A study of cytological findings in cervical (PAP) smears in correlation with cervical biopsies

Uterine cervix carcinoma is one of the main causes of cancer death in women worldwide. The use of the cervical (PAP) smear as a screening tool has significantly reduced the incidence of cervical cancer. Cytohistopathologic correlation of Pap smears is a widely accepted method of internal quality assurance of laboratories and to assess its utility as a screening test. The aim of our study was to compare the cytological findings in cervical smears with the histopathological features of the corresponding biopsies and calculate the sensitivity, specificity and accuracy rates of cervical cytological examination and to analyse the errors that occurred on reporting of cervical smears. Cervical smears with corresponding histopathological biopsies were reviewed. Accuracy rates of cytological examination, before and after blind review, were calculated and compared. Slides with cyto-histopathologic discrepancies were then reviewed unblinded to analyze the type of errors. Out of 2911 cases sent for cervical cytological evaluation, 210 were available for cytohistopathological correlation. Positive correlation was found in 198 cases and discrepancies in 12 cases. This gave an overall accuracy rate of 94.3%, sensitivity of 68% and specificity of 99.5%. On blind review, cytological diagnosis changed in seven cases. The 12 discrepant cases were analysed again, revealed that on initial reporting there were four screening errors, three interpretive errors, and five sampling errors. Cytohistopathological correlation is essential to assess the quality of cervical cytological examination in the laboratory. Cautious reporting is essential when smears are associated with artifacts due to radiotherapy, inflammation and in smears with low cellularity.