Prediction of Resection Weight in Reduction Mammaplasty: Validation of the Galveston Scale

Introduction: Reduction mammaplasty (RM) is one of the most commonly performed plastic surgery procedures. Prediction of resection weight is important both for the patients and the surgeons. Current literature suggests that symptomatic relief does not correlate with resection weight, however, insurance carriers are still using resection weight as one of the most important criteria to determine medical necessity. A number of prediction scales have been proposed, none of which are validated and widely accepted. The purpose of our study was to assess the validity of current scales in our patient population and propose a more accurate tool for prediction of resection weight.

Methods: A retrospective chart review of patient that underwent reduction mammaplasty at the University of Texas Medical Branch from 2012-2017 was performed. Multiple regression analyses were applied to all patients operated by the senior author (L.G.P.) to identify independent predictive factors for breast resection weight. A new prediction scale was created. Established prediction scales and the new Galveston scale were then applied to patients operated by different surgeons, excluding the patients that were operated by the senior author, and compared for accuracy of prediction of resection weight. Results were analyzed through linear regression analysis and p-values <0.05 were considered statistically significant.

Results: A total of 184 patients were used for the initial single-surgeon analysis. The new Galveston scale included BMI, breast measurements and age as independent predictive values. 130 patients were included in the multiple-surgeon group for validation of the new scale. There was no overlap in patients between the single- and multiple-surgeon groups. The mean age was 39.2 years, mean BMI 34.5 and the average resection weight was 907 grams. 62.3% of the patients underwent inferior pedicle wise pattern reduction, 26.9% medial or superiomedial pedicle and 10.8% amputation style with free nipple grafting. Galveston scale demonstrated the best predictive value (R² = 0.71). Appel and Descamps performed worse with R²=0.69 and R²=0.68 respectively. Schnur scale demonstrated the poorest prediction value with R²=0.28.

Conclusion: Prediction of resection weight in RM remains important for patient counseling and as an adjunct tool for the plastic surgeon preoperatively and intraoperatively, as a guide to estimate the amount of tissue to be resected. We recommend a patient-specific and surgeon-specific approach, instead of the “one-scale-fits-all” paradigm. We propose the Galveston scale for older patients with higher BMI and breasts requiring large resections. Symptomatic relief does not correlate with amount of tissue removed and medical necessity should be based on patient symptoms, physical examination and the physician’s clinical judgment.