Utility of flow cytometry of cerebrospinal fluid as a screening tool in the diagnosis of central nervous system lymphoma

Abstract

Context: Experiences at our institution show that flow cytometry analysis (FCA) has become routine clinical practice in the workup of patients with altered mental status, even if risk factors are low.

Objective: To assess diagnostic accuracy of combined FCA and cytology in the diagnosis of central nervous system lymphoma in an unselected patient population with neurologic symptoms, including patients with no history of lymphoma or suspicious radiology.

Design: Between 2001 and 2011, cerebrospinal fluid was submitted from 373 patients for lymphoma screening by FCA. The medical records were reviewed for patient symptomatology, history of malignancy, brain imaging, FCA results, cytology results, brain biopsy, and clinical follow-up.

Results: A lymphoid malignancy was detected by FCA in 4% of cases. A positive diagnosis was more likely in patients with either a history of hematologic malignancy and/or a suspicious radiology result (P = .009). All patients with no history of lymphoma and no suspicious radiology (n = 102) had negative cytology, and none had a correspondingly positive FCA result. The positive and negative predictive values of combined cytology and FCA in the patients with history of lymphoma and/or abnormal imaging results were 92% and 89%, respectively, when compared with open brain tissue biopsy, and 89% and 86%, respectively, when compared with clinical follow-up. When low-risk patients were included, the positive predictive value remained at 92%, but the negative predictive value dropped to 52% with the open brain biopsy as the reference, and values did not change significantly for the group with clinical follow-up.

Conclusions: Concurrent FCA and cytology are most useful in the appropriate clinical setting, and we propose a triage algorithm for how FCA on cerebrospinal fluid is best used.