Dietrich, Charles S. III MD; Yancey, Michael K. MD; Miyazawa, Kunio MD; Williams, David L. DO; Farley, John MD
Department of Obstetrics and Gynecology, Tripler Army Medical Center, TAMC, Hawaii.
Address reprint requests to: John Farley, MD, Department of Obstetrics and Gynecology, Tripler Army Medical Center, Jarrett White Road, TAMC, HI 96859–5000; E‐mail: firstname.lastname@example.org.
The views expressed herein are those of the authors and do not reflect the official policy or opinion of the Department of Defense or the United States Army.
Received June 29, 2001. Received in revised form October 5, 2001. Accepted October 18, 2001.
OBJECTIVE: To evaluate risk factors for early cytologic abnormalities and recurrent cervical dysplasia after loop electrosurgical excision procedure (LEEP).
METHODS: A retrospective analysis was performed of all pathology records for LEEPs performed at our institution from January 1996 through July 1998. Follow‐up cytology from 2 through 12 months after LEEP was reviewed. Patients with abnormal cytology were referred for further colposcopic evaluation. Statistical analysis using χ2 test for trend, proportional hazards model test, Fisher exact tests, and life table analysis were performed to identify risk factors for early cytologic abnormalities after LEEP and to determine relative risk of recurrent dysplasia.
RESULTS: A total of 298 women underwent LEEP during the study period, and 29% of these had cytologic abnormalities after LEEP. Grade of dysplasia, ectocervical marginal status, endocervical marginal status, and glandular involvement with dysplasia were not found to be independent risk factors for early cytologic abnormalities. However, when risk factors were analyzed cumulatively, the abnormal cytology rate increased from 24% with no risk factors to 67% with three risk factors present (P = .037). Of patients with abnormal cytology after LEEP, 40% developed subsequent dysplasia, and the mean time to diagnosis was approximately 6 months. The relative risk of subsequent dysplasia ranged from a 20% increase to twice the risk if post‐LEEP cytology was low‐grade squamous intra‐epithelial lesion or high‐grade squamous intraepithelial lesion, respectively.
CONCLUSION: Based on these results, consideration should be given for early colposcopic examination of patients who have evidence of marginal involvement or endocervical glandular involvement with dysplasia. These patients are at increased risk for abnormal cytology and recurrent dysplasia. This initial visit should occur at 6 months, as the mean time to recurrence of dysplasia was 6.5 months.
Loop electrosurgical excision procedure (LEEP) of the transformation zone has become an established modality in the diagnosis and treatment of cervical intraepithelial neoplasia.1–3 The procedure can be quickly and safely performed in a clinical setting and is well tolerated by the patient. Relative to cone biopsy, morbidity is lowered by shorter operative times, elimination of general or regional anesthesia, and reduced blood loss.4,5 In addition, a pathologic specimen is provided for histologic review and confirmation. Often, this specimen is smaller, yet adequate for evaluation, when compared with other conization techniques.4,5 Reported success rates in the treatment of dysplasia with LEEP are high, ranging from 63–97% and are comparable with older ablative procedures.5–8
Incomplete excision of dysplasia has been reported to increase the failure rate.4–11 Controversy exists, however, on the best way to predict residual disease. Although some authors have found that severity of dysplasia or marginal status can be used to predict persistent disease, others have shown no correlation with residual dysplasia and have found margins difficult to interpret.5,8–11 Furthermore, no standardized follow‐up after LEEP has been established, and protocols vary from institution to institution. Some centers only use cytology, whereas others extensively employ colposcopy.8–10
The purpose of our study was to evaluate risk factors for early cytologic abnormalities and recurrent cervical dysplasia after LEEP in a large managed care environment. Additionally, we sought to determine the optimal follow‐up management schema for patients after LEEP.
