Ever since 1879 when Paul Erlich first described the prominent red granular staining of a unique subset of granulocytes with the acid aniline dye eosin, physicians have sought to understand the role of eosinophik in immune responses where they are prominent.1 While an association between eosinophilia and helminthic parasite infections has been well described,1"3 many other conditions such as drug reactions, asthma, atopic disorders, and some lymphoproliferative and autoimmune diseases also can be associated with eosinophilia, making it a nonspecific finding for any particular diagnosis.1'4 The evaluation of a patient with eosinophilia is challenging due to the diverse set of diseases and disorders associated with this laboratory finding. Uncovering an etiology requires thoughtful application of all the tools available to the clinician.
Eosinophilia associated with parasite infections is associated specifically with tissue invasion and migration of helminthic parasites, and the degree of peripheral eosinophilia is usually proportional to the extent of tissue invasion.2'4 Conversely, parasite infections that do not invade host tissues, and most protoioal infections, generally do not result in increased eosinophilia.3 This article deals mainly with eosinophilia in the context of helminth infections. Because symptoms are unreliable to define who should be evaluated for parasites,2 eosinophil counts and stool examinations often are recommended for screening travelers and immigrants.5 Important issues related to travelers and patients with eosinophilia will be discussed to provide a framework for cogent evaluation of these patients.
DEFINITION AND BACKGROUND
Eosinophilia generally is defined as having >500 eosinophils per mm3 in the peripheral blood and not based on a particular percentage of the white blood cell differential.3,6 This points out the important principle of comparing total cellular counts rather than percentages on a differential count. Because some automated complete blood cell (CBC) count instruments have great difficulty differentiating eosinophils from neutrophils (PMN), total counts approximated by multiplying the total white blood cell (WBC) count by the percentage of eosinophils are prone to significant error.6,7 A more accurate method involves enumeration of the absolute number of eosinophils per mm3 using special stains that uniquely identify eosinophils. If you are certain your lab uses recently developed instruments that reliably differentiate between eosinophils and PMN, then multiplying the total leukocyte count by the percent eosinophils may be a useful approximation of the absolute eosinophil count.7
Under normal circumstances, the absolute eosinophi! count shows a diurnal variation, with the highest levels near midnight and the lowest around noontime.6 Some investigators believe this is reflective of the normal diurnal cortisol cycle because exogenous steroid medications are known to lower eosinophil counts significantly.4 In addition, acute pyogenic and viral infections may be associated with a transient, but at times profound, reduction in absolute eosinophil counts.2
While circulating peripheral blood eosinophils are readily obtainable, they represent only a small portion of the entire eosinophil population and hence provide only a limited view of their role in the immune response.3 Under normal circumstances, eosinophils are predominantly tissue-based cells that spend only about 2 to 3 days of their approximate 4-week lifespan in the peripheral circulation.1 The tissue-to-blood ratio is approximately 100 to 200:1. 1>8 Moreover, the distribution is not uniform in all tissues.3'8 Mature eosinophils are found most prominently in the skin and epithelial lining of the respiratory and gastrointestinal tracts, representing the epithelial surfaces most directly exposed to the external environment. Occasionally, eosinophils are found in inflammatory fluid collections such as bronchoalveolar lavage (BAL) fluid,9 pleural fluid,2 or abnormal cerebrospinal fluid,10 and essentially always are associated with a pathologic process, although not necessarily parasitic. Excellent reviews of various tissue fluid eosinophilias are available and will not be reviewed further in this article.
