Victoria Kong, Jun Jiang, Tobias Niederegger, Thomas Schaschinger, Giuseppe Sofo, Miguel Aranda Wildberger, Adriano Fabi, Thierry Schweizer, Elisabeth A Kappos, Daniel Schmauss, Hans-Guenther Machens, Fortunay Diatta, Martin Kauke-Navarro, Samuel A Knoedler
Diabetes is a significant but heterogeneous risk factor in ABR, with insulin-treated patients at particularly elevated perioperative risk, supporting treatment-specific risk stratification.
BACKGROUND: Autologous breast reconstruction (ABR) offers superior quality of life compared with implant-based reconstruction or mastectomy alone but carries substantial perioperative risk. Diabetes mellitus is a known predictor of postoperative morbidity but is often treated as a binary variable, obscuring differences between patients on oral antihyperglycemic agents (diabetes managed with oral antihyperglycemic agents [DM-ORAL]) and insulin (diabetes requiring insulin therapy [DM-INS]).
MATERIALS AND METHODS: Using the National Surgical Quality Improvement Program, we identified female breast cancer patients undergoing immediate ABR (2008-2021). Diabetes status was categorized as nondiabetic, DM-ORAL, or DM-INS. Primary outcomes were 30-day adverse events, reoperation, unplanned readmission, and surgical and medical complications. Groups were compared and multivariable regression used to adjust for confounders.
RESULTS: Of 8121 patients, 559 (6.9%) had diabetes, including 457 (81.8%) DM-ORAL and 102 (18.2%) DM-INS. Diabetes prevalence increased over time with a 0.3 percentage-point annual rise (R2 = 0.58; P = .003), driven mainly by DM-ORAL (R2 = 0.52; P = .005), while DM-INS showed a nonsignificant trend (R2 = 0.28; P = .07). Compared with nondiabetic patients, diabetics had higher overall complication rates (34.7% vs. 24.1%), including medical (6.4% vs. 2.8%) and surgical complications (23.6% vs. 15.2%) (all P < .0001). Within diabetics, DM-INS demonstrated higher morbidity, including unplanned readmission (14.7% vs. 6.3%, P = .008) and superficial infection (10.8% vs. 5.3%, P = .04) compared to DM-ORAL. On multivariable analysis, diabetes independently predicted complications (OR 1.4), with higher risk in DM-INS (OR 1.9) than DM-ORAL (OR 1.3); DM-INS more than doubled surgical complication risk (OR 2.5).
CONCLUSION: Diabetes is a significant but heterogeneous risk factor in ABR, with insulin-treated patients at particularly elevated perioperative risk, supporting treatment-specific risk stratification.