ObjectiveTo determine the diagnostic value of clinical factors and biochemical or microbiological measures for diagnosing a drain-associated ventriculitis, we summarized the available evidence.MethodsWe performed a systematic review and meta-analysis of studies of patients with external ventricular CSF drains who developed drain-associated ventriculitis by searching MEDLINE, EMBASE, and CENTRAL electronic database. We reported the occurrence of abnormal test results in patients with and without drain-associated ventriculitis. For continuous variables, we recalculated mean values presented in multiple studies.ResultsWe identified 42 articles published between 1984 and 2018 including 3,035 patients with external CSF drains of whom 697 (23%) developed drain-associated bacterial ventriculitis. Indications for drain placement were subarachnoid, intraventricular or cerebral hemorrhage or hemorrhage not further specified (69%), traumatic brain injury (13%), and obstructive hydrocephalus secondary to a brain tumor (10%). Fever was present in 116 of 162 patients with ventriculitis (72%) compared with 80 of 275 (29%) patients without ventriculitis. The CSF cell count was increased for 74 of 80 patients (93%) with bacterial ventriculitis and 30 of 95 patients (32%) without ventriculitis. CSF culture was positive in 125 of 156 episodes classified as ventriculitis (80%), and CSF Gram stain was positive in 44 of 81 patients (54%). In patients with ventriculitis, PCR on ribosomal RNA was positive on 54 of 78 CSF samples (69%).ConclusionClinical factors and biochemical and microbiological measures have limited diagnostic value in differentiating between ventriculitis and sterile inflammation in patients with external CSF drains. Prospective well-designed diagnostic accuracy studies in drain-associated ventriculitis are needed.