Agreement between cytology, colposcopy and cervical histopathology - Revista Brasileira de Ginecologia e Obstetrícia

PURPOSE: to evaluate the agreement between conventional cytology using the Papanicolaou test, repeated at the time of colposcopy, with colposcopic and histopathological findings. METHODS: the study was carried out at the central public health laboratory of the state of Pernambuco between January and July, 2008, involving 397 women referred for colposcopic evaluation following an abnormal cervical smear test. Cytology was repeated at the time of colposcopy using conventional method, with particular attention being paid to the presence of abnormal colposcopic findings. The nomenclature used for cytology was the 2001 Bethesda system terminology, while that used for histology was the World Health Organization 1994 classification. Cytology performed at the time of colposcopy was compared with colposcopy and with histopathology obtained by colposcopy-directed biopsy. The Kappa coefficient was used to evaluate the agreement between methods, as well as the χ2 test, with the level of significance set at 5%. RESULTS: poor agreement was found between cytology performed at the time of colposcopy and colposcopic findings (K=0.33; 95%CI=0.21-0.45) and between colposcopy and histopathology (K=0.35; 95%CI=0.39-0.51). Cytology performed at the time of colposcopy compared with histopathology revealed a Kappa of 0.41 (95%CI=0.29-0.530), which was considered to reflect moderate agreement. CONCLUSIONS: agreement was better between cytology and histopathology than between colposcopy and cytology or between colposcopy and histopathology.