- Email: [email protected]
Recurrence of dysplasia after loop electrosurgical excision procedures with long-term follow-up
Recurrence of dysplasia after loop electrosurgical excision procedures with long-term follow-up Capt Dagoberto I. Gonzalez, Jr, USAF, MC, Lt Col Christopher M. Zahn, USAF, MC, Maj Matthew G. Retzloff, USAF, MC, Maj William F. Moore, USAF, MC, Lt Col Edward R. Kost, USAF, MC, and Col Russell R. Snyder, USAF, MC(retired) San Antonio, Texas OBJECTIVE: The aim of this study was to determine the rates of recurrent dysplasia with longer follow-up durations and to determine whether margin status and other variables were associated with recurrence. STUDY DESIGN: A retrospective chart review was performed for all women who underwent a loop electrosurgical excision procedure at Wilford Hall Medical Center, Lackland Air Force Base, Texas, between January 1993 and December 1994. Extracted information included age, parity, indication for the loop electrosurgical excision procedure, histologic classification of the loop electrosurgical excision procedure specimen, margin status, and whether a “deep” (endocervical) pass had been performed. Follow-up data included findings of repeated cytologic examination, colposcopy, and biopsy if performed. RESULTS: The mean duration of follow-up for all women was 24 months. Margins were positive in 28%, with 73% of these being endocervical. The overall recurrent dysplasia rate was 31%, with a mean time to recurrence of 11.9 months. Participants with any positive margins had a higher recurrence rate than did those with negative margins (47% vs 26%; P = .009). High-grade lesions at the margin were more commonly associated with recurrence than were low-grade lesions relative to those with clear margins (high-grade lesion vs negative margins, 55% vs 26%; P = .003; low-grade lesion vs negative margins, 36% vs 26%; P = .34). Recurrence was not associated either with the performance of an endocervical pass or with the histologic diagnosis of the loop electrosurgical excision procedure specimen. CONCLUSION: With comprehensive long-term follow-up, positive margins on loop electrosurgical excision procedure specimens were shown to be a risk factor for recurrence of cervical dysplasia, particularly when high-grade lesions were seen at the margin. Recurrence was also considerable among women with negative margins. Women should be counseled regarding this risk, and the importance of follow-up should be emphasized. (Am J Obstet Gynecol 2001;184:315-21.)
Key words: Loop electrosurgical excision procedure, positive margins, negative margins
Loop electrosurgical excision procedures (LEEP), including large loop excision of the transformation zone procedures, have been used extensively in Great Britain and have gained acceptance in the United States as a management option for the treatment of cervical dysplasia. These procedures have been used as a treatment alternative, and their efficacy as a diagnostic tool, including as an alternative to cold-knife cone biopsy, has also been fairly well established.1-17 Efficacy rates for LEEP are comparable to those of ablative techniques, such as laser therapy.6, 15 An advantage of LEEP relative to ablative techniques is the acquisition of a tissue specimen for histopathologic diagnosis, which has been suggested to
From the Department of Obstetrics and Gynecology, Wilford Hall United States Air Force Medical Center and Brooke Army Medical Center. Received for publication December 27, 1999; revised May 16, 2000; accepted July 7, 2000. Reprint requests: Capt Dagoberto I. Gonzalez, Jr, USAF, MC, 2200 Bergquist Dr, Suite 1, Lackland Air Force Base, TX 78236. 6/1/109937 doi:10.1067/mob.2001.109937
be potentially more accurate than punch biopsy of the cervix.9, 17, 18 In addition, LEEP have been reported to allow diagnosis of unsuspected cancer that might otherwise have been inappropriately treated with an ablative technique.1, 3, 6-8, 13, 17, 18 Although the overall efficacy of LEEP in the treatment of cervical dysplasia has been reported to be >90%, data are difficult to compare because of differences among studies. Variations include definitions of cure, length of follow-up, margin reporting and status, and method of follow-up, including cytologic examination only or cytologic examination with colposcopy and biopsy as indicated. In addition, several studies have reported efficacy rates for a “see and treat” approach without previous biopsy, whereas others have reported efficacy rates after a biopsy-confirmed diagnosis. The purpose of this study was to evaluate recurrence of dysplasia after LEEP in women with previous biopsyconfirmed dysplasia with a prolonged follow-up interval. We attempted to identify factors associated with an increased risk of recurrence, such as margin status, perfor315
316 Gonzalez et al
February 2001 Am J Obstet Gynecol
Table I. Indications for all LEEP performed Indication
No.
%
Persistent cervical intraepithelial neoplasia I Cervical intraepithelial neoplasia II Cervical intraepithelial neoplasia III Unexplained Papanicolaou test result Positive result of endocervical curettage Inadequate colposcopy
11 88 46 1 32 6
6 48 25 0.5 17 3
mance of a deep endocervical pass, status of endocervical curettage, and grade of dysplasia in the LEEP specimens. Material and methods This study received investigative review board approval at the participating institution. A retrospective chart review was performed on all women who underwent large loop excision of the transformation zone procedures and LEEP at Wilford Hall Medical Center, Lackland Air Force Base, Texas, between January 1993 and December 1994. One hundred ninety-nine women were eligible for the study. Fifteen participants were excluded because of inadequate follow up, significant glandular atypia, or uninterpretable margins. All procedures were performed either by obstetrics and gynecology residents with staff supervision or by gynecology or gynecologic oncology staff. Extracted data included age, parity, indication for LEEP, colposcopic examination at the time of the procedure, LEEP diagnosis, performance of a deep endocervical pass, margin status, follow-up diagnosis, and duration of follow-up. As shown in Table I, most procedures were performed as treatment for high-grade dysplasia. Approximately 20% of the procedures were performed for diagnostic indications traditionally reserved for coldknife cervical conization. Colposcopic evaluation was performed after application of 5% acetic acid solution to the cervix. Lugol iodine solution was applied at the discretion of the staff physician. Local anesthetic was then administered; 4 to 5 mL of 1% lidocaine hydrochloride with epinephrine at a 1:100,000 dilution mixed with 4 to 5 mL of 1% lidocaine hydrochloride without epinephrine was injected directly into the cervix in a circumferential pattern. Loop procedures were performed with a CooperSurgical model 6000 (CooperSurgical, Inc, Shelton, Conn) device and disposable stainless-steel wire loops (CIRCON Corporation, Santa Barbara, Calif). The procedures were performed with a blend setting and a power output of 40 W. Ninetyfive percent of the excisions were done with a 2.0 × 0.8-cm loop, and a 1.0 × 1.0-cm loop was used for deep passes when performed. The larger loop was used for a central pass that removed the entire transformation zone. Additional ectocervical passes were done with appropriately sized loops as dictated by colposcopic findings. A deep pass of the endocervix was performed for women with a
positive result of endocervical curettage, visible lesions extending into the cervical canal, or an inadequate colposcopic examination. A deep pass was also performed in additional cases at the discretion of the staff physician. An endocervical curettage was then performed with a Kevorkian endocervical curette. A 5-mm cautery ball with a power setting of 50 W coagulation was used to achieve hemostasis. Either Monsel solution or silver nitrate or both of these were applied as needed. All specimens were marked for orientation with a delayed absorbable suture at the 12 o’clock position. Specimens were marked fresh with indelible ink (Becton Dickinson Microbiology Systems, Sparks, Md) to identify appropriate margins. If deep passes were obtained, the true endocervical margin was also inked. Specimens were then placed in a 10% formalin solution and submitted for histopathologic examination. Specimens were cut into 5-µm sections with 3 levels through the block and stained with hematoxylin and eosin. A single investigator (Christopher M. Zahn) reviewed >95% of the specimens. A margin was considered positive when human papillomavirus (HPV) change or dysplasia, regardless of grade, was present at the inked margin or at the cauterized margin if ink was not identified. The result of endocervical curettage was considered positive if any signs of dysplasia or any histopathologic changes consistent with HPV were present. All women underwent a postoperative examination at 4 to 6 weeks and were then followed up at 4- to 6-month intervals. Follow-up visits, except for the postoperative examination, included a Papanicolaou smear, colposcopy, and biopsy as indicated. If subsequent evaluations showed normal results for 1 year, women were no longer seen at the colposcopy clinic and were followed up with yearly Papanicolaou smears. In cases of residual abnormalities, colposcopic examinations with directed biopsies were continued, and treatment procedures such as repeated LEEP, laser ablation, or cold-knife conization were performed as indicated. Diagnoses of the LEEP specimens and any follow-up tissue samples were recorded as normal, as HPV changes, or as mild, moderate, or severe dysplasia. The abnormal diagnoses were additionally classified as either low-grade lesions (HPV changes or mild dysplasia) or high-grade lesions (moderate or severe dysplasia). Normal diagnoses included reactive changes, reactive atypia, and inflammation. Recurrence or persistence of dysplasia was recorded for the entire study group. Abnormalities at follow-up were also compared between women categorized according to margin status, performance of a deep endocervical pass, and the diagnosis (low-grade vs high-grade lesion) of the original LEEP specimens. Statistical analysis was performed with the χ2 test or the Fisher exact test as appropriate. Statistical significance was defined as P < .05.
