What makes UpToDate so powerful?

  • over 11000 topics
  • 22 specialties
  • 5,700 physician authors
  • evidence-based recommendations
See more sample topics
Find Print
0 Find synonyms

Find synonyms Find exact match

Approach to the patient with unexplained eosinophilia
Official reprint from UpToDate®
www.uptodate.com ©2017 UpToDate, Inc. and/or its affiliates. All Rights Reserved.
The content on the UpToDate website is not intended nor recommended as a substitute for medical advice, diagnosis, or treatment. Always seek the advice of your own physician or other qualified health care professional regarding any medical questions or conditions. The use of this website is governed by the UpToDate Terms of Use ©2017 UpToDate, Inc.
Approach to the patient with unexplained eosinophilia
View in Chinese
All topics are updated as new evidence becomes available and our peer review process is complete.
Literature review current through: Nov 2017. | This topic last updated: May 03, 2016.

INTRODUCTION — Peripheral blood eosinophilia can be caused by numerous allergic, infectious, and neoplastic disorders, which require a variety of different treatments. A major goal of the initial evaluation is to identify disorders that require specific treatments (eg, parasitic infection, drug hypersensitivity, leukemia, non-hematologic cancer).

The peripheral blood eosinophil count does not accurately predict the risk of organ damage in all patients. Although complications of eosinophilia are more common in patients with higher eosinophil counts (eg, >1500 eosinophils/microL), a patient with mild peripheral blood eosinophilia may have significant organ involvement by eosinophils. Thus, it is also important to consider the clinical status of the patient and to determine whether there is evidence of end-organ involvement.

This topic presents our approach to determining the underlying cause of unexplained peripheral blood eosinophilia. Disorders with eosinophilic involvement of specific organs are presented separately. (See "Eosinophil biology and causes of eosinophilia" and "Causes of pulmonary eosinophilia" and "The significance of urinary eosinophils" and "Eosinophilic meningitis".)

TERMINOLOGY — We use the following terms:

Absolute eosinophil count – The absolute eosinophil count refers to the number of circulating eosinophils in the peripheral blood (in cells/microL). It is determined by multiplying the total white blood cell (WBC) count by the percentage of eosinophils. Some laboratories report this calculated number directly, while others require the clinician to make the calculation.

Eosinophilia – Eosinophilia refers to an absolute eosinophil count in the peripheral blood of ≥500 eosinophils/microL; this is considered abnormal in most laboratories [1,2]. The degree of eosinophilia can also be categorized as mild (500 to 1500 eosinophils/microL), moderate (1500 to 5000 eosinophils/microL), or severe (>5000 eosinophils/microL).

The typical percentage of eosinophils in an individual without eosinophilia is less than 5 percent, but the presence of eosinophilia cannot be determined based on the eosinophil percentage alone, because this percentage is a relative number that varies with the total WBC count and the relative percentages of other WBCs (eg, neutrophils, lymphocytes).

Hypereosinophilia (HE) – Hypereosinophilia (HE) is defined as moderate to severe eosinophilia (ie, ≥1500 eosinophils/microL). End-organ manifestations may be present, but are not required for the HE designation. When the cause of the eosinophilia is unknown and clinical manifestations are absent, patients with hypereosinophilia are considered to have hypereosinophilia of unknown significance (HEUS) [3]. (See "Hypereosinophilic syndromes: Clinical manifestations, pathophysiology, and diagnosis", section on 'Hypereosinophilia of undetermined significance'.)

Hypereosinophilic syndrome (HES) – Hypereosinophilic syndrome (HES) is used to describe a group of heterogeneous clinical syndromes; it does not necessarily imply a primary hematologic or neoplastic disorder. Only the following two features are required [4-6]:

Hypereosinophilia (ie, absolute eosinophil count ≥1500/microL) on at least two occasions

Signs of organ dysfunction attributable to the eosinophilia

Historically, the terms "HES" and "idiopathic HES" have been used to describe eosinophilia ≥1500/microL associated with end organ dysfunction attributable to the eosinophilia in which the cause is unknown. From a practical standpoint, however, it is important to recognize that 1) end organ manifestations of eosinophilia can be identical irrespective of the cause; 2) the etiology of the eosinophilia may not be evident at presentation but may be determined later; and 3) some individuals with moderate to severe eosinophilia remain asymptomatic or develop signs of organ dysfunction many years after the eosinophilia is first noted [3,7]. Furthermore, with the availability of new diagnostic tests and targeted therapies, the ability to identify specific etiologies of HES continues to grow. (See "Hypereosinophilic syndromes: Clinical manifestations, pathophysiology, and diagnosis", section on 'Definitions'.)

