Assessment of cervical intraepithelial neoplasia (CIN) with colposcopic biopsy and efficacy of loop electrosurgical excision procedure (LEEP)

Abstract

Purpose: Conization for suspected high grade cervical intraepithelial neoplasia (CIN) is often performed based on abnormal cytology only. Loop electrosurgical excision procedure (LEEP) is a very common technique in this context. The present study analyses the accuracy of preoperative assessment of CIN with cytology plus colposcopic biopsy and assesses the efficacy of LEEP for the treatment of CIN.

Methods: Two-hundred and sixty-six consecutive patients treated with LEEP for suspected CIN at our center were retrospectively analyzed. Cytology, HPV-DNA testing, colposcopically directed cervical biopsy and/or endocervical curettage were performed to assess cervical lesions before and 3-6 months after surgery.

Results: Median age of the patients was 34 years. Median follow-up was 50 months. Preoperative HPV testing was positive for high risk types in 77.9%. All patients underwent LEEP without further ablative procedures. Complete excision of the lesion could be achieved in 84.3%; in 13.5% margins were not securely cleared and in 2.2% the lesion was not excised entirely. Overall complication rate was 5.4% (mainly postoperative bleeding and pain). Overall concordance of colposcopic biopsy and cone histology was 85.8%. The concordance rate was higher for CIN 2/3 (95.1%) compared with CIN 1 (63.2%). Nine patients (3.4%) had persistent disease after 3 months, 4 (1.5%) developed disease recurrence and underwent re-conization. HPV testing at 3-6 months after surgery was negative in 78.5%; 2 of the patients developing disease recurrence had a persistent HPV infection after surgery.

Conclusions: Assessment of cervical lesions with colposcopic biopsy is an accurate method (concordance with cone histology 85.8%). Surgical treatment of high grade CIN with LEEP is a safe procedure with low recurrence rates, resulting in a clearance of cervical HPV infection in the majority of cases.