[Quality control of stereotaxic cytologic test of non-palpable lesions of the breast].
The authors compare the accuracy and other indicators of efficiency of ten centers (1,784 total cases) performing stereotaxic cytology and adhering to a multicentric study aimed at validating a quality control system for this diagnostic procedure. The results from single centers were compared with average results. No significant differences were observed for most centers from the average sensitivity (86%), specificity (91%), positive predictive value of a dubious (57%) or positive (96%) cytologic report, inadequacy rate on malignant (6%) or benign (11%) lesions, and benign/malignant biopsy ratio (0.63). Some significant differences from the average values indicated the need for some centers to review different phases of the diagnostic process, namely to verify the accuracy of sampling sites (low sensitivity with good specificity and predictivity), to review the criteria for cytologic diagnosis (specificity less than 90%, low predictive value or higher predictive value for dubious compared with positive reports), to optimize the impact of cytology on the final decision (lack of reduction of benign/malignant biopsy ratio). Periodic check of the above parameters is proposed as a routine quality control of this diagnostic procedure.