Background: Since its introduction more than 45 years ago, the pelvic wash has gained widespread acceptance and is used routinely use at most centers. However, widely varying figures have been reported regarding its sensitivity for peritoneal involvement in ovarian tumors. In the current study, the authors evaluated a consecutive group of pelvic (peritoneal or abdominopelvic) washings performed in the evaluation of adnexal masses to determine whether histologic subtype significantly affects the tumor detection rate using this procedure.
Methods: Reports from all washes performed over a 5-year period in the evaluation of adnexal masses were evaluated and correlated with those of the synchronously obtained histologic specimens. The sensitivity for each histologic subtype was calculated, with ovarian surface involvement and/or tumoral involvement of any peritoneal surface defined as the criterion standard. Cases with cytologic and histologic concordance were defined as true-positive or true-negative. Statistical significance was determined using the Fisher exact test.
Results: In the current study, 185 of 846 (21.9%) total washes were associated with malignant (n = 161) or borderline (n = 24) tumors involving the ovary. For the malignancies, the overall cytology detection rate was 25%. A comparison of the cytology detection rates for the individual histologic subtypes with the overall rate demonstrated that the serous carcinomas were more likely (P = 0.0144) and the clear cell carcinomas were less likely (P = 0.0452) to be detected in pelvic washings. Cytology detection rates for mucinous, endometrioid, and undifferentiated carcinomas did not appear to differ significantly (P > 0.05) from the average detection rate. The cytohistologic correlation rate (efficiency), sensitivity, and specificity for the 5 most common histologic subtypes (n = 130) were 79.23%, 50.77%, and 93.33%, respectively. Differences also were observed in the calculated sensitivity for each subtype: serous (n = 57), 71.4%; endometrioid (n = 30), 58.33%; clear cell (n = 19), 20%; mucinous (n = 13), 50%; and undifferentiated (n = 11), 50%. Borderline tumors demonstrated a sensitivity and specificity of 80% and 100%, respectively.
Conclusions: In the current study, the pelvic wash was found to be a specific, but only moderately sensitive, technique for detecting peritoneal involvement in ovarian tumors. The histologic subtype of the underlying ovarian tumor was found to have an effect on the likelihood of detection of peritoneal involvement using this diagnostic assay.
Copyright 2004 American Cancer Society.