include malaria, bacterial infections (typhoid fever, plague, relapsing fever, acute brucellosis, tularemia, septicemic melioidosis, septicemic glanders), chlamydial and rickettsial diseases (psittacosis, scrub typhus, epidemic typhus), and viral illnesses (dengue, yellow fever, poliomyelitis, and the hemorrhagic fevers—Lassa fever, Crimean-Congo fever, Omsk hemorrhagic fever). The aseptic meningoencephalitides (Venezuelan equine encephalitis, lymphocytic choriomeningitis, East African trypanosomiasis) will also present within 3 weeks. In contrast, chronic brucellosis, tuberculosis, Q fever, malaria, schistosomiasis, trichinosis, filariasis, visceral leishmaniasis, amebic abscess, cysticercosis, and West African trypanosomiasis can become apparent months or years after the patient has left an endemic region. Hepatitis (A, B, C, E) and HIV infection may also become manifest more than 3 weeks after exposure.
The existence of a periodicity to the patient’s fever should be determined; however, with the exception of relapsing fever (borreliosis), in which the patient is typically febrile for 3 to 4 days and afebrile for a similar length of time, characteristic patterns are uncommon. Fever that occurs every other day or every third day is highly suggestive of malaria, but these cycles are usually absent in the traveler with imported infection. On occasion, the fever associated with visceral leishmaniasis (kala-azar) may have a periodicity mimicking that of malaria.
A complete travel history should be obtained, including a list of the countries visited and the specific regions toured; the importance of this information is illustrated by the fact that malaria is generally not associated with travel to Mexico, but the disease is present outside cities and resorts along Mexico’s southwestern coast. Moreover, the significance of short visits must not be overlooked; for example, voyagers to highly endemic areas can contract malaria during transient stops in airports or seaports, and bartonellosis (Oroya fever) can develop in travelers to Peru, Ecuador, or Colombia following brief excursions into the Andes mountains.
Many infections are acquired by ingesting contaminated food, and so the patient’s dietary practices and preferences should be detailed. The traveler to less well-developed countries who enjoys indigenous food and drink is at risk to acquire salmonellosis, toxoplasmosis, trichinosis, or cysticercosis from inadequately cooked meat; salmonellosis, shigellosis, leptospirosis, amebiasis, giardiasis, dracunculiasis, or hepatitis A from contaminated ice cubes or drinking water; brucellosis, salmonellosis, campylobacteriosis, or tuberculosis from unpasteurized milk or milk products, including cheese and ice cream; and salmonellosis, hepatitis A, fish tapeworm infection, angiostrongyliasis, or infection with liver or lung flukes from raw or undercooked fish or shellfish.
Although food-borne disease is often associated with the poor sanitation prevalent in Third World countries, visitors to industrialized nations and to urban areas of developing countries are also at risk. For example, an outbreak of typhoid fever has been reported among vacationers at a tourist resort in Switzerland, and brucellosis remains a threat to visitors to Mexico and Spain who eat unpasteurized goat cheese.
Because arthropods represent common vectors for the transmission of infections, a history of insect contact or bites should be secured. Among the arthropods associated with the transmission of disease are mosquitoes (malaria, dengue, filariasis, yellow fever, Rift Valley fever), fleas (plague, murine typhus), lice (epidemic typhus, relapsing fever), blackflies (onchocerciasis), mites (rickettsialpox, scrub typhus), and ticks (tularemia, relapsing fever, Crimean-Congo hemorrhagic fever, and the spotted fever group of rickettsial infections, including Rocky Mountain spotted fever, boutonneuse fever, African tick-bite fever and Queensland tick typhus). Contact with rodents—and, in some cases, potential arthropod vectors—suggests plague, leptospirosis, lymphocytic choriomeningitis, Lassa fever, Venezuelan equine encephalitis, rickettsialpox, and both the spotted fever and typhus groups of rickettsial diseases. Exposure to farm animals indicates the possibility of brucellosis, tuberculosis, salmonellosis, tularemia, glanders, anthrax, psittacosis, and Q fever.
Wading, bathing, swimming, or boating in fresh water in endemic regions (the Caribbean, South America, Africa, the Middle East, Japan, China, and the Philippines) suggests a diagnosis of schistosomiasis. Many travelers with diseases transmitted by these arthropod and environmental vectors do not provide a history of exposure; thus, the absence of highly suggestive historical information does not exclude the possibility of the illnesses.
