Risk stratification of surgical-site outcomes by BMI and flap type in autologous breast reconstruction.

[INTRODUCTION] Afflicting 2 million lives annually worldwide, breast cancer remains devastating. This study utilized a continuously updated network of electronic medical records (TriNetX Inc, Cambridge, MA) for analysis of 90-day postoperative outcomes of autologous breast reconstruction by increasing body mass index (BMI).

[METHODS] The deidentified electronic medical records (EMRs) of 29,453,000 females, age 18-99 years, were retrospectively screened from 45 healthcare organizations. A combined cohort of 7136 patients undergoing autologous breast reconstruction via transverse rectus abdominus muscle (TRAM), deep inferior epigastric perforator (DIEP), or latissimus flap was categorized by BMI into 5 subgroups: normal (n = 3568), overweight (n = 1239), class I (n = 1166), class II (n = 807), and class III (n = 356) obesity. The normal BMI cohort was then compared with each elevated BMI cohort. BMI strata were analyzed for risk of surgical-site occurrences within 90 days of surgery using CPT codes. Stringent propensity score matching was performed.

[RESULTS] For the combined group (N = 7136), significant linear increases in risk were observed with increasing BMI for infection (risk ratio [RR] 1.39-2.91,p < 0.05) and dehiscence (RR 2.65-5.17, p < 0.05). Similar linear increases were observed for the abdominally based group (N = 5454) for infection (RR 1.45-2.47, p < 0.05) and dehiscence (RR 2.54-4.77, p < 0.05). For DIEP (N = 4874), near-linear increases were observed for infection (RR 1.60-2.79, p < 0.05) and dehiscence (RR 1.57-5.59, p < 0.05). For TRAM (N = 714), significant increases were observed for seroma, infection, dehiscence, deep vein thrombosis (DVT), sepsis, and PE while increased risks of seroma, DVT, PE, and hernia were observed for latissimus (N = 1380).

[CONCLUSIONS] Regardless of flap type, our analysis suggests that a BMI> 39.9 is the inflection point beyond which it may be beneficial not to perform autologous breast reconstruction. Limitations include this study's retrospective nature; thus, future prospective studies would be beneficial.