The US Food and Drug Administration approved sacral neuromodulation (SNM) for treatment of various urinary incontinence and retention disorders. Sacral neuromodulation implantation is recommended for patients with urinary symptoms who fail conservative therapy. After implantation of the device, many patients obtain considerable improvement in symptoms. However, wound infections have been reported after implantation in 5% to 11% of patients, which typically require removal of the device. It has been hypothesized that perioperative antibiotic selection is related to the risk for infections, but little is known about antibiotics or other potential risk factors of the development of infections.
The aims of this retrospective cohort study were to determine the rate of postoperative wound infections after SNM implantation and to evaluate antibiotics and various other potential predictive risk factors of infection. All patients underwent SNM implantation performed by 1 of 3 surgeons at a single medical center from 2007 to 2010. The choice of preoperative antibiotics was based on surgeon preference; the patients received cefazolin alone, vancomycin alone, or vancomycin with gentamicin. Multivariate models were used to assess the association between potential risk factors and the development of postoperative wound infections. In addition to preoperative antibiotic regimens, variables examined included sex, surgeon, surgical site (outpatient vs inpatient), body mass index, comorbidities (diabetes mellitus, immunosuppression), current smoking, history of urinary tract infections, and preoperative skin preparation.
Of the 136 identified patients who underwent SNM implantation, 8 (5.9 %) experienced infections that required removal of the device. Cefazolin alone was less effective in preventing infection than the other antibiotic regimens (P = 0.03). Multivariate analysis showed that the odds of having an infection in the cefazolin-treated patients were more than 7 times higher than that of the patients treated with another antibiotic regimen (odds ratio, 7.3; 95% confidence interval, 1.2–54.8). Seven of the 8 explanted devices grew Staphylococcus aureus on culture that was resistant to cephalosporins. None of the other examined variables made a statistically significant contribution to infection.
These data show that preoperative antibiotic selection with cefazolin is the only significant risk factor for subsequent infection and explantation of SNM devices.