A RANDOMIZED PROSPECTIVE-STUDY COMPARING 3 TECHNIQUES OF CONIZATION -COLD KNIFE, LASER, AND LEEP

Three different techniques of cervical excision, cold knife conization , laser conization, and loop electrosurgical excisional procedure (LEE P) were prospectively compared with respect to treatment reliability, effectiveness, and safety. One hundred ten women with CIN1-2 and the s quamnocolumnar junction not seen or CIN3 at the original diagnosis wer e randomized to treatment with cold knife conization (n = 37), laser c onization (n = 37), or LEEP (n = 36). All three treatments were perfor med with local anesthesia on an outpatient basis. The mean age, histol ogic features (original and histology of the conization), endocervical involvement, and ectocervical extension were similar in the three gro ups. Blood loss and operating time were less (P < 0.01) in the LEEP gr oup (5.4 cc of mean blood loss and 5.4 min mean duration time) than in the two other groups (16.2 cc and 14.0 min for cold knife conization, 21.5 cc and 15.6 min for laser conization). Volumes of the cones were evaluated: LEEP cones and laser cones were smaller than the cold knif e cones (P < 0.001). During the pathological review of the conization, the major problem was difficulty in evaluating the lesion and its mar gins due to the coagulation induced by the laser or the LEEP. This alt eration was present in 53% of the LEEP conization specimens and in 51% of the laser conization specimens. In the majority of the cases the c oagulation was mild, but in one case (LEEP group) and in two cases (la ser group) the conization was totally altered by the coagulation, and in 31% of all the LEEP conizations and 38% of all the laser conization s, evaluation of the entire margin was not possible due to coagulation of the tissue. During postoperative follow-up, the number of complica tions was the same in the three groups (two episodes of post-operative bleeding in each of the three groups). Two months after the treatment the cervix was evaluated: the os was diminished in the cold knife gro up compared to the two other groups and as a result, the squamnocolumn ar junction was not seen in entirety in 50% of cold knife cases, in 19 % of LEEP cases, and in 20% of laser cases. These results suggest that in our hands: (1) laser conization is relatively costly and time cons uming and alters the tissues significantly, and (2) the choice between cold knife and LEEP is more difficult-cold knife gives a sample adequ ate for histological evaluation (including evaluation of the margins), while the LEEP procedure is technically easier and less time consumin g but sometimes induces electrocautery artifact so that evaluation of the margins is not possible. (C) 1994 Academic Press, Inc.