Gonzalez et al 317
Volume 184, Number 3 Am J Obstet Gynecol
Table IV. Performance of a deep pass and dysplasia at follow-up
Table II. Relationship between margin status (all margins) and dysplasia at follow-up Positive margins Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
Negative margins
No.
%
No.
%
Statistical significance
24 17 7
47 33 14
29 23 6
26 21 5.5
P = .009 P = .09 P = .12
Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
With deep pass
Without deep pass
No.
%
No.
%
Statistical significance
18 14 4
30 23 7
37 27 10
30 22 8
P = .94 P = .88 P > .99
Table III. Relationship between status of endocervical margin and dysplasia at follow-up
Table V. Comparison of low-grade versus high-grade lesions on LEEP specimens and dysplasia at follow-up
Positive margins
Low-grade lesions
High-grade lesions
No.
%
No.
%
Statistical significance
20 17 3
27 23 4
33 23 10
38 27 12
P = .12 P = .55 P = .08
Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
Negative margins
No.
%
No.
%
Statistical significance
18 12 6
49 32 16
35 28 7
28 23 5.6
P = .02 P = .22 P = .08
Results A total of 184 women met the criteria for the study. The mean age was 28.4 years (range, 16-66 years), and the mean length of follow-up was 24 months (range, 3-59 months). Nulliparous women accounted for 42% of the study group. Subsequent procedures were performed in 16 women (9%), including repeated LEEP in 10 cases, cold-knife cone biopsy in 5 cases, and laser ablation in 1 case. An endocervical curettage was done at the time of the LEEP, and results were positive in only 4 cases (2%). Inadequate colposcopy at the time of the LEEP was noted in only 3 cases; 1 of these women was found to have mild dysplasia on subsequent follow-up evaluation. The LEEP specimen had normal findings in 23 participants (12.5%). These women were therefore excluded from margin status consideration as well as from analysis of recurrent dysplasia according to the diagnosis of the LEEP specimen. In this group of women with negative LEEP histologic results, there were 4 recurrences (17%), 3 of which were low grade and 1 of which was high grade. Among the remainder of the LEEP specimens, the vast majority were considered to show dysplasia. Although HPV changes were included in the low-grade lesion category, of the 161 LEEP specimens with pathologic findings, only 7 were considered to show HPV changes only. Recurrence of dysplasia in the total study group varied according to the length of follow-up. At 6 months the recurrence rate was only 9.2%; during the total length of follow-up for the study, however, recurrent dysplasia was identified in 57 women (31%). Of the cases of recurrent dysplasia, 14 (25%) were high-grade lesions. The interval to recurrence for the overall group ranged from 2 to 48 months, with a mean of 11.9 months.
Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
Positive margins were identified in 51 cases (28%). Although both positive ectocervical and positive endocervical margins were included, 73% (37 of 51 women) had positive endocervical margins. Among those with positive margins, 57% had high-grade lesions in the margins, and only 3 specimen margins were considered positive on the basis of HPV changes only. The relationship between recurrent dysplasia and margin status of the LEEP specimen is demonstrated in Tables II and III. Positive margins were further categorized as demonstrating either low-grade or high-grade lesions. Relative to those with negative margins, recurrent dysplasia was more commonly associated with high-grade positive margins (55% vs 26%; P = .003) than with low-grade positive margins (36% vs 26%; P = .34). For those with recurrent dysplasia, the interval to recurrence was also calculated according to margin status. The mean time to recurrence for those with positive margins was 11.5 months; the mean time to recurrence for those with negative margins was 12.6 months. A deep pass was performed in 54 cases (29%). Comparisons of recurrence rates with the performance of a deep pass and with the diagnosis of the LEEP specimens are shown in Tables IV and V, respectively. Neither of these factors was associated with a statistically significant increased risk of recurrence. In addition, performance of a deep pass was not related to margin status. The proportion of women with positive margins was similar regardless of whether a deep pass was performed (deep pass vs no deep pass, 26% vs 28%; P = .75). Comment The significance of margin status of LEEP specimens has been controversial. Positive margins have been described as being insignificant and probably not pre-
318 Gonzalez et al
February 2001 Am J Obstet Gynecol
Table VI. References regarding LEEP and follow-up, including margin status when reported Recurrence (%) Follow-up
Study Hanau and Bibbo20 Gold et al11 Bigrigg et al12 Wright et al13 Naumann et al14 Herzog et al17 Alvarez et al15 Felix et al16 Lund et al21 Oyesanya et al1 Murdoch et al2 Shafi et al22 Spitzer et al3 Keijser et al9 Whiteley and Olah4 Prendiville et al5 Gunasekera et al6 Luesley et al7 Murdoch et al8 Flannelly et al10
Patients with appropriate data n = 122 n = 123 n = 774 n = 275 n = 141 n = 65 n = 153 n = 185 n = 57 n = 122 n = 43 n = 721 n = 174 n = 176 n = 424 n = 80 n = 102 n = 98 n = 557 n = 600 n = 317
Time (mo) 10.9* 17.9* 22† 12.4* 6-12‡ 12* 3-6‡ 3§ 12§ NA 12§ 3§ 6§ 11.5* 57.6† 6§ 19.2* 6§ 6§ 3§ 48
Method
Overall success (%)
Positive margin (%)
Positive margin (No.)