In view of the clinical heterogeneity of HE and HES, there have been many attempts to develop a classification system in order to better direct clinical management [4-6,8,9]. For example, some experts subdivide HES into myeloproliferative syndrome-type HES (M-HES), in which clinical features point towards a myeloproliferative disorder; and lymphoid HES (L-HES), in which T cell production of cytokines, such as IL-5, leads to overproduction of eosinophils [2]. However, none of the available classification systems is perfect, and the majority of patients with HES (eg, up to two-thirds) do not fit neatly into a specific category.

From the perspective of patient management, a more important distinction is whether the treatment should be directed at the eosinophilia or at another condition that is stimulating eosinophil production (eg, parasitic infection, drug hypersensitivity, non-hematologic cancer). (See 'Overview of our approach' below.)

Primary (or neoplastic) hypereosinophilic syndrome (HES) – Primary (neoplastic) HES refers to a myeloproliferative neoplasm that produces a predominance of mature eosinophils. (See "Eosinophil biology and causes of eosinophilia", section on 'Neoplastic diseases'.)

PATHOPHYSIOLOGY — Eosinophils are primarily tissue-dwelling cells; they are several hundredfold more abundant in tissues than in blood. Importantly, the degree of peripheral blood eosinophilia does not always accurately predict the risk of organ damage. Thus, organ involvement and end-organ damage cannot be predicted by a high eosinophil count, nor can they be excluded by a low eosinophil count.

Target organs — Common target organs of eosinophils in disease include the skin, lung, and gastrointestinal tract. However, cardiac and nervous system damage can also occur, and can be more concerning and potentially life-threatening.

Eosinophil biology in health and disease is reviewed elsewhere. (See "Eosinophil biology and causes of eosinophilia", section on 'Eosinophil biology'.)

MAJOR CAUSES OF EOSINOPHILIA — Eosinophilia can be caused by a number of conditions (table 1). The degree of eosinophilia is rarely helpful for identifying the cause, except at extremes of eosinophil counts (eg, very mild eosinophilia may be associated with asthma or allergic rhinitis; very severe eosinophilia [ie, ≥20,000 eosinophils/microL] is more likely to be caused by a myeloproliferative neoplasm). The disorders that can cause eosinophilia are best distinguished by the patient's history, clinical presentation, and specific laboratory testing. A more detailed discussion of the causes of eosinophilia is found separately. (See "Eosinophil biology and causes of eosinophilia", section on 'Major causes of eosinophilia'.)

OVERVIEW OF OUR APPROACH — The most important steps in evaluating an individual with eosinophilia are assessing the presence and degree of tissue/organ involvement, which determines the urgency of the evaluation; and determining the etiology, which has major implications for treatment.

Acutely ill patient or extremely high eosinophil count — An acutely ill patient with eosinophilia, or an individual with an extremely high eosinophil count (eg, ≥100,000 eosinophils/microL), requires hospitalization and urgent evaluation for the cause of the eosinophilia; these cases are rare. (See "Hypereosinophilic syndromes: Clinical manifestations, pathophysiology, and diagnosis".)

If the acute illness appears to be due to organ dysfunction (possibly caused by tissue eosinophil infiltration), or if an appropriate alternate therapy is not identified, urgent therapy directed at reducing eosinophilia (eg, high dose glucocorticoids) should be initiated. Ideally, baseline laboratory testing to determine the potential cause of the eosinophilia is sent before initiating urgent therapy, because therapy may obscure the underlying cause. However, urgent therapy should not be delayed while attempting to obtain, or awaiting results of, this testing. (See 'Initial testing' below and "Hypereosinophilic syndromes: Treatment", section on 'Immediate treatment for severe disease'.)

The risk of dissemination of strongyloides in the setting of glucocorticoid therapy necessitates that individuals with potential exposure to strongyloides infection be treated empirically for strongyloides with ivermectin (200 mcg/kg daily for two days). This applies to all individuals with potential exposure, even if the serology is negative, because the reliability of serologic testing is variable. (See "Strongyloidiasis", section on 'Epidemiology' and "Strongyloidiasis", section on 'Treatment'.)

Outpatient with signs of organ involvement — Peripheral blood eosinophilia does not correlate well with the extent of tissue damage except at the extremes of absolute eosinophil counts. Thus, signs of organ dysfunction should be addressed promptly, regardless of the level of blood eosinophilia. Helminth infections are among the most common causes worldwide. In the developed world, allergies and asthma predominate overall, and drug hypersensitivity is common in those with eosinophils counts >1500/microL.

For a patient with signs of organ involvement who is not acutely ill or hospitalized, the pace of the evaluation and the need for specialist referral depends on the specific organ involved and the degree of organ dysfunction. In many cases, initial testing can be done in the outpatient setting. (See 'Laboratory evaluation and other testing' below and 'When to refer' below.)

Incidental finding, healthy individual — Eosinophilia discovered as an incidental finding on a complete blood count in an otherwise healthy individual can be evaluated on an outpatient basis. The initial evaluation should include a complete history, physical examination, and baseline laboratory testing. (See 'Laboratory evaluation and other testing' below.)