The patient’s history should include a review of the travel precautions that were taken, including vaccinations and chemoprophylaxis for malaria or traveler’s diarrhea. Few of the usually recommended vaccines are totally effective in preventing disease; for example, the typhoid vaccines are 60% to 80% effective in protecting against the disease, and they do not confer long-term immunity. Similarly, even strict adherence to malaria chemoprophylaxis, which occurs infrequently, does not guarantee that Plasmodium species will not return with the international traveler. Chloroquine may not be adequate to prevent infection with Plasmodium vivax or Plasmodium ovale, and the drug does not protect against chloroquine-resistant Plasmodium falciparum, which is prevalent in regions of Panama, South America, India, Southeast Asia, East Africa, and the Indonesian archipelago.
The patient’s sexual and gastrointestinal histories are other important elements of the initial evaluation. A complaint of persistent diarrhea would suggest intestinal parasites, such as Entamoeba histolytica, and invasive bacterial pathogens, including Salmonella species, Shigella species, and Aeromonas hydrophila, which has been identified as a cause of traveler’s diarrhea that can persist for months. In addition, diarrhea can be a prominent symptom in patients with malaria. The exposure of a group of travelers to a common source of infection can result in a cluster of patients with disease; therefore, travel agents and other members of tour groups should be contacted to determine if fellow travelers have become ill. Information derived in this manner may not only facilitate the evaluation of the traveler with enigmatic fever but may also be of importance to public health agencies.
Meticulous and repeated physical examinations are essential in evaluating the febrile traveler. A pulse that is inappropriately slow for the degree of concomitant temperature elevation (i.e., pulse-temperature deficit) can be a manifestation of typhoid fever, Q fever, psittacosis, scrub typhus, or anicteric leptospirosis. Conversely, a tachycardia that persists when the patient is afebrile suggests viral, rickettsial, or parasitic myocarditis—the latter caused by trichinosis, toxoplasmosis, Chagas’ disease, or African trypanosomiasis. Bilateral periorbital swelling is a clue to the diagnosis of trichinosis and Rocky Mountain spotted fever, and unilateral periorbital swelling in the traveler to South America indicates Chagas’ disease. Conjunctival injection may indicate leptospirosis or Rocky Mountain spotted fever; an ulcerating, purulent conjunctivitis in a patient with regional lymphadenitis suggests oculoglandular tularemia. A severe tonsillitis with pseudomembranes and lymphadenopathy in the traveler who consumed meat from wild animals is a sign of tularemia.
A cutaneous ulcer is characteristic of tularemia, anthrax, glanders, mycotic disease (blastomycosis, sporotrichosis), mycobacterial infection (Mycobacterium ulcerans, Mycobacterium marinum), cutaneous leishmaniasis, and dracunculiasis. An eschar is suggestive of a rickettsial illness (scrub typhus, rickettsialpox, boutonneuse fever, Queensland tick typhus, North Asian tick typhus). Skin nodules are often seen in patients with bartonellosis, cysticercosis, or onchocerciasis; travelers with cutaneous myiasis have nodules that drain purulent material, but these patients usually do not have fever or other constitutional symptoms. A skin rash is characteristic of secondary syphilis, relapsing fever, plague, meningococcemia, gonococcemia, rat-bite fever, leptospirosis, glanders, rickettsial disease, toxoplasmosis, and a number of viral illnesses, including rubella, measles, acute HIV infection, the hemorrhagic fevers, dengue, and denguelike illnesses (chikungunya, O’nyong-nyong fever). The rash associated with syphilis, rubella, and measles can occur on the palms and soles. Rose spots, which are small (2 to 4 mm) erythematous maculopapular lesions, suggest typhoid fever; unfortunately, rose spots are infrequently present (10%) in patients with the illness. A generalized vesicular rash is expected in chickenpox or rickettsialpox. Multiple pustular or necrotic cutaneous lesions in a patient with chest radiographic findings compatible with tuberculosis suggest melioidosis. Recurrent or upper-extremity cellulitis that spreads in a centrifugal fashion is a cardinal manifestation of paroxysmal inflammatory filariasis.
