OBJECTIVE: To analyze the performance of intraoperative cytological diagnosis and its contribution to enhance the diagnostic concordance of intraoperative procedures with definitive histological diagnosis in ovarian lesions. Descriptions of cytologic features of some ovarian lesions as seen in imprint and smear specimens are documented. Advantages and limitations of cytological specimens are underscored. STUDY DESIGN: We performed a retrospective review of intraoperative consultations of ovarian neoplastic and non-neoplastic lesions of 337 cases identified from the pathologic records of our laboratory. All cases evaluated had intraoperative cytological imprint or smear specimens. The intraoperative diagnosis transmitted to the surgeon was the combined result of a thorough macroscopic study of the surgical specimens, frozen section analysis of tissue samples obtained from the most representative areas, evaluation of smear and imprint cytological specimens, and cyto-histological correlation. The intraoperative diagnosis was compared with the final histological diagnosis. Blinded review of cytological slides in discrepant cases was carried out. RESULTS: Final histological diagnosis showed 268 benign lesions, 49 malignant tumors and 20 borderline epithelial tumors. Diagnostic accuracy was 97%, sensitivity 93%, specificity 98%, positive predictive value 91% and negative predictive value 98%. After the cytological review of the discrepant cases, accuracy increased to 98.5%, specificity to 100%, and positive predictive value to 100%. The sensitivity and negative predictive values did not change, remaining 93 and 98%, respectively. The proportion of false positives dropped from 7 (2.1%) to 0, and the proportion of false negatives from 9 (2.7%) to 5 (1.5%). The new values showed significant differences for specificity (p=0.0401), positive predictive value (p=0.0479) and for the proportion of false positives (p=0.0226). Cytologic evaluation contributed with wider sampling of tumors and excellent cyto-morphological details. CONCLUSION: Adding cytological evaluation of imprint and smear specimens and cyto-histologic correlation to the traditional gross examination and frozen sections results in better correlation of the intraoperative consultation with the definitive histological diagnosis. In this series cytology was more accurate than frozen sections in some cases. Cytological evaluation of imprints and smears should be considered an important complementary tool in the setting of intraoperative consultation.
OBJECTIVE: To analyze the performance of intraoperative cytological diagnosis and its contribution to enhance the diagnostic concordance of intraoperative procedures with definitive histological diagnosis in ovarian lesions. Descriptions of cytologic features of some ovarian lesions as seen in imprint and smear specimens are documented. Advantages and limitations of cytological specimens are underscored. STUDY DESIGN: We performed a retrospective review of intraoperative consultations of ovarian neoplastic and non-neoplastic lesions of 337 cases identified from the pathologic records of our laboratory. All cases evaluated had intraoperative cytological imprint or smear specimens. The intraoperative diagnosis transmitted to the surgeon was the combined result of a thorough macroscopic study of the surgical specimens, frozen section analysis of tissue samples obtained from the most representative areas, evaluation of smear and imprint cytological specimens, and cyto-histological correlation. The intraoperative diagnosis was compared with the final histological diagnosis. Blinded review of cytological slides in discrepant cases was carried out. RESULTS: Final histological diagnosis showed 268 benign lesions, 49 malignant tumors and 20 borderline epithelial tumors. Diagnostic accuracy was 97%, sensitivity 93%, specificity 98%, positive predictive value 91% and negative predictive value 98%. After the cytological review of the discrepant cases, accuracy increased to 98.5%, specificity to 100%, and positive predictive value to 100%. The sensitivity and negative predictive values did not change, remaining 93 and 98%, respectively. The proportion of false positives dropped from 7 (2.1%) to 0, and the proportion of false negatives from 9 (2.7%) to 5 (1.5%). The new values showed significant differences for specificity (p=0.0401), positive predictive value (p=0.0479) and for the proportion of false positives (p=0.0226). Cytologic evaluation contributed with wider sampling of tumors and excellent cyto-morphological details. CONCLUSION: Adding cytological evaluation of imprint and smear specimens and cyto-histologic correlation to the traditional gross examination and frozen sections results in better correlation of the intraoperative consultation with the definitive histological diagnosis. In this series cytology was more accurate than frozen sections in some cases. Cytological evaluation of imprints and smears should be considered an important complementary tool in the setting of intraoperative consultation.
Authors: Nithya D G Ratnavelu; Andrew P Brown; Susan Mallett; Rob J P M Scholten; Amit Patel; Christina Founta; Khadra Galaal; Paul Cross; Raj Naik Journal: Cochrane Database Syst Rev Date: 2016-03-01