If signs of organ involvement are found on the initial evaluation or develop during this evaluation, the patient may need to be seen more urgently by a physician experienced in evaluating eosinophilic disorders and/or admitted to the hospital. (See 'When to refer' below.)

The patient with no signs of organ involvement should undergo periodic monitoring for the development of organ involvement at six month intervals. (See 'Initial testing' below.)

In some cases, eosinophilia resolves without treatment. Potential explanations for resolution of eosinophilia include removal of an offending agent (ie, transient exposure); clearance of an infection; and downregulation of host responses (as has been reported in the setting of chronic helminth infection). The expected time course of resolution of eosinophilia after a transient exposure is unknown, although in the case of drug hypersensitivity and removal of the offending agent, resolution can take many months.

The optimal evaluation of the asymptomatic traveler or immigrant with eosinophilia is uncertain. Up to 50 percent of such patients never have a cause of their eosinophilia identified despite exhaustive evaluation [10].

Children — The likely causes of eosinophilia in children and our approach to the evaluation of children are identical to that in adults with a few exceptions. Asthma and atopic disease remain the most common causes of mild to moderate eosinophilia in children; however, close attention should be paid to the frequency and etiology of infections, since immunodeficiency syndromes typically present in childhood and can be associated with atopic disease and peripheral blood and tissue eosinophilia. (See "Primary humoral immunodeficiencies: An overview" and "Combined immunodeficiencies" and "Approach to the child with recurrent infections".)

Food allergy and eosinophilic esophagitis are also more common causes of eosinophilia in the pediatric age group and can be missed if an appropriate history is not elicited; the only clue to the diagnosis of eosinophilic esophagitis in a child may be occasional vomiting. (See "Clinical manifestations of food allergy: An overview" and "History and physical examination in the patient with possible food allergy" and "Clinical manifestations and diagnosis of eosinophilic esophagitis".)

Since helminth infections require appropriate exposure, some infections are more or less prevalent in children. As an example, visceral larva migrans, which requires ingestion of eggs from soil contaminated with animal feces, is seen almost exclusively in young children; whereas filariasis, which requires repeated exposure to the bites of infected insect vectors, increases with age and is rare in children under four years of age. (See "Toxocariasis: Visceral and ocular larva migrans".)

Some forms of leukemia and lymphoma (eg, pre-B cell acute lymphoblastic leukemia [ALL]) are more common in childhood and can present with asymptomatic eosinophilia, sometimes several years before the malignancy is detected. In contrast, solid tumors are rarely the cause of eosinophilia in children. Although definitive data are lacking, the prevalence of other rare causes of eosinophilia, including FIP1L1/PDGFRA-positive myeloproliferative neoplasms, eosinophilic granulomatosis and polyangiitis, and episodic angioedema and eosinophilia, appear to be comparable in children and adults.

HISTORY — All patients should have a thorough history that addresses symptoms of organ involvement, medical conditions, exposures (medications, foods, over-the-counter remedies, travel, and occupational and recreational exposures), and prior eosinophil counts. Family history may also be helpful in rare cases of familial hematologic syndromes. (See 'Family history' below.)

Symptoms — Symptoms of specific organ system involvement may suggest a potential cause of eosinophilia and/or the consequences of eosinophil-induced tissue damage. We assess the following:

Fever, weight loss, fatigue


Nasal involvement

Wheezing/cough/chest congestion

Gastrointestinal involvement/diarrhea


Nervous system symptoms

Symptoms attributable to lymphadenopathy or hepatosplenomegaly

Symptoms of cardiac dysfunction or myocarditis

We also ask about medical conditions that may be associated with eosinophilia (eg, asthma, rheumatologic conditions, malignancy) and whether there has been a recent change in disease symptoms that could represent disease progression or a new diagnosis.

Exposures — Exposure history should include occupational and recreational activities, medications and supplements, foods, and travel.

Occupational/recreational exposures — Some occupations and/or recreational activities are associated with specific exposure risks. Examples include a risk for strongyloides infection in miners, a risk for ascariasis in slaughterhouse workers, and a risk for schistosomiasis in river rafters.

Medications and over the counter remedies — Eosinophilia can be caused by almost any prescription or nonprescription drug, herbal remedy, or dietary supplement (table 2). Thus, a detailed review of past, current, and over-the-counter medications and remedies should be elicited. The temporal relationship between administration of the medication or herbal remedy may be helpful, but the time course of eosinophilia does not always follow a consistent pattern. Although drug-induced eosinophilia is often accompanied by fever, rash, or other clinical manifestations, signs and symptoms may be absent.

