Abstract & Commentary
Synopsis: Eosinophilia in returning travelers has limited predictive value for travel-related infections, but Helminthic infections are the most common diagnoses, especially when the eosinophilia is moderate-to-marked in degree.
Source: Schulte C, et al. Diagnostic significance of blood eosinophilia in returning travelers. Clin Infect Dis. 2002;34: 407-411.
A retrospective analysis was conducted on 14, 298 returned travelers seen in the Department of Infectious Diseases and Tropical Medicine at the University of Munich, Germany, from January 1995 through December 1999. The majority of patients (96.8%) had traveled to developing countries. Eosinophilia in this study was defined as at least 8% of the white blood cell count. The evaluation of patients with eosinophilia included: microscopic examination of stool, urine, blood, wounds, skin; rectal mucosal snips; 24-hour terminal urinalysis for Schistosoma ova; skin snips for Onchocerca volvulus; serology for fascioliasis, filariasis, hydatid disease, amebiasis, schistosomiasis, toxocariasis, trichinosis; antigen-capture ELISAs for Giardia lamblia and Entamoeba histolytica.
A total of 689 patients (4.8%) were found to have eosinophilia, with males more frequently affected (male-to-female ratio = 1.77). The mean age of patients was 34.3 years, and the majority were Europeans. The duration of travel ranged from 3 days to 32 years, with a median stay of 35 days. Those who had traveled to west Africa had the highest risk (RR = 2.95), whereas travelers to Latin America, Southeast Asia, and the Indian subcontinent had reduced risks of developing eosinophilia (RR = 0.39-0.91).
Although some patients did present with fatigue (24.4%), diarrhea (21.3%), and skin lesions (17.1%), 33% of the patients with eosinophilia were asymptomatic. A definitive diagnoses was made in only 36% of patients, and only 18.9% were found to have a specific helminthic infection. The positive predictive value of eosinophilia for helminthic infections was 18.9%, whereas the negative predictive value was 98.7%. The probability of obtaining a definite diagnosis increased as the degree of eosinophilia increased, reaching more than 60% when eosinophils were greater than 16%. In patients with more pronounced levels of eosinophilia, the positive predictive value for helminthic infection reached 46.6%.
The highest percent eosinophil counts occurred among patients diagnosed with helminthic infections. A total of 52.4% of all definite diagnoses made were helminthic infections. On the other hand, only 41.5% of patients found to have helminthic infections actually showed eosinophilia at presentation, consistent with the concept that certain parasites cause eosinophilia only during their migration through tissues.
Comment by Lin H. Chen, MD
Eosinophilia is usually defined as > 450 eosinophils/mm3, and can be associated with a wide variety of diseases including infectious, allergic, neoplastic, and idiopathic causes. Eosinophils are leukocytes produced in the bone marrow, and the development of eosinophils is controlled by cytokines, especially IL-5.1 Eosinophil levels show a diurnal pattern, being highest in the early morning, and the levels also decrease with an increase in endogenous and exogenous steroids.1 Following exposure to helminths, the eosinophil response tends to be greater in travelers than in those with chronic exposure.2 Additionally, eosinophilia can precede patent infections.1 Moreover, eosinophilia counts can transiently increase after treatment of parasitic infections such as schistosomiasis,3 lymphatic filariasis,3 and onchocerciasis.4 Eosinophilia can also last for 3-6 months after treatment of some infections such as loiasis.5
Numerous helminth infections are associated with eosinophilia, but most protozoa are not. Two reported exceptions are Isospora belli and Dientamoeba fragilis. When other protozoa are identified in patients with eosinophilia, one should suspect and look for helminth infections. The helminth infections commonly associated with eosinophilia include:1,8 Angiostrongylus cantonensis, ascariasis, clonorchiasis, fascioliasis, fasciolopsiasis, filaria (Wuchereria bancrofti, Brugia malayi, Brugia timori, Loa loa, Onchocerca volvulus), gnathostomiasis, hookworm (Necator americanus, Ancylostoma duodenale), flukes (Nanophyetus salmincola, Heterophyes heterophyes, Metagonimus yokogawai, Paragonimus westermani, schistosomiasis), strongyloidiasis, trichinellosis, and toxocariasis. Other causes of eosinophilia that may be encountered by travelers include scabies, myiasis, coccidioidomycosis, chronic indolent tuberculosis, HIV, and drug reactions, especially to antibiotics.1,8
Any search for parasites should depend on the specific risks encountered by the traveler. Therefore, a detailed exposure history is crucial in the evaluation of eosinophilia in a traveler. Especially important are pre-existing allergies, medications, travel itinerary, specific areas visited, duration of travel, food, and water sanitation, accommodations, exposures to fresh water, animals, insects, and sexual contacts. Physical signs and symptoms such as skin lesions, pruritus, wheeze, cough, hepatomegaly, abdominal pain, and neurologic findings may suggest more specific investigations.
