Diagnostic Evaluation of Newly Arrived Asymptomatic Refugees with Eosinophilia

Transcription

1 MAJOR ARTICLE Diagnostic Evaluation of Newly Arrived Asymptomatic Refugees with Eosinophilia Lorna M. Seybolt, 1,2,a Demian Christiansen, 3 and Elizabeth D. Barnett 1,2 1 Boston Medical Center, 2 Boston University School of Medicine, and 3 Boston University School of Public Health, Boston, Massachusetts (See the editorial commentary by Nutman on pages 368 9) Background. Refugees may arrive for resettlement with asymptomatic parasitic infections, and eosinophilia may be the only clue to the presence of infection. Our aim was to determine the prevalence of eosinophilia and develop a standardized approach to the evaluation of asymptomatic refugees with eosinophilia. Methods. We reviewed the medical records of refugees seen from October 1998 through May 2002 at Boston Medical Center. Data examined included age, country of origin, absolute eosinophil count, results of stool ova and parasite testing, and results of serological testing for Strongyloides stercoralis, Schistosoma species, and filaria. Results. Eosinophilia defined as an absolute eosinophil count of 450 cells/ml was present in 266 (12%) of 2224 refugees. Patients with eosinophilia ranged in age from 2 months to 81 years and had arrived from Africa, Eastern Europe, Southeast Asia, South America, the Caribbean, and the Middle East. Absolute eosinophil counts ranged from 450 to 3224 cells/ml. Pathogens were identified in stool samples of 76 (29%) of 265 patients. Serological testing for S. stercoralis, Schistosoma species, and/or filaria was done for 120 (45%) of 266 patients. Results of serological testing were positive for S. stercoralis in 45 (39%) of 115 patients, for Schistosoma species in 15 (22%) of 67 patients, and for filaria in 18 (51%) of 35 patients. Serological evidence of parasitic infection was seen at all levels of eosinophilia and in patients with and without pathogens identified in their stool samples. Conclusions. Systematic evaluation for parasites in asymptomatic, newly arrived refugees with eosinophilia should include stool ova and parasite examination, serological examination for S. stercoralis for all patients, and serological examination for Schistosoma species and filaria in patients from regions where these organisms are endemic. Many refugees arrive in the United States with asymptomatic, undiagnosed parasitic infections [1 8] that may become individual and/or public health problems if left undiagnosed and untreated [2]. Previous studies have shown that 5% 63% of asymptomatic, newly arrived refugees had intestinal parasitic pathogens found by ova and parasite (O&P) examination; prevalence varied by age and country of origin [4 9]. Most studies that assess parasitic infection in asymptomatic persons have focused on the detection of pathogens in the stool rather than evaluation of eosinophilia. Although eosinophilia has been considered as a marker for parasitic disease in returning expatriates, and it has been found Received 3 June 2005; accepted 8 September 2005; electronically published 28 December Reprints or correspondence: Dr. Lorna M. Seybolt, Maine Pediatric Specialty Group, 887 Congress St., Ste. 320, Portland, ME (seybol@mmc.org). a Present affiliation: Maine Medical Center, Portland, Maine Clinical Infectious Diseases 2006; 42: by the Infectious Diseases Society of America. All rights reserved /2006/ $15.00 to have poor positive predictive value for strongyloidiasis, schistosomiasis, and filariasis [10], this conclusion may not be applicable in refugees for whom exposure to parasites is more intense and prolonged. Therefore, some experts recommend that persons with eosinophilia who emigrate from regions where parasites are endemic be evaluated aggressively for the presence of parasitic infections [9]. In a study of Southeast Asian refugees with persistent eosinophilia and no diagnosis after a comprehensive screening evaluation, a significant number were subsequently found to have infection with hookworm and Strongyloides stercoralis [11]. As eradication efforts for diseases such as filariasis continue worldwide, we in the developed world are compelled to contribute to this progress by improving our ability to diagnose and treat these diseases in emigrants. Without a standard approach for screening and management, we likely are missing asymptomatic infections in persons who would benefit from detection and treatment. We studied the prevalence and management of eosinophilia and associated parasitic infec- Eosinophilia in Asymptomatic Refugees CID 2006:42 (1 February) 363

