Assessment and AI-Based Prediction of Complications in Reduction Mammoplasty: A Combined Statistical and Fuzzy Inference Approach in a 10-Year Cohort from Tehran Hospitals (2013–2023)
摘要
Gigantomastia causes physical, psychological, and dermatological issues, often with ptosis; reduction mammoplasty effectively relieves symptoms but carries minor (self-limiting or medically managed) and major (surgically requiring) complications, influenced by factors such as BMI and smoking.
MethodThis retrospective cohort study included 439 patients (878 breasts) who underwent reduction mammoplasty by a single surgeon in Tehran hospitals from 2013 to 2023, with data extracted from medical records across three domains: surgical/anthropometric details, past medical history, and patient characteristics. Complications were categorized as major or minor and analyzed using conventional statistics and an artificial intelligence-based fuzzy inference system to predict postoperative outcomes.
ResultOur analysis of 439 patients (878 breasts) revealed that suprasternal notch-to-nipple distance (SNN), BMI, smoking, underlying comorbidities, and pedicle type were significantly associated with postoperative complications, while age and weight of tissue resected (WTR) showed no significant effect. The free nipple graft technique and greater SNN were linked to the highest complication rates, with only 14 major complications reported overall, including 4 cases of complete nipple–areola necrosis.
ConclusionOur study identifies high BMI, smoking, comorbidities, increased SNN distance, and free nipple graft/inferior pedicle techniques as key predictors of complications in reduction mammoplasty, supporting risk-adapted surgical planning.
Level of Evidence IIIThis journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.