The laboratory evaluation of eosinophilia should be guided by the level of suspicion of specific pathogens. The initial tests usually include a complete blood count with differential, an absolute eosinophil count in order to characterize the degree of eosinophilia, chemistries, urinalysis, a PPD skin test, 3 stool samples for ova and parasites, and chest x-ray. IgE may not be helpful because of its lack of specificity. If the history indicates possible exposure, serologies for strongyloidiasis, schistosomiasis, toxocariasis, and filariases would be useful. Further evaluation can be sought with additional serology, skin snips, biopsy of tissue (skin, rectum, bladder, liver, muscle, cyst), and examination of tissues and fluids for ova and parasites. For those patients identified with strongyloidiasis, treatment is recommended to avoid possible hyperinfection syndrome at a later time. When no definite diagnosis is reached, common practice is to reevaluate in 3-6 months. If eosinophilia persists, repeat blood, stool, and urine studies should be done, and empiric treatment of strongyloidiasis or hookworm with, respectively, ivermectin or albendazole can be considered.1
It is often challenging to make a definite diagnosis in returned travelers presenting with eosinophilia, and this study confirms the low yield in establishing one. A previous study by Libman and colleagues9 concluded that eosinophil counts had a limited role in screening asymptomatic expatriates for schistosomiasis, filariasis, and strongyloidiasis. The sensitivity of eosinophil count as a screening test for these parasites in the Libman study was 38%, and the positive predictive value of eosinophilia for these parasites was 9%. The Schulte study observed that travelers who visited West Africa had the highest risk for developing eosinophilia. Although eosinophilia only had a positive predictive value of 18.9% for all helminth infections, more than half of the definite diagnoses made were helminth infections. Finally, helminth infections were more likely to be identified when the eosinophil count reached a higher level (> 16%). Given the difficulty in establishing definite diagnoses in travelers presenting with eosinophilia, further study on the etiology and epidemiology of eosinophilia in travelers would be valuable for those who provide post-travel evaluations.
Dr. Chen is Clinical Instructor, Harvard Medical School and Travel/Tropical Medicine Clinic, Lahey Clinic Medical Center, Boston, Mass.
1. Moore TA, Nutman TB. Eosinophilia in the returning traveler. Infect Dis Clin North Am. 1998;12:503-521.
2. Nutman TB, et al. Loa loa infection in temporary residents of endemic regions: Recognition of a hyperresponsive syndrome with characteristic clinical manifestations. J Infect Dis. 1986;154:10.
3. Ottesen EA, Weller PF. Eosinophilia following treatment of patients with schistosomiasis mansoni and Bancroft’s filariasis. J Infect Dis. 1979;139:343.
4. Limaye AP, et al. Interleukin-5 and the post-treatment eosinophilia in patients with onchocerciasis. J Clin Invest. 1991;88:1418.
5. Klion AD, Ottesen EA, Nutman TB. Effectiveness of diethylcarbamazine in treating loiasis acquired by expatriate visitors to endemic regions: Long-term follow-up. J Infect Dis. 1994;169:604.
6. DeHovitz JA, et al. Clinical manifestations and therapy of Isospora belli infection in patients with the acquired immunodeficiency syndrome. N Engl J Med. 1986; 315:87.
7. Cuffari C, Oligny L, Seidman EG. Dientamoeba fragilis masquerading as allergic colitis. J Pediatr Gastroenterol Nutr. 1998;26:16.
8. Wilson, ME, Weller PF. Eosinophilia. In: Guerrant RL, Walker DH, Weller PF, eds. Tropical Infectious Diseases. Philadelphia, PA: Churchill Livingstone; 1999:1400-1419.
9. Libman MD, MacLean JD, Gyorkos TW. Screening for schistosomiasis, filariasis, and strongyloidiasis among expatriates returning from the tropics. Clin Infect Dis. 1993;17:353-359.