While often considered to be essentially an unusual type of granulocyte, eosinophils have many unique functions and their production is regulated separately. Increased numbers of eosinophils in the blood are the result of enhanced bone marrow production since they do not marginate, recirculate, or undergo any further cell division after leaving the marrow space.1 Eosinophil development and differentiation is regulated by the cytokines interleukin-3 (II-r3), interleukin5 (IL-5), and granulocyte- macrophage colonystimulating factor (GM-CSF).11 While all three play a role,11-15 uniquely promotes eosinophilia while 11-3 and GM-CSF also aid in the development of other white blood cell precursors.8,11 Increased levels of IL-5 have been found in the serum of patients with parasitic infections and other conditions associated with eosinophilia.12,13 Interleukin-5 also plays a role in chemotaxis of eosinophils into tissues where they participate in the local immune reaction.14
Increased tissue eosinophilia is either a result of the usual egress in the fece of enhanced marrow production or facilitated specifically by chemotactic factors such as ILS,11-2, the complement component C5a, platelet activating factor (PAF), or the chemokine RANTES produced in local inflammatory reactions.8,14 Eosinophils also express some unique surface proteins, such as VLA-4, which assist them to disperse into tissues apart from PMNs by specific binding to endothelial vascular cell adhesion molecule- 1 (VCAM1).15
PROPOSED IMMUNE MECHANISMS
The role of eosinophils in inflammatory immune responses against parasites is incompletely understood. The role largely has been studied in terms of eosinophils as proinflammatory effector cells.8 As with many aspects of the immune system, the effector response of eosinophils can have both helpful (parasite elimination) or detrimental (tissue destruction) results. Overall, the effector functions are related to secretion of their highly toxic and inflammatory granule products, production of lipid mediators, and products of oxygen metabolism in the tissues where they accumulate. Their granules containing eosinophils cationic protein (ECP) and major basic protein (MBP) are powerful helminthotoxins. Eosinophil peroxidase and oxygen metabolites also are known to be directly toxic to helminths. While the lipid mediators leukotriene C4 and PAF are not toxic to parasites, they promote eosinophil infiltration and enhance the inflammatory reaction through increased vascular permeability.8 Eosinophils are known to exert an antibody-dependent cellular cytotoxicity toward in vitro preparations of various helminth parasites.1 This cytotoxiciry has either an itnmunog' lobulin (Ig) G- or IgE-assisted mechanism.
Eosinophilia Not Associated With Parasitic Infections
Helminth Parasites Associated
As previously mentioned, the cytokines IIj-3, 11-5, and GM-CSF are important for inducing eosinophilopoesis. In addition, these cytokines serve to prolong the viability of the cells, activate the eosinophils to enhance their effector role, and induce expression of class II MHC and other activation molecules on the cell surface,8 Some recent evidence has shown that class II MHC bearing eosinophils can present antigen to CD4+ T cells.16'17 Eosinophils also can synthesize cytokines such as IL-3,11-5, GM-CSF, TNF-α and TGF-ot.18 These data all suggest a further role for eosinophiis in the afferent arm of the immune response to parasite infections.
EOSINOPHILIA AS A SCREENING TOOL
The use of eosinophilia as a screening test for parasites has never been fully evaluated in a randomized, prospective study. In reviewing the available literature, a bias toward using eosinophilia as a screening tool to decide who to evaluate further for possible parasite infection was prevalent.19'22 However, those same articles revealed a sensitivity of only 27% to 39% for a parasitologic diagnosis among expatriate workers and travelers evaluated on return from areas where parasites were prevalent. In a rigorous study of expatriate North Americans, the positive predictive value of eosinophiiia for parasite infection was 14%.20 In a prospective, thorough evaluation of Southeast Asian refugees, eosinophilia carried a 67% sensitivity and 55% positive predictive value.19 These data reflect the reality that even in refugee populations, many nonparasitic diseases and disorders are associated with eosinophilia (Table i ). It is also useful to look at the specificity for ruling out parasitic infections in the population who have a normal (ie, <500/mm3) eosinophil count. In the only study in which the data allowed calculation of a specificity, the absence of eosinophils was 91% specific for the lack of a parasitic diagnosis, and the negative predictive value was 96%.20 These data on North American expatriates may have been biased since patients with initial eosinophilia were evaluated more extensively, but nevertheless indicate that a normal eosinophil count may be useful diagnostically. The negative predictive value in the review of refugees, however, was only 73%. 19
To put the eosinophilia of parasitosis into perspective, several points must be considered. First, as noted above, eosinophilia is largely due to the tissue migration of helminthic parasites including nematodes, trematodes, and cestodes as summarized in Table 2. Protozoal infections are almost never associated with a significant eosinophilia.1'3 Second, while foreign travel is often where helminthic parasites are acquired, several parasites occasionally can be found in people who have never traveled outside the United States (Table 3).2 Third, many other causes of significant eosinophilia are known, and many are more common than parasite infections (Table 1). An interesting summary of 418 patients seen at the Mayo Clinic in the 1940s with >20% eosinophilia revealed parasites to account for only 4% of the total number of patients.23 While better diagnostic tests in recent years may increase the percentage of parasitic diagnoses, the overall trend is unlikely to change dramatically.
