Do Prophylactic Postoperative Antibiotics Prevent Sacral Neuromodulation Infections?

Abstract

Importance There are no definitive guidelines for use of postoperative antibiotics after sacral neuromodulation (SNM) leading to practice pattern variation among health care professionals. Objectives The objectives of this study were to determine if additional antibiotics beyond preoperative intravenous antibiotics and surgical preparation decrease postoperative infections and to determine if additional antibiotics are associated with other postoperative complications. Study Design This was a single-center, retrospective cohort study of all SNM procedures from 2014 to 2023 performed by 12 surgeons. Cohorts were stratified between patients who received preoperative antibiotics only (OnlyPreAbx) and patients who received both preoperative antibiotics and postoperative antibiotics (PrePostAbx) after either insertion of tined lead and/or after insertion of a neurostimulator. Results There were 212 patients included in this study: 70 (33.0%) in the OnlyPreAbx group and 142 (67.0%) in the PrePostAbx group. Of patients receiving postoperative antibiotics, 76 (53.5%) received cephalexin, 49 (34.5%) received sulfamethoxazole-trimethoprim, and 17 (12.0%) received alternative antibiotics. Six patients overall (2.8%) experienced infections after SNM. There was no difference in the rates of infection between groups (4 [1.9%] PrePostAbx vs 2 [0.9%] OnlyPreAbx, P = 0.99). Of the 4 patients with postoperative infections in the PrePostAbx group, 2 had cellulitis requiring antibiotics and 2 required full explantation. Of the 2 patients with postoperative infections in the OnlyPreAbx group, both patients required explantation. In a subanalysis comparing infected and noninfected patients, infected patients (n = 6) had higher rates of hypertension (n = 6, 100%; P = 0.02) and diabetes mellitus (n = 3, 50%; P = 0.05). Conclusions Additional postoperative antibiotics did not decrease infection rates in patients undergoing SNM. Similar comparative analyses should be performed with larger sample sizes.