Negative margin (No.)
Cytologic study Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy NA Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy
64 82 94 80 94 86 NA 93 77 60 93 95 93 91 81 91-95 97 95 95-96 95.5 90
25 22 NA NA NA 13 46 NA 28 45 16 44 18 17 NA 15 NA NA NA 17 NA
55 29 33 NA NA 31 About 14 NA 69 47 14 10 53 20 NA 25 NA NA NA NA 14
38 17 NA NA NA 6.5 <1 NA 5 34 14 NA 25 6.3 NA NA NA NA NA NA 6
Statistical significance NS P < .05
P < .05
P < .05
P < .05
NA, Not applicable. *Mean. †Median. ‡Range. §Total.
dictive of residual or recurrent disease.19 As shown in Table VI,1-17, 20-22 however, many reports have demonstrated a substantial percentage of women with positive margins. In those studies in which margin status and recurrence were reported, several demonstrated a significantly increased risk of dysplasia among patients with positive margins. Analysis of our data demonstrates that women with positive margins were significantly more likely to have recurrent dysplasia, with rates of recurrence comparable to those listed in Table VI. This association is maintained whether any positive margin or endocervical margins only are considered. When the recurrence was further classified as a low-grade lesion or high-grade dysplasia, however, statistical significance was not reached, possibly because of inadequate numbers of patients in this subanalysis and the potential for type II error. Alternatively, when the positive margins were further categorized as showing low-grade or high-grade lesions, recurrence was significantly associated only with margins with highgrade lesions relative to women with negative margins. Only 4 women (2% of the original 199 participants) were excluded from analysis because of the inability to assess margins, which is comparable to a low rate of uninterpretable margins reported in other series.1, 4, 7, 9, 16 This high rate of success in assessment is probably because the LEEP were performed as a single central pass with the large loop, with additional passes as indicated.
This technique is in contrast to the performance of multiple small passes, which is more likely to lead to the fragmentation and orientation difficulties that, along with thermal artifact, have been suggested as contributing to the inability to adequately assess margins.14, 23 Perhaps a more interesting finding is the recurrence of dysplasia among women with negative margins and the overall efficacy rate of 69% for the total length of followup. This is in sharp contrast to the findings of the studies listed in Table VI, most of which reported efficacy rates >90%. However, several factors must be considered. Of the 20 studies listed, only 6 reported follow-up periods of ≥12 months; 7 of the studies reported ≤6 months of follow-up. Indeed, our efficacy rate at 6 months of follow-up is comparable to the rates listed in Table VI for ≥6 months of observation. The methods of surveillance for recurrence were quite varied. In several studies only cytologic smears were used, whereas in others both cytologic examination and colposcopy were performed, with recurrence documented by tissue sampling. Several reports have cited a higher number of enrolled study subjects; however, critical review demonstrates the actual number of participants with follow-up data to be substantially lower in some cases.* In addition, criteria used to define recurrence also vary. In several studies only high-grade *References 3, 7, 11, 12, 14, 15, 20, and 21.
Gonzalez et al 319
Volume 184, Number 3 Am J Obstet Gynecol
Comments Only 87 in margin group Follow-up, high-grade dysplasia only; margins, endocervix only Small loop Large loop “Indeterminate” margins included in recurrence “See and treat”; 33% negative for dysplasia
“See and treat”; 6.8% pathologic reports normal “See and treat”; 7.5% pathologic reports normal Recurrence counted only if high-grade dysplasia “See and treat”; 5% pathologic reports normal “See and treat”; 32.8% pathologic reports normal Cumulative 4-y follow-up; margin data at 1 y
(moderate to severe) dysplasia was considered as a treatment failure, with no data given regarding the recurrence of mild dysplasia.6, 11 Our data are more consistent with the reports by Hanau and Bibbo,20 Lund et al,21 and Shafi et al.22 In the series by Hanau and Bibbo20 a total of 109 women was included, but only 87 women were included for margin analysis. In the report by Lund et al21 the follow-up interval was not specified. Shafi et al22 reported fairly substantial rates of recurrence among those with both positive and negative margins; however, the overall success rate at 6 months was still >90%. The mean time to recurrence in our series was slightly <1 year; this is consistent with the findings of Flannelly et al,10 which demonstrated that approximately 95% of abnormal follow-up studies after LEEP occurred in the first 2 years. It is interesting that the time to recurrence among the women who had recurrent dysplasia did not vary greatly between those with positive margins and those with negative margins, even though the risk of recurrence was higher for those with positive margins. Another important consideration is those studies reporting on the “see and treat” approach, in which women with normal histologic characteristics of the LEEP specimen or those with only low-grade changes (HPV or mild dysplasia) might be considered “overtreated.” Inclusion of women with LEEP specimens classified as normal would certainly affect the apparent efficacy rate. In the
“see and treat” studies, the prevalence of normal results of histologic examination of the LEEP specimen was reported to range from 5% to as high as 33%.4, 7-9, 15 Although the consideration of HPV changes alone at a margin as constituting a positive margin may be viewed as controversial, there were only 3 women in this category, and inclusion of these patients had minimal effect on overall recurrence according to margin status. In fact, if HPV changes alone at a margin were considered in the negative margin group rather than in the positive margin group, the recurrence rate among those with negative margins would actually be slightly higher (27% rather than 26%). Explanation of the relatively high rate of recurrence in our series, especially among those with negative margins, is speculative but may include a longer follow-up interval. Evidence to support this contention includes lower recurrence rates at 6 months of follow-up, comparable to other reports, but higher recurrence rates with longer observation. In addition, with a mean time to recurrence of slightly <1 year regardless of margin status, a higher recurrence rate and thus a lower efficacy rate might be expected with a longer follow-up interval, as was demonstrated in our series. Another factor to potentially explain the high recurrence in this series is the performance of a comprehensive evaluation, including cytologic examination, colposcopy, and biopsy as indicated, for every woman at each visit for ≥1 year after the LEEP. The potential importance of colposcopy, rather than just cytologic examination, was demonstrated in a series by Flannelly et al10 in which 47% of patients with abnormal colposcopic results and biopsy-confirmed dysplasia after LEEP had normal cytologic results. Alternatively, biases must also be considered. As with any retrospective study, the possibility of selection bias exists. However, all women who underwent the procedure during the specified period were initially included, and few were excluded for reasons cited previously. In addition, our institutional practice pattern for all women undergoing LEEP included the same comprehensive postprocedure surveillance plan, which makes it unlikely that patients with recurrence had been preferentially selected. Procedural performance must also be considered as a possible bias; however, only 4 patients were excluded because the margins could not be assessed. In addition, positive margins were identified in 28% of the study group, which is well within the range for positive margins in the reports listed in Table VI. Although endocervical curettage was performed at the end of each procedure (before cauterization of the tissue bed) in all patients undergoing LEEP, the result of endocervical curettage was positive in only 4 cases. These considerations taken together imply that it is unlikely that a significant amount of cervical disease remained after the procedure was performed. Therefore it is unlikely that suboptimal performance of the procedure was a significant factor.