Lymphadenopathy is common in patients with secondary syphilis, HIV infection, plague, tularemia, tuberculosis, glanders, histoplasmosis, South American blastomycosis, scrub typhus, lymphogranuloma venereum, rubella, Chagas’ disease, toxoplasmosis, visceral leishmaniasis (kala-azar), filariasis, African trypanosomiasis, and onchocerciasis. Posterior auricular or posterior cervical adenopathy and a skin rash are highly suggestive of rubella and toxoplasmosis. Liver tenderness, hepatomegaly, or splenomegaly is an expected finding in patients with typhoid fever, brucellosis, relapsing fever, leptospirosis, Rocky Mountain spotted fever, chronic Q fever, psittacosis, viral hepatitis (A, B, C, E), yellow fever, malaria, toxoplasmosis, visceral leishmaniasis (kala-azar), amebic abscess, schistosomiasis, flukes, and visceral larva migrans.
An abnormal mental status or signs of meningitis may be important clues to the presence of relapsing fever, leptospirosis, rickettsial disease (especially Rocky Mountain spotted fever), psittacosis, tuberculosis, viral encephalitis, malaria, or African trypanosomiasis. Seizures are characteristic of cysticercosis, and they can occur in patients with paragonimiasis or schistosomiasis caused by Schistosoma japonicum.
The preliminary laboratory evaluation should include a CBC count with differential analysis. A marked anemia with fragmented RBCs on the peripheral smear might indicate hemolysis secondary to malaria or bartonellosis (Oroya fever). A hypochromic microcytic anemia can result from chronic gastrointestinal blood loss produced by hookworm infestation. An eosinophilia suggests schistosomiasis, filariasis, liver or lung flukes, or intestinal helminths (hookworm, Strongyloides, Ascaris, Trichuris); a marked eosinophilia (>50%) indicates extraintestinal parasitic disease (trichinosis, visceral larva migrans, or the autoinfection syndrome of strongyloidiasis). Although eosinophilia occurs between paroxysmal attacks of filariasis, the finding may be absent during acute episodes. Liver function studies may reveal hepatocellular dysfunction, which is found in typhoid fever, leptospirosis, relapsing fever, Q fever, viral hepatitis (A, B, C, E), yellow fever, malaria, and toxoplasmosis. The liver function tests may also reveal an elevated alkaline phosphatase or other evidence of obstruction, characteristic of secondary syphilis, amebic abscess, hydatid cyst, liver flukes, schistosomiasis, and visceral larva migrans. Renal failure with hepatitis and aseptic meningitis indicates a possibility of leptospirosis.
An infiltrate on chest radiograph is expected in patients with pneumonic melioidosis, Q fever, psittacosis, tuberculosis, fungal infection (histoplasmosis, blastomycosis), or Legionnaires’ disease, which can be acquired through travel abroad. Pneumonia in the patient with synovitis, orchitis, or meningoencephalitis may be a sign of brucellosis. The presence of pulmonary infiltrates in a patient with eosinophilia suggests paragonimiasis, strongyloidiasis, or ascariasis. Plague and anthrax are very unusual and highly lethal causes of pneumonia in travelers. An isolated, right-sided pleural effusion is an important clue to the presence of an amebic liver abscess.
A sputum analysis of patients with abnormal chest radiographic findings might include a wet preparation to detect yeast (Histoplasma capsulatum, Paracoccidioides brasiliensis) and acid-fast staining to identify mycobacteria (Mycobacterium tuberculosis). The sputum from travelers to regions where melioidosis is endemic (Southeast Asia and adjacent countries) should be carefully reviewed for the presence of Burkholderia (Pseudomonas) pseudomallei, which is a small, poorly staining, gram-negative bacillus. Lower respiratory tract secretions from voyagers to Korea, Japan, Taiwan, central China, the Philippines, West Africa, and certain regions of Thailand and South America should be searched for the characteristic operculated egg of Paragonimus westermani, which is the most common species associated with human paragonimiasis. Both melioidosis and paragonimiasis can closely mimic pulmonary tuberculosis.
Identified in 30% to 40% of cases, malaria represents the most common cause of pyrexia among febrile travelers returning from the tropics. Obviously, blood smears are an essential component of the initial laboratory evaluation of any voyager who has visited a region in which malaria is endemic. Several smears should be obtained during a 2- to 3-day period, and they should be carefully reviewed by an experienced hematologist, pathologist, or tropical medicine expert. A blood smear can reveal bacteremia (Bartonella bacilliformis or Yersinia pestis) or parasitemia (Plasmodium species, Trypanosoma species, Leishmania species, Wuchereria bancrofti, or Brugia malayi). Because some microfilaria exhibit nocturnal periodicity, blood for smears from patients who travel to endemic regions of filariasis (regions of the Caribbean and South America, Southeast Asia, and East and West Africa) should be obtained after sundown; patients from the South Pacific who are infected with W. bancrofti have maximal parasitemia during daytime hours. Concentration techniques or diethylcarbamazine provocation may be necessary to detect microfilaria.