MATERIALS AND METHODS
A retrospective analysis was performed of all pathology records for LEEPs performed at our institution from January 1996 through July 1998. Routine indications for LEEP at our institution included high‐grade dysplasia, positive endocervical curettage, significant discordance between cytology and histology, and persistent low‐grade lesions. Colposcopic examination immediately before LEEP was performed with a Wallach ZoomScope colposcope to identify the extent of the lesion. Then, LEEP was performed with variable‐sized loop electrodes attached to the electrocautery with a cut setting at 50 W. After removal of the specimen, the base of the cervix was typically cauterized with a ball electrode with coagulation set at 50 W to ensure hemostasis. All LEEPs were performed by staff gynecologists or by residents in Obstetrics and Gynecology or Family Practice under the direct supervision of an attending gynecologist or gynecologic oncologist. The specimen(s) was oriented, placed in formalin, and then reviewed by staff pathologists. Histologic grading of dysplasia was done using standard terminology of mild, moderate, or severe. Endocervical curettage was routinely obtained after LEEP to assess endocervical margin. Patients were instructed to return in 3 months for follow‐up cytologic evaluation. Residents or staff gynecologists performed Papanicolaou smears using a wooden spatula and an endocervical brush with direct application onto glass slide and subsequent fixation.
The preoperative indications for LEEP and the maximum severity of dysplasia on the submitted specimen were determined. In addition, the presence of endocervical glandular involvement with dysplasia and the endocervical and ectocervical marginal status were noted. Follow‐up cytology beginning at 2 months after LEEP and continuing for 12 months after LEEP was then reviewed. Cytology was described using the Bethesda system. Abnormal cytology was defined as atypical squamous cells of undetermined significance (ASCUS), ASCUS favor squamous intraepithelial lesion (SIL), atypical glandular cells of undetermined significance (AGUS), low‐grade squamous intraepithelial lesion, or high‐grade squamous intraepithelial lesion. Patients with abnormal cytology were referred for further colposcopic evaluation at our institution. Patients were excluded from the study if no follow‐up records could be located, or if the initial follow‐up extended beyond 4 months. Patients were also excluded if marginal status could not be determined. Statistical analysis was performed with χ2 test for trend and Fisher exact tests for independence and trend to identify potential risk factors for early cytologic abnormalities after LEEP. Proportional hazards model test was used to determine the risk ratio of recurrent dysplasia. Life table analysis was performed to determine the relative risk of recurrent dysplasia after LEEP in those patients with early cytologic abnormalities.
A total of 509 patients underwent LEEP at our institution during the study period, and 208 (40%) patients were excluded because they were lost to follow‐up. Eighty‐nine (43%) of those lost to follow‐up were excluded because no follow‐up records could be found, and 119 (57%) were excluded because the initial follow‐up examination occurred beyond a 3‐month interval after LEEP. Three more patients were excluded because of an indeterminate marginal status. The study group, therefore, was composed of 298 patients, and histologic margins were involved with dysplasia in 23% (69 of 298) of LEEP specimens (Table 1). The overall rate of cytologic abnormalities after LEEP in our study was 29% (86 of 298) (Table 2). When early follow‐up abnormal cytology was stratified based on the grade of dysplasia present in LEEP specimens, no statistical significance was found between groups (Table 2).
Of the LEEP specimens, 47% (140 of 298) were noted to have the histologic risk factors for recurrent dysplasia of marginal (ectocervical and/or endocervical) or glandular involvement with dysplasia. When the rate of follow‐up abnormal cytology was stratified based on the presence of dysplasia at the histologic LEEP margin or endocervical glandular involvement with dysplasia, no statistical significance was found (P = .097) (Table 1). Ectocervical marginal status, endocervical marginal status, and glandular involvement were not independent risk factors for early cytologic abnormalities. However, the greater the cumulative number of risk factors identified in the LEEP specimen, the greater the risk for abnormal post‐LEEP cytology. This abnormal cytology rate increased from 24% when no risk factors were identified in the LEEP specimen to 67% when all three risk factors were present. This finding was statistically significant (P = .037) (Table 3).
A total of 254 (85%) patients in the study group had further follow‐up data that could be located beyond the initial 3‐month Papanicolaou smear. Subsequent dysplasia was found in 46 of 254 (18.1%) patients (Table 4). This recurrent dysplasia could represent residual or recurrent dysplasia because of the relatively short time of follow‐up. In those patients with normal post‐LEEP cytology, 8.0% had documented dysplasia on follow‐up with the mean time to diagnosis of approximately 8 months (range 2–12 months). Patients who had abnormal cytology on initial follow‐up cytology were found to have persistent or recurrent dysplasia 40.0% of the time (Table 4). The percentage of patients with recurrent dysplasia increased from 30% of patients diagnosed with ASCUS at initial follow‐up cytology to 82% of those with high‐grade squamous intraepithelial lesion on initial follow‐up cytology. The mean time to diagnosis was approximately 6.46 months (range 2–12 months). The rate of recurrence was significantly quicker among those patients with low‐grade squamous intraepithelial lesion and high‐grade squamous intraepithelial lesion on cytology (P < .001). There was no significant increased relative risk for the development of dysplasia among patients who had ASCUS or low‐grade squamous intra‐epithelial lesion cytology (P = .14). If, however, post‐LEEP cytology was high‐grade squamous intraepithelial lesion, there was an increased relative risk of 2.16 for the subsequent development of dysplasia (P < .01) (Table 4).