In a prospective cohort study of 824 patients receiving prolonged intravenous antibiotic therapy as outpatients, 25 percent developed eosinophilia (peak absolute eosinophil count ranging from 500 to 8610/mL, median 726/mL) after a median of 15 days of therapy [11]. Patients with eosinophilia had a significantly higher likelihood of rash (15 versus 6 percent; adjusted hazard ratio [HR], 4.16) or renal injury (15 versus 10 percent; adjusted HR, 2.13), compared with patients without eosinophilia. Liver injury was comparable between the two groups. DRESS (drug rash with eosinophilia and systemic symptoms) syndrome or possible DRESS, occurred in seven patients, four of whom were receiving vancomycin.

Medication-associated eosinophilia per se does not mandate cessation of medication administration but rather requires careful consideration of the following questions: a) are there clinical manifestations in association with the eosinophilia that would dictate changing the medication, b) are there alternative medications (eg, antibiotics, seizure meds) that could be substituted, or is the likely medication the preferred or required therapy. It is also important to realize that eosinophilia may not resolve for weeks to months after discontinuation of the offending agent.

Foods — Dietary history should ascertain the ingestion of raw or undercooked meat because incompletely cooked pork containing encysted larvae is a source of trichinellosis. Ingestion of soil or vegetables contaminated by excrement from infected dogs or cats can cause toxocariasis, and undercooked or raw crab or crayfish can transmit paragonimiasis. (See "Trichinellosis" and "Toxocariasis: Visceral and ocular larva migrans" and "Paragonimiasis".)

Travel — A history of residence in or recent travel to a parasite-endemic area may be helpful in suggesting a parasitic etiology of eosinophilia. However, a lack of recent travel should not eliminate parasitic infection as a potential cause of eosinophilia, especially for helminths with a worldwide distribution and/or a long latency period. As an example, foreign military service, even decades earlier, could be a source of strongyloides infection, which occurs via penetration of the skin upon contact with soil or water contaminated by human feces. This history may be omitted unless it is specifically queried. (See "Eosinophil biology and causes of eosinophilia", section on 'Parasites and other infections'.)

Since some helminth parasites have a limited geographic distribution, a detailed travel history is especially important. As an example, Loa loa infection is endemic only in Central and West Africa, whereas gnathostomiasis has a worldwide distribution (table 3). Consequently, a patient presenting with migratory angioedema without a history of travel to Africa should be evaluated for some causes of migratory angioedema (eg, gnathostomiasis), but not for loiasis (eg, such an individual does not require a midday blood smear or filarial serology). (See "Skin lesions in the returning traveler" and "Loiasis (Loa loa infection)".)

In contrast to helminthic parasites, most pathogens responsible for travelers' diarrhea, including bacteria and protozoan parasites (eg, Giardia, Entamoeba), generally do not cause eosinophilia. Exceptions are Dientamoeba fragilis, Isospora belli, and Sarcocystis species, which can cause eosinophilia. These organisms have a worldwide distribution, with geographic concentrations (eg, I. belli in tropical/subtropical areas and India). (See "Dientamoeba fragilis" and "Epidemiology, clinical manifestations, and diagnosis of Cystoisospora (Isospora) infections" and "Sarcocystosis".)

Prior eosinophil counts — New onset of eosinophilia suggests a new diagnosis, although it does not help narrow the diagnosis. Persistent eosinophilia without symptoms is reassuring and suggests that the evaluation can be done less urgently.

Waxing and waning eosinophilia can result from the following:

Use of steroids can temporarily decrease eosinophil count regardless of the cause.

Intercurrent bacterial or viral infections can transiently decrease the eosinophil count regardless of the cause.

Intermittent leakage of fluids from walled-off echinococcal cysts can transiently stimulate blood eosinophilia, although this is rare [12].

Family history — Cases of familial clustering of eosinophilic syndromes have been described (eg, eosinophilic esophagitis), but familial clustering is uncommon in most cases of eosinophilia [13]. Familial hypereosinophilia is rare, but has been described. (See "Hypereosinophilic syndromes: Clinical manifestations, pathophysiology, and diagnosis", section on 'Diagnosis of familial HES'.).

PHYSICAL EXAMINATION — The physical examination should focus on identifying lesions that suggest a possible cause of eosinophilia, and on determining the presence of organ involvement. We specifically assess the skin, eyes, nose, lymph nodes, gastrointestinal, cardiac, respiratory, and neurologic systems, and the presence of splenomegaly.


Whom to test — The initial tests for etiology and organ involvement listed below should be performed in the following patients:

All patients with unexplained eosinophilia of ≥1500 eosinophils/microL on two separate occasions

Patients with persistent eosinophils ≥500/microL and signs or symptoms of organ involvement (eg, rash, wheezing, abnormalities on chest radiography, elevated troponin)

Patients with eosinophilia between 500 and 1500 eosinophils/microL who have a history of travel to a parasite endemic area, symptoms attributable to eosinophilia (with the exception of allergic symptoms), or other clinical findings of concern

Patients not in these categories (eg, asymptomatic, no concerning travel history, mild eosinophilia [absolute eosinophil count <1500/microL]) may be observed. The eosinophil count may be repeated to determine if it is increasing.

