Positive Margins After Conization and Risk of Persistent Lesion

Abstract

It has been suggested that when pathologic studies show positive margins after conization, it may not be necessary to perform additional surgery in all patients. This study investigated an alternative method of treatment for patients with persistent lesions after conization. Five hundred five women underwent conization by the same surgeon between 1976 and 1998. Histological examination of the cone specimen in these 505 patients showed positive margins in 71 (14 percent). Fifty specimens had involvement of the endocervical margin, and 21 had involvement of the exocervical margin. Twelve women were treated with hysterectomy within 2 months of conization, 10 for microinvasive carcinoma and 2 for carcinoma in situ (CIN). Fifty-nine women underwent follow-up monitoring with routine clinical examination and cervical cytologic tests at 3-month intervals for 1 year after the initial cone biopsy. Twelve of these patients, all of whom had normal findings on examination at 3 months, failed to return after the first follow-up visit, leaving 47 patients who completed the 1-year period. Indications for conization in the 71 study patients included CIN3 (48 women), CIN2 (20 women), and CIN1, condylomata, and microinvasive carcinoma (1 woman each). One patient had a histological diagnosis of condylomata, 10 had CIN1 or 2 (14.1 percent ), 47 had CIN3 (66.2 percent ), and 12 had microinvasive carcinoma (16.9 percent ). These diagnoses agreed with the preoperative biopsy results in 80 percent of cases. Average length of follow-up was 38.5 months. None of the 12 patients who had hysterectomy as the initial treatment had a recurrence of the tumor. Nine of the 47 patients (19 percent ) followed by monitoring developed a recurrence of their disease an average of 18 months after cone biopsy. Two of these women underwent laser ablation for exocervical lesions, three underwent a second conization for endocervical lesion, and two underwent hysterectomy for endocervical lesions. Two patients, one with CIN2 and human immunodeficiency virus and one with CIN3, refused further treatment and were followed with monitoring. Two of the seven patients who underwent additional surgery because of a recurrence had a second recurrence. One woman with CIN3 that first recurred at 15 months was diagnosed with CIN1 42 months after laser treatment. She received no further treatment and continued to be monitored. The other patient, who also had CIN 3, had a recurrence 8 months after conization, for a first recurrence at 13 months. She refused hysterectomy and instead underwent cervical amputation. Nineteen of the 47 patients had persistent lesions, that is, they continued to have abnormal cervical smears during monitoring. One patient refused additional treatment, three had a second conization, six had laser treatments, and nine underwent hysterectomy. In all, 25 of the 47 patients being monitored underwent a second treatment, 7 for recurrent lesions and 18 for persistent lesions. Histological examination of the surgical specimen in the 17 patients who were treated surgically (8 were treated with laser) found no persistent lesion in 5 patients (29 percent). Of the 12 women who underwent immediate hysterectomy, no lesions were found in 8 patients (66.7 percent). This difference in the number of persistent lesions in monitored women compared with immediately treated women is significant (P = .04). In the group of 47 women who were monitored, a comparison of the 28 patients who had persistent or recurrent lesions on follow-up with the 19 patients who had consistently normal findings showed that positive endocervical margins were significantly associated with persistent or recurrent disease. Seventy-nine percent of the women with positive endocervical lesions had a recurrence or a persistent lesion, compared with 48 percent of the women with normal findings on follow-up. Gynecol Oncol 2000;76:311–314