Fourth, as global travel and international relocation become more common, primary care physicians often are faced with patients from areas where parasite infections are prevalent. In a recent review of Southeast Asian adult and pediatrie refugees, approximately half of the 1 194 patients screened had eosinophilia.19 Of those, about half were found to have parasites, although not always clearly pathogenic species. An important implication of the presence of nonpathogenic organisms such as Entamoeba colt or Bìastocystìs hominis in the gut is that there has been contamination of the gastrointestinal tract in the same manner that pathogenic organisms are acquired. Fifth, the eosinophilia associated with parasites is dynamic depending on the Hie cycle of the parasite, especially concerning the duration and type of tissue migration.2 Finally, the diagnostic evaluation of eosinophilic patients must recognize that parasitic life-cycle delays often will require repeated evaluations and serologie diagnostic techniques when available.2'5
Eoslnopnllla-lnducing Parasite Infections That Can Be Acquired In the United States
Careful consideration of parasitic life cycles and the natural dynamics pf parasite-associated eosinophilia will aid in the cogent evaluation of patients with possible helminth infections.2'23 Because the tissue migratory phase of the life cycle precedes the localization of most helminths in the gastrointestinal tract by several weeks to months, it follows that diagnostic eggs or larva will not appear in the stool for weeks to months after acquiring the parasite.
Parasite-associated eosinophilia is a dynamic process, and the pattern may be helpful in narrowing the diagnostic possibilities. A sustained high-grade eosinophilia is unusual and typically seen only during active infection with visceral larva migrans or trichinellosis, which remain in tissue throughout their life cycle. High-grade eosinophilia, which may persist but usually at a low to moderate level, is consistent with schistosomiasis, Strongyloidiasis, filariasis, or hookworm infection. Eosinophilia that varies directly with episodic tissue movement of the parasite during its life is found with loiasis, dracunculiasis, or gnathostomiasis. When eosinophilia characterizes only the tissue migratory stage, diseases like ascariasis or trichuriasis can see resolution of the eosinophilia despite persistent gut infection once they have completed their pulmonary migration. Parasites that usually only elicit eosinophilia after death or cyst disruption and corresponding release of antigen include the cestode infections echinococcosis and cysticercosis. Finally, the absence of eosinophilia is common in protozoal infections such as giardiasis, amebiasis, malaria, and cryptosporidiosis, even though some degree of local tissue invasion may occur.1'3
A point underscored by the review of Southeast Asian refugees is the importance of using serology to evaluate patients fully. Strongyloidiasis was found in 38% of refugees who underwent a previous screen (history, physical, and stool ova and parasites (O & P] X 3) and were referred for persistent, unexplained eosinophilia. Twenty-nine percent of those were found only by serologie testing.19 In filariasis, where diagnostic microfilaria are not produced until 6 to 12 months after infection, only serology and classical symptoms in a patient exposed in an endemic area will lead to the correct diagnosis in the early phases of the disease.24 Similarly, the life cycle of Toxocara species (animal ascaride) that are accidentally acquired by humans lead to persistent tissue migration, hence the name visceral larva migrans. Without any readily obtainable diagnostic forms, serology is required to establish a diagnosis.2
Because some parasites live for up to 10 years and the resolution of eosinophilia without treatment may represent the natural course of a persistent infection, a normal eosinophil count should not be considered evidence that parasites are not present. In general, it can be said that when present, eosinophilia can be a modestly helpful diagnostic tool. Yet given the low sensitivity and high but variable specificity of eosinophil counts, determining exposure history is much more important when considering who to evaluate more extensively.
The evaluation of an individual with eosinophilia should begin with a history to look for clues pointing to either a parasitic or nonparasitic etiology. Drug ingestion is one of the key elements and needs to go beyond a standard request for current medications. Our experience with L-tryptophan points out the need to question patients about all products they ingest because many do not consider vitamins and dietary supplements to be medications. In addition, eosinophilia can be the sole manifestation of drug sensitivity, apart from skin rashes, fever, or urticaria.2
Most parasites have defined geographic distributions, making a detailed travel history obviously important to determine potential exposure. In addition to determining the places a traveler has visited, the exposure history is also helpful to determine the likelihood of parasite infection. Travelers who only stay in expensive hotels in urban areas and always wear shoes and insect repellents are unlikely to become infected with parasites transmitted by soil contact or arthropod vectors. Hookworm and strongyloidiasis typically are contracted only by walking barefoot in soil contaminated with human feces. In contrast, development workers and missionaries who often spend extended periods in rural settings have a much greater risk of acquiring parasites. While significant parasite infections have been acquired in the face of minimal exposure, many infections, such as fitariasis, usually require repeated bites due to low vector infection rates. Therefore, determining the length of stay in an area becomes important to evaluating the likelihood of infection. Schistosomiasis infection rates also are related to the extent and duration of exposure to freshwater in areas of disease transmission. Dietary review can be useful to determine exposure to trichinosis, tapeworm (including cysticercosis), and toxocariasts.
