- Email: [email protected]
Outcomes after Cervical Cold Knife Conization with Complete and Incomplete Excision of Abnormal Epithelium: A Review of 699 Cases
GYNECOLOGIC ONCOLOGY ARTICLE NO.
67, 34–38 (1997)
GO974817
Outcomes after Cervical Cold Knife Conization with Complete and Incomplete Excision of Abnormal Epithelium: A Review of 699 Cases K. Mohamed-Noor,* MRACOG, M. A. Quinn,† MGO, MRCOG, FRACOG, MRCP, CGO, and J. Tan,† MRCOG, FRACOG *Office of the Registrar and †Oncology/Dysplasia Unit, Royal Women’s Hospital, Melbourne, Victoria 3053, Australia Received January 30, 1997
early invasion, glandular abnormalities detected on cytology, and a discrepancy between cytology and/or colposcopy and/ or histology. Conization is also considered to be an important diagnostic procedure, as excised specimens are examined for histological evidence of invasive disease, thus identifying patients for whom more radical surgery may be more appropriate (3). Cervical cone biopsy may, in some cases, result in incomplete excision of neoplastic epithelium. A number of studies have reported patient outcomes in cases of both complete and incomplete excision, and these reports demonstrate not only that cure of CIN may result even when the margins of the conization specimen include neoplastic epithelium (4), but also that recurrent disease may occur despite histologically proven complete excision of high-grade CIN lesions (6). This study aims to examine patient outcomes after conization, both complete and incomplete, as treatment for CIN at the Royal Women’s Hospital (RWH) from 1966, when a database of Dysplasia Clinic attendances and interventions was begun.
A retrospective analysis was undertaken of 699 cone biopsies performed at the Royal Women’s Hospital in Melbourne from 1966 to 1992. In 572 cases (82%), abnormal epithelium was assessed as having been completely excised, and in 127 (18%) excision was incomplete. There were no significant differences in age, parity, cytology, histology, or indications for conization between patients in whom excision was incomplete and those in whom complete excision was achieved. Of the patients whose cone biopsy histology showed complete excision of abnormal epithelium, 96.7% were found to have been cured of disease on the basis of normal followup or normal histology of hysterectomy specimens. The overall cure rate after incomplete cone biopsy was found to be 77%, but was influenced by the site of incomplete excision. The cure rate for incomplete excision at the ectocervical margin was 86%; incomplete excision at the endocervix was 68% and only 40% if excision was incomplete at both edges. Cone biopsy undertaken for cervical intraepithelial neoplasia is likely to be curative when the lesion is completely excised, but recurrent disease may occur and adequate follow-up is an essential part of patient management after conization, regardless of histological findings in the conization specimen. Most cases of incompletely excised cervical intraepithelial neoplasia will also be cured, especially if the incomplete margin is ectocervical, and cytological and colposcopic follow-up may be an acceptable alternative to repeat cone biopsy or hysterectomy in the management of such cases. q 1997 Academic Press
MATERIALS AND METHODS
Patients A retrospective analysis was performed on patients who underwent cervical cone biopsy between February 1966 and February 1992. These patients were identified from the RWH Dysplasia Clinic datafile. For each patient, the following details were obtained: age, parity, preoperative cytology and histology, indications for conization, histology of conization specimens, completeness of excision, and results of subsequent follow-up and management.
