Predicting Breast Reduction Resection Weights and Attainment of Insurance-Mandated Minimums.

Background: Despite evidence that reduction mammaplasty alleviates macromastia symptoms independent of resection weights, resection minimums are commonly used to grant insurance coverage. Multiple formulas have been published to predict resection weights, but very few have evaluated predictive performance relative to attaining common insurance minimums.

Methods: This was a retrospective single-center review of 268 patients from 2007 to 2020. Multiple linear regression and exponential models were created to predict resection weights and attainment of the Schnur, 350-g, and 500-g minimums. Accuracy was compared against published Appel, Descamps, and Galveston equations. Body mass index subgroup analyses were performed.

Results: The exponential model possessed the lowest resection weight prediction error overall (172.8 ± 211.5 g). All equations performed significantly better than surgeons in predicting attainment of the 500-g minimum. None performed better than the surgeons' predictive accuracy in attaining a 350-g minimum. Multiple linear regression and exponential models performed significantly better than surgeons in predicting attainment of the Schnur minimum. Only the exponential model performed significantly better than surgeons for both the Schnur (82% versus 71%; P < 0.01) and the 500-g minimums (89% versus 68%; P < 0.01). On body mass index subgroup analyses, all three minimums were biased in favor of obese women-the least egregious of these was the 350-g minimum.

Conclusions: All minimums were biased against nonobese women. Our exponential model equation based on preoperative sternal notch-to-nipple and nipple-to-inframammary fold distances accurately predicts resection weights and improves on our surgeons' predictive accuracy in attaining the Schnur or 500-g minimums. This may prove useful in the preoperative setting to better counsel patients.