2 tions in newly arrived, asymptomatic refugees. Our goal was to use these data to develop a standardized approach to the diagnosis and treatment of parasitic infections in asymptomatic refugees with eosinophilia. METHODS Patient population. We conducted a retrospective chart review of refugees who had a 2-visit health evaluation as part of the Refugee Health Assessment Program at Boston Medical Center from October 1998 through May Patients with eosinophilia (defined as an absolute eosinophil count [AEC] of 450 cells/ml) were identified, and their medical records were reviewed. Standard laboratory evaluation for all refugees included a complete blood cell count with WBC differential count and a stool O&P examination. Subjects whose records did not include a complete blood cell count and/or a WBC differential count were excluded from review. Variables of interest. Data abstracted from the medical records included age, sex, country of origin (which we converted to region of origin: Africa, Eastern Europe, Southeast Asia, Middle East, and Caribbean/Latin America), AEC, results of stool O&P examination, and the results of serological testing for antibodies to S. stercoralis, Schistosoma species, and/or filaria. Screening laboratory tests were performed on the first visit; AEC and results of stool O&P examination were available prior to the second visit, 1 month later. During the study period, no protocols were in place for the evaluation of eosinophilia; further testing to investigate of the cause of eosinophilia was undertaken at the discretion of individual care providers. Stool pathogens. Organisms identified in stool samples and considered to be pathogens included Giardia lamblia, Entamoeba histolytica/dispar, Trichuris trichiura, Ascaris lumbricoides, hookworm, Schistosoma species, and Dientamoeba fragilis. The following were not considered pathogens: Endolimax nana, Entamoeba coli, Entamoeba hartmanni, Blastocystis hominis, Iodamoeba butschlii, and Chilomastix mesnili. Serological testing. Serological tests for strongyloidiasis and schistosomiasis were performed by the Centers for Disease Control and Prevention (Atlanta, GA). For strongyloidiasis, an enzyme immunoassay with a sensitivity of 90% was used. Antibodies to Schistosoma species were detected by ELISA with a sensitivity for Schistosoma mansoni of 99% and a sensitivity for Schistosoma hematobium of 95%. Serological tests for filaria were conducted by the Laboratory of Parasitic Diseases at the National Institutes of Health (Bethesda, MD). Statistical analysis. Distribution of the variables within the study population was examined. Results of serological testing were compared with the degree of severity of eosinophilia and with the presence of pathogens in the stool. Data were evaluated with x 2 test, Fisher s exact test, and logistic regression. Analyses were performed using SAS software, version 8.2 (SAS Institute). The Institutional Review Board of Boston University Medical Center approved this study. RESULTS Patient population. We identified 2271 eligible subjects; 47 persons were excluded because their records lacked a complete blood cell count and/or a WBC differential count. Eosinophilia was present in 266 (12%) of Of those with eosinophilia, 154 (58%) of 266 were male. Age ranged from 2 months to 81 years with a median age of 17.3 years. Almost 30% were!10 years of age. Subjects with eosinophilia arrived from Africa, Eastern Europe, Southeast Asia, the Caribbean and/or Latin America, and the Middle East (table 1). Eosinophilia. AECs ranged from 450 to 3224 cells/ml. The median count was cells/ml, and 175% of subjects had an AEC of!1000 cells/ml. Stool pathogens. Pathogens were detected on stool O&P examination for 76 (29%) of 265 subjects. More than half of the pathogens identified (47 of 84) were G. lamblia. Seven subjects had 11 pathogen detected (table 2). There was no difference in the prevalence of pathogens according to region of origin of the subject (data not shown). Thirty-seven subjects with no pathogens found on initial stool examination had multiple specimens submitted for O&P examination (19 subjects submitted 2 specimens and 18 subjects submitted 3 specimens). The results of the second examination showed that 6 of the 37 subjects had pathogens identified. Of the 18 who submitted 3 specimens, 3 subjects had pathogens found in both the second Table 1. Demographic characteristics of 266 refugees with eosinophilia seen at Boston Medical Center, October 1998 May Characteristic Value Age in years Median 17.3 Range Sex Male 154 (58) Female 112 (42) Region of origin Africa 118 (44) Eastern Europe 50 (19) Southeast Asia 35 (13) Caribbean/Latin America 26 (10) Middle East 37 (14) NOTE. Data are no. (%) of patients, unless otherwise indicated. 364 CID 2006:42 (1 February) Seybolt et al.