INTRODUCTION AND AIMS
Carcinoma in situ (CIS) was originally considered to be a malignant lesion with potential for invasion in all cases, and hysterectomy was recommended as the treatment of choice (1). However, with recognition of CIS as part of the spectrum of cervical intraepithelial neoplasia (CIN), and because cases of CIS far exceed cases of cervical cancer, ablative techniques and cervical conization have subsequently replaced hysterectomy as the preferred treatment of CIS (2). Current indications for cone biopsy include lesions seen as going into the endocervical canal in which the upper limit cannot be viewed, lesions considered compatible with
Gyn 4817
/
6d23$$$141
All cone biopsies were performed under general anesthetic. Throughout the study period, colposcopy was performed prior to conization and Lugol’s iodine was used to delineate abnormal epithelium. Standard conization was performed by 34
0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
Operative Techniques
09-16-97 05:32:17
goa
AP: GYN
OUTCOME AFTER CERVICAL COLD KNIFE CONIZATION
the cold knife technique and all specimens were marked for correct orientation. Hemostasis was achieved by either Sturmdorf sutures or interrupted vertical sutures. Histopathological Techniques Until 1980, conization specimens were examined by the sagittal–coronal method. The apex was sliced into one or more coronal blocks and the truncated cone serially sectioned into several sagittal blocks. Beginning in 1980, a ‘‘whole embedding’’ method was used, in which conization specimens are bisected with a longitudinal median cut through the endocervical canal, each half then being embedded in wax in toto and serially sectioned at 200-mm intervals to produce 60–100 sections (6). In cases of hysterectomy being performed for recurrent or suspected residual CIN, the cervical portions of the hysterectomy specimens were processed by these procedures. Patient Follow-up Following cone biopsy, patients are, whenever possible, followed up at the Dysplasia Clinic by initial review at 3 months posttreatment and at 6-month intervals thereafter for 2 years. At each follow-up visit, ectocervical and endocervical cytological smears and colposcopic evaluation are performed, and targeted biopsies are taken if indicated by cytological or colposcopic findings. After 2 years of follow-up, if there has been no cytological or colposcopic evidence of recurrence, patients may be discharged from the clinic to have annual cervical cytology smears performed by their general practitioners, or they may be followed up indefinitely at the clinic. RESULTS
Seven hundred and fifty-three cone biopsies were identified as being performed during the study period. Fifty-four cases were excluded, as data was incomplete for 30, 22 had no recorded follow-up, and histology of the conization specimen was missing in 2 cases. Six hundred and ninetynine cases were available for analysis. Patient Profiles and Cyto/histological Abnormalities The age range of patients was 17–83 (mean 40.2, SD { 11.4), and parity ranged from 0–9 (mean 2.5, SD { 1.6). Four hundred and forty-six patients (64%) had major cytological findings of CIN III, CIS, adenocarcinoma in situ (ACIS), and minimally invasive or invasive changes, 206 (29%) had other degrees of CIN, and 47 (17%) had other cytological abnormalities. Colposcopically directed targeted biopsy was performed in 422 cases (60% of sample). Major histological abnormalities (CIN III or greater) were found in 309 (73%) biopsy specimens, other degrees of CIN in 76 (18%), and minor abnormalities in 38 (9%).
AID
Gyn 4817
/
6d23$$$142
09-16-97 05:32:17
35
Most cone biopsies were performed because abnormal epithelium was out of colposcopic range (500, 72%). Other indications for conization were cyto/histological discrepancy (119, 17%), colposcopic suspicion of invasion (61, 9%), histological evidence of microinvasive disease (15, 2%), and cytological suspicion of ACIS (4, 0.6%). Major histological abnormalities (CIN III or greater) were found in 521 (75%) conization specimens, other degrees of CIN in 105 (15%), human papillomavirus (HPV) infection in 10 (1.4%), normal epithelium in 29 (4%), and other findings (atypia, inflammation, and metaplasia) in 34 (5%). Table 1 shows initial cytological changes and histological findings in colposcopic biopsies and conization specimens. Completeness of Conization All 699 conization specimens were examined histologically. In 572 specimens (82%), abnormal epithelium was assessed as having been completely excised, and in 127 (18%) excision was incomplete. There were no significant differences in age, parity, cytology, histology, or indications for conization between patients in whom excision was incomplete and those in whom complete excision was achieved. In addition, there were no statistically significant differences in rates of incomplete excision for the periods prior to 1980 (71 of 291 cases, 24.4%) and after 1980 (56 of 281, 19.9%). Patient Follow-up Follow-up periods after conization ranged from 2 months to 21 years (mean 62.2 months, SD { 62.2). Forty percent of patients were followed up for less than 2 years, and 60% were seen at the Dysplasia Clinic over at least a 2-year period. Patient Outcomes after Complete Excision Of the 572 patients in whom abnormal epithelium was considered to have been completely excised, 7 had immediate hysterectomy (within 3 months) because of invasive histology (3), minimally invasive histology (2), and CIS with margin of excision close to abnormal epithelium (2). Of the 3 cases who underwent hysterectomy because of invasive lesions, none had residual disease in the hysterectomy specimen and remained normal at follow-up. One patient with minimally invasive histology had occult adenocarcinoma in the hysterectomy specimen and died within 5 years of diagnosis despite adequate initial treatment. Hysterectomy specimens from the other three patients were free of residual disease. The other 565 patients whose lesions were completely excised were followed up at the Dysplasia Clinic. Of these, 19 developed cytological recurrences, in some cases confirmed by biopsy. Treatment of recurrent disease was by hysterectomy in 13 cases, repeat cone biopsy in 3, laser
