Validating machine learning approaches for prediction of donor related complication in microsurgical breast reconstruction: a retrospective cohort study

Autologous reconstruction using abdominal flaps remains the most popular method for breast reconstruction worldwide. We aimed to evaluate a prediction model using machine-learning methods and to determine which factors increase abdominal flap donor site complications with logistic regression. We evaluated the predictive ability of different machine learning packages, reviewing a cohort of breast reconstruction patients who underwent abdominal flaps. We analyzed 13 treatment variables for effects on the abdominal donor site complication rates. To overcome data imbalances, random over sampling example (ROSE) method was used. Data were divided into training and testing sets. Prediction accuracy, sensitivity, specificity, and predictive power (AUC) were measured by applying neuralnet, nnet, and RSNNS machine learning packages. A total of 568 patients were analyzed. The supervised learning package that performed the most effective prediction was neuralnet. Factors that significantly affected donor-related complication was size of the fascial defect, history of diabetes, muscle sparing type, and presence or absence of adjuvant chemotherapy. The risk cutoff value for fascial defect was 37.5 cm2. High-risk group complication rates analyzed by statistical method were significant compared to the low-risk group (26% vs 1.7%). These results may help surgeons to achieve better surgical outcomes and reduce postoperative burden.

The “Hug Flap”: Surgical Technique to Enhance the Aesthetic Breast Projection in Autologous Breast Reconstruction

Background Obtaining a natural breast mound shape contributes profoundly to a symmetrical and successful outcome in breast reconstruction. Objectives The authors sought to describe a new and efficient technique that enhances breast projection in delayed breast reconstruction employing abdominal free flaps and compare it with the current standard methods utilized. Methods The charts of 490 consecutive patients who underwent delayed breast reconstruction employing free abdominal perforator flaps were reviewed between 2007 and 2017. Three methods of breast reconstruction were compared: undermining, de-epithelialization, and the “hug flap” (HF). In the newly described technique, the caudal mastectomy skin was de-epithelialized, and then the medial and lateral thirds were undermined and folded over to cover the central part. The rates of complications and secondary corrections were analyzed between the 3 groups. Results There were 570 free abdominal flaps performed. The de-epithelization technique was the most commonly utilized (328 cases) followed by the undermining technique (153 cases). The HF technique was employed in 89 cases. The majority of HFs were performed in unilateral breast reconstruction. Bilateral cases were conducted in only 12 patients. The need for additional fat grafting was significantly (P = 0.003) less required in the HF group compared with the undermining and de-epithelializing groups (12% vs 28% and 21%, respectively). Conclusions Although all breast-enhancing options can be mixed and matched based on the surgeon’s preference and experience as well as each patient’s needs, the HF can be considered as an adjunct tool to provide adequate flap projection and enhance breast symmetry. Level of Evidence: 4