The overall rate of early cytologic abnormalities after LEEP in this study approaches 30%. This is consistent with that reported in the literature.10 Hanau et al10 looked at 162 LEEP patients and found an overall abnormal cytology rate of 33.3% with a mean length of follow‐up of 10.9 months. Margins were positive in 32.4% of their LEEP specimens.7,10 However, margins could not be accurately evaluated in approximately 30% of their patients, whereas in our study, only three (1%) patients were excluded because of indeterminate margins. The authors in this study also did not look at the rate of persistent histologically proven dysplasia. They concluded that prediction of follow‐up by marginal status was dubious.
The number of LEEP specimens with dysplasia involving the endocervical or ectocervical margins or endocervical glandular involvement was 50% and within the variable rate of 29–50% reported in the literature.4,5,7,9 Variability in the literature is probably a result of technique, the definition of positive margin used by the authors, and whether they included endocervical glandular or endocervical and ectocervical involvement with dysplasia. Although we were not able to identify a single statistically significant risk factor for abnormal initial cytology, we demonstrated that an increasing rate of residual dysplasia was associated with an increasing cumulative number of risk factors found on histologic evaluation of the LEEP specimen. We defined these risk factors as positive endocervical margin for dysplasia, positive ectocervical margin for dysplasia, or endocervical glandular involvement with dysplasia. The presence of these risk factors logically represents the impinging proximity of the lesion to the excision margin. Previous studies have documented the decreased specimen volume obtained with LEEP compared with conization.4,6 Previous reports have also cited severity of dysplasia in LEEP specimen and marginal status as risk factors for recurrent dysplasia.4,6,11 Histologic severity of dysplasia on the LEEP specimen, however, and individual marginal involvement were not independently associated with an increased risk of abnormal cytology or recurrent dysplasia in our study. Those patients, however, who developed abnormal cytology at initial follow‐up visit, were found to have up to twice the risk of persistent/ recurrent dysplasia after LEEP. The mean time to the development of recurrent dysplasia was also significantly shorter among those patients who had initial abnormal cytology after LEEP when compared with patients with normal cytology, 6 months versus 8 months, respectively.
The findings in this study are consistent with some of the previously published conclusions in the literature. Murdoch et al7 reviewed 721 LEEP patients. They noted that a histologic report of incomplete excision increased the risk of residual disease and was greatest when both endocervical and ectocervical margins were involved. Patients at their center who have positive margins are initially followed colposcopically for this reason.7 If initial follow‐up colposcopy is negative, they are discharged to routine care. We, however, noted no independent association between any single risk factor and abnormal initial follow‐up cytology or recurrent dysplasia. Also, the mean time to diagnosis of recurrent dysplasia in our population was 6 to 8 months after LEEP for those patients who had initial normal or abnormal cytology, respectively. As a result, a significant number of women who would eventually develop recurrent dysplasia would be missed if patients with normal cytology were eliminated from follow‐up based on an initial 3‐month cytologic evaluation. Particularly, patients with positive endocervical glandular involvement as their only risk factor for recurrence would be missed with this follow‐up schema. Livasy et al evaluated 248 patients after LEEP and found that positive margins, positive glands, and multiquadrant disease were all predictors of residual/ recurrent dysplasia and warranted close follow‐up with colposcopy.11 Our study's findings are most consistent with the latter; however, we did not find any association between severity of dysplasia and recurrence. Additionally, they did not comment on the interval of initial follow‐up after LEEP. We also did not observe any individual correlation between positive endocervical or ectocervical marginal status, endocervical glandular involvement, and abnormal initial follow‐up cytology or recurrent dysplasia. However, the presence of one or more risk factors found on LEEP specimen to include positive margins or endocervical glandular involvement with dysplasia increased the risk for recurrence.