goa
AP: GYN
36
MOHAMED-NOOR, QUINN, AND TAN
TABLE 1 Initial Cytological Changes and Histological Findings at Colposcopy and at Conization Cyto/histological abnormality
At initial cytology (n Å 699)
Invasive disease Minimal invasion ACISa CISb CIN IIIc CIN II CIN I Dysplasia, unspecified Other d
45 20 4 287 90 51 22 133 47
At colposcopic biopsy (n Å 422)
(6.4%) (2.9%) (0.6%) (41.1%) (12.9%) (7.3%) (3.1%) (19.0%) (6.7%)
11 4 1 218 74 26 11 39 38
(2.6%) (0.9%) (0.2%) (51.6%) (17.5%) (6.2%) (2.6%) (9.2%) (9.2%)
At conization (n Å 699) 9 43 6 379 84 62 33 10 73
(1.3%) (6.2%) (0.9%) (54.2%) (12.0%) (8.9%) (4.8%) (1.4%) (10.4%)
a
ACIS, adenocarcinoma in situ. CIS, carcinoma in situ. c CIN, intraepithelial neoplasia. d Other, atypia, atrophy, metaplasia, inflammation,and HPV changes. b
ablation in 2, and excision of the abnormal area in 1. Recurrent disease occurred from 3 months to 17 years after cone biopsy (mean 43.8 months, SD 56.4). Histopathological findings in 12 of the recurrent lesions were identical to the original cone biopsy histology. In another 6 cases, recurrent lesions were within one histological grade of findings at cone histology, and in 1 case normal histology of the hysterectomy specimen was found despite persistent cytological changes suggestive of CIN I and HPV infection following cone biopsy for ACIS. All patients re-treated for recurrent disease remained normal at subsequent follow-up, apart from one patient who was pregnant at the time of both initial and repeat cone biopsies. This patient had persistent abnormal cytology and subsequently underwent hysterectomy which showed evidence of CIS in the hysterectomy specimen. There were no invasive lesions in the recurrence group. Thirty-four patients had hysterectomy for other reasons (fibroids, menorrhagia) during the follow-up period, and no recurrent lesions were found in hysterectomy specimens from these patients. Five hundred twelve patients remained cytologically and colposcopically normal throughout the follow-up period. Of the 572 patients whose cone biopsy histology showed complete excision of abnormal epithelium, 553 (96.7%) were found by this study to have been cured of disease on the basis of normal follow-up or normal histology of hysterectomy specimens. Figure 1 outlines the follow-up and outcomes for this group. Patient Outcomes after Incomplete Excision In 127 patients, abnormal epithelium was considered to have been incompletely excised; 79 (62%) cases were at the level of the ectocervix, 38 (30%) at the level of the endocervix, and 10 (8%) at both margins. Of these patients, 33 had immediate hysterectomy (within 3 months) because of high-
AID
Gyn 4817
/
6d23$$$143
09-16-97 05:32:17
degree histological abnormalities in conization specimens, 17 showing CIS, 11 showing minimally invasive disease and 3 cases of invasive carcinoma. Despite incomplete excision, in only 18 (55%) cases was there evidence of residual lesions in hysterectomy specimens. Of these 18 cases of residual disease, 3 (30%) had been incompletely excised at the ectocervix, 11 (61%) at the endocervix, and 4 (8%) at both margins. The other 94 patients whose lesions were incompletely excised were followed up at the Dysplasia Clinic. Of these, 13 developed cytological or colposcopic recurrences of disease, and invasive carcinoma occurred in 3 patients who had had CIS diagnosed at cone biopsy. One of these patient had not attended for follow-up after 6 months following cone biopsy and subsequently presented 19 years later with Stage IV cancer of the cervix. The other 2 patients with recurrent invasive disease were diagnosed at 20 months and 7 years after cone biopsy. Both patients were followed up in the Dysplasia Clinic and the recurrences were detected by abnormal smears. Histology of hysterectomy specimens from both cases showed microcarcinoma and subsequent follow-up has been normal. Overall, recurrent disease of any grade occurred from 6 months to 10 years after cone biopsy (mean 44 months, SD 39.6). Of the other 10 patients with cytological recurrences, histology of hysterectomy specimens confirmed CIN III in 7 and CIN II in 1. No abnormal epithelium was found in hysterectomy specimens from the other 2 patients. In the 11 patients with proven recurrent disease after incomplete conization, abnormal epithelium was considered to have been incompletely excised in 1 at the endocervix, in 9 at the ectocervix, and in 3 at both margins. Eight patients had hysterectomy for other reasons during the follow-up period after incomplete conization, and no recurrent lesions were found in hysterectomy specimens
goa
AP: GYN
OUTCOME AFTER CERVICAL COLD KNIFE CONIZATION
FIG. 1.
