Cryptococcal Meningitis Developing in a Patient with Neurosarcoidosis

Abstract

Cryptococcal meningitis is a critical disease that occasionally involves immunosuppressed patients. We herein report a 79-year-old Japanese man who received low-dose prednisolone therapy for neurosarcoidosis and panhypopituitarism. He presented a 10-day history of a fever and altered mental status. The FilmArray^Ⓡ Meningitis/Encephalitis Panel and serum cryptococcal antigen tests were both negative, but the cerebrospinal fluid sample became positive for Cryptococcus neoformans after seven-day incubation. After the diagnosis of cryptococcal meningitis, we successfully treated the patient with a recommended treatment regimen. When an immunocompromised patient presents with a subacute fever accompanying any central nervous symptoms, cryptococcal meningitis should be screened for.

Introduction

Cryptococcosis is a fungal infection caused by Cryptococcus neoformans or Cryptococcus gattii, typically involving patients with immunocompromising factors, such as human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS), malignancies, transplantation recipients, and use of immunosuppressing agents (1-3). The causative pathogens are ubiquitously found in the natural environment, including soil, bird excrement, and tree bark (4), and healthy individuals can be infected as well. Cryptococcal infection causes various clinical presentations, ranging from asymptomatic antigenemia, and localized pulmonary disease, to disseminated infections of the skin, bone, lymph node, or central nervous system. Among these, cryptococcal meningitis is a life-threatening condition (5), and an early and appropriate diagnosis of the disease is essential.

Patients with sarcoidosis are vulnerable to cryptococcosis, and some of them develop neurosarcoidosis (NS) that clinically mimics cryptococcal meningitis. Cryptococcal antigen (CrAg) testing is recommended to differentiate these difficult-to-diagnose cases (6). However, although the sensitivity and specificity of serum and cerebrospinal fluid (CSF) CrAg were reportedly over 94% in HIV-infected patients (7), the testing sensitivity of the serum CrAg is relatively low at 78% in non-HIV patients (8). Recently, the FilmArray^Ⓡ meningitis/encephalitis panel (BioFire Diagnostics/Biomerieux, Salt Lake City, USA) has become available as a multiplex polymerase chain reaction (PCR) assay for the diagnosis of common pathogens in central nervous system infections (9,10). However, Cryptococcus species are known to be underdiagnosed due to false negative results (11).

We herein report a case of cryptococcal meningitis involving an older Japanese man with underlying NS in which those diagnostic approaches failed to determine the disease.

Case Report

A 79-year-old Japanese man presented to our institution with a 10-day history of a subacute fever, altered mental status, and dysconsciousness. He had received hormone replacement therapy and low-dose prednisolone therapy (4 mg per day on consultation) for central diabetes insipidus and hypofunction of the anterior pituitary gland by hypophysitis due to NS that had been diagnosed 17 years prior.

On admission, we initially suspected adrenal insufficiency; however, corticosteroid replacement therapy did not improve his condition. A lumbar puncture examination revealed mild meningeal inflammation with an elevated cell count (9 /μL) and protein level (70 mg/dL) and a reduced glucose level (43 mg/dL; 121 mg/dL in the plasma). Considering an infectious etiology, the FilmArray^Ⓡ Meningitis/Encephalitis Panel and serum CrAg tests were performed, both of which revealed negative results. We then examined the interleukin 6 (IL-6) level in the CSF, considering NS as the patient's background, and found that it was highly elevated at 3,833 pg/mL. Therefore, we considered exacerbation of NS as a plausible cause of the patient's condition.

One week later, however, the CSF sample inoculated into enrichment broth became positive for C. neoformans, and an additional test for CrAg using CSF revealed positivity. Under the diagnosis of cryptococcal meningitis, we administered a combination of amphotericin B (225 mg every day, 5 mg/kg) with flucytosine (1,000 mg every 12 hours, 22 mg/kg). This antifungal therapy was continued for 3 weeks until the patient became afebrile, followed by oral fluconazole 400 mg daily as consolidation therapy for 8 weeks and oral fluconazole 200 mg daily as maintenance therapy for 6 months.

Discussion

Cryptococcal meningitis is a critical disease in immunosuppressed patients, with reported mortality rates of 15-27% (5,12). Sarcoidosis and steroid use are both considered risk factors for the development of cryptococcosis (2,3,8). Steroid use in non-HIV patients, especially at a dosage of ≥20 mg/day, is reportedly associated with cryptic disease, with an odds ratio of 5.3 [95% confidential interval (CI): 1.6-18.3] (13). Although there is no specific study on the risk of cryptococcal meningitis among NS patients, a multivariate regression model study revealed that the odds ratio of extrapulmonary sarcoidosis was higher than that of systemic corticosteroid use [4.31 (80% CI; 1.6-11.4) vs. 3.23 (80% CI; 1.2-8.6)] (3). In the present case, the daily dose of prednisolone was relatively small (4 mg/day), so NS itself was presumed to have triggered the onset of cryptococcal meningitis.

A physical diagnosis of cryptococcal meningitis may be challenging because of the slow onset and nonspecific presentation. CrAg testing was developed as an easily-available laboratory method for the diagnosis of cryptococcosis. The diagnostic accuracy of the CrAg testing has been widely evaluated among HIV patients (6) ; however, the sensitivity of serum CrAg testing in non-HIV cases is lower than 80%, and cryptococcal meningitis cannot be ruled out by a negative result on serum CrAg testing alone (7). A previous study recommended submitting CrAg testing in CSF samples when cryptococcal meningitis was suspected (8). In our case, CrAg was not initially tested for CSF, which was one of the reasons for the delayed diagnosis of the disease.

A definitive diagnosis of NS is extremely challenging, and various diagnostic tools have been suggested (14). An elevated IL-6 level in the CSF is one clinically available biomarker that clinicians have recently applied to patients to aid in diagnosing NS (15). However, IL-6 elevation in CSF has low specificity and is reportedly not suitable for determining the etiology of central nervous system infection (16,17). Recently, the FilmArray^Ⓡ meningitis/encephalitis panel has been used increasingly frequently as a multiplex PCR test for common pathogens of central nervous system infection, including Cryptococcus species (9-11), and has shown sufficient sensitivity and specificity for the differentiation of NS (8,11). Negative results of FilmArray^Ⓡ and serum CrAg testing led us to negate the possibility of cryptococcal meningitis.

Owing to the highly elevated IL-6 level in the CSF, we first diagnosed this case as exacerbation of NS rather than the cryptococcal meningitis. When an immunocompromised patient presents with a subacute fever accompanying any central nervous symptoms, it is important to include cryptococcal meningitis as a differential diagnosis and carefully conduct diagnostic testing.

Our case highlights the difficulty of diagnosing cryptococcal meningitis in patients with NS. We encourage clinicians to take note of the fact that the FilmArray^Ⓡ panel and serum CrAg testing are useful but can fail to correctly diagnose such cases. In addition, elevated IL-6 levels in the CSF are clinically non-specific and do not necessarily suggest NS alone.

The authors state that they have no Conflict of Interest (COI).