An increasing eosinophil count or development of new signs or symptoms should prompt reevaluation.

Initial testing — In the patient groups listed above, we suggest the following initial tests for possible causes of eosinophilia and signs of organ involvement (table 4) (see 'Whom to test' above):

Complete blood count with differential to determine abnormalities of other cell lines. Absolute numbers of other white blood cells must be determined because relative percentages may be lowered in the setting of increased eosinophils.

Review of the peripheral blood smear (for immature white blood cells, dysplastic features that would suggest primary hematologic disorder)

Serum chemistries, creatinine, urinalysis (for evidence of renal insufficiency, and in rare cases adrenal insufficiency)

Serum B12 level (elevated in myeloproliferative neoplasms and autoimmune lymphoproliferative syndrome [ALPS])

Liver function tests (for evidence of hepatic involvement)

Troponin (for evidence of subclinical eosinophil-associated myocarditis); those with elevated troponin should have electrocardiography and echocardiography.

Chest radiography (for evidence of pulmonary involvement)

Serologic testing for Strongyloides. Positive serology in an untreated patient is presumptive evidence of infection; however, serology remains positive after treatment, so it is not useful for documenting cure or reinfection.

Flow cytometry for lymphocyte subsets (may show clonality in lymphocytic hypereosinophilic syndrome, or lymphoid malignancy (table 5); selective deficiencies in immunodeficiency syndromes (table 6)).

Evidence of organ dysfunction should prompt initiation of therapy in order to reduce the risk of permanent and/or life-threatening organ damage in most patients. This is discussed in detail separately. (See "Hypereosinophilic syndromes: Treatment" and "Eosinophil biology and causes of eosinophilia", section on 'Disorders with eosinophilic involvement of specific organs'.)

For those without signs of organ involvement, we periodically repeat this testing, along with complete blood count and differential (eg, at six month intervals). Any temporal relationships with exposures or illnesses should be noted. Of note, intermittent administration of steroids can produce a false appearance of episodic eosinophilia. (See 'Prior eosinophil counts' above.)

Additional testing in selected patients — The following may be also appropriate in selected patients, depending on individual patient characteristics and the results of the initial evaluation (table 4):

Serologic testing for strongyloides should be performed in patients with a history of potential exposure, and empiric therapy with ivermectin should be considered if reliable serologic testing is not available. (See "Strongyloidiasis".)

Patients with risk factors for tuberculosis should also be screened for latent tuberculosis prior to glucocorticoid treatment due to the risk of reactivation. (See "Latent tuberculosis infection in children".)

Individuals with an appropriate exposure history should have testing for additional parasites and other infections. (See 'Testing for other parasites and infections' below.)

Patients with recurrent infections or clinical features suggestive of autoimmune disease (eg, hemolytic anemia, Hashimoto's thyroiditis, inflammatory arthritis) or immunodeficiency should have measurement of serum immunoglobulins and immunoglobulin isotypes, and screening for ALPS (eg, enumeration of CD4 and CD8 double-negative T cells) and other immune deficiencies depending on the clinical presentation. (See "Overview of IgG4-related disease" and "Primary humoral immunodeficiencies: An overview" and "Severe combined immunodeficiency (SCID): An overview" and "Autoimmune lymphoproliferative syndrome (ALPS): Clinical features and diagnosis".)

Individuals with features of myeloproliferative disease (eg, splenomegaly, anemia or thrombocytopenia, dysplastic eosinophils); skin lesions consistent with urticaria pigmentosa; or a history of anaphylaxis should have measurement of random serum vitamin B12 and tryptase levels. (See "Mastocytosis (cutaneous and systemic): Epidemiology, pathogenesis, and clinical manifestations" and "Mastocytosis (cutaneous and systemic): Evaluation and diagnosis in adults".)

Individuals with signs of adrenal dysfunction (eg, malaise, weakness, weight loss, orthostatic hypotension, skin hyperpigmentation) and those at risk for adrenal hemorrhage or necrosis should have early morning cortisol levels and/or dynamic cortisol testing. (See "Clinical manifestations of adrenal insufficiency in adults" and "Diagnosis of adrenal insufficiency in adults".)

In patients with rhinosinusitis, asthma, and/or vasculitis, anti-neutrophil cytoplasmic antibody (ANCA) testing can be done to evaluate for eosinophilic granulomatosis with polyangiitis (EGPA), although the sensitivity of this testing is low. (See "Clinical features and diagnosis of eosinophilic granulomatosis with polyangiitis (Churg-Strauss)".)

Those with suspected systemic disorders, such as lymphoma, should have evaluation of peripheral blood lymphocyte subsets and clonality; and computed tomography (CT) scans of the chest, abdomen, and pelvis; with biopsy of an appropriate lymph node (eg, rapidly increased in size, persistent for several weeks, unusual location, associated with fever, sweats, weight loss). (See "Evaluation of peripheral lymphadenopathy in adults" and "Peripheral lymphadenopathy in children: Evaluation and diagnostic approach".)

