A subjective rating scale for evaluating the appearance outcome of autologous breast reconstruction

Subjective evaluations of the appearance outcome of autologous breast reconstruction are usually performed by surgeons and not by patients. Such surgeon-based evaluations are rarely reproducible and show little interobserver agreement. Among existing patient-based subjective scales, none has been tested for reliability, and no study to date has evaluated the reliability when both surgeons and patients use the same scale.

The authors developed a new instrument for assessing the appearance of autologous breast reconstruction. The instrument's use by four plastic surgeons and 36 transverse rectus musculocutaneous flap patients was assessed for test-retest reliability, internal consistency, surgeon-patient and surgeon-surgeon interobserver agreement, and interitem correlation.

The instrument demonstrated high overall internal consistency when used by patients (Cronbach alpha = 0.92). Test-retest reliability on each aesthetic subitem in the scale was higher among patients than among surgeons (weighted kappa range, 0.57 to 0.88 versus 0.25 to 0.66). Interrater agreement was poor among both patients and surgeons (weighted kappa, 0.0 to 0.39). Poor correlation was found between surgeons' evaluations of aesthetic subitems and patients' overall appearance and overall satisfaction scores.

The instrument both demonstrates better internal consistency and is more reliable when used by patients to evaluate their own reconstructions. By contrast, the instrument's use by surgeons is not as internally consistent and reproducible. The poor interobserver agreement among surgeons and the weak correlation between surgeon and patient evaluations suggest that patient input regarding item-specific criteria should be included in evaluations of breast reconstructions.