Finally, it is important to note that 40% of our patients were not included in the final analysis of this study. However, the majority of the patients, 57%, were excluded because their initial follow‐up cytology occurred after the 3‐month interval prescribed by the study. These patients were not “lost to follow‐up” as they did have an evaluation within 6 months of their LEEP. Given that the study was performed to determine the optimal time and type of follow‐up analysis, it was necessary to exclude these patients. Therefore, those patients who were truly lost to follow‐up accounted for only 18% (89 of 509) of our population. The literature reports a lost to follow‐up rate of 20–30% after LEEP.10,11 Even a prospective study evaluating histologic follow‐up after LEEP had 13% of patients lost to follow‐up.8 The percentage of patients lost to follow‐up in our study was probably associated with the transient nature of the population involved in this study.
Based on our study, we propose the following scheme for the postprocedure management of women undergoing LEEP. Those patients who have cumulative evidence of endocervical marginal, ectocervical marginal, and endocervical glandular involvement with dysplasia on their LEEP specimens should have colposcopic evaluation with screening cytology at their initial follow‐up visit. This initial visit should take place at 6 months after LEEP, given the mean time for development of dysplasia in our study was 6.5 months. These patients are at increased risk for abnormal cytology and persistent dysplasia. As a result, performing colposcopic evaluation with screening cytology at initial 6‐month follow‐up quickly identifies these histologic abnormalities, and would save them an additional office visit. This decrease in office visits could positively impact compliance by decreasing any time away from job or family, and also decreasing the possibility of any discomfort that might be caused by the visits. Additionally, as in our population, if adequate follow‐up after cytologic evaluation cannot be assured, then colposcopic evaluation and treatment of high‐risk patients in a timely, organized, and prudent manner would ensure that patients with recurrent dysplasia would not go untreated. Thin‐prep cytology technique could also be considered for initial follow‐up evaluation of patients after LEEP, given the 14% of patients who had ASCUS on initial follow‐up cytology (Table 4). Future studies can be directed at developing a prospective randomized trial comparing cytological evaluation with colposcopic evaluation in patients undergoing LEEP.
1. Leusley D, Cullimore J, Redman C, Lawton F, Emens J, Rollason T, et al. Loop diathermy excision of the cervical transformation zone in patients with abnormal cervical smears. BMJ 1990;300:1690–3.
2. Brigg A, Haffenden D, Sheehan A, Codling B, Read M. Efficacy and safety of large-loop excision of the transformation zone. Lancet 1994;343:32–4.
3. Ferris D, Hainer B, Pfenninger J, Zuber T, DeWitt D, Line R. Electrosurgical loop excision of the cervical transformation zone: The experience of family physicians. J Fam Pract 1995;41:337–44.
4. Mathevet P, Dargent D, Roy M, Beau G. A randomized prospective study comparing three techniques of conization: Cold knife, laser, and LEEP. Gynecol Oncol 1994; 54:175–9.
5. Simmons JR, Anderson L, Hernandez E, Heller P. Evaluating cervical neoplasia: LEEP as an alternative to cold knife conization. J Reprod Med 1998;43:1007–13.
6. Huang L, Hwang J. A comparison between loop electrosurgical excision procedure and cold knife conization for treatment of cervical dysplasia: Residual disease in a subsequent hysterectomy specimen. Gynecol Oncol 1999;73:12–5.
7. Murdoch JB, Morgan PR, Lopes A, Monaghan JM. Histological incomplete excision of CIN after large loop excision of the transformation zone (LLETZ) merits careful follow up, not retreatment. Br J Obstet Gynaecol 1992;99:990–3.
8. Baldauf J, Dreyfus M, Ritter J, Cuenin C, Tissier I, Meyer P. Cytology and colposcopy after loop electrosurgical excision: Implications for follow-up. Obstet Gynecol 1998;92:124–9.
9. Felix J, Muderspach L, Duggan B, Roman L. The significance of positive margins in loop electrosurgical cone biopsies. Obstet Gynecol 1994;84:996–1000.
10. Hanau C, Bibbo M. The case for cytologic follow-up after LEEP. Acta Cytol 1997;41:731–6.
11. Livasy C, Maygraden S, Rajaratnam C, Novotny D. Predictors of recurrent dysplasia after loop electrocautery excision procedure for CIN-3: A study of margin, endocervical gland and quadrant involvement. Mod Pathol 1999;12:233–8.