Granisetron for Postoperative Nausea and Vomiting Prophylaxis in Microvascular Decompression Despite Prominent...

Background Microvascular decompression (MVD) is associated with a high incidence of postoperative nausea and vomiting (PONV). Unlike in general surgery, where PONV is predominantly serotonin-mediated, MVD additionally activates non-serotonergic emetic mechanisms, including vestibular stimulation and direct brainstem perturbation. Whether selective 5-HT₃ receptor antagonists can effectively prevent PONV in MVD remains unclear. Materials and methods This retrospective, exploratory, quasi-experimental before-after study included 40 consecutive MVD patients at a single institution, divided into a granisetron group (n = 20; 1 mg intravenously at dural closure) and a control group (n = 20) based on an institutional protocol change. Primary outcomes included PONV incidence, nausea severity, vomiting episodes, and rescue antiemetic use. Results Using the predefined moderate-to-severe threshold (nausea severity score ≥4, vomiting, or rescue antiemetic use), the 24-hour PONV incidence was lower in the granisetron group than in the control group (35% vs. 80%), corresponding to an absolute risk reduction (ARR) of 45% and a relative risk reduction (RRR) of 56% (p = 0.010). At six hours, the granisetron group showed lower nausea severity scores (4.0 (0-6.0) vs. 8.5 (6.75-9.0); p = 0.002) and fewer vomiting episodes (0 (0-1.0) vs. 1 (0.75-2.0); p = 0.022). Early mobilization and oral intake were also improved. Because the analysis was exploratory and the sample size limited, the estimates should be interpreted as hypothesis-generating. Conclusions In this exploratory, single-center, quasi-experimental before-after study of 40 patients, prophylactic granisetron was associated with an RRR of 56% and an ARR of 45% (number needed to treat ≈ 2.2) for moderate-to-severe PONV following MVD, consistent with a substantial serotonergic contribution to PONV in this setting. The residual PONV rate of 35% suggests that non-serotonergic emetic mechanisms remain unaddressed. Given the small sample size and single-center design, these findings should be regarded as hypothesis-generating rather than confirmatory; nonetheless, they support the evaluation of multimodal antiemetic strategies and may inform the design of adequately powered future studies on optimal antiemetic strategies for MVD.