3 Table 2. Pathogens detected in refugees with eosinophilia by the initial stool ova and parasite test. Pathogen detected Value Specific pathogen, no. (%) of instances of detection Giardia lamblia a 47 (56) Entamoeba histolytica/dispar a 11 (13) Trichuris trichiura 10 (12) Ascaris lumbricoides 6(7) Hookworm 4 (5) Schistosoma mansoni 3(4) Dientamoeba fragilis 3(4) Any 1 pathogen, no. (%) of patients 76 (29) 2 Pathogens, no. (%) of patients 6 (2) 3 Pathogens, no. (%) of patients 1 (0.4) a Generally does not cause eosinophilia. and third specimen (the same organisms were identified in subsequent specimens). Serological testing. Serological testing for antibodies to S. stercoralis, Schistosoma, and/or filaria was performed for almost half of the subjects with eosinophilia (120 [45%] of 266 subjects). S. stercoralis serological examination was performed for 115 (43%) of 266 subjects, Schistosoma serological examination was performed for 67 (25%) of 266, and filaria serological examination was performed for 35 (13%) of 266. After adjusting for region of origin of the subject, we found the odds of a subject having serological testing performed increased as AEC increased; subjects whose AEC was in the highest quartile (range, cells/ml) were almost 8 times more likely to have serological testing done, compared with subjects whose AEC was in the lowest quartile (range, cells/ml) (table 3). The results of serological examination were positive for S. stercoralis in 45 (39%) of 115 subjects, for Schistosoma species in 15 (22%) of 67, and for filaria in 18 (51%) of 35. Subjects who had more-severe eosinophilia (AEC of 1000 cells/ml) were more likely to test positive for filaria than were subjects with less-severe eosinophilia (AEC of cells/ml) (67% vs. 31%, P p.04) (table 4). There was no association between the severity of eosinophilia and results of testing for either S. stercoralis or Schistosoma species. No association between the presence of pathogens in stool and the performance of serological testing was found. Serological examination was performed for 88 (46%) of 190 patients who had no pathogens isolated in their stool samples, compared with 32 (42%) of 76 patients with 11 pathogen found by stool O&P examination ( P p.53). There was no association between the presence of pathogens in the stool and the results of serological testing for S. stercoralis and filaria (table 4). No subjects with pathogens identified in their stool had a positive test result for Schistosoma infection, although 3 persons who had Schistosoma mansoni identified on stool O&P examination did not have serological testing done. DISCUSSION Eosinophilia was present in 12% of newly arrived refugees screened at our clinic. Most subjects with eosinophilia were children, and more than half were male. Although there is little information in the published literature regarding the prevalence of eosinophilia in asymptomatic refugees, prior studies of refugees and immigrants have shown that younger persons and males are more likely to harbor parasites [4, 12]. The most common cause of eosinophilia around the world is infection with helminthic parasites [13]. Although we cannot comment on the overall prevalence of S. stercoralis, Schistosoma species, or filaria in our study population, between a quarter and a half of those tested had serological evidence of infection. We chose to focus on strongyloidiasis, schistosomiasis, and filariasis, because they are common diseases in the regions of the world from which refugees emigrate, and reliable serological tests are available [11]. Persons may be infected and yet be asymptomatic for many years before serious illness and disability become apparent [14]. Strongyloidiasis, for example, may be fatal in persons who become immunocompromised from either another illness or from immunosuppression caused by the treatment of another illness [15]. Serious disability and disfigurement may result from lymphatic filariasis or onchocerciasis [16, 17]. Left untreated, S. haematobium infection may lead to infertility or urinary tract malignancy [18]. Safe and effective treatments are available for these patients once the parasites are detected [19]. The best method for the diagnosis and treatment of parasitic infections in asymptomatic immigrants from regions where helminth infection is endemic is still subject to debate. Several approaches to the evaluation of eosinophilia exist. One strategy is to use multiple stool O&P examinations; however, although the testing of multiple specimens increases the likelihood of finding a pathogen, the optimal number of tests has not been determined from a cost-benefit and patient burden perspective [20, 21]. S. stercoralis is notoriously difficult to detect in stool Table 3. Odds of patients having serological testing done, stratified by absolute eosinophil count (AEC) and adjusted for subjects regions of origin. AEC, cells/ml OR 95% CI P Reference !.0001 Eosinophilia in Asymptomatic Refugees CID 2006:42 (1 February) 365

