Pathologic Correlation between Abnormal Endocervical Curettage (ECC) and the Follow-Up Loop Electrosurgical Excision Procedure (LEEP) and Cold Knife Conization
X Gao, JG Lake, G Birdsong. Emory University of Medicine, Grady Memorial Hospital, Atlanta, GA; Emory University School of Medicine, Atlanta, GA
Background: Diagnosis of ECC specimens can be challenging for pathologists because specimens are typically small, poorly orientated tissue fragments. Nevertheless, positive results have significant clinical implications. This study compares the diagnoses of positive ECCs and the corresponding follow-up LEEP/conizations specimens.
Design: 170 specimens accessioned between 2004 and 2008 were selected. Selection criteria were positive ECC, no concurrent ectocervical biopsy during the colposcopy, and follow-up with LEEP/conization within 1 year of the ECC. ECC diagnoses include squamous dysplasia-cannot be graded, CIN I-III, and invasive lesions. Exactly matching ECC and LEEP were categorized as consistent. CIN I ECC followed by CIN II or higher LEEP was regarded as underscored; CIN II or higher ECC followed by a lower grade or negative LEEP was regarded as overscored. A diagnosis of "squamous dysplasia, cannot be graded" was arbitrarily interpreted as CIN II-III. Significant discordancy was designated as CIN I in the ECC with CIN II or higher LEEP, or CIN II or higher ECC with negative or CIN I LEEP. Bivariate significance tests were used to analyze the possible relationship between discordant results and patient age, time between ECC and LEEP, the grade of dysplasia in ECC, and whether the same pathologist reviewed the ECC and LEEP.
Results: 86 of 170 (51.6 %) patients had discordant results; of those, 54 were overscored and 32 underscored. Significantly discordant overscored diagnoses were found in 24 of 54 (44.4%) ECC; of those, the majority patients were 40 or older (16 of 24). Of 32 underscored ECC patients, significant discordancy was found in 24 (75%), and the majority were < 40 years old (18 of 24). An ECC with CIN I or II was more likely to be discordant than an ECC with CIN III. (p < 0.05). In addition, ECCs from patients 40 or younger were more likely to be discordant and to be underscored (p < 0.05). The other parameters motioned above were not significantly associated with discordant diagnoses.
Conclusions: Diagnoses for ECCs and subsequent LEEP/conization specimens are discordant in a substantial percentage of cases. ECCs with lower grade dysplasia or from younger patients are associated with higher rate of discordant diagnoses. ECCs from younger patients are also at higher risk to be underscored than those from older patients.
Category: Gynecologic & Obstetrics
Tuesday, March 23, 2010 9:30 AM
Poster Session III # 172, Tuesday Morning