Because many parasitic conditions associated with increased eosinophilia are asymptomatic, one cannot rely on a history of symptoms to determine who to evaluate. The most important information remains the extent and duration of exposure, in the context of the geographic areas visited.2
In most cases of eosinophilia, the vital signs and physical examination are normal.2 Fever can suggest trichinosis when associated with generalized systemic symptoms and malaise, and lymphatic fìlariasis when associated with focal lymphangitis. Gastrointestinal symptoms are seen with fascioliasis, visceral larva migrans, hookworm infestation (both human and canine), and ascariasis. Urticaria is seen with several tissue- in vasi ve helminths, but may point to infection with Onchocerca volvulus, Loa loa, Strongyloides stercaraUs, hookworm, or Trichineüaspiraks. Rarely, wheezing is seen during the pulmonary migration of some helminths such as hookworm, ascaris, and toxocara.
As noted above, patients with significant exposure histories need a thorough evaluation. More than one absolute eosinophil count will be helpful to look for trends and avoid transiently normal results due to early infection, natural fluctuations, or intercurrent viral infections.3 In feet, given the relatively high negative predictive value, repeated normal eosinophil counts are reassuring to exclude helminthic infections in returning travelers. Liver function tests and urinalysis may point to important organ involvement. If Schistosoma hematobiitm exposure is suspected (freshwater contact in the Middle East or Africa), the urine may be evaluated further for ova from this parasite. A chest radiograph may reveal findings consistent with the pulmonary migratory phase of several helminths.
Stools collected on three occasions, preferably 48 hours apart to account for the intermittent nature of egg release in gut helminths, should be tested for O & P. Special concentration techniques may be necessary to find schistosomiasis or strongyloidiasis. Further stool tests may be prudent in the weeks after finding eosinophilia given the life-cycle-driven delay between the onset of eostnophilia and diagnostic eggs or larva appearing in the stool. If stools are negative, a duodenal aspirate may be helpfiil to diagnose strongyloidiasis, hookworm, or liver flukes (Ckmorch« smensis or Fasciola hepática). Rectal snips are sometimes necessary to find the eggs of Schistosoma mansoni or japomcum.
When filaria! infections are suspected, anticoagulated blood should be filtered from the time of day associated with the highest microfilarial concentration (noon for L loa, MansoneUa perstans, and Mansonelia azzardi; midnight for Wuchereria bancroftì and Bragia malati)·24 For onchocerciasis, skin snips need to be tested since the microfilaria reside in die skin rather than the circulation. As previously noted, the life cycle of many parasites means that diagnostic forms will be difficult to find in early infection, or they are not accessible, and hence serology should be obtained when classical specimen evaluation is unrevealing. Currently, serology is available for filariasis, schistosomiasis, strongyloidiasis, visceral larva migrans, echinococcosis, cysticercosis, and trichinellosis. In a recent review on screening for filariasis, schistosomiasis, and strongyloidiasis, stool examinations and serology were the most cost-effective tools to diagnose these important parasitic infections.20 Eosinophilia had a limited role in the evaluation of these infections.
Eosinophilia is defined as an absolute count of >500 eosinophils per mm3 of peripheral blood. Eosinophiíia is associated with many disorders, limiting its usefulness as a diagnostic tool in screening expatriates for parasite infections. In addition, only tissueinvasive helminthic parasites cause eosinophilia, which limits its general application as a screening tool for parasitic infections. Because eosinophilia may resolve spontaneously over time, the life cycle of parasites must be considered when evaluating eosinophilic patients, and repeated stool examinations or appropriate serology may be necessary to make the correct diagnosis. Future work on the risks associated with subclinical parasite infections would be helpful to place eosinophitia and other screening tests in proper perspective. Referral of difficult cases to specialists in travel medicine should be considered because detailed information about the geographic distribution and life cycle of helminthic parasites is often crucial to making the correct diagnosis.
1. Spry CJF. Eosmopteliu; A Guidi to tht Scientific end Medicai Latratore. Oxford, England: Oxford University Press 1986.
2. Wilson ME, Eosinophilia. In: Wilson ME, ed. A Wfarid Quids to fafeetums: Diseases, Diirribucon. Dii^noiu. New Yoik, NY: Oxford University Press; 1991:164-175.
3. Mahmoud AAF. Eosinophilia. Irt Warren KS, Mahmoud AAF, eds. Trapica! and Geographic Medicine. 2nd éd. New Îbfk, NY: McGiaw-Hill; 1990:70-75.
