Development and Validation of a Diagnostic Histopathological Scoring System for Capsular Contracture Based on 720 Breast Implant Capsules.

BACKGROUND

Capsular contracture is traditionally evaluated with the Baker classification, but this has notable limitations regarding reproducibility and objectivity.

OBJECTIVES

The aim of this study was to develop and validate procedure-specific histopathological scoring systems to assess capsular contracture severity.

METHODS

Biopsies of breast implant capsules were used to develop histopathological scoring systems for patients following breast augmentation and breast reconstruction. Ten histological parameters were evaluated by multivariable logistic regression to identify those most associated with capsular contracture. Significant parameters (P < .05) were selected for the scoring systems and assigned weighted scores (1-10). Validation was assessed from the area under the curve (AUC) and the mean absolute error (MAE).

RESULTS

A total of 720 biopsies from 542 patients were included. Four parameters were selected for the augmentation scoring system, namely, collagen layer thickness, fiber organization, inflammatory infiltration, and calcification, providing a combined maximum score of 26. The AUC and MAE for the augmentation scoring system were 81% and 0.8%, which is considered strong. Three parameters were selected for the reconstruction scoring system, namely, fiber organization, collagen layer cellularity, and inflammatory infiltration, providing a combined maximum score of 19. The AUC and MAE of the reconstruction scoring system were 72% and 7.1%, which is considered good.

CONCLUSIONS

The new histopathological scoring systems provide an objective, reproducible, and accurate assessment of capsular contracture severity. We propose these novel scoring systems as a valuable tool for confirming capsular contracture diagnosis in the clinical setting, for research, and for implant manufacturers and insurance providers in need of a confirmed capsular contracture diagnosis.