COLPOSCOPICALLY DIRECTED PUNCH BIOPSY - A POTENTIALLY MISLEADING INVESTIGATION

Objective-To determine the relation between the histology of an initial colposcopically directed punch biopsy and a subsequent diathermy loop excision biopsy of the transformation zone, and the effect of lesion size on this relation. Design-Prospective observational study of loop diathermy excision biopsies of the cervical transformation zone. Setting-Academic unit colposcopy clinics at The Birmingham and Midland Hospital for Women and Dudley Road Hospital, Birmingham. Subjects-243 women managed by colposcopy and directed punch biopsy followed by loop diathermy excision. Outcome measures-The histology of the punch biopsy and the excised transformation zone and the size of the abnormal lesion. Results-In 132 (54%) of the 243 women the histology of the punch biopsy and loop excision specimen did not agree. In 62 (47%) of these 132 women a more severe lesion was found in the excised transformation zone, including three unsuspected adenocarcinoma in situ and one stage Ia1 cancer. In 39 (41%) of the 96 women in whom a lesion of CIN 3, or greater severity, was found in the loop excision specimen, the paired punch biopsy had suggested a lesion of lesser severity. In small area lesions, the punch biopsy was more likely to show more severe disease than the loop excision specimen (P = 0.0014). Conclusions-The results suggest that directed punch biopsy in an inadequate endpoint by which to judge the severity of an epithelial lesion. The findings have important implications for patient management and the design of trials and cast doubt on the results of studies using punch biopsy as an endpoint.