Eosinophilic meningitis: A case series report and review of the literature

Military Medicine, Oct 2003 by Hughes, Pamela A, Magnet, Andrew D, Fishbain, Joel T

Prompted by a case of eosinophilic meningitis (EM), a review of the literature was performed to evaluate the strength of the diagnoses associated with EM and compares these results with our hospital's experience. Articles were critically reviewed for supporting evidence, method of diagnosis, and established standards for specific diagnosis. EM has been defined as >10 eosinophils per mm^sup 3^ or >10% eosinophils of total cell count. Sixty-two cases of EM were found at our institution and reviewed. The results of this case series review concur with those found in the literature. It also suggests the importance of considering infectious and noninfectious etiologies when faced with eosinophils in the cerebrospinal fluid. This review and case study analysis provides the clinician with a critically established set of differential diagnoses and a concise definition of EM that may assist the physician in the evaluation of patients presenting with eosinophils in the cerebrospinal fluid.

Introduction

Eosinophilic meningitis (EM) is a rare finding on spinal fluid analysis. EM is defined as the presence of > or =10 eosinophils per mm^sup 3^ of cerebrospinal fluid (CSF) or pleocytosis with > or =10% eosinophils.1 Finding EM in the clinical setting suggests a variety of diagnoses. Eosinophilic pleocytosis of spinal fluid has been documented in parasitic infections, coccidioidomycosis, patients with ventriculoperitoneal (VP) shunts, malignancies, and some medications. Although many diseases have been associated with EM, systematic and critical review of these findings has been lacking in the literature, and the clinician will likely find difficulty focusing specific diagnostic strategies and therapy based on the available literature.

A patient with a presumed case of Angiostrongylus cantonensis infection was admitted to our institution with EM. In an attempt to differentiate the etiology of his process, a literature review revealed a wide range of diagnoses attributable to EM. A. cantonensis infection occurs following the ingestion of some species of raw snails and may be difficult to prove histologically or serologically. This infection can also occur following the ingestion of contaminated fresh vegetables.2 Based on this case and the difficulties in proving A. cantonensis infection, we undertook this study to evaluate the differential diagnosis of EM on spinal fluid analysis, the validity or strength of the supporting data for each diagnosis, and how our institution's experience compares with the differential diagnosis found in the literature.

Methods

The Human Use Committee at Tripler Army Medical Center approved the study protocol with exemption status. Investigators adhered to the policies for protection of human subjects as prescribed in 45 CFR 46.

The hospital's computer laboratory database (Composite Health Care System) was searched for all patients from 1989 through January 2000. Cases were identified using CSF results and had to include a differential white blood cell count analysis. EM was defined using previously published standards as > or =10 eosinophils per mm^sup 3^ or > or =10% eosinophils of the total cell count.1 The laboratory at our institution analyzes the presence of eosinophils through staining preparations considered to be the most accurate method for identifying eosinophils in the CSF.3 Patients meeting this definition were included in the study. Discharge diagnoses were obtained from the medical records or computerized discharge summaries. Patients were excluded from analysis if no discharge information was available.

A MEDLINE (National Library of Medicine, Bethesda, Maryland) search was performed through November, 2002 using the keywords eosinophilic meningitis. All English language articles were analyzed to produce the literature database of diagnoses. We critically reviewed the articles for specific diagnoses associated with EM. Supporting evidence and methods of diagnosis were tabulated for each diagnosis. Cases were stratified into three categories; well-defined, possible, and unlikely/circumstantial. Well-defined cases were identified if there was significant pathology (tissue confirmation, parasite identification), well-established criteria for that particular diagnosis (paired serology, positive cultures), or multiple reported cases in an endemic location (particularly A. cantonensis). Possible cases were defined by the presence of a single case report in the literature, but the case possessed well-established criteria for a diagnosis, or a single serology was reported consistent with recent infection. Unlikely/circumstantial cases were defined as a nonclassic clinical presentation for that disease process, the data was incomplete, other etiologies were possible and could not be excluded by the clinical data provided, or the reviewing author of foreign language articles considered the data circumstantial.3

Case Presentation

A 21-year-old active duty Marine, stationed in Hawaii, ingested a raw snail while on a field exercise. One week after ingestion, he presented with intermittent nausea, vomiting, loose watery stools, frontal headaches, anorexia, rhinorrhea, congestion, and subjective fevers with chills and sweats. He denied abdominal pain and neck stiffness. He was admitted with a diagnosis of gastroenteritis with volume depletion. He was treated with intravenous fluid hydration and ibuprofen with significant improvement in his symptoms. Two weeks later, the patient presented with mental status changes.

On this presentation, physical examination revealed an alert, but confused and occasionally combative man. He was afebrile with normal vital signs. Other than mild cervical muscle tenderness and pain on full neck flexion and extension, his examination was unremarkable. Fourteen hours after admission, a repeat examination revealed left cranial nerve VI and VII palsies that persisted throughout his hospital course.