Follow-up and outcomes after complete conization.
from these patients. Seventy-three patients remained cytologically and colposcopically normal throughout the followup period. Of the 127 patients whose cone biopsy histology showed incomplete excision of abnormal epithelium, 98 (77%) were found by this study to have been cured of disease on the basis of normal follow-up or normal histology of hysterectomy specimens. The cure rate for incomplete excision at the ectocervical margin was 86% (68 of 79 cases), 68% (26 of 38 cases) for incomplete excision at the endocervix, and 40% (4 of 10 cases) if excision was incomplete at both edges. Figure 2 outlines follow-up and outcomes for this group. DISCUSSION
This review of 699 conizations confirms the therapeutic effectiveness of cone biopsy with an overall cure rate of
FIG. 2.
AID
Gyn 4817
/
6d23$$$143
37
93%, a rate comparable to that of 87% reported by Bjerre et al. (7) and 93% by Ahlgren et al. (8). However, despite colposcopic evaluation and use of iodine staining prior to conization, 18% of cone biopsies performed at RWH during the study period resulted in incomplete excision. This figure is comparable to that reported by White et al. (20%) (4) and by Lopes et al. (24%) (9), and is lower than that reported by Buxton et al. (45%) (10). In the study reported here, factors which could have influenced accuracy of colposcopic evaluation and ease of operative procedure, such as increasing patient age and parity, were not associated with an increased likelihood of incomplete excision. It seems likely that, regardless of technique and patient factors, a fairly constant rate of incomplete excisions of about 20% can be expected for cervical conization. After complete conization, the cure rate in this study was
Follow-up and outcomes after incomplete conization.
09-16-97 05:32:17
goa
AP: GYN
38
MOHAMED-NOOR, QUINN, AND TAN
found to be 96.7%, comparable to previously reported figures of 98% by Ahlgren et al. (8). There is no ready explanation for the occurrence of persistent or recurrent disease in the remaining 3.3% of patients who were reported to have completely excised lesions. It has been postulated by White et al. that recurrences after complete conization may be due to multifocal disease, inadequate examination of the conization specimen, or true recurrence because of ongoing exposure to HPV infection (4). There may, therefore, be factors other than histopathological ones involved in determining the true completeness of a cone biopsy, and these are not apparent in the results obtained in this study. The results obtained here, however, clearly demonstrate the importance of adequate follow-up even in cases of histologically complete excision of CIN lesions. After incomplete excision, this study demonstrated a lower cure rate of 77% comparable to previously reported figures of 70% by Ahlgren et al. (8) and 56% by White et al. (4), and that the likelihood of developing persistent disease may be determined by which margin was incompletely excised. Incomplete excision at both margins is more likely to occur if there is a large area of abnormal epithelium or a multifocal abnormality, and this may partly explain the higher recurrence rate when this occurs. White et al., in analyzing cases of persistent disease after incomplete conization, found a positive correlation with smoking. Unfortunately, in this study, smoking and other patient factors were not able to be evaluated, but may be helpful in future prospective studies in predicting patients who may be more likely to have persistent disease. Seventy-seven percent of patients who had incomplete excision of CIN in this study were cured. Possible explanations for this phenomenon include spontaneous regression and inflammatory responses of healing, which may play a role in destroying areas of residual CIN at the margin of resection. This latter theory may be supported by findings in studies of laser conization, where the surface of the cone bed is unavoidably ablated. These studies describe cure rates of 88 (11) and 94% (12) after incomplete laser conization. Because the time from conization to detection of recurrence is quite variable, regardless of completeness of excision, no conclusion can be drawn as to the optimum duration of colposcopic follow-up after conization.
