Eosinophilic Meningitis in a Returned Traveler From Santo Domingo: Case Report and Review

Sebastiano Leone MD, Michele De Marco MD, Piero Ghirga MD, Emanuele Nicastri MD, Mario Esposito MD, Pasquale Narciso MD
DOI: http://dx.doi.org/10.1111/j.1708-8305.2007.00152.x 407-410 First published online: 1 November 2007


We describe one case of eosinophilic meningitis (EM) in a traveler returning from Santo Domingo, presumably caused by Angiostrongylus cantonensis. Treatment with mebendazole and steroids was effective. The presence of persistent headache, fever, and eosinophilia in travelers who return from developing countries should alert clinicians to the possibility of EM.

Case report

A 30‐year‐old man coming from an emergency department of another hospital with diagnosis of meningitis was admitted to our infectious diseases ward. The patient showed a 10‐day‐long history of progressive headache, fever (37.6°C), and vomiting. He was previously healthy and had spent the past 2 years in Santo Domingo. Cerebrospinal fluid (CSF) showed 500 cells/μL, a CSF glucose level of 0.46 g/dL, and a CSF protein level of 129 mg/dL. His white blood cell count was 8,850/μL, with 14% eosinophils. Results of serum electrolyte, blood urea nitrogen, glucose, and liver function tests were normal. On examination, the man was tachycardic, pale, and feverish and had headache, neck pain, and generalized paresthesias. He was alert, collaborating, and had nuchal rigidity, but he did not present accentuated deep tendon reflexes in both legs, nor motor weakness. His sounds were normal and the abdomen showed no hepatosplenomegaly. Chest radiograph findings were normal, and computed tomography (CT) scan of the brain was unremarkable. We started empirical treatment with ceftriaxone, ampicillin, and acyclovir. Weil‐Felix test, Widal–Wright and Listeria monocytogenes serodiagnosis, and treponema pallidum hemagglutination assay were negative. Tests for the detection of serum antibodies to Borrelia burgdorferi, Rickettsia conorii, Dengue virus, cytomegalovirus, herpesvirus simplex, varicella‐zoster virus, and other neurotropic strains were negative. The blood smear for plasmodium species was also negative. Microscopic examination of stool did not reveal the presence of parasitic eggs. After 4 days, acyclovir was interrupted, and because persistent headache remained, a short course (4 d) of steroid therapy was associated with antibiotics, resulting in an improvement of symptomatology. For the increased blood eosinophilia and persistent neurological symptoms, we performed a second lumbar puncture. CSF showed 600 cells/μL, with eosinophilic pleocytosis, a CSF glucose level of 0.36 g/dL, and a CSF protein level of 136 mg/dL (Figure 1). The results of microscopic examination of the CSF were negative for the search of bacteria and fungi. Polymerase chain reaction tests for detection of Mycobacterium tuberculosis and L monocytogenes were performed, but these were also negative. Malignant cells and parasitic larvae were not identified in the CSF specimen. The serologic test for Entamoeba histolytica, Taenia solium, Trichinella spp., Schistosoma spp., Coccidioides immitis, and Blastomyces dermatitidis was also negative. Magnetic resonance imaging (MRI) of brain and spinal cord was performed, and it showed no abnormalities. Clinical, laboratory, and imaging data supported the diagnosis of eosinophilic meningitis (EM) caused by helminths. Treatment with mebendazole (400 mg once daily) was started, but an accentuation of the symptomatology was noticed after the first administration, and the patient was switched to steroid therapy. In agreement with other data reported in the literature, we initiated a 2‐week course of prednisolone, 60 mg/d, which is effective against central nervous system (CNS) infections due to Angiostrongylus  cantonensis. The patient was discharged after 2 weeks of steroid therapy, and at 1‐month follow‐up, the general state was good, but paresthesia persisted only to the face.

Figure 1

Microscopic examination of the cerebrospinal fluid.