4. Weller PF. Eosinophitia. 1 Allergy CUn Immurai. I9S4;73:1-!0.
5. Wolfe MS. Protection travelers. In: Mandell OL. Douglas RG Jr, Bennetr )E. eds. Principles and Practice of In/ecnoui Diseases . 3rd ed. New York. NY: Churchill Livingstone, 1990:234-1MO.
6. Philpott J, Keystone JS. Eosinophilia: an approach to the problem in the returning traveler. Traneler Mediane lntemaaonai. 1987;15:51-56.
7. Terstappen LWMM, de Graoth BG, Vischer K, van Kouterik FA, Greve J. Four-parameter white blood cell differential counting based on light scattering measurements. CyiomeTry. 1988;9:39-43.
8. T. Weller PF. The immunobiology of eosinophils. N Engl Med. 1991i321:lllO-llI8.
9. Allen JN, Davis WB, Pacht ER. Diagnostic significance ci increased branchoalveotar lavage fluid eosinophils. Am Rev Resp'r Du. 1990;142:642-647.
10. Kuberski T. Eosinophils in the cerebiospinal fluid. Ann iniem Med. 1979;9J:70-75.
11. Clutterbuck EJ, Hirst EM, Sanderson CJ. Human interleukin-5 (1L5) regulates the production of eosinophils in human bone marrow cultures: comparison and interaction with IU, IU3, ILnS. and GMCSE Bbod. 1989;73:1 504-15 12.
12. Umaic AP, Abrams JS, Silver JE. Cttesen EA. Nutman TB. Regulation of parasite- induced eosinophilia: selectively increased inrerteukin 5 production in helminrh-infected patients. } ?? Med- 1990;! 72=399-402.
13. Owen WF, Rothenberg ME, Petetson J, et al. In ter lenk in- 5 and phenatypically altered eosinophils in the blood of patients with the idiopathic hypereosinophilic syndrome. J Exp Med. 1989; 170:343-348.
14. Meurer R, Van Ripei G, Feeney W, et al. Formation of eosinophilic and monocytic intradermal inflammatory sites in the dog by injection of human RANTES but not human monocyte chemoatrractant protein- 1, human macrophage inflammatory protein-?a, or human interleukin-8. ) EJC(J Mal. 1993:1787:1913-1921.
15. Walsh GM, Mermcd JJ, Marmeli A. Kay BA, Wardlaw AJ, Human eosinophil. but not neutrophil, adherence to IL-I -stimulated human umbilical vascular endothelial cells is (very late antigen-4) dependent. J InunwioJ. 1991; 146:34 19-3423.
16. Hansel TT, De Vries M, Carballido JM, et al. lnducrion and function of eosinophil imracellutar adhesion molecule-1 and HLA-DR. J lmmaaot. 1991; 149:2 130-2 136.
17. Mawhorter SD. Kaiura JK, Boom WH. Human eosinophils as antigen presenting cells: relative efficiency for superantigen- and antigen- induced CD4 + T cell proliferation. Immunology. In press.
18. Kita H, Ohnishi T, Okubo Y, Weiler D. Abrams JS, Gleich GJ. Granulocyte/ macrophage colony-stimulating (actor and interleukin 3 release from human peripheral blood eosinoph ils and neutroph iIs. J ?f Med. 1991:174:745-748·
19. Markell EK. Eosinophilia. In: Marke!! EK. Voge M, eds. Medical Pataxtotogy. Philadelphia, Pa : WB SaundetsCo; 1981:300-301.
20. Nutman TB, Otresen EA, Gam A, et al. Eosinophilia in Southeast Asian refugees: evaluation at a referral center. J infect Dis. 1987:155:309-313.
21. Libman MD, MacLean JD, Oyotkus TW. Screening for schistosomiasis, filariasis. and strongyloidiasis among expatriates returning from the oopics. Clin infect Du. IW3-.17-.353-359.
22. Fryatt RJ, Teng J, Harries AD. Siorvanes L, Hall AP. Intestinal helminihiasis in ex- patriares reluming to Britain from the tropics. Trop Geogr Med. 1990:42:119-122.
23. Harries AD M^rs B, Bhattacharrya D. Eosinophilia in Caucasians returning from the tropics. Trans R Soc Trop Mtd Hyg. 1980:80:327-328.
24. Ottesen EA. Filaria! infections. Inferi Dis CIm North Am. 1993:7:619-633.
Eosinophilia Not Associated With Parasitic Infections
Helminth Parasites Associated
Eoslnopnllla-lnducing Parasite Infections That Can Be Acquired In the United States