320 Gonzalez et al
The performance of a deep pass has been specifically addressed in 3 reports. Wright et al13 specifically stated that no deep pass was done; Naumann et al14 reported that a deep pass was not done routinely. The series by Felix et al16 was the only study in which all patients underwent a deep pass, although no data were presented that specifically addressed outcome according to the performance of a deep pass. In our series performance of a deep pass was not associated with either positive margins or recurrence of dysplasia. Initial review may suggest that this result is biased, because most procedures were done as a treatment option and few had inadequate colposcopic results at the time of the LEEP. However, this same reasoning would also imply that there should be a lower rate of recurrence and fewer cases with positive margins, both of which are refuted by our results. The grade of dysplasia in the LEEP specimen in relation to recurrence was addressed only in the series by Hanau and Bibbo,20 in which low-grade lesions in the LEEP specimens were more commonly associated with recurrent dysplasia than were high-grade lesions; this difference was statistically significant (55% vs 31%; P = .008). In contrast, our series showed that high-grade lesions in the LEEP specimens were more commonly associated with recurrence; however, the comparison with low-grade lesions was not statistically significant. Some studies regarding cone biopsy and recurrence have concluded that follow-up cytologic examination, colposcopy, or endocervical curettage may be more reliable than margin status in predicting recurrence and that this consideration may be extrapolated to LEEP.10, 14, 20, 24 In our series, although follow-up cytologic examination and colposcopy were not individually analyzed in relation to recurrence, the result of endocervical curettage was not a reliable predictor of recurrence. In two other series of women undergoing LEEP, the result of endocervical curettage also did not reliably predict recurrence.13, 21 Although our recurrence rate was considerable regardless of margin status, most recurrences were low-grade lesions, and there were no cases of carcinoma. We therefore agree with conclusions from other reports regarding LEEP and cone biopsy that positive margins do not require immediate further diagnostic procedures but do require close follow-up with additional procedures as indicated.2, 14, 17, 20, 24, 25 In conclusion, LEEP are an acceptable treatment modality for management of cervical dysplasia. With comprehensive long-term follow-up, however, recurrence of dysplasia was substantial among those with positive margins, particularly those with high-grade lesions at the margin. Recurrence was also considerable among women with negative margins. Neither performance of a deep pass nor the grade of dysplasia in the LEEP specimen was
February 2001 Am J Obstet Gynecol
correlated with recurrence. Women should be counseled regarding the potential for recurrent dysplasia even in the setting of complete excision, and the importance of follow-up should be emphasized. REFERENCES
1. Oyesanya OA, Amerasinghe C, Manning EA. A comparison between loop diathermy conization and cold-knife conization for management of cervical dysplasia associated with unsatisfactory colposcopy. Gynecol Oncol 1993;50:84-8. 2. 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. BJOG 1992;99:990-3. 3. Spitzer M, Chernys AE, Seltzer VL. The use of large-loop excision of the transformation zone in an inner-city population. Obstet Gynecol 1993;82:731-5. 4. Whiteley PF, Olah KS. Treatment of cervical intraepithelial neoplasia: experience with low-voltage diathermy loop. Am J Obstet Gynecol 1990;162:1272-7. 5. Prendiville W, Cullimore J, Norman S. Large loop excision of the transformation zone (LLETZ): a new method of management for women with cervical intraepithelial neoplasia. BJOG 1989;96:1054-60. 6. Gunasekera PC, Phipps JH, Lewis BV. Large loop excision of the transformation zone (LLETZ) compared to carbon dioxide laser in the treatment of CIN: a superior mode of treatment. BJOG 1990;97:995-8. 7. Luesley DM, Cullimore J, Redman CW, Lawton FG, Emens JM, Rollason TP, et al. Loop diathermy excision of the cervical transformation zone in patients with abnormal cervical smears. BMJ 1990;300:1690-3. 8. Murdoch JB, Grimshaw RN, Monaghan JM. Loop diathermy excision of the abnormal cervical transformation zone. Int J Gynecol Cancer 1991;1:105-11. 9. Keijser KGG, Kenemans P, van der Zanden PH, Schijf CP, Vooijs GP, Rolland R. Diathermy loop excision in the management of cervical intraepithelial neoplasia: diagnosis and treatment in one procedure. Am J Obstet Gynecol 1992;166:1281-7. 10. Flannelly G, Langhan H, Jandial L, Mann E, Campbell M, Kitchener H. A study of treatment failures following large loop excision of the transformation zone for the treatment of cervical intraepithelial neoplasia. BJOG 1997;104:718-22. 11. Gold M, Dunton CJ, Murray J, Macones G, Hanau C, Carlson JA. Loop electrocautery excisional procedure: therapeutic effectiveness as an ablation and a conization equivalent. Gynecol Oncol 1996;61:241-4. 12. Bigrigg A, Haffenden DK, Sheehan AL, Codling BW, Read MD. Efficacy and safety of large-loop excision of the transformation zone. Lancet 1994;343:32-4. 13. Wright TC Jr, Gagnon S, Richart RM, Ferenczy A. Treatment of cervical intraepithelial neoplasia using the loop electrosurgical excision procedure. Obstet Gynecol 1992;79:173-8. 14. Naumann RW, Bell MC, Alvarez RD, Edwards RP, Partridge EE, Helm CW, et al. LLETZ is an acceptable alternative to diagnostic cold-knife conization. Gynecol Oncol 1994;55:224-8. 15. Alvarez RD, Helm CW, Edwards RP, Naumann RW, Partridge EE, Shingleton HM, et al. Prospective randomized trial of LLETZ versus laser ablation in patients with cervical intraepithelial neoplasia. Gynecol Oncol 1994;52:175-9. 16. Felix JC, Muderspach LI, Duggan BD, Roman LD. The significance of positive margins in loop electrosurgical cone biopsies. Obstet Gynecol 1994;84:996-1000. 17. Herzog TJ, Williams S, Adler LM, Rader JS, Kubiniec RT, Camel M, et al. Potential of cervical electrosurgical excision procedure for diagnosis and treatment of cervical intraepithelial neoplasia. Gynecol Oncol 1995;57:286-93. 18. Chappatte OA, Byrne DL, Raju KS, Nayagam M, Kenney A. Histologic differences between colposcopic directed biopsy and loop excision of the transformation zone (LLETZ): a cause for concern. Gynecol Oncol 1991;43:46-50.