AID
Gyn 4817
/
6d23$$$143
09-16-97 05:32:17
CONCLUSIONS
Cone biopsy is very likely to cure CIN when the lesion is completely excised. However, even after complete conization, recurrent disease may occur in a small percentage of patients, and adequate follow-up remains an essential part of patient management after conization, regardless of histological findings in the conization specimen. Most cases of incompletely excised CIN will also be cured, especially if the incomplete margin is ectocervical, and cytological and colposcopic follow-up may be an acceptable alternative to hysterectomy in the management of such cases. REFERENCES 1. Gusberg SB, Marshall D: Intraepithelial carcinoma of the cervix: A clinical reappraisal. Obstet Gynecol 19:713–716, 1962 2. Richart RM: Cervical intraepithelial neoplasia. Pathol Annal 8:301– 305, 1973 3. Matseoane S, Williams SB, Navarro C, Hedriana H, Mushayandebvu T: Diagnostic value of conization of the uterine cervix in the management of cervical neoplasia: A review of 756 consecutive patients. Gynecol Oncol 47:287–291, 1992 4. White CD, Cooper WL, Williams RR: Cervical intraepithelial neoplasia extending to the margins of resection in conization of the cervix. J Reprod Med 36:635–638, 1991 5. Grace HF, Pattison NS: Cone biopsy of the cervix: Does surgical technique influence cytological follow-up? Aust N Z J Obstet Gynaecol 27:216–219, 1987 6. Ostor AG: Studies on 200 cases of early squamous cell carcinoma of the cervix. Int J Gynecol Pathol 12:193–207, 1993 7. Bjerre B, Eliasson G, Linell F, Soderberg H, Sjoberg NO: Conization as only treatment of carcinoma in situ of the uterine cervix. Am J Obstet Gynecol 125:143–152, 1976 8. Ahlgren M, Ingemarsson I, Lindberg LG, Nordqvist RB: Conization as treatment of carcinoma in situ of the uterine cervix. Obstet Gynecol 46:135–139, 1975 9. Lopes A, Morgan P, Murdoch J, Piura B, Monaghan JM: The case for conservative management of ‘‘incomplete excision’’ of CIN after laser conization. Gynecol Oncol 49:247–249, 1993 10. 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 70:529–532, 1987 11. Grundsell H, Alm P, Larsson G: Cure rates after laser conization for early cervical neoplasia. Ann Chir Gynaecol 72:218–222, 1983 12. Andersen ES, Nielsen K, Larsen G: Laser conization: Follow-up in patients with cervical intraepithelial neoplasia in the cone margin. Gynecol Oncol 39:328–331, 1990.
goa
AP: GYN
67, 34–38 (1997)
GO974817
Outcomes after Cervical Cold Knife Conization with Complete and Incomplete Excision of Abnormal Epithelium: A Review of 699 Cases K. Mohamed-Noor,* MRACOG, M. A. Quinn,† MGO, MRCOG, FRACOG, MRCP, CGO, and J. Tan,† MRCOG, FRACOG *Office of the Registrar and †Oncology/Dysplasia Unit, Royal Women’s Hospital, Melbourne, Victoria 3053, Australia Received January 30, 1997
early invasion, glandular abnormalities detected on cytology, and a discrepancy between cytology and/or colposcopy and/ or histology. Conization is also considered to be an important diagnostic procedure, as excised specimens are examined for histological evidence of invasive disease, thus identifying patients for whom more radical surgery may be more appropriate (3). Cervical cone biopsy may, in some cases, result in incomplete excision of neoplastic epithelium. A number of studies have reported patient outcomes in cases of both complete and incomplete excision, and these reports demonstrate not only that cure of CIN may result even when the margins of the conization specimen include neoplastic epithelium (4), but also that recurrent disease may occur despite histologically proven complete excision of high-grade CIN lesions (6). This study aims to examine patient outcomes after conization, both complete and incomplete, as treatment for CIN at the Royal Women’s Hospital (RWH) from 1966, when a database of Dysplasia Clinic attendances and interventions was begun.
A retrospective analysis was undertaken of 699 cone biopsies performed at the Royal Women’s Hospital in Melbourne from 1966 to 1992. In 572 cases (82%), abnormal epithelium was assessed as having been completely excised, and in 127 (18%) excision was incomplete. There were no significant differences in age, parity, cytology, histology, or indications for conization between patients in whom excision was incomplete and those in whom complete excision was achieved. Of the patients whose cone biopsy histology showed complete excision of abnormal epithelium, 96.7% were found to have been cured of disease on the basis of normal followup or normal histology of hysterectomy specimens. The overall cure rate after incomplete cone biopsy was found to be 77%, but was influenced by the site of incomplete excision. The cure rate for incomplete excision at the ectocervical margin was 86%; incomplete excision at the endocervix was 68% and only 40% if excision was incomplete at both edges. Cone biopsy undertaken for cervical intraepithelial neoplasia is likely to be curative when the lesion is completely excised, but recurrent disease may occur and adequate follow-up is an essential part of patient management after conization, regardless of histological findings in the conization specimen. Most cases of incompletely excised cervical intraepithelial neoplasia will also be cured, especially if the incomplete margin is ectocervical, and cytological and colposcopic follow-up may be an acceptable alternative to repeat cone biopsy or hysterectomy in the management of such cases. q 1997 Academic Press
MATERIALS AND METHODS
Patients A retrospective analysis was performed on patients who underwent cervical cone biopsy between February 1966 and February 1992. These patients were identified from the RWH Dysplasia Clinic datafile. For each patient, the following details were obtained: age, parity, preoperative cytology and histology, indications for conization, histology of conization specimens, completeness of excision, and results of subsequent follow-up and management.