4 Table 4. Results of serological testing for antibodies to selected parasites, by degree of severity of eosinophilia and presence of stool pathogens. Organism tested for, by patient AEC No. (%) of subjects with a positive serological test result Pathogen present in stool sample No. (%) of subjects with a positive stool O&P result Strongyloides species cells/ml 25 (68) Yes 8 (33) cells/ml 20 (49) No 37 (45) Schistosoma species cells/ml 9 (29) Yes 0 (0) cells/ml 6 (21) No 15 (32) Filaria cells/ml 5 (31) a Yes 2 (33) cells/ml 13 (67) No 16 (54) NOTE. a AEC, absolute eosinophil count; O&P, ova and parasite examination. P p.04 for comparison of presence of antibodies by AEC. specimens [22], and other parasitic infections, such as filaria, cannot be diagnosed by stool examination. Our data do not support a strategy of using only stool O&P tests, because subjects with pathogens in their stool were found to have serological evidence of infection with other parasites. Additional stool O&P examinations may be helpful as one part of a diagnostic strategy because, although only a few subjects in our study gave multiple stool specimens, we found that several of those subjects had pathogens identified in the second sample. Another diagnostic approach is serological testing for antibodies to common parasites. Some experts advocate serological testing for patients from regions where helminth infection is endemic who have unexplained asymptomatic eosinophilia [1, 9, 23]. Selection of parasites for which to test is based on the following principles of screening: (1) the presence of an asymptomatic infection, (2) the potential for individual and/or public health impact if the disease is left untreated, and (3) the availability of accurate and reliable screening tests and treatment. Knowledge of geographic distribution of various parasites would guide the selection as well. Those who have recommended this approach suggest that parasites to be considered include S. stercoralis, Schistosoma species, filaria, and, in children, Toxocara species [23]. A third strategy is empirical treatment with albendazole either prior to departure from the region where helminth infection is endemic or after arrival, when eosinophilia is detected. Although this may be effective in treating many parasites, albendazole would not be the drug of choice to treat infection due to S. stercoralis, Schistosoma species, or filaria. Some refugees do receive a dose of albendazole prior to departure [24]. Therefore, the prevalence of pathogens in stool may be declining at least in persons coming from those areas where predeparture treatment is being administered. In Massachusetts, there has been a decrease in the detection of intestinal helminths among African refugees since the introduction of predeparture albendazole treatment [25]. The utility of the stool O&P examination in determining the etiology of eosinophilia in persons from these areas would be expected to decline as the prevalence of pathogens in the stool declines because of empirical treatment. Limitations of this study include the retrospective chart review design. Serological testing was not done routinely for patients with eosinophilia, although, after adjusting for region of origin of our subjects, we found that our data show providers were more likely to perform serological tests for patients with more-severe eosinophilia. When S. mansoni was found in stool samples, patients were treated and no further testing was done. Several patients without Schistosoma species found in stool had positive serological test results for both S. stercoralis and Schistosoma species; those subjects not tested may have had coinfection that was missed, which may have led to an underestimate of the prevalence of S. stercoralis infection in refugees. Serological testing for other parasites that may cause eosinophilia, such as Toxocara species, was rarely performed. We cannot report on long-term outcomes of the subjects, such as resolution of eosinophilia after treatment or progression to clinical disease in patients in whom infection was not identified, because primary care physicians at other sites provided followup care. We relied solely on the results of serological testing in most cases, and we do not know which patients had past infections that had resolved and which would have progressed to clinically relevant disease in the absence of treatment. The presence of antibodies in persons living in areas of endemicity indicates infection at some time but does not necessarily indicate current infection. Cross reactions do occur among helminths, which limits the specificity of some serological tests. Clinicians should be familiar with the laboratory used to conduct the tests and with the sensitivity and specificity of any testing performed. This study focused on refugees, and it may not be appropriate to generalize the results to other immigrant populations. However, immigrants who come from areas where helminth infection is endemic and who have suboptimal living conditions may have more in common with refugee populations in terms of exposure and would benefit from similar screening strategies. Primary care providers, the first clinicians to evaluate many new immigrants, can expect to see patients who may have had sporadic medical care and are not required to have the same overseas evaluation as departing refugees. Providers will need to obtain detailed exposure histories from patients who recently immigrated to determine if they might benefit from screening. As increasing numbers of immigrants arrive from areas where parasitic diseases are endemic, providers will encounter in- 366 CID 2006:42 (1 February) Seybolt et al.