Those with suspected vasculitis (eg, eosinophilic granulomatosis with polyangiitis [EGPA]) should have chest CT and other studies appropriate to their symptoms, and biopsy of involved tissue if feasible. (See "Overview of and approach to the vasculitides in adults" and "Vasculitis in children: Evaluation".)

Individuals with suspected hematologic malignancy should have bone marrow evaluation with cytogenetic and molecular testing. In rare cases (eg, profound leukocytosis, bone marrow cannot be obtained), peripheral blood rather than bone marrow may be used for molecular testing to diagnose primary hematologic disorders, as discussed below. (See 'Bone marrow examination' below.)

Testing for other parasites and infections — In addition to serologies for Strongyloides species, which should be performed in all patients, additional testing for parasitic or other infections may be indicated depending on the exposure history (table 4). (See 'Exposures' above and "Strongyloidiasis", section on 'Whom to test'.)

Toxocara serology – Toxocara canis and cati are endemic worldwide and can be ingested with soil or food contaminated by dog or cat feces. This is typically a concern with children who may ingest contaminated soil. Visceral larva migrans is almost exclusively a disease of children, and children with toxocara are acutely ill. Toxocara serology is not useful in adults, due to the high rate of seropositivity in the adult population. (See "Toxocariasis: Visceral and ocular larva migrans".)

Trichinella serology – Trichinella species have been reported worldwide; human infections are seen most commonly in China, Thailand, Mexico, Argentina, Bolivia, and parts of Eastern and Central Europe. Infection results from ingestion of undercooked meat, especially pork. (See "Trichinellosis".)

Other serologies – Depending on the travel history, serology for other helminths may be appropriate (eg, schistosomiasis, filariasis). (See "Epidemiology, pathogenesis, and clinical manifestations of schistosomiasis" and "Epidemiology, pathogenesis, and clinical manifestations of lymphatic filariasis".)

Stool studies – Individuals with a travel or exposure history that suggests a possible helminthic infection should also have stool testing for ova and parasites. The parasite eggs and/or larvae that may be identified in stool include Strongyloides, hookworm (ie, Ancylostoma duodenale, Necator americanus), and flukes (ie, Fasciola, Clonorchis, Opisthorchis species). These generally do not cause diarrhea. (See "Hookworm infection" and "Intestinal flukes" and "Liver flukes: Fascioliasis" and "Liver flukes: Clonorchis, Opisthorchis, and Metorchis".)

Stool studies should also be performed in individuals with watery stools or diarrhea, to look for diagnostic trophozoites of D. fragilis and oocysts of I. belli and Sarcocystis species, all of which can cause peripheral blood eosinophilia; Sarcocystis spp. can also cause myositis [14].

Of note, while stool studies may be helpful in some cases and should be performed, they are neither sensitive nor specific, and results may be highly variable depending on the laboratory. Repeat testing may increase sensitivity.

Several helminths cannot be identified reliably by stool studies, including the following:

Strongyloides stercoralis




Toxocara species

Testing for these organisms is typically done by serology; in some cases organisms may be identified on tissue biopsy.

Viral serologies – Depending on exposures, HIV serology and HTLV-I and II screening may be indicated. (See "Screening and diagnostic testing for HIV infection".)

Testing of body fluids – Urine sediment can be tested for Schistosoma haematobium if urinary symptoms are present and the history of exposure is consistent. Cerebrospinal fluid can be tested for eosinophils and serology if eosinophilic meningitis is suspected; infectious causes include helminths and coccidiomycosis. (See "Diagnosis of schistosomiasis" and "Eosinophilic meningitis" and "Coccidioidal meningitis".)

TRIAL OF DRUG DISCONTINUATION — If a drug-induced cause of eosinophilia is suspected based on an implicated medication and lack of other explanation, a trial of drug discontinuation (and substitution of an alternative if necessary) may be helpful. The most likely drug(s) can be discontinued first if the patient is not acutely ill; if the cause remains undetermined or the patient becomes acutely ill, discontinuation of all non-essential medications is appropriate.

Blood eosinophilia by itself need not necessitate cessation of drug therapy, but should lead to an evaluation of organs likely to be involved in eosinophil-associated drug reactions (table 2). (See 'Initial testing' above.)

If organ dysfunction develops, cessation of the inciting drug is necessary. Of note, eosinophilia may require months to resolve following drug discontinuation [13]. (See "Drug allergy: Classification and clinical features" and "An approach to the patient with drug allergy".)


Specialist referral — Referral to a specialist is appropriate for clinical evaluation and biopsy of a potentially affected tissue. As examples, pulmonary evaluation is appropriate for individuals with evidence of lung involvement (eg, symptoms, radiographic findings); allergy testing is appropriate for those with suspected eosinophilic esophagitis. (See "Causes of pulmonary eosinophilia" and "Allergy testing in eosinophilic esophagitis".)