INTRODUCTION AND AIMS
Carcinoma in situ (CIS) was originally considered to be a malignant lesion with potential for invasion in all cases, and hysterectomy was recommended as the treatment of choice (1). However, with recognition of CIS as part of the spectrum of cervical intraepithelial neoplasia (CIN), and because cases of CIS far exceed cases of cervical cancer, ablative techniques and cervical conization have subsequently replaced hysterectomy as the preferred treatment of CIS (2). Current indications for cone biopsy include lesions seen as going into the endocervical canal in which the upper limit cannot be viewed, lesions considered compatible with
Gyn 4817
/
6d23$$$141
All cone biopsies were performed under general anesthetic. Throughout the study period, colposcopy was performed prior to conization and Lugol’s iodine was used to delineate abnormal epithelium. Standard conization was performed by 34
0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
Operative Techniques
09-16-97 05:32:17
goa
AP: GYN
OUTCOME AFTER CERVICAL COLD KNIFE CONIZATION
the cold knife technique and all specimens were marked for correct orientation. Hemostasis was achieved by either Sturmdorf sutures or interrupted vertical sutures. Histopathological Techniques Until 1980, conization specimens were examined by the sagittal–coronal method. The apex was sliced into one or more coronal blocks and the truncated cone serially sectioned into several sagittal blocks. Beginning in 1980, a ‘‘whole embedding’’ method was used, in which conization specimens are bisected with a longitudinal median cut through the endocervical canal, each half then being embedded in wax in toto and serially sectioned at 200-mm intervals to produce 60–100 sections (6). In cases of hysterectomy being performed for recurrent or suspected residual CIN, the cervical portions of the hysterectomy specimens were processed by these procedures. Patient Follow-up Following cone biopsy, patients are, whenever possible, followed up at the Dysplasia Clinic by initial review at 3 months posttreatment and at 6-month intervals thereafter for 2 years. At each follow-up visit, ectocervical and endocervical cytological smears and colposcopic evaluation are performed, and targeted biopsies are taken if indicated by cytological or colposcopic findings. After 2 years of follow-up, if there has been no cytological or colposcopic evidence of recurrence, patients may be discharged from the clinic to have annual cervical cytology smears performed by their general practitioners, or they may be followed up indefinitely at the clinic. RESULTS
Seven hundred and fifty-three cone biopsies were identified as being performed during the study period. Fifty-four cases were excluded, as data was incomplete for 30, 22 had no recorded follow-up, and histology of the conization specimen was missing in 2 cases. Six hundred and ninetynine cases were available for analysis. Patient Profiles and Cyto/histological Abnormalities The age range of patients was 17–83 (mean 40.2, SD { 11.4), and parity ranged from 0–9 (mean 2.5, SD { 1.6). Four hundred and forty-six patients (64%) had major cytological findings of CIN III, CIS, adenocarcinoma in situ (ACIS), and minimally invasive or invasive changes, 206 (29%) had other degrees of CIN, and 47 (17%) had other cytological abnormalities. Colposcopically directed targeted biopsy was performed in 422 cases (60% of sample). Major histological abnormalities (CIN III or greater) were found in 309 (73%) biopsy specimens, other degrees of CIN in 76 (18%), and minor abnormalities in 38 (9%).