Volume 184, Number 3 Am J Obstet Gynecol
19. Wright TC, Kurman RJ, Ferenczy A. Precancerous lesions of the cervix. In: Kurman RJ, editor. Blaustein’s pathology of the female genital tract. 4th ed. New York: Springer-Verlag; 1994. p. 229-77. 20. Hanau CA, Bibbo M. The case for cytologic follow-up after LEEP. Acta Cytol 1997;41:731-6. 21. Lund ED, Robinson WR, Adams J, O’Quinn AG. The predictive value of LEEP specimen margin status for residual/recurrent cervical intraepithelial neoplasia [abstract 72]. Gynecol Oncol 1997;64:307. 22. Shafi MI, Dunn JA, Buxton EJ, Finn CB, Jordan JA, Luesley DM. Ab-
Gonzalez et al 321
normal cervical cytology following large loop excision of the transformation zone: a case controlled study. BJOG 1993;100:145-8. 23. Montz FJ, Holschneider CH, Thompson LD. Large-loop excision of the transformation zone: effect on pathologic interpretation of resection margins. Obstet Gynecol 1993;81:976-82. 24. Buxton EJ, Luesley DM, Wade-Evans T, Jordan JA. Residual disease after cone biopsy: completeness of excision and follow-up cytology as predictive factors. Obstet Gynecol 1987;70:529-32. 25. Lapaquette TK, Dinh TV, Hannigan EV, Doherty MG, Yandell RB, Buchanan VS. Management of patients with positive margins after cervical conization. Obstet Gynecol 1993;82:440-3.
Key words: Loop electrosurgical excision procedure, positive margins, negative margins
Loop electrosurgical excision procedures (LEEP), including large loop excision of the transformation zone procedures, have been used extensively in Great Britain and have gained acceptance in the United States as a management option for the treatment of cervical dysplasia. These procedures have been used as a treatment alternative, and their efficacy as a diagnostic tool, including as an alternative to cold-knife cone biopsy, has also been fairly well established.1-17 Efficacy rates for LEEP are comparable to those of ablative techniques, such as laser therapy.6, 15 An advantage of LEEP relative to ablative techniques is the acquisition of a tissue specimen for histopathologic diagnosis, which has been suggested to
From the Department of Obstetrics and Gynecology, Wilford Hall United States Air Force Medical Center and Brooke Army Medical Center. Received for publication December 27, 1999; revised May 16, 2000; accepted July 7, 2000. Reprint requests: Capt Dagoberto I. Gonzalez, Jr, USAF, MC, 2200 Bergquist Dr, Suite 1, Lackland Air Force Base, TX 78236. 6/1/109937 doi:10.1067/mob.2001.109937
be potentially more accurate than punch biopsy of the cervix.9, 17, 18 In addition, LEEP have been reported to allow diagnosis of unsuspected cancer that might otherwise have been inappropriately treated with an ablative technique.1, 3, 6-8, 13, 17, 18 Although the overall efficacy of LEEP in the treatment of cervical dysplasia has been reported to be >90%, data are difficult to compare because of differences among studies. Variations include definitions of cure, length of follow-up, margin reporting and status, and method of follow-up, including cytologic examination only or cytologic examination with colposcopy and biopsy as indicated. In addition, several studies have reported efficacy rates for a “see and treat” approach without previous biopsy, whereas others have reported efficacy rates after a biopsy-confirmed diagnosis. The purpose of this study was to evaluate recurrence of dysplasia after LEEP in women with previous biopsyconfirmed dysplasia with a prolonged follow-up interval. We attempted to identify factors associated with an increased risk of recurrence, such as margin status, perfor315
316 Gonzalez et al
February 2001 Am J Obstet Gynecol
Table I. Indications for all LEEP performed Indication
No.
%
Persistent cervical intraepithelial neoplasia I Cervical intraepithelial neoplasia II Cervical intraepithelial neoplasia III Unexplained Papanicolaou test result Positive result of endocervical curettage Inadequate colposcopy
11 88 46 1 32 6
6 48 25 0.5 17 3
mance of a deep endocervical pass, status of endocervical curettage, and grade of dysplasia in the LEEP specimens. Material and methods This study received investigative review board approval at the participating institution. A retrospective chart review was performed on all women who underwent large loop excision of the transformation zone procedures and LEEP at Wilford Hall Medical Center, Lackland Air Force Base, Texas, between January 1993 and December 1994. One hundred ninety-nine women were eligible for the study. Fifteen participants were excluded because of inadequate follow up, significant glandular atypia, or uninterpretable margins. All procedures were performed either by obstetrics and gynecology residents with staff supervision or by gynecology or gynecologic oncology staff. Extracted data included age, parity, indication for LEEP, colposcopic examination at the time of the procedure, LEEP diagnosis, performance of a deep endocervical pass, margin status, follow-up diagnosis, and duration of follow-up. As shown in Table I, most procedures were performed as treatment for high-grade dysplasia. Approximately 20% of the procedures were performed for diagnostic indications traditionally reserved for coldknife cervical conization. Colposcopic evaluation was performed after application of 5% acetic acid solution to the cervix. Lugol iodine solution was applied at the discretion of the staff physician. Local anesthetic was then administered; 4 to 5 mL of 1% lidocaine hydrochloride with epinephrine at a 1:100,000 dilution mixed with 4 to 5 mL of 1% lidocaine hydrochloride without epinephrine was injected directly into the cervix in a circumferential pattern. Loop procedures were performed with a CooperSurgical model 6000 (CooperSurgical, Inc, Shelton, Conn) device and disposable stainless-steel wire loops (CIRCON Corporation, Santa Barbara, Calif). The procedures were performed with a blend setting and a power output of 40 W. Ninetyfive percent of the excisions were done with a 2.0 × 0.8-cm loop, and a 1.0 × 1.0-cm loop was used for deep passes when performed. The larger loop was used for a central pass that removed the entire transformation zone. Additional ectocervical passes were done with appropriately sized loops as dictated by colposcopic findings. A deep pass of the endocervix was performed for women with a
positive result of endocervical curettage, visible lesions extending into the cervical canal, or an inadequate colposcopic examination. A deep pass was also performed in additional cases at the discretion of the staff physician. An endocervical curettage was then performed with a Kevorkian endocervical curette. A 5-mm cautery ball with a power setting of 50 W coagulation was used to achieve hemostasis. Either Monsel solution or silver nitrate or both of these were applied as needed. All specimens were marked for orientation with a delayed absorbable suture at the 12 o’clock position. Specimens were marked fresh with indelible ink (Becton Dickinson Microbiology Systems, Sparks, Md) to identify appropriate margins. If deep passes were obtained, the true endocervical margin was also inked. Specimens were then placed in a 10% formalin solution and submitted for histopathologic examination. Specimens were cut into 5-µm sections with 3 levels through the block and stained with hematoxylin and eosin. A single investigator (Christopher M. Zahn) reviewed >95% of the specimens. A margin was considered positive when human papillomavirus (HPV) change or dysplasia, regardless of grade, was present at the inked margin or at the cauterized margin if ink was not identified. The result of endocervical curettage was considered positive if any signs of dysplasia or any histopathologic changes consistent with HPV were present. All women underwent a postoperative examination at 4 to 6 weeks and were then followed up at 4- to 6-month intervals. Follow-up visits, except for the postoperative examination, included a Papanicolaou smear, colposcopy, and biopsy as indicated. If subsequent evaluations showed normal results for 1 year, women were no longer seen at the colposcopy clinic and were followed up with yearly Papanicolaou smears. In cases of residual abnormalities, colposcopic examinations with directed biopsies were continued, and treatment procedures such as repeated LEEP, laser ablation, or cold-knife conization were performed as indicated. Diagnoses of the LEEP specimens and any follow-up tissue samples were recorded as normal, as HPV changes, or as mild, moderate, or severe dysplasia. The abnormal diagnoses were additionally classified as either low-grade lesions (HPV changes or mild dysplasia) or high-grade lesions (moderate or severe dysplasia). Normal diagnoses included reactive changes, reactive atypia, and inflammation. Recurrence or persistence of dysplasia was recorded for the entire study group. Abnormalities at follow-up were also compared between women categorized according to margin status, performance of a deep endocervical pass, and the diagnosis (low-grade vs high-grade lesion) of the original LEEP specimens. Statistical analysis was performed with the χ2 test or the Fisher exact test as appropriate. Statistical significance was defined as P < .05.