Referral to a clinician who specializes in eosinophilia is appropriate if a thorough evaluation has been conducted and the cause of persistent eosinophilia is not found. This may be an infectious diseases expert, hematologist, or allergist, depending on the institution. Individuals with expertise in tropical diseases should be consulted rather than attempting empiric antihelminthic therapy, which may delay appropriate diagnostic testing and lead to complications of a therapy that was not indicated or that was used incorrectly.  

Bone marrow examination — Hematologic evaluation with bone marrow examination (aspiration and biopsy) is appropriate for any individual with a potential primary hematologic cause of eosinophilia. Examples include the following:

Acutely ill patient or eosinophil count >100,000/microL without an obvious causative condition. (See 'Acutely ill patient or extremely high eosinophil count' above.)

Eosinophil count ≥1500/microL or signs of organ involvement by eosinophils without an obvious cause identified from initial testing. (See 'Initial testing' above.)

Abnormal features on the peripheral blood smear that suggest an underlying hematologic condition (eg, immature or dysplastic white blood cells, thrombocytopenia, unexplained anemia).

Bone marrow morphology should be evaluated for abnormalities of eosinophilic precursors (picture 1), with special stains for reticulin and/or mast cells if indicated. Detection of mast cells in the bone marrow is discussed separately. (See "Mastocytosis (cutaneous and systemic): Evaluation and diagnosis in adults", section on 'Bone marrow examination'.)

Bone marrow should undergo cytogenetic and/or specific molecular testing for hematologic neoplasms associated with eosinophilia, including the following:

Testing for chronic myeloid leukemia using cytogenetics, fluorescence in situ hybridization (FISH) for the t(9:22) translocation, or reverse transcriptase polymerase chain reaction (RT-PCR) for the BCR-ABL fusion product. (See "Clinical manifestations and diagnosis of chronic myeloid leukemia".)

Testing for abnormalities of the platelet derived growth factor receptor (PDGFR) alpha or beta, fibroblast growth factor receptor 1 (FGFR1), and Fip1-like 1 (FIP1L1) genes. Testing for PDGFRA-FIP1L1 is done using FISH for the CHIC2 locus (which is normally located between PDGFRA and FIP1L1 and is therefore absent in individuals with the FIP1L1/PDGFRA fusion), and/or RT-PCR for the fusion transcript (commercially available in the United States through Quest Laboratories and Mayo Clinic).

Multicolor flow cytometry for evidence of clonal lymphocytic proliferation, including analyses for (at a minimum) CD3, CD4, and CD8 (to detect CD3 negative, CD4 positive subsets). T cell receptor rearrangement studies on peripheral blood and/or bone marrow (available at specialty academic centers) can be done to diagnose lymphocytic variant HES.

Depending upon the clinical presentation and the urgency for initiating therapy, investigations may be performed sequentially, starting with those that appear more likely to be informative. For example, if a patient presents with splenomegaly, mucosal ulcers of the oral cavity, and signs of heart failure, testing for the FIP1L1/PDGFRA fusion indicative of myeloproliferative hypereosinophilic syndrome (HES) is indicated. Ideally, diagnostic testing for myeloproliferative or lymphocytic etiologies for HES should be performed prior to initiating therapies that might suppress clonal cell populations.


Eosinophils are primarily tissue-dwelling cells. Importantly, the degree of peripheral blood eosinophilia does not always accurately predict the risk of organ damage. (See 'Terminology' above and 'Pathophysiology' above.)

It is important to consider the possibility of a variety of causes of eosinophilia during the initial evaluation because the timing and degree of eosinophilia can be quite variable for many conditions. Major causes of eosinophilia include neoplastic disorders, parasitic and other infections, allergic disorders, adrenal insufficiency, and rheumatologic conditions (table 1 and table 2). (See "Eosinophil biology and causes of eosinophilia".)

The most important steps in evaluating an individual with eosinophilia are assessing the presence and degree of tissue/organ involvement, which determines the urgency of the evaluation; and determining the etiology, which has major implications for treatment. (See 'Overview of our approach' above.)

An acutely ill patient with eosinophilia, or an individual with an extremely high eosinophil count (eg, ≥100,000 eosinophils/microL), requires hospitalization and urgent evaluation for the cause of the eosinophilia. (See 'Acutely ill patient or extremely high eosinophil count' above.)

For a patient with signs of organ involvement who is not acutely ill or hospitalized, the pace of the evaluation and the need for specialist referral depends on the specific organ involved and the degree of organ dysfunction. (See 'Outpatient with signs of organ involvement' above and 'When to refer' above.)

Eosinophilia discovered as an incidental finding on a complete blood count in an otherwise healthy individual can be evaluated on an outpatient basis. (See 'Incidental finding, healthy individual' above.)