AID
Gyn 4817
/
6d23$$$142
09-16-97 05:32:17
35
Most cone biopsies were performed because abnormal epithelium was out of colposcopic range (500, 72%). Other indications for conization were cyto/histological discrepancy (119, 17%), colposcopic suspicion of invasion (61, 9%), histological evidence of microinvasive disease (15, 2%), and cytological suspicion of ACIS (4, 0.6%). Major histological abnormalities (CIN III or greater) were found in 521 (75%) conization specimens, other degrees of CIN in 105 (15%), human papillomavirus (HPV) infection in 10 (1.4%), normal epithelium in 29 (4%), and other findings (atypia, inflammation, and metaplasia) in 34 (5%). Table 1 shows initial cytological changes and histological findings in colposcopic biopsies and conization specimens. Completeness of Conization All 699 conization specimens were examined histologically. In 572 specimens (82%), abnormal epithelium was assessed as having been completely excised, and in 127 (18%) excision was incomplete. There were no significant differences in age, parity, cytology, histology, or indications for conization between patients in whom excision was incomplete and those in whom complete excision was achieved. In addition, there were no statistically significant differences in rates of incomplete excision for the periods prior to 1980 (71 of 291 cases, 24.4%) and after 1980 (56 of 281, 19.9%). Patient Follow-up Follow-up periods after conization ranged from 2 months to 21 years (mean 62.2 months, SD { 62.2). Forty percent of patients were followed up for less than 2 years, and 60% were seen at the Dysplasia Clinic over at least a 2-year period. Patient Outcomes after Complete Excision Of the 572 patients in whom abnormal epithelium was considered to have been completely excised, 7 had immediate hysterectomy (within 3 months) because of invasive histology (3), minimally invasive histology (2), and CIS with margin of excision close to abnormal epithelium (2). Of the 3 cases who underwent hysterectomy because of invasive lesions, none had residual disease in the hysterectomy specimen and remained normal at follow-up. One patient with minimally invasive histology had occult adenocarcinoma in the hysterectomy specimen and died within 5 years of diagnosis despite adequate initial treatment. Hysterectomy specimens from the other three patients were free of residual disease. The other 565 patients whose lesions were completely excised were followed up at the Dysplasia Clinic. Of these, 19 developed cytological recurrences, in some cases confirmed by biopsy. Treatment of recurrent disease was by hysterectomy in 13 cases, repeat cone biopsy in 3, laser
goa
AP: GYN
36
MOHAMED-NOOR, QUINN, AND TAN
TABLE 1 Initial Cytological Changes and Histological Findings at Colposcopy and at Conization Cyto/histological abnormality
At initial cytology (n Å 699)
Invasive disease Minimal invasion ACISa CISb CIN IIIc CIN II CIN I Dysplasia, unspecified Other d
45 20 4 287 90 51 22 133 47
At colposcopic biopsy (n Å 422)
(6.4%) (2.9%) (0.6%) (41.1%) (12.9%) (7.3%) (3.1%) (19.0%) (6.7%)
11 4 1 218 74 26 11 39 38
(2.6%) (0.9%) (0.2%) (51.6%) (17.5%) (6.2%) (2.6%) (9.2%) (9.2%)
At conization (n Å 699) 9 43 6 379 84 62 33 10 73
(1.3%) (6.2%) (0.9%) (54.2%) (12.0%) (8.9%) (4.8%) (1.4%) (10.4%)
a
ACIS, adenocarcinoma in situ. CIS, carcinoma in situ. c CIN, intraepithelial neoplasia. d Other, atypia, atrophy, metaplasia, inflammation,and HPV changes. b
ablation in 2, and excision of the abnormal area in 1. Recurrent disease occurred from 3 months to 17 years after cone biopsy (mean 43.8 months, SD 56.4). Histopathological findings in 12 of the recurrent lesions were identical to the original cone biopsy histology. In another 6 cases, recurrent lesions were within one histological grade of findings at cone histology, and in 1 case normal histology of the hysterectomy specimen was found despite persistent cytological changes suggestive of CIN I and HPV infection following cone biopsy for ACIS. All patients re-treated for recurrent disease remained normal at subsequent follow-up, apart from one patient who was pregnant at the time of both initial and repeat cone biopsies. This patient had persistent abnormal cytology and subsequently underwent hysterectomy which showed evidence of CIS in the hysterectomy specimen. There were no invasive lesions in the recurrence group. Thirty-four patients had hysterectomy for other reasons (fibroids, menorrhagia) during the follow-up period, and no recurrent lesions were found in hysterectomy specimens from these patients. Five hundred twelve patients remained cytologically and colposcopically normal throughout the follow-up period. Of the 572 patients whose cone biopsy histology showed complete excision of abnormal epithelium, 553 (96.7%) were found by this study to have been cured of disease on the basis of normal follow-up or normal histology of hysterectomy specimens. Figure 1 outlines the follow-up and outcomes for this group. Patient Outcomes after Incomplete Excision In 127 patients, abnormal epithelium was considered to have been incompletely excised; 79 (62%) cases were at the level of the ectocervix, 38 (30%) at the level of the endocervix, and 10 (8%) at both margins. Of these patients, 33 had immediate hysterectomy (within 3 months) because of high-