Several blood cultures should be performed. In general, the bacteria that produce blood-borne infection in the international traveler can usually be isolated with commercially available culture media. To increase the probability that relatively fastidious bacteria, such as Brucella species, will be isolated, the clinical microbiology laboratory should hold blood cultures for 4 to 6 weeks. A number of bacterial pathogens cannot be recovered by using routine microbiologic media; for example, a semisolid medium (e.g., Fletcher’s medium) is usually required for the isolation of Leptospira species, and selective media (e.g., cysteine-glucose-blood agar) and special containment facilities are necessary for the recovery of Francisella tularensis. A careful microscopic examination of stools must be performed to detect intestinal parasites. An evaluation of a minimum of three fresh fecal specimens is usually required to exclude the presence of ova or parasites. Catharsis-induced or “purged” stools are superior to other specimens for the detection of these organisms. Stool culture should also be performed to detect the presence of bacterial pathogens.
Serologic tests can be pivotal to the diagnosis of a number of exotic diseases, including brucellosis, relapsing fever, melioidosis, glanders, leptospirosis, tularemia, Q fever, psittacosis, rickettsial disease, lymphogranuloma venereum, viral infection, South American blastomycosis, toxoplasmosis, amebiasis, trichinosis, cysticercosis, schistosomiasis, and visceral larva migrans. Accordingly, 5 to 10 mL of serum should be obtained early in the patient’s course and stored frozen; a second aliquot should be obtained 2 to 4 weeks later, and both samples are then sent for testing. A serologic diagnosis usually requires the demonstration of a seroconversion or a fourfold rise in antibody titer.
The subsequent medical evaluation is guided by the results of the history, physical examination, and preliminary laboratory data. A bone marrow biopsy for culture and histologic evaluation should be considered early in the assessment of the traveler with persistent fever, progressive weight loss, or other significant constitutional symptoms. The bone marrow culture may identify Salmonella species or Brucella species in patients with sterile blood cultures, and the histologic review may reveal findings consistent with mycobacterial infection (tuberculosis), fungal disease (histoplasmosis), or a parasitic disorder (leishmaniasis). The other tissues from which biopsy specimens may be needed to establish a diagnosis include lymph nodes (tuberculosis, toxoplasmosis, leishmaniasis), liver (tuberculosis, schistosomiasis, leishmaniasis), muscle (trichinosis, toxoplasmosis), and rectum (schistosomiasis). In addition, the biopsy of cutaneous lesions, such as ulcers, may be required to obtain sufficient material for histologic and microbiologic evaluation. Biopsy specimens of skin lesions in patients with suspected Rocky Mountain spotted fever can be stained with an immunofluorescent antibody technique that permits rapid diagnosis.
As with other patients with persistent and perplexing fever, indiscriminate antimicrobial trials should be avoided. Nevertheless, in the critically ill patient, empiric therapy may be life-saving; such therapy must be based on the patient’s travel history, which should provide epidemiologic clues to the most likely offending agents. The traveler to a region of endemic P. falciparum must receive treatment for malaria; in general, these patients should be assumed to be infected with chloroquine-resistant strains. Chloramphenicol or ciprofloxacin should be included in most empiric antimicrobial regimens for the toxic traveler; although perhaps they are not the drugs of choice, these agents possess activity against a variety of pathogens capable of producing a fatal illness, such as Salmonella species (typhoid fever), Brucella species (brucellosis), Borrelia recurrentis (relapsing fever), B. bacilliformis (Oroya fever), Burkholderia (Pseudomonas) pseudomallei (melioidosis), Bacillus anthracis (anthrax), Y. pestis (plague), F. tularensis (tularemia), Rickettsia rickettsii (Rocky Mountain spotted fever), Rickettsia prowazekii (epidemic typhus), and Rickettsia tsutsugamushi (scrub typhus). The initiation of empiric therapy does not eliminate the need to identify the source of the patient’s illness. The gravity of the clinical situation that precipitates the empiric use of antimicrobials demands a specific diagnosis; thus, the medical evaluation of the seriously ill traveler must be expedited.
