Antibiotic Prophylaxis in Two Stage Prepectoral Alloplastic Breast Reconstruction: One Prescription Does Not Fit All
Pooja S. Yesantharao, MS1; Lawrence Cai, MD2; Franca Kraenzlin, MD1; Gordon K. Lee, MD2; Brian Thornton, MD3; Rahim S. Nazerali, MD MHS2 1.Department of Plastic & Reconstructive Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA 2.Division of Plastic & Reconstructive Surgery, Stanford University School of Medicine, Stanford, California, USA 3.ThorntonMD Plastic Surgery, Louisville, Kentucky, USA
Background: Surgical site infection remains a challenge in alloplastic breast reconstruction. While postoperative antibiotic prophylaxis is routinely used, no standardized consensus guidelines exist, especially for prepectoral reconstruction. While prepectoral placement is less-invasive than subpectoral placement, it involves less vascularized tissue coverage of the tissue expander/implant and relies more heavily on dermal matrices, which have been linked to infection in prior literature. Thus, a more nuanced investigation of best practices with regards to antibiotic stewardship in the context of prepectoral expander/implant-based breast reconstruction is necessary. This study evaluated the impact of postoperative antibiotic prophylaxis duration on clinical outcomes in patients undergoing two-stage prepectoral expander/implant breast reconstruction.
Methods: This was an Institutional Review Board-approved, multi-institutional, retrospective investigation of 447 patients (573 breasts), stratified by duration of postoperative antibiotic prophylaxis (1 day, 14 days, or 21 days). Patient data were abstracted from medical records. Multivariable-adjusted logistic regression with robust variances was used to identify predictors of postoperative surgical site infections. All statistical analyses were completed using Stata v.15, and the two-tailed threshold for statistical significance was alpha values less than 0.05.
Results: Patients across study cohorts were well-matched with regards to demographic/clinical factors. Patients were separated into high risk (e.g. smoker, diabetic, etc.) and standard risk groups (infection rates: 22.2% versus 9.7%, respectively, chi square: p=0.02). Amongst standard risk patients, infection rates significantly differed by antibiotic duration (1 day - 15.2%, 14 days - 7.9% and 21 days - 5.4%, respectively, chi square: p=0.003). Upon univariable and multivariable analyses, the 14-day cohort (adjusted odds ratio: 0.5, 95% confidence interval: 0.2–1.0, p=0.05) had significantly reduced odds of infection compared to the 24-hour cohort. Twenty-one days of antibiotics, however, did not provide any additional benefit over 14 days. In contrast, amongst "high-risk" patients, duration of postoperative antibiotics did not influence infection rates (chi square: p=0.98). In fact, upon univariable and multivariable analyses adjusting for baseline differences amongst patient cohorts, neither 14-day (adjusted odds ratio: 1.2, 95% confidence interval: 0.1–9.5, p=0.89) nor 21-day cohorts (adjusted odds ratio: 1.1, 95% confidence interval: 0.1–10.0, p=0.90) had significantly reduced odds of postoperative infection among "high-risk" patients, when compared to the 24-hour cohort. Across all patients, incidence of postoperative complications other than infection and its sequelae was 20.7%. Overall complication rates did not significantly differ between study cohorts (chi square: p=0.51).
Conclusion: Two stage tissue expander/implant-based techniques remain the predominant form of post-mastectomy breast reconstruction. Thus, it is important to determine ways to continue improving outcomes for patients undergoing this type of breast reconstruction. In standard risk patients, extending postoperative antibiotic prophylaxis beyond 1 day significantly reduced infection rates after prepectoral tissue expander placement, though continuing antibiotics prophylaxis beyond 14 days did not confer additional clinical benefit. Among high risk patients, however, alternative strategies for infection control need to be investigated (Figure).
Back to 2021 Abstracts