Gonzalez et al 317
Volume 184, Number 3 Am J Obstet Gynecol
Table IV. Performance of a deep pass and dysplasia at follow-up
Table II. Relationship between margin status (all margins) and dysplasia at follow-up Positive margins Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
Negative margins
No.
%
No.
%
Statistical significance
24 17 7
47 33 14
29 23 6
26 21 5.5
P = .009 P = .09 P = .12
Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
With deep pass
Without deep pass
No.
%
No.
%
Statistical significance
18 14 4
30 23 7
37 27 10
30 22 8
P = .94 P = .88 P > .99
Table III. Relationship between status of endocervical margin and dysplasia at follow-up
Table V. Comparison of low-grade versus high-grade lesions on LEEP specimens and dysplasia at follow-up
Positive margins
Low-grade lesions
High-grade lesions
No.
%
No.
%
Statistical significance
20 17 3
27 23 4
33 23 10
38 27 12
P = .12 P = .55 P = .08
Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
Negative margins
No.
%
No.
%
Statistical significance
18 12 6
49 32 16
35 28 7
28 23 5.6
P = .02 P = .22 P = .08
Results A total of 184 women met the criteria for the study. The mean age was 28.4 years (range, 16-66 years), and the mean length of follow-up was 24 months (range, 3-59 months). Nulliparous women accounted for 42% of the study group. Subsequent procedures were performed in 16 women (9%), including repeated LEEP in 10 cases, cold-knife cone biopsy in 5 cases, and laser ablation in 1 case. An endocervical curettage was done at the time of the LEEP, and results were positive in only 4 cases (2%). Inadequate colposcopy at the time of the LEEP was noted in only 3 cases; 1 of these women was found to have mild dysplasia on subsequent follow-up evaluation. The LEEP specimen had normal findings in 23 participants (12.5%). These women were therefore excluded from margin status consideration as well as from analysis of recurrent dysplasia according to the diagnosis of the LEEP specimen. In this group of women with negative LEEP histologic results, there were 4 recurrences (17%), 3 of which were low grade and 1 of which was high grade. Among the remainder of the LEEP specimens, the vast majority were considered to show dysplasia. Although HPV changes were included in the low-grade lesion category, of the 161 LEEP specimens with pathologic findings, only 7 were considered to show HPV changes only. Recurrence of dysplasia in the total study group varied according to the length of follow-up. At 6 months the recurrence rate was only 9.2%; during the total length of follow-up for the study, however, recurrent dysplasia was identified in 57 women (31%). Of the cases of recurrent dysplasia, 14 (25%) were high-grade lesions. The interval to recurrence for the overall group ranged from 2 to 48 months, with a mean of 11.9 months.
Diagnosis of recurrence All dysplasia Low-grade changes High-grade dysplasia
Positive margins were identified in 51 cases (28%). Although both positive ectocervical and positive endocervical margins were included, 73% (37 of 51 women) had positive endocervical margins. Among those with positive margins, 57% had high-grade lesions in the margins, and only 3 specimen margins were considered positive on the basis of HPV changes only. The relationship between recurrent dysplasia and margin status of the LEEP specimen is demonstrated in Tables II and III. Positive margins were further categorized as demonstrating either low-grade or high-grade lesions. Relative to those with negative margins, recurrent dysplasia was more commonly associated with high-grade positive margins (55% vs 26%; P = .003) than with low-grade positive margins (36% vs 26%; P = .34). For those with recurrent dysplasia, the interval to recurrence was also calculated according to margin status. The mean time to recurrence for those with positive margins was 11.5 months; the mean time to recurrence for those with negative margins was 12.6 months. A deep pass was performed in 54 cases (29%). Comparisons of recurrence rates with the performance of a deep pass and with the diagnosis of the LEEP specimens are shown in Tables IV and V, respectively. Neither of these factors was associated with a statistically significant increased risk of recurrence. In addition, performance of a deep pass was not related to margin status. The proportion of women with positive margins was similar regardless of whether a deep pass was performed (deep pass vs no deep pass, 26% vs 28%; P = .75). Comment The significance of margin status of LEEP specimens has been controversial. Positive margins have been described as being insignificant and probably not pre-
318 Gonzalez et al
February 2001 Am J Obstet Gynecol
Table VI. References regarding LEEP and follow-up, including margin status when reported Recurrence (%) Follow-up
Study Hanau and Bibbo20 Gold et al11 Bigrigg et al12 Wright et al13 Naumann et al14 Herzog et al17 Alvarez et al15 Felix et al16 Lund et al21 Oyesanya et al1 Murdoch et al2 Shafi et al22 Spitzer et al3 Keijser et al9 Whiteley and Olah4 Prendiville et al5 Gunasekera et al6 Luesley et al7 Murdoch et al8 Flannelly et al10
Patients with appropriate data n = 122 n = 123 n = 774 n = 275 n = 141 n = 65 n = 153 n = 185 n = 57 n = 122 n = 43 n = 721 n = 174 n = 176 n = 424 n = 80 n = 102 n = 98 n = 557 n = 600 n = 317
Time (mo) 10.9* 17.9* 22† 12.4* 6-12‡ 12* 3-6‡ 3§ 12§ NA 12§ 3§ 6§ 11.5* 57.6† 6§ 19.2* 6§ 6§ 3§ 48
Method
Overall success (%)
Positive margin (%)
Positive margin (No.)
Negative margin (No.)
