Endocervical curettage at conization to predict residual cervical adenocarcinoma in situ.

OBJECTIVE: To determine if performing an endocervical curettage (ECC) at the time of conization is a useful diagnostic tool for predicting residual cervical adenocarcinoma in situ (AIS) among women who might wish to preserve their fertility. METHODS: All patients diagnosed with AIS from 1995 to 2000 at four institutions were identified. Data were retrospectively extracted from clinical records. Women included in the statistical analysis were (1) younger than 40 years, (2) had an ECC performed at the time of the initial cone biopsy, (3) had a clearly demarcated surgical margin pathologically, and (4) underwent a second surgical procedure. RESULTS: Twenty-nine (24%) of 123 AIS patients met criteria for inclusion. The median age was 33 years (range, 17 to 39) and 13 (46%) were nulliparous. Initial surgery was a cold-knife conization (n = 17) or loop electrosurgical excision procedure (n = 12). Twelve (41%) ECCs and 15 (52%) cone margins were histologically positive. Sixteen patients underwent a repeat conization; 13 underwent hysterectomy. Thirteen (45%) patients had residual AIS at the time of their second surgical procedure. ECC had a superior positive predictive value (100% vs 47%; P < 0.01) and negative predictive value (94% vs 57%; P = 0.01) compared to cone margin in predicting residual AIS. None of the women undergoing fertility-sparing surgery developed recurrent AIS or adenocarcinoma. CONCLUSION: ECC performed at the time of conization may be a useful tool for predicting residual AIS in women considering fertility preservation.

Cytologic and biopsy findings leading to conization in adenocarcinoma in situ of the cervix.

OBJECTIVE: To investigate the utility of currently available screening tests in preoperatively detecting adenocarcinoma in situ of the cervix. METHODS: Patients with a cone biopsy diagnosis of adenocarcinoma in situ from 1987 to 2000 at our institution were identified. Results from Papanicolaou smears, cervical biopsies, and endocervical curettages preceding the diagnostic cone biopsy were collected from medical records and referring providers. Fisher exact test (two-tail) was used for statistical analysis. RESULTS: The preoperative screening results preceding a cone biopsy containing adenocarcinoma in situ were available in 118 patients. Among 94 Papanicolaou smears, 65 (69%) glandular lesions and 29 (31%) squamous or unspecified lesions were reported. Biopsy and/or endocervical curettage after the 29 squamous or unspecified lesions on Papanicolaou smear detected 15 additional glandular lesions, totaling 80 (85%) of 94 cases of glandular disease detected before conization. Among all 118 cases with some form of preoperative data available, glandular disease was predicted in 100 cases (85%). In cases of suspected glandular disease, 86% were treated with cold knife cone compared with 22% in cases of suspected squamous abnormalities (P <.001). CONCLUSION: The sensitivity of detecting a glandular abnormality before a cone biopsy containing adenocarcinoma in situ is 69% with the Papanicolaou smear and 85% with the addition of biopsy and endocervical curettage. This underscores the importance of using preoperative assessment to appropriately plan treatment for a suspected glandular lesion.