Disseminated tuberculosis confounding a co-morbid primary CNS lymphoma

10-Year Retrospective Review of the Etiologies for Meningitis With Elevated Adenosine Deaminase in Cerebrospinal Fluid: Etiologies Other Than TB

Purpose

An elevated adenosine deaminase (ADA) level in the cerebrospinal fluid (CSF) is considered a reliable marker of tuberculous meningitis (TBM). However, CSF-ADA levels can also be elevated in other diseases. We aimed to find the most common diagnosis of patients with elevated CSF-ADA levels for the last 10 years.

Methods

We retrospectively investigated the diagnoses of all patients with elevated CSF-ADA (ADA ≥ 10 IU/L) levels between 2010 and 2019 at the Samsung Medical Center. Definite TBM was defined based on microbiological evidence. Clinical TBM was defined based on the brain imaging and response to the standard TB treatment. We compared the laboratory characteristics of the three most common diagnoses.

Results

CSF-ADA levels were elevated in 137 (5.6%) of 2,600 patients. The most common diagnoses included hematologic malignancy (HM; n = 36, 26.2%), TBM (n = 26, 19.0%), and viral meningitis (VM; n = 25, 18.2%). CSF-ADA levels did not differ significantly between TBM [median (interquartile range (IQR)), 20.2 IU/L (13.8–29.3)] and HM [16.5 (12.8–24.0)]. However, CSF-ADA levels were lower in VM [14.0 (11.0–16.1)] than in TBM (p = 0.027). Lymphocyte-dominant pleocytosis was more common in VM [77.0% (70.8–81.5)] than in TBM [16.0 (3.0–51.0), p = 0.015] or HM [36.0 (10.0–72.0); p = 0.032]. Interestingly, the CSF characteristics of clinical TBM were similar to those of VM but not definite TBM.

Conclusion

The most common diagnoses with elevated CSF-ADA levels were HM, followed by TBM and VM. Clinicians should carefully consider the differential diagnoses in patients with elevated CSF-ADA levels, especially those in the early stage of meningitis without microbiological evidence for TBM.

A Difficult Case of Pyrexia of Unknown Origin

This is the case of a 15-y-old boy who presented with fever and back pain with MRI features of spondylitis. A CT-guided vertebral biopsy showed acute and chronic inflammatory cells and grew Pseudomonas aeruginosa on aerobic culture. The child was treated for 2 wk with antibiotics with no response. Meanwhile, he developed new lung, liver, and splenic lesions on CT imaging. Empiric antitubercular therapy was then started and continued for 8 wk during which time there was progressive clinical deterioration. At this time the patient underwent bronchoscopy with lavage and endoscopic ultrasound-guided subcarinal lymph node and lung biopsy. The Xpert MTB/Rif ULTRA was "trace positive" in the bronchoscopic lavage with indeterminate rifampicin resistance, while it was negative in lymph node and lung biopsy. The lymph node and lung biopsy histopathology showed nonspecific inflammatory changes with no granulomas or malignant cells. In view of the positive Xpert ULTRA with indeterminate rifampicin resistance and no response to first-line drugs, treatment with second-line antitubercular drugs was initiated. The clinical condition continued to deteriorate; here the imaging findings were reviewed again and repeat aspiration cytology and biopsy from intra-abdominal nodes was carried out. This yielded the diagnosis of Hodgkin lymphoma. The patient had stage IVB disease. He responded well to standard chemotherapy and is currently in remission. The case illustrates the need to avoid empiric therapy, repeat invasive procedures if so needed, choose the site/method of biopsy appropriately and interpret results of investigations carefully when evaluating a patient with pyrexia of unknown origin.