Eosinophilic Meningitis After Eating Freshwater Snails

Clinical Presentation & History

VT, age 26, presented with fever, reduced level of consciousness, and a first generalized tonic-clonic seizure. VT’s parent witnessed the seizure and described it as lasting 30 seconds with uprolling of eyeballs and urinary incontinence. A day prior to presentation, VT had vomiting and a severe throbbing headache. VT’s last known alcohol consumption was the day prior, and family reported VT consumed alcohol heavily for the past 6 years. VT worked as a manual laborer at a palm oil plantation. Other past medical and surgical history was unremarkable.

Physical Examination & Laboratory Testing

On examination, VT’s temperature was 40.2 °C, blood pressure 141/106 mm Hg, pulse rate 98/minute, respiratory rate 18/minute, and oxygen saturation of 100%. VT appeared drowsy and had a Glasgow Coma Scale (GCS) score of 10 (eye opening, 3; verbal response, 2; motor response, 5). VT was not oriented to time, place, or person and was unable to follow simple commands. Kernig and Brudzinski signs were negative and their pupils were equal and reactive bilaterally. VT showed no evidence of head injury and had no skin rashes. Extremity mobility, tone, and reflexes were normal, including flexor plantar responses with negative clonus. Complete neurologic examination with gait and balance were difficult to perform owing to VT’s reduced level of consciousness. Ocular examination revealed ocular bobbing (See Video 1). Other cranial nerve functions were grossly intact.

Initial laboratory tests (Table 1) revealed elevated inflammatory markers with a white blood cell count of 20,000/ mcL composed of neutrophils (86%), eosinophils (0.32%), and absolute eosinophil count of 60/mcL. Electrolyte imbalances were considered secondary to vomiting, poor oral intake, and chronic alcoholism. Liver function panels were consistent with alcoholic hepatitis, including elevated aspartate aminotransferase (119 U/L) compared with alanine aminotransferase (27 U/L). Serum creatine kinase was 2,289 U/L, considered secondary to seizure. Screening for hepatitis B and C, HIV, and syphilis were negative.

Initial Management

VT was admitted to the hospital and empirically treated with intravenous thiamine for Wernicke encephalopathy with a differential of acute meningoencephalitis. Ceftriaxone and acyclovir treatment was initiated for a possible acute central nervous system (CNS) infection. Levetiracetam was given for seizure control. Treatment included intravenous fluids and electrolytes, including correction of sodium over a 5-day period, which resolved the observed electrolyte and liver enzyme abnormalities.

Brain Imaging and Cerebrospinal Fluid Analysis

Head CT with and without contrast were normal (Figure 1) as were brain and spinal MRIs (Figure 2). A diagnostic lumbar puncture was performed and a normal opening pressure of 9 cm H₂O observed. Analysis of cerebrospinal fluid (CSF) showed pleocytosis of 103 cells, with differentials of 79% lymphocytes, 20% eosinophils and 1% of polymorphic cells (Table 2). CSF glucose was 2 mmol/L compared with serum glucose level of 4.9 mmol/L for a CSF-to-serum glucose ratio of 0.40. Tests for viral, fungal, and bacterial infections were negative. Peripheral blood smear did not reveal any peripheral eosinophilia.

Diagnosis

Because CSF analysis raised concern for a possible parasitic infection of the CNS, a subsequent collaborative history was taken from VT’s parent and revealed that VT habitually consumed exotic food, including freshwater snails, raw fishes, and partially cooked monitor lizards. Stool testing for ova and cysts was negative, abdominal CT revealed no parasitic infestations, and no parasites were found in the CSF. Serum antibody testing against the 31 kD antigen of Angiostrongylus cantonensis using the western blot technique were considered but could not be obtained because these are unavailable in Malaysia and the COVID-19 pandemic precluded sending samples to nearby countries with those capabilities. Consequences of hyponatremia overcorrection could cause some of the clinical symptoms observed; however, this is unlikely because the electrolyte balance was restored over days, not hours, and symptom improvement rather than worsening occurred as the electrolyte balance was restored. Additionally, MRI findings on day 5 of admission were not consistent with sodium overcorrection. Serology for neurocysticercosis, taenia, toxoplasmosis, and toxocariasis was negative (Table 1). Review of CSF cytology, peripheral blood smear, brain and abdomen imaging, and the absence of constitutional symptoms excluded several other possible causes of eosinophilic meningitis (eg, vasculitis or lymphoma). Although clinical findings and dietary history supported a diagnosis of angiostrongyliasis, it was not possible to definitively prove this diagnosis because of the limited serologic testing for Angiostrongylus cantonensis. In the absence of such serology, laboratory-based diagnosis of angiostrongyliasis remains challenging.

Treatment

Intravenous antibiotic and antiviral medications were discontinued on day 5 of admission after blood, sputum, urine, and CSF cultures were reported as negative for viral and bacterial infection (Tables 1 and 2). Because hematologic causes had been ruled out, treatment with oral prednisolone 1 mg/ kg/day (60 mg/day) was initiated, and VT’s symptoms gradually improved. VT became more alert and had no more seizures and markedly reduced ocular bobbing (See Video 2). VT’s detailed ophthalmologic examination was normal. Results of repeated lumbar puncture and CSF analysis after 2 weeks of steroid treatment (Table 2) were normal. VT was discharged 15 days after admission when inflammatory markers and CSF results normalized (Tables 1 and 2).