EM is a distinct clinical entity that may have infectious and noninfectious causes. 1 Eosinophilic pleocytosis is rarely found in CSF, and the most common causes among the infectious agents are the parasites. Globally, A cantonensis is the main cause. 2 The parasite is distributed throughout tropical and subtropical countries between Madagascar and Tahiti, although the majority of human cases have been reported in the south Pacific islands and in Southeast Asia, especially in Taiwan and Thailand. 3–5 Recently, an outbreak of the disease was reported in the southern coastal area of the Chinese mainland. 6 However, with the increase in world travel and the ship‐borne dispersal of infected rats, the parasite has been extending its range. Cases of A cantonensis infection have been reported in Australia, North America, and, more recently, in Europe. 7–9 This nematode naturally parasitizes the pulmonary artery of wild rats. The larvae of A cantonensis are excreted in the feces of rats and picked up by intermediate or transport hosts such as snails, vegetables, slugs, and small molluscs. Humans become infected after eating these hosts or fresh produce contaminated by them. Upon ingestion, the larvae penetrate the vasculature of the gastrointestinal tract and usually migrate into the CNS via the bloodstream, causing eosinophilic meningoencephalitis. Angiostrongyliasis often presents as a subacute infectious disease of the CNS, although fatal cases are reported. 10 After an incubation period of 1 to 3 weeks, signs and symptoms such as headache, visual disturbance, photophobia, vomiting, nuchal rigidity, hyperesthesia, paresthesia, and fever (less common) may develop. 11 A history of severe frontal or bitemporal headache is characteristic. 12 Paresthesias are the most distinctive neurological finding in adults with angiostrongyliasis; symptoms may be present on the extremities, trunk, or face, and they can persist for weeks after other symptoms have resolved. 11,13 Ocular migration of larvae has been reported in countries where A cantonensis is endemic. 6,14 Delirium, seizures, and persistent cognitive dysfunction have also been observed. 12 Analysis of CSF samples obtained from infected patients has revealed pleocytosis with eosinophilia. 6,14 Eosinophilic pleocytosis exceeds 10% in more than 95% of patients. 12 Slom and colleagues showed that the mean eosinophil count was significantly higher at the midpoint of the clinical course than during the acute (p = 0.03) and convalescent (p = 0.03) phases. 8 An eosinophil count of at least 0.6 × 109/L in peripheral blood may also be observed. 8 Other parasites potentially responsible for EM are Gnathostoma spinigerum, Baylisascaris procyonis, Toxocara canis, T solium, Trichinella spp., Schistosoma spp., Paragonimus spp. The absence of focal lesions on CT or MRI of the brain helps distinguish A cantonensis meningitis from other CNS helminthic infections. 2,3 Coccidioidomycosis, cryptococcosis, neurosyphilis, tuberculosis, rickettsiae, viruses, malignancies, drugs, and idiopathic hypereosinophilic syndrome can occasionally cause mild CSF eosinophilia. 3 Because larvae are rarely detected in the CSF and immunologic techniques are available only at specialized centers, a definitive diagnosis is very difficult, being often based only on the identification of contaminated foods that the patients may have ingested and based on the symptoms and signs they present. 1,12,15 We believe that the EM of our patient was probably caused by A cantonensis, on the basis of epidemiological and clinical data and of the response to therapy. The patient has sojourned for a long period in Santo Domingo, which is an endemic zone for A cantonensis, and he had consumed freshwater shrimps. The clinical symptomatology gradually initiated with headache and generalized paresthesia, the latter being a characteristic neurological sign of EM caused by A cantonensis. Serologic tests were useful for differential diagnosis from other CNS parasitic infections, neurosyphilis, tuberculosis, and fungal and viral infections. The absence of focal lesions, present in other helminthic infections, on the CT and the MRI supports the diagnosis from A cantonensis. We were not able to make a definitive diagnosis not only because of the negative microscopic examination of the stool and the search for the larvae in the CSF but also because of the lack of specific tests. The administration of mebendazole caused a worsening of the symptoms, possibly because of the inflammatory reaction to antigens released by dying worms, while the administration of corticosteroids at the dosage of 60 mg determined the disappearance of the headache. 16 Treatment of EM caused by A cantonensis is an unresolved issue because there is no proven specific therapy for angiostrongyliasis in humans. Treatment consists primarily of supportive therapy using analgesic agents and preventing volume depletion. There are several reports in which repeated lumbar puncture provided symptomatic relief for patients with persistent headaches caused by increased intracranial pressure, but this improvement is often transient. 8,17,18 Anthelmintic agents (mebendazole and albendazole) have been used for treatment of EM caused by A cantonensis. 4,19 Du and colleagues observed that combined treatment with interleukin‐12 and mebendazole lessens the severity of experimental EM in mice. 20 In a recent experimental study, Wang and colleagues evidenced that pathological changes in the brains of the treated rabbits are more severe than in brains of those without albendazole treatment, suggesting that the drug may not be very suitable for the treatment of cerebral angiostrongyliasis. 21 Corticosteroid treatment is controversial because a number of anecdotal reports of steroid use have shown conflicting results. 11,22 The possible benefits for corticosteroid treatment of EM were recently investigated by Chotmongkol and colleagues in a prospective, placebo‐controlled, double‐blind study. 23 The authors evidenced that there were significant differences between the treatment and the control groups, with regards to the number of patients who still had headache after 14 days of therapy (p = 0.00004), the median length of time until complete disappearance of headache (p = 0.00000), and the number of patients who had repeated the lumbar puncture (p = 0.002). 23 Recently, however, several reports evaluated the efficacy of combined treatment with steroids and anthelmintic agents for EM caused by A cantonensis. 24,25 Chotmongkol and colleagues underlined that a 2‐week course of prednisolone (60 mg/d) and albendazole (15 mg/kg/d) determined a complete disappearance of headaches in 88.5% of patients. 24 The same author group evidenced that the combination of prednisolone (60 mg/d) and mebendazole (10 mg/kg/d) given for a 2‐week course was safe and beneficial in relieving headaches in EM. 25 Last, it is important to emphasize that most patients with EM due to A cantonensis have a self‐limited course and recover completely. 26

We conclude that neurohelminthiasis should always be considered in patients returning from developing countries with headache, neurological signs, and blood and CSF eosinophilia because it is the most frequent cause of EM. Although there is no proven therapy, the most recent data in the literature suggest that the therapy with anthelmintic and corticosteroids should be promptly started in case of EM due to A cantonensis or other helminths.

Declaration of interests

The authors state that they have no conflicts of interest.


We thank the patient who agreed to participate in this study.


View Abstract