Diagnostic Yield, Morbidity, and Mortality of Intraventricular Neuroendoscopic Biopsy: Systematic Review and Meta-Analysis.

OBJECTIVE

Neuroendoscopic techniques for biopsy of intraventricular tumors are increasingly used, although published data have demonstrated a wide range of outcomes. We performed a systematic review and meta-analysis to investigate the diagnostic yield, morbidity, and mortality of neuroendoscopic biopsy.

METHODS

Medline and Embase were searched for original data on outcomes of neuroendoscopic biopsy. Summary estimates were achieved by applying a random effects model as per DerSimonian-Laird. Measures of heterogeneity and publication bias were also assessed. Meta-regression was used to assess the relative effect of rigid versus flexible endoscopy on the outcomes of interest. Data on study demographics, operative variables, histopathologies of identified lesions, and clinical features of intraventricular tumors were also collected.

RESULTS

A total of 30 studies with 2069 total biopsies were included. Neuroendoscopic biopsies were performed concurrently with at least 1 other procedure in 82.7% (n = 1252/1513) of procedures. Germ cell tumors, astrocytomas, and non-neoplastic lesions accounted for most of reported intraventricular lesions at 26.6% (n = 423), 25.5% (n = 406), and 12.4% (n = 198), respectively. The combined diagnostic yield of 28 studies reporting 1995 total biopsies was 87.9% (95% confidence interval [CI] 84.1%-90.9%) with moderate heterogeneity (I(2) = 68.0%). The combined major morbidity of 17 studies reporting 592 total biopsies was 3.1% (95% CI 1.9%-5.1%). The combined mortality of 22 studies reporting 991 total biopsies was 2.2% (95% CI 1.3%-3.6%). There was no significant heterogeneity for major morbidity and mortality (both I(2) = 0). Among included studies, 50% (n = 14) reported using a rigid endoscope exclusively. The results of meta-regression demonstrated no significant differences in diagnostic yield when comparing studies using rigid versus flexible endoscopes exclusively.

CONCLUSIONS

These results indicate that neuroendoscopic biopsy has a very good diagnostic yield and reasonably low complication rate. The procedure seems most advantageous for diagnosis of intraventricular lesions where cerebrospinal fluid diversion is an additional therapeutic requirement.