Follow-up Care

At discharge, VT was able to ambulate independently but still had residual ocular bobbing for which outpatient follow-up continued. Within 1 month of hospital discharge and steroid taper, VT returned to baseline premorbid function with resolved ocular bobbing (See Video 3) and seizure freedom. VT ceased consuming alcohol and exotic food as per advice. Levetiracetam was stopped and results of repeated lumbar puncture and CSF analysis at day 60 after presentation were completely normal (Table 2).

We notified the public health department and gathered specimens of the freshwater snails routinely consumed by the people in VT’s neighborhood (Figure 3). Unfortunately, testing to confirm Angiostrongylus cantonensis was not possible because that service was temporarily suspended owing to the COVID-19 pandemic. Nevertheless, preventive measures were taken by advising the public not to consume raw or partially cooked snails, monitor lizards, or freshwater fishes.

Discussion

Eosinophilic meningitis is a rare CNS condition characterized by meningeal inflammation with at least 10% eosinophils in the total CSF white blood cell count.¹ Although there are many causes of eosinophilic meningitis, including infectious, inflammatory, drug-induced and neoplastic processes (Table 3),² the most reported are Angiostrongylus cantonensi and Gnathostoma spinigerum—both related to human dietary habits.³ Humans become infested when they ingest raw or partially cooked snails, monitor lizards, freshwater fishes, frogs, or unwashed produce, such as lettuce.⁴ The ingested larvae penetrate the gut wall and enter the bloodstream, allowing ova and larvae to reach the brain and meninges where they usually die. The host response to the larvae in the CNS produces a spectrum of clinical manifestations, which can be self-limiting or life threatening.

Symptoms typically occur 1 week to 1 month after exposure and encompass mild disease to severe forms of meningoencephalitis.⁵ The most common presenting features are headache, neck stiffness, and fever.⁶ In the case presented, symptoms included high grade fever, vomiting, altered level of consciousness, and recent seizure exactly 1 week after the last consumption of a food that is a typical carrier of Angiostrongylus cantonensis. A thorough dietary history and focused neurologic examination with CSF findings of eosinophilia formed the basis for a clinical diagnosis of eosinophilic meningitis.⁷ Determining the underlying etiology, however, was challenging because serologic testing for Angiostrongylus cantonensis is available only in countries where it is endemic. Adding to the complexity of diagnosis, the larvae do not produce focal lesions on brain imaging and are rarely detected in the CSF.²

Owing to the rarity of this entity and lack of knowledge regarding the pathophysiology of the disease, no specific treatment guidelines are available for management of eosinophilic meningitis.^3,4 Most cases have been reported from Southeast Asia and Pacific regions.^6,8 To our knowledge, this is the first case of eosinophilic meningitis in Malaysia successfully treated with steroids and supportive treatment.

The use of anthelmintic treatment with or without corticosteroids remains controversial.^9,10 We adopted a treatment proposed by Chotmongkol et al,³ adding a 2-week course of oral prednisolone, which helps to relieve the headache and appears to shorten the illness duration. In another study, Chotmongkol et al also demonstrated there is no significant difference in terms of clinical outcomes when patients receive prednisolone alone or in combination with anthelmintic treatment.¹¹ Prociv et al postulated anthelmintic treatment can cause more damage by increasing the inflammatory response by inducing massive parasite death in the CNS.¹² Our patient fulfilled the clinical and laboratory criteria for eosinophilic meningitis and demonstrated remarkable clinical improvement after receiving corticosteroids alone.

It is also imperative for clinicians to be aware of noninfectious and nonparasitic etiologies of eosinophilic meningitis, particularly hematologic disorders (eg, hypereosinophilic syndromes, Hodgkin lymphoma, nonHodgkin lymphoma, acute lymphocytic leukaemia, disseminated glioblastoma, paraneoplastic manifestation of solid tumors, or leptomeningeal metastasis from other solid tumors).² Neoplastic disorders carry different pathogenesis and need to be investigated and managed accordingly without delay. Although rare, eosinophilic infiltration of CSF should prompt clinicians to look for secondary causes as early appropriate treatment can be lifesaving.

Conclusion

The diagnostic challenges and difficulty proving Angiostrongylus cantonensis as the causative agent in this case are acknowledged. This particular individual had a clear dietary history, neurologic findings, and presence of CSF eosinophilia without antigen testing or isolation of parasites in the CSF. Other potential causes were ruled out and VT showed remarkable improvement clinically and biochemically after a tapering course of steroids. Serial lumbar punctures performed during and after treatment showed completely normalized CSF with no further relapse.

Foodborne parasitic eosinophilic meningitis is an emerging disease that clinicians need to be aware of, especially given the diagnostic challenges. CSF cell differentiation analysis must be performed on anyone with history of possible exposure and clinical features of meningitis. Recognizing the triad of exotic food consumption, clinical features of meningitis, and CSF eosinophilia and ensuring a vigorous search for underlying etiology is essential for early and appropriate treatment and, hopefully, better prognosis. Increased awareness in areas where consumption of exotic food is common may result in earlier diagnoses and improve understanding. It is also important to notify such cases to the public health department for appropriate preventive measures to be taken to increase public awareness on possible risk of exposure.

Acknowledgments

We thank Dr Humberto Saavedra for his valuable comments and suggestions.