Abstract 21: Optimizing Perforator Selection: A Comprehensive, Multivariate Analysis of Fat Necrosis and Abdominal Morbidity in DIEP Flap Breast Reconstruction

Abstract

PURPOSE: Studies have suggested several factors that affect free flap fat necrosis and abdominal morbidity after DIEP flap breast reconstruction. We aimed, by conducting the most inclusive and current multivariate analysis of these factors to date, to clarify how patient specific perforator selection can affect fat necrosis and abdominal morbidity, thus guiding future operative decision-making. METHODS: Retrospective review of a prospectively maintained, 866 free-flap database was performed for patients undergoing breast reconstruction at one institution from 2010–2016. 29 potential predictors, including patient demographic factors and intraoperative parameters, were included in a multivariate analysis for outcomes of fat necrosis, abdominal wounds, and abdominal bulge or hernia. These results were reported as Odds Ratios (OR) with 95% Confidence Intervals. Univariate analyses were utilized to confirm that potentially confounding pre-surgical and intraoperative factors amongst each of the three primary outcome variables were sufficiently equivalent. Any wound requiring local wound care after 2 weeks was characterized as a minor abdominal wound, whereas major abdominal wounds were defined as wounds requiring return to the operating room. RESULTS: 409 total DIEAP flaps were included with an average 18.5-month follow-up: 14.4% had fat necrosis, 21.2% had a minor or major abdominal wound, and 6% had an abdominal bulge or hernia. Analysis showed increased odds of fat necrosis with increasing flap weight (OR 1.002 per 1g increase, p<.001), and earlier year of surgery (OR 2.324 for 2010–2013 vs. 2014–2016, p=.02), and decreased odds of fat necrosis with lateral or both-row perforators vs. medial row (OR.303, .0229, p-value=.0013), and neoadjuvant chemotherapy (OR.384, p=.016). Perforator flow rate/caliber and number of perforators did not affect fat necrosis. However, upon subgroup analysis on flaps with fat necrosis, we found that there was a significant difference in weight between single perforator flaps and multi-perforator flaps (789g vs. 983g respectively, p= .048). There was an increased odds of having abdominal wounds with smoking (OR 1.869, p=.02), hypertension (OR 1.720, p=.04), and increasing flap weight (OR 1.001 per 1g increase, p<.01). BMI was not a significant factor for abdominal wounds when controlled by flap weight in the multivariate analysis. Increased odds of abdominal bulge/hernia were seen with a return to the OR same hospital stay (OR 5.922, p<.01), and with lateral/both row perforators vs. medial row (OR 3.01, p=.058). CONCLUSIONS: Our analysis of DIEP flaps shows surgeon experience reduced the odds of fat necrosis, while heavier flap weights increased these odds. Moreover, adding lateral row perforators may decrease fat necrosis at the potential cost of increasing abdominal bulges. While perforator number was not a significant predictor of fat necrosis in the multivariate analysis, the subgroup analysis may indicate that there is a higher allowable threshold of flap weight before fat necrosis occurs with multi-perforator versus single perforator DIEP flaps. A.S. Hembd: None. M. Cho: None. C. Venutolo: None. S. Teotia: None. N. Haddock: None.