Length of Stay Following Breast Reconstruction with Free Tissue Transfer: A Doubly Robust Nationwide Analysis
Vikas S. Kotha1, Kenneth L. Fan, M.D.1, Richard Amdur, PhD2, David H. Song, M.D., M.B.A1.
1Georgetown University, Washington, DC, USA, 2George Washington University, Washington, DC, USA.
Background: The aim of this study was to use the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database to identify predictors and causes of extended length of stay (LOS) after breast reconstruction with free tissue transfer.
Methods: Consecutive cases of breast reconstruction with free tissue transfer were retrieved from the ACS-NSQIP (2005-2017) database using Current Procedural Terminology (CPT) code 19364. Extended LOS was defined as postoperative LOS >5 days. Missing data were analyzed for patterns and handled using multiple imputation. Covariate propensity-adjusted binary logistic regression models adjusted for demographic confounders (operation year, BMI, race/ethnicity) were used to identify significant preoperative, intraoperative, and postoperative associations with extended LOS. Statistical significance was defined as P <0.05. All analyses were performed with SPSS 25 (IBM Corp.).
Results: 9,686 cases were studied. The incidence of extended LOS was 34%. Recent reconstructions (2017) were at less risk of extended LOS (OR 0.2-0.3, p<0.0001). Preoperative influencers of LOS were overweight (OR 1.4, p=0.0001) and obese (OR 1.5-1.8, p=0.0001) body mass index, diabetes (OR 1.4, p=0.0001), recent malignancy (OR 2.2), and platelet level >350 mg/dL. Reconstructions lasting >500 minutes were at significant increased risk of extended LOS(OR 2.8, p<0.0001). Reoperation (OR 3.2) and postoperative transfusion (OR 2.4, p=0.0001) significant postoperative causes of extended LOS.
Conclusion: This study characterizes extended length of stay after free tissue transfer breast reconstruction via predictive analysis of a large prospective nationwide database. Contemporary health systems are financially overburdened and interested in performance-based cost-reduction strategies. Such strategies commonly incorporate postoperative length of stay a surrogate of cost and performance efficiency. Given the rise in popularity of free flap breast reconstruction and persisting high costs compared to alloplastic reconstructive modalities, there is an interest to characterize the perioperative conditions associated with prolonged recovery and resource utilization. Although the validity of large databases have been questioned, the goals of this study favorably suit such data. Because ABR is a heterogenous undertaking involving biosocially-diverse patients, various surgical techniques varies, and medical facilities of varying infrastructure, small-scale analyses may lack broad relevance. Therefore, we utilized the American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) to further our understanding of postoperative LOS. The results of this study further define the boundaries of perioperative risk-stratification to optimize preoperative management and operative planning.
Back to 2019 Abstracts