RESUMO
BACKGROUND AND OBJECTIVES: Ventriculoperitoneal shunt (VPS) is usually the primary choice for cerebrospinal fluid shunting for most neurosurgeons, while ventriculoatrial shunt (VAS) is a second-line procedure because of historical complications. Remarkably, there is no robust evidence claiming the superiority of VPS over VAS. Thus, we aimed to compare both procedures through a meta-analysis. METHODS: Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines, the authors systematically searched the literature for articles comparing VAS with VPS. The included articles had to detail one of the following outcomes: revisions, infections, shunt-related mortality, or complications. In addition, the cohort for each shunt model had to encompass more than 4 patients. RESULTS: Of 1872 articles, 16 met our criteria, involving 4304 patients, with 1619 undergoing VAS and 2685 receiving VPS placement. Analysis of revision surgeries showed no significant difference between VAS and VPS (risk ratio [RR] = 1.10, 95% CI: 0.9-1.34; I2 = 84%, random effects). Regarding infections, the analysis also found no significant difference between the groups (RR = 0.67, 95% CI: 0.36-1.25; I2 = 74%, random effects). There was no statistically significant disparity between both methods concerning shunt-related deaths (RR = 2.11, 95% CI: 0.68-6.60; I2 = 56%, random effects). Included studies after 2000 showed no VAS led to cardiopulmonary complications, and only 1 shunt-related death could be identified. CONCLUSION: Both methods show no significant differences in procedure revisions, infections, and shunt-related mortality. The literature is outdated, research in adults is lacking, and future randomized studies are crucial to understand the profile of VAS when comparing it with VPS. The final decision on which distal site for cerebrospinal shunting to use should be based on the patient's characteristics and the surgeon's expertise.