Cytologic study Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy NA Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy Cytologic study; colposcopy
64 82 94 80 94 86 NA 93 77 60 93 95 93 91 81 91-95 97 95 95-96 95.5 90
25 22 NA NA NA 13 46 NA 28 45 16 44 18 17 NA 15 NA NA NA 17 NA
55 29 33 NA NA 31 About 14 NA 69 47 14 10 53 20 NA 25 NA NA NA NA 14
38 17 NA NA NA 6.5 <1 NA 5 34 14 NA 25 6.3 NA NA NA NA NA NA 6
Statistical significance NS P < .05
P < .05
P < .05
P < .05
NA, Not applicable. *Mean. †Median. ‡Range. §Total.
dictive of residual or recurrent disease.19 As shown in Table VI,1-17, 20-22 however, many reports have demonstrated a substantial percentage of women with positive margins. In those studies in which margin status and recurrence were reported, several demonstrated a significantly increased risk of dysplasia among patients with positive margins. Analysis of our data demonstrates that women with positive margins were significantly more likely to have recurrent dysplasia, with rates of recurrence comparable to those listed in Table VI. This association is maintained whether any positive margin or endocervical margins only are considered. When the recurrence was further classified as a low-grade lesion or high-grade dysplasia, however, statistical significance was not reached, possibly because of inadequate numbers of patients in this subanalysis and the potential for type II error. Alternatively, when the positive margins were further categorized as showing low-grade or high-grade lesions, recurrence was significantly associated only with margins with highgrade lesions relative to women with negative margins. Only 4 women (2% of the original 199 participants) were excluded from analysis because of the inability to assess margins, which is comparable to a low rate of uninterpretable margins reported in other series.1, 4, 7, 9, 16 This high rate of success in assessment is probably because the LEEP were performed as a single central pass with the large loop, with additional passes as indicated.
This technique is in contrast to the performance of multiple small passes, which is more likely to lead to the fragmentation and orientation difficulties that, along with thermal artifact, have been suggested as contributing to the inability to adequately assess margins.14, 23 Perhaps a more interesting finding is the recurrence of dysplasia among women with negative margins and the overall efficacy rate of 69% for the total length of followup. This is in sharp contrast to the findings of the studies listed in Table VI, most of which reported efficacy rates >90%. However, several factors must be considered. Of the 20 studies listed, only 6 reported follow-up periods of ≥12 months; 7 of the studies reported ≤6 months of follow-up. Indeed, our efficacy rate at 6 months of follow-up is comparable to the rates listed in Table VI for ≥6 months of observation. The methods of surveillance for recurrence were quite varied. In several studies only cytologic smears were used, whereas in others both cytologic examination and colposcopy were performed, with recurrence documented by tissue sampling. Several reports have cited a higher number of enrolled study subjects; however, critical review demonstrates the actual number of participants with follow-up data to be substantially lower in some cases.* In addition, criteria used to define recurrence also vary. In several studies only high-grade *References 3, 7, 11, 12, 14, 15, 20, and 21.
Gonzalez et al 319
Volume 184, Number 3 Am J Obstet Gynecol
Comments Only 87 in margin group Follow-up, high-grade dysplasia only; margins, endocervix only Small loop Large loop “Indeterminate” margins included in recurrence “See and treat”; 33% negative for dysplasia
“See and treat”; 6.8% pathologic reports normal “See and treat”; 7.5% pathologic reports normal Recurrence counted only if high-grade dysplasia “See and treat”; 5% pathologic reports normal “See and treat”; 32.8% pathologic reports normal Cumulative 4-y follow-up; margin data at 1 y
(moderate to severe) dysplasia was considered as a treatment failure, with no data given regarding the recurrence of mild dysplasia.6, 11 Our data are more consistent with the reports by Hanau and Bibbo,20 Lund et al,21 and Shafi et al.22 In the series by Hanau and Bibbo20 a total of 109 women was included, but only 87 women were included for margin analysis. In the report by Lund et al21 the follow-up interval was not specified. Shafi et al22 reported fairly substantial rates of recurrence among those with both positive and negative margins; however, the overall success rate at 6 months was still >90%. The mean time to recurrence in our series was slightly <1 year; this is consistent with the findings of Flannelly et al,10 which demonstrated that approximately 95% of abnormal follow-up studies after LEEP occurred in the first 2 years. It is interesting that the time to recurrence among the women who had recurrent dysplasia did not vary greatly between those with positive margins and those with negative margins, even though the risk of recurrence was higher for those with positive margins. Another important consideration is those studies reporting on the “see and treat” approach, in which women with normal histologic characteristics of the LEEP specimen or those with only low-grade changes (HPV or mild dysplasia) might be considered “overtreated.” Inclusion of women with LEEP specimens classified as normal would certainly affect the apparent efficacy rate. In the
“see and treat” studies, the prevalence of normal results of histologic examination of the LEEP specimen was reported to range from 5% to as high as 33%.4, 7-9, 15 Although the consideration of HPV changes alone at a margin as constituting a positive margin may be viewed as controversial, there were only 3 women in this category, and inclusion of these patients had minimal effect on overall recurrence according to margin status. In fact, if HPV changes alone at a margin were considered in the negative margin group rather than in the positive margin group, the recurrence rate among those with negative margins would actually be slightly higher (27% rather than 26%). Explanation of the relatively high rate of recurrence in our series, especially among those with negative margins, is speculative but may include a longer follow-up interval. Evidence to support this contention includes lower recurrence rates at 6 months of follow-up, comparable to other reports, but higher recurrence rates with longer observation. In addition, with a mean time to recurrence of slightly <1 year regardless of margin status, a higher recurrence rate and thus a lower efficacy rate might be expected with a longer follow-up interval, as was demonstrated in our series. Another factor to potentially explain the high recurrence in this series is the performance of a comprehensive evaluation, including cytologic examination, colposcopy, and biopsy as indicated, for every woman at each visit for ≥1 year after the LEEP. The potential importance of colposcopy, rather than just cytologic examination, was demonstrated in a series by Flannelly et al10 in which 47% of patients with abnormal colposcopic results and biopsy-confirmed dysplasia after LEEP had normal cytologic results. Alternatively, biases must also be considered. As with any retrospective study, the possibility of selection bias exists. However, all women who underwent the procedure during the specified period were initially included, and few were excluded for reasons cited previously. In addition, our institutional practice pattern for all women undergoing LEEP included the same comprehensive postprocedure surveillance plan, which makes it unlikely that patients with recurrence had been preferentially selected. Procedural performance must also be considered as a possible bias; however, only 4 patients were excluded because the margins could not be assessed. In addition, positive margins were identified in 28% of the study group, which is well within the range for positive margins in the reports listed in Table VI. Although endocervical curettage was performed at the end of each procedure (before cauterization of the tissue bed) in all patients undergoing LEEP, the result of endocervical curettage was positive in only 4 cases. These considerations taken together imply that it is unlikely that a significant amount of cervical disease remained after the procedure was performed. Therefore it is unlikely that suboptimal performance of the procedure was a significant factor.