The likely causes of eosinophilia in children with eosinophilia are identical to adults with a few exceptions. Children may be more likely to have asthma and atopic disease; immunodeficiency syndromes; food allergy and/or eosinophilic esophagitis; and some forms of leukemia. Visceral larva migrans is seen almost exclusively in young children, whereas filariasis increases with age and is rare in children under four years. (See 'Children' above.)

All patients should have a thorough history that addresses symptoms of organ involvement, medical conditions, exposures (medications (table 2), over-the-counter remedies, travel, foods, and occupational and recreational exposures), and prior eosinophil counts. The physical examination should focus on the skin (table 3), eyes, nose, lymph nodes, gastrointestinal, cardiac, respiratory, and neurologic systems, and the presence of splenomegaly. (See 'History' above and 'Physical examination' above.)

We suggest a series of initial laboratory tests for all patients with unexplained eosinophilia of ≥1500 eosinophils/microL on two separate occasions, any patient with signs of organ involvement, or any individual with a history of travel to a parasite endemic area (table 4). (See 'Initial testing' above.)

Additional testing or a trial of drug discontinuation may be warranted in selected patients (eg, additional parasite serologies depending on exposure history; serum immunoglobulin levels for those with recurrent infections or signs of autoimmune disease; serum tryptase levels in those with findings of myeloproliferative disorders, skin lesions of urticaria pigmentosa, or a history of anaphylaxis) (table 4). Referral to a clinician who specializes in eosinophilia is appropriate if a thorough evaluation has been conducted and the cause of persistent eosinophilia is not found. (See 'Additional testing in selected patients' above and 'Testing for other parasites and infections' above and 'Trial of drug discontinuation' above and 'When to refer' above.)

Bone marrow aspiration and biopsy is appropriate for individuals with findings that suggest a potential primary hematologic cause of eosinophilia (eg, hypereosinophilia, organ involvement, dysplastic or immature white blood cells on the peripheral blood smear). If bone marrow testing is done, the sample should be evaluated for abnormalities of eosinophilic precursors (picture 1), with special stains for reticulin and/or mast cells if indicated, along with cytogenetic and/or specific molecular testing for hematologic neoplasms associated with eosinophilia. (See 'Bone marrow examination' above.)

Use of UpToDate is subject to the  Subscription and License Agreement.


  1. Tefferi A. Blood eosinophilia: a new paradigm in disease classification, diagnosis, and treatment. Mayo Clin Proc 2005; 80:75.
  2. Roufosse F, Weller PF. Practical approach to the patient with hypereosinophilia. J Allergy Clin Immunol 2010; 126:39.
  3. Chen YY, Khoury P, Ware JM, et al. Marked and persistent eosinophilia in the absence of clinical manifestations. J Allergy Clin Immunol 2014; 133:1195.
  4. Klion AD, Bochner BS, Gleich GJ, et al. Approaches to the treatment of hypereosinophilic syndromes: a workshop summary report. J Allergy Clin Immunol 2006; 117:1292.
  5. Simon HU, Rothenberg ME, Bochner BS, et al. Refining the definition of hypereosinophilic syndrome. J Allergy Clin Immunol 2010; 126:45.
  6. Valent P, Klion AD, Rosenwasser LJ, et al. ICON: Eosinophil Disorders. World Allergy Organ J 2012; 5:174.
  7. Helbig G, Hus M, Francuz T, et al. Characteristics and clinical outcome of patients with hypereosinophilia of undetermined significance. Med Oncol 2014; 31:815.
  8. Valent P, Klion AD, Horny HP, et al. Contemporary consensus proposal on criteria and classification of eosinophilic disorders and related syndromes. J Allergy Clin Immunol 2012; 130:607.
  9. Gotlib J. World Health Organization-defined eosinophilic disorders: 2012 update on diagnosis, risk stratification, and management. Am J Hematol 2012; 87:903.
  10. Libman MD, MacLean JD, Gyorkos TW. Screening for schistosomiasis, filariasis, and strongyloidiasis among expatriates returning from the tropics. Clin Infect Dis 1993; 17:353.
  11. Blumenthal KG, Youngster I, Rabideau DJ, et al. Peripheral blood eosinophilia and hypersensitivity reactions among patients receiving outpatient parenteral antibiotics. J Allergy Clin Immunol 2015; 136:1288.
  12. Georgiou S, Maroulis J, Monastirli A, et al. Anaphylactic shock as the only clinical manifestation of hepatic hydatid disease. Int J Dermatol 2005; 44:233.
  13. Klion A. Hypereosinophilic syndrome: current approach to diagnosis and treatment. Annu Rev Med 2009; 60:293.
  14. Esposito DH, Stich A, Epelboin L, et al. Acute muscular sarcocystosis: an international investigation among ill travelers returning from Tioman Island, Malaysia, 2011-2012. Clin Infect Dis 2014; 59:1401.
Topic 5691 Version 20.0

Topic Outline



All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.