C. T. Marcinak1, K. S. Ahmed1, S. M. Issaka1, K. J. Kelly1, S. M. Ronnekleiv-Kelly1, P. R. Varley1, D. E. Abbott1, S. M. Weber1, R. M. Minter1, S. Zafar1 1University Of Wisconsin, Department Of Surgery, Madison, WI, USA
Introduction:
Wound complications, including surgical site infections (SSI) and wound disruption, remain a major cause of morbidity after abdominal surgery. For patients undergoing pancreatectomy, several small trials have demonstrated lower rates of wound complications with the use of negative pressure wound therapy (NPWT) or an intraoperative wound protector (WP). However, these interventions have not been studied side-by-side in a large cohort. We sought to compare the effect of NPWT and WP on wound complications using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database.
Methods:
We performed a retrospective analysis using pancreatectomy data collected in the ACS-NSQIP between 2019 and 2021. Documentation of NPWT use in ACS-NSQIP began in 2019, and the analysis was limited to patients who underwent Whipple-type or total pancreatectomy. Predictor variables were the use of postoperative NPWT and/or intraoperative WP. The primary outcome was any postoperative wound complication, including superficial SSI, deep SSI, or wound disruption. Patients with either the predictor variables or primary outcome of interest missing were excluded, as were those with incompletely closed incisions. The independent association between use of NPWT or WP and wound complications was tested using multivariable Cox proportional hazards analysis incorporating patient demographics, relevant comorbid conditions, recent chemo- or radiotherapy, and American Society of Anesthesiologists physical status.
Results:
The final analysis included 13,083 patients, of whom 53.5% were male. The median age at the time of surgery was 67 years (IQR, 59 to 73 years). The cohort included 12,640 (96.6%) patients who underwent a Whipple-type procedure and 443 (3.4%) who underwent total pancreatectomy. NPWT alone and WP alone were used in 1,128 (8.6%) and 3,499 (26.7%) patients, respectively. Both interventions were used in 437 (3.3%) patients. Wound complications occurred in 1,046 (8.0%) patients, including 97 (8.6%) NPWT-only patients, 216 (6.2%) WP-only patients, and 31 (7.1%) patients who underwent both therapies. On multivariable analysis, WP alone was associated with lower odds of wound complication (OR, 0.70; 95% CI, 0.60-0.82; p < 0.001), while neither NPWT alone (p = 0.81) nor the use of both interventions (p = 0.33) demonstrated decreased odds of wound complication.
Conclusion:
The present analysis demonstrates lower odds of wound complications with the use of WP, but it fails to demonstrate a benefit with the use of NPWT alone or WP plus NPWT. As hospitals and databases continue to aggregate statistics on the use of these therapies, investigators will be able to better define their efficacy in high-risk populations.