320 Gonzalez et al
The performance of a deep pass has been specifically addressed in 3 reports. Wright et al13 specifically stated that no deep pass was done; Naumann et al14 reported that a deep pass was not done routinely. The series by Felix et al16 was the only study in which all patients underwent a deep pass, although no data were presented that specifically addressed outcome according to the performance of a deep pass. In our series performance of a deep pass was not associated with either positive margins or recurrence of dysplasia. Initial review may suggest that this result is biased, because most procedures were done as a treatment option and few had inadequate colposcopic results at the time of the LEEP. However, this same reasoning would also imply that there should be a lower rate of recurrence and fewer cases with positive margins, both of which are refuted by our results. The grade of dysplasia in the LEEP specimen in relation to recurrence was addressed only in the series by Hanau and Bibbo,20 in which low-grade lesions in the LEEP specimens were more commonly associated with recurrent dysplasia than were high-grade lesions; this difference was statistically significant (55% vs 31%; P = .008). In contrast, our series showed that high-grade lesions in the LEEP specimens were more commonly associated with recurrence; however, the comparison with low-grade lesions was not statistically significant. Some studies regarding cone biopsy and recurrence have concluded that follow-up cytologic examination, colposcopy, or endocervical curettage may be more reliable than margin status in predicting recurrence and that this consideration may be extrapolated to LEEP.10, 14, 20, 24 In our series, although follow-up cytologic examination and colposcopy were not individually analyzed in relation to recurrence, the result of endocervical curettage was not a reliable predictor of recurrence. In two other series of women undergoing LEEP, the result of endocervical curettage also did not reliably predict recurrence.13, 21 Although our recurrence rate was considerable regardless of margin status, most recurrences were low-grade lesions, and there were no cases of carcinoma. We therefore agree with conclusions from other reports regarding LEEP and cone biopsy that positive margins do not require immediate further diagnostic procedures but do require close follow-up with additional procedures as indicated.2, 14, 17, 20, 24, 25 In conclusion, LEEP are an acceptable treatment modality for management of cervical dysplasia. With comprehensive long-term follow-up, however, recurrence of dysplasia was substantial among those with positive margins, particularly those with high-grade lesions at the margin. Recurrence was also considerable among women with negative margins. Neither performance of a deep pass nor the grade of dysplasia in the LEEP specimen was
February 2001 Am J Obstet Gynecol
correlated with recurrence. Women should be counseled regarding the potential for recurrent dysplasia even in the setting of complete excision, and the importance of follow-up should be emphasized. REFERENCES
1. Oyesanya OA, Amerasinghe C, Manning EA. A comparison between loop diathermy conization and cold-knife conization for management of cervical dysplasia associated with unsatisfactory colposcopy. Gynecol Oncol 1993;50:84-8. 2. 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. BJOG 1992;99:990-3. 3. Spitzer M, Chernys AE, Seltzer VL. The use of large-loop excision of the transformation zone in an inner-city population. Obstet Gynecol 1993;82:731-5. 4. Whiteley PF, Olah KS. Treatment of cervical intraepithelial neoplasia: experience with low-voltage diathermy loop. Am J Obstet Gynecol 1990;162:1272-7. 5. Prendiville W, Cullimore J, Norman S. Large loop excision of the transformation zone (LLETZ): a new method of management for women with cervical intraepithelial neoplasia. BJOG 1989;96:1054-60. 6. Gunasekera PC, Phipps JH, Lewis BV. Large loop excision of the transformation zone (LLETZ) compared to carbon dioxide laser in the treatment of CIN: a superior mode of treatment. BJOG 1990;97:995-8. 7. Luesley DM, Cullimore J, Redman CW, Lawton FG, Emens JM, Rollason TP, et al. Loop diathermy excision of the cervical transformation zone in patients with abnormal cervical smears. BMJ 1990;300:1690-3. 8. Murdoch JB, Grimshaw RN, Monaghan JM. Loop diathermy excision of the abnormal cervical transformation zone. Int J Gynecol Cancer 1991;1:105-11. 9. Keijser KGG, Kenemans P, van der Zanden PH, Schijf CP, Vooijs GP, Rolland R. Diathermy loop excision in the management of cervical intraepithelial neoplasia: diagnosis and treatment in one procedure. Am J Obstet Gynecol 1992;166:1281-7. 10. Flannelly G, Langhan H, Jandial L, Mann E, Campbell M, Kitchener H. A study of treatment failures following large loop excision of the transformation zone for the treatment of cervical intraepithelial neoplasia. BJOG 1997;104:718-22. 11. Gold M, Dunton CJ, Murray J, Macones G, Hanau C, Carlson JA. Loop electrocautery excisional procedure: therapeutic effectiveness as an ablation and a conization equivalent. Gynecol Oncol 1996;61:241-4. 12. Bigrigg A, Haffenden DK, Sheehan AL, Codling BW, Read MD. Efficacy and safety of large-loop excision of the transformation zone. Lancet 1994;343:32-4. 13. Wright TC Jr, Gagnon S, Richart RM, Ferenczy A. Treatment of cervical intraepithelial neoplasia using the loop electrosurgical excision procedure. Obstet Gynecol 1992;79:173-8. 14. Naumann RW, Bell MC, Alvarez RD, Edwards RP, Partridge EE, Helm CW, et al. LLETZ is an acceptable alternative to diagnostic cold-knife conization. Gynecol Oncol 1994;55:224-8. 15. Alvarez RD, Helm CW, Edwards RP, Naumann RW, Partridge EE, Shingleton HM, et al. Prospective randomized trial of LLETZ versus laser ablation in patients with cervical intraepithelial neoplasia. Gynecol Oncol 1994;52:175-9. 16. Felix JC, Muderspach LI, Duggan BD, Roman LD. The significance of positive margins in loop electrosurgical cone biopsies. Obstet Gynecol 1994;84:996-1000. 17. Herzog TJ, Williams S, Adler LM, Rader JS, Kubiniec RT, Camel M, et al. Potential of cervical electrosurgical excision procedure for diagnosis and treatment of cervical intraepithelial neoplasia. Gynecol Oncol 1995;57:286-93. 18. Chappatte OA, Byrne DL, Raju KS, Nayagam M, Kenney A. Histologic differences between colposcopic directed biopsy and loop excision of the transformation zone (LLETZ): a cause for concern. Gynecol Oncol 1991;43:46-50.
Volume 184, Number 3 Am J Obstet Gynecol
19. Wright TC, Kurman RJ, Ferenczy A. Precancerous lesions of the cervix. In: Kurman RJ, editor. Blaustein’s pathology of the female genital tract. 4th ed. New York: Springer-Verlag; 1994. p. 229-77. 20. Hanau CA, Bibbo M. The case for cytologic follow-up after LEEP. Acta Cytol 1997;41:731-6. 21. Lund ED, Robinson WR, Adams J, O’Quinn AG. The predictive value of LEEP specimen margin status for residual/recurrent cervical intraepithelial neoplasia [abstract 72]. Gynecol Oncol 1997;64:307. 22. Shafi MI, Dunn JA, Buxton EJ, Finn CB, Jordan JA, Luesley DM. Ab-
Gonzalez et al 321
normal cervical cytology following large loop excision of the transformation zone: a case controlled study. BJOG 1993;100:145-8. 23. Montz FJ, Holschneider CH, Thompson LD. Large-loop excision of the transformation zone: effect on pathologic interpretation of resection margins. Obstet Gynecol 1993;81:976-82. 24. Buxton EJ, Luesley DM, Wade-Evans T, Jordan JA. Residual disease after cone biopsy: completeness of excision and follow-up cytology as predictive factors. Obstet Gynecol 1987;70:529-32. 25. Lapaquette TK, Dinh TV, Hannigan EV, Doherty MG, Yandell RB, Buchanan VS. Management of patients with positive margins after cervical conization. Obstet Gynecol 1993;82:440-3.