5 creasing numbers of persons with asymptomatic infections that may be suggested by the presence of eosinophilia. CONCLUSIONS Serological evidence of parasitic infection was seen in patients who had different degrees of severity of eosinophilia, as well as in patients with and without pathogens in their stool. We recommend that systematic evaluation for parasites in asymptomatic refugees with eosinophilia include stool O&P examination and serological testing for S. stercoralis for all patients, and serological testing for Schistosoma species and filaria for patients from regions where these pathogens are endemic. Acknowledgments Potential conflicts of interest. References All authors: no conflicts. 1. Stauffer WM, Kamat D, Walker PF. Screening of international immigrants, refugees, and adoptees. Prim Care 2002; 29: Muennig P, Pallin D, Sell RL, Chan M. The cost effectiveness of strategies for the treatment of intestinal parasites in immigrants. N Engl J Med 1999; 340: Bass JL, Mehta KA, Eppes B. Parasitology screening of Latin American children in a primary care clinic. Pediatrics 1992; 89: Lifson AR, Thai D, O Fallon A, Mills WA, Hang K. Prevalence of tuberculosis, hepatitis B virus, and intestinal parasitic infections among refugees to Minnesota. Public Health Reports 2002; 117: Meropol SB. Health status of pediatric refugees in Buffalo, NY. Arch Pediatr Adolesc Med 1995; 149: Hayes EB, Talbot SB, Matheson ES, Pressler HM, Hannah AB, Mc- Carthy CA. Health status of pediatric refugees in Portland, ME. Arch Pediatr Adolesc Med 1998; 152: Ryan N, Plackett M, Dwyer B. Parasitic infections of refugees. Med J Aust 1988; 148: Catanzaro A, Moser RJ. Health status of refugees from Vietnam, Laos, and Cambodia. JAMA 1982; 247: Walker PF, Jaranson J. Refugee and immigrant health care. Med Clin North Am 1999; 83: Libman MD, Maclean JD, Gyorkos TW. Screening for schistosomiasis, filariasis, and strongyloidiasis among expatriates returning from the tropics. Clin Infect Dis 1993; 17: Nutman TB, Ottesen EA, Ieng S, et al. Eosinophilia in Southeast Asian refugees: evaluation at a referral center. J Infect Dis 1987; 155: Godue CB, Gyorkos TW. Intestinal parasites in refugee claimants: a case study for selective screening? Can J Public Health 1990; 81:191 5, 13. Rothenberg ME. Eosinophilia. N Engl J Med 1998; 338: Gyorkos TW, MacLean JD, Viens P, Cheang C, Kokoskin-Nelson E. Intestinal parasite infection in the Kampuchean refugee population 6 years after resettlement in Canada. J Infect Dis 1992; 166: Scowden EB, Schaffner W, Stone W. Overwhelming strongyloidiasis an unappreciated opportunistic infection. Medicine 1978; 57: Burnham G. Onchocerciasis. Lancet 1998; 351: Melrose WD. Lymphatic filariasis: new insights into an old disease. Int J Parasitol 2002; 32: Helling-Giese G, Kjetland EF, Gundersen SG, et al. Schistosomiasis in women: manifestations in the upper reproductive tract. Acta Tropica 1996; 62: Drugs for parasitic infections. Med Lett Drugs Ther August 2004; 46: Available at: Accessed 31 May Gyorkos TW, MacLean JD, Law CG. Absence of significant differences in intestinal parasite prevalence estimates after examination of either one or two stool specimens. Am J Epidemiol 1989; 130: Marti H, Koella JC. Multiple stool examinations for ova and parasites and rate of false-negative results. J Clin Microbiol 1993; 31: Sato Y, Kobayashi J, Toma H, Shiroma Y. Efficacy of stool examination for detection of Strongyloides infection. Am J Trop Med Hyg 1995; 53: Klion AD. Eosinophilia. In: Keystone JS, Kozarsky PE, Freedman DO, Nothdurft HD, Connor BA, eds. Travel medicine. Philadelphia: Mosby, 2004: Centers for Disease Control and Prevention. Enhanced medical assessment strategy for Barawan Somali refugees Kenya MMWR Morb Mortal Wkly Rep 1998; 46: Geltman PL, Cochran J, Hedgecock C. Intestinal parasites among African refugees resettled in Massachusetts and the impact of an overseas pre-departure treatment program. Am J Trop Med Hyg 2003; 69: Eosinophilia in Asymptomatic Refugees CID 2006:42 (1 February) 367