Stringent infection control measures may be necessary in the hospitalized febrile traveler. In most circumstances, specific findings will determine which infection control guidelines are necessary; these findings include pulmonary infiltrates and skin rashes suggestive of communicable disorders, such as chickenpox, measles, and the viral hemorrhagic fevers. When the cause of the patient’s illness is unknown and the travel history includes exposure to communicable diseases (typhoid fever, viral hepatitis, dengue, yellow fever), the patient should be placed in a private room, and health care workers must adhere to the appropriate precautions. In selected cases, the patient will require respiratory (“droplet”) or strict isolation, at least until the cause of the illness is known or until the possibility of readily transmissible disorders is excluded. (A.L.E.)
Boreham RE, Relf WA. Imported malaria in Australia. Med J Aust 1991;155:754.
In this review of 146 cases of imported malaria, the authors find that only 11.6% of the patients were taking the recommended prophylactic drugs.
Castellani PM, et al. Six cases of travel-associated Legionnaires’ disease involving four countries. Infection 1992;20:73.
Legionnaires’ disease can be acquired as a consequence of travel to Europe and elsewhere.
Centers for Disease Control. Imported dengue—United States, 1995. MMWR Morb Mortal Wkly Rep 1995;45:988.
The authors note a substantial increase in the incidence of dengue in the Caribbean, Central America, and Mexico and a corresponding rise in the number of imported cases.
Cossar JH, et al. A cumulative review on travelers, their experience of illness and the implications of these findings. J Infect 1990;21:27.
In this extensive analysis, the authors note that many travelers lacked protective antibodies against a number of infectious illnesses, including typhoid and diphtheria, and that the travel agent was the most frequently consulted source of pretravel health advice.
Doherty JF, et al. Fever as the presenting complaint of travelers returning from the tropics. OJM 1995;88:277.
In a review of 587 febrile patients, the authors report that malaria (42%), diarrheal illness (6.5 %), and dengue (6%) represented the most common diagnoses; in 25% of the cases, however, a specific diagnosis could not be established.
Liu LX, Weller PF. Approach to the febrile traveler returning from Southeast Asia and Oceania. Curr Clin Top Infect Dis 1992;12:138–164.
A comprehensive review of the illnesses leading to pyrexia in patients returning from these regions.
Magill AJ. Fever in the returned traveler. Infect Dis Clin North Am 1998;12:445.
A very well-referenced review.
Markowitz LE, et al. International measles importations United States, 1980–1985. Int J Epidemiol 1988;17:187.
Nonimmune travelers who acquire measles abroad can precipitate outbreaks of the disease in the United States.
Ryan CA, Hargrett-Bean NT, Blake PA. Salmonella typhi infections in the United States, 1975–1984: increasing role of foreign travel. Rev Infect Dis 1989;11:1.
In a review of 2,666 reported cases of typhoid fever, the authors note that 62% were imported, most frequently from Mexico or India.
Strickland GT. Hunter’s tropical medicine, 6th ed. Philadelphia: WB Saunders, 1984.
A classic reference textbook in tropical medicine.
Svenson JE, et al. Imported malaria. Clinical presentation and examination of symptomatic travelers. Arch Intern Med 1995;155:861.
In this retrospective review of imported malaria at two hospital-based tropical disease centers, the authors note that chemoprophylaxis was prescribed for only 46% of the patients and that only half of those travelers took the medication as prescribed. They also report that almost 90% of the patients with P. falciparum malaria presented within 6 weeks following their return from travel; in contrast, almost one third of the patients with P. vivax malaria presented 6 months following their trips.
Warren KS, Mahmoud AAF. Tropical and geographical medicine, 2nd ed. New York: McGraw-Hill, 1990.
This comprehensive textbook can provide substantial assistance in the evaluation of the febrile traveler.
Wilson ME. A world guide to infections: diseases, distribution, diagnosis. New York: Oxford University Press, 1991.
The unique format of this detailed textbook provides succinct discussions, by specific geographic region, of many infectious diseases.
Wyler DJ. Evaluation of cryptic fever in a traveler to Africa. Curr Clin Top Infect Dis 1992;12:329.
An excellent discussion of the assessment of the patient with enigmatic fever following a trip to Africa, plus concise descriptions of some exotic illnesses.