AID
Gyn 4817
/
6d23$$$143
09-16-97 05:32:17
degree histological abnormalities in conization specimens, 17 showing CIS, 11 showing minimally invasive disease and 3 cases of invasive carcinoma. Despite incomplete excision, in only 18 (55%) cases was there evidence of residual lesions in hysterectomy specimens. Of these 18 cases of residual disease, 3 (30%) had been incompletely excised at the ectocervix, 11 (61%) at the endocervix, and 4 (8%) at both margins. The other 94 patients whose lesions were incompletely excised were followed up at the Dysplasia Clinic. Of these, 13 developed cytological or colposcopic recurrences of disease, and invasive carcinoma occurred in 3 patients who had had CIS diagnosed at cone biopsy. One of these patient had not attended for follow-up after 6 months following cone biopsy and subsequently presented 19 years later with Stage IV cancer of the cervix. The other 2 patients with recurrent invasive disease were diagnosed at 20 months and 7 years after cone biopsy. Both patients were followed up in the Dysplasia Clinic and the recurrences were detected by abnormal smears. Histology of hysterectomy specimens from both cases showed microcarcinoma and subsequent follow-up has been normal. Overall, recurrent disease of any grade occurred from 6 months to 10 years after cone biopsy (mean 44 months, SD 39.6). Of the other 10 patients with cytological recurrences, histology of hysterectomy specimens confirmed CIN III in 7 and CIN II in 1. No abnormal epithelium was found in hysterectomy specimens from the other 2 patients. In the 11 patients with proven recurrent disease after incomplete conization, abnormal epithelium was considered to have been incompletely excised in 1 at the endocervix, in 9 at the ectocervix, and in 3 at both margins. Eight patients had hysterectomy for other reasons during the follow-up period after incomplete conization, and no recurrent lesions were found in hysterectomy specimens
goa
AP: GYN
OUTCOME AFTER CERVICAL COLD KNIFE CONIZATION
FIG. 1.
Follow-up and outcomes after complete conization.
from these patients. Seventy-three patients remained cytologically and colposcopically normal throughout the followup period. Of the 127 patients whose cone biopsy histology showed incomplete excision of abnormal epithelium, 98 (77%) were found by this study to have been cured of disease on the basis of normal follow-up or normal histology of hysterectomy specimens. The cure rate for incomplete excision at the ectocervical margin was 86% (68 of 79 cases), 68% (26 of 38 cases) for incomplete excision at the endocervix, and 40% (4 of 10 cases) if excision was incomplete at both edges. Figure 2 outlines follow-up and outcomes for this group. DISCUSSION
This review of 699 conizations confirms the therapeutic effectiveness of cone biopsy with an overall cure rate of
FIG. 2.
AID
Gyn 4817
/
6d23$$$143
37
93%, a rate comparable to that of 87% reported by Bjerre et al. (7) and 93% by Ahlgren et al. (8). However, despite colposcopic evaluation and use of iodine staining prior to conization, 18% of cone biopsies performed at RWH during the study period resulted in incomplete excision. This figure is comparable to that reported by White et al. (20%) (4) and by Lopes et al. (24%) (9), and is lower than that reported by Buxton et al. (45%) (10). In the study reported here, factors which could have influenced accuracy of colposcopic evaluation and ease of operative procedure, such as increasing patient age and parity, were not associated with an increased likelihood of incomplete excision. It seems likely that, regardless of technique and patient factors, a fairly constant rate of incomplete excisions of about 20% can be expected for cervical conization. After complete conization, the cure rate in this study was
Follow-up and outcomes after incomplete conization.
09-16-97 05:32:17
goa
AP: GYN
38
MOHAMED-NOOR, QUINN, AND TAN
found to be 96.7%, comparable to previously reported figures of 98% by Ahlgren et al. (8). There is no ready explanation for the occurrence of persistent or recurrent disease in the remaining 3.3% of patients who were reported to have completely excised lesions. It has been postulated by White et al. that recurrences after complete conization may be due to multifocal disease, inadequate examination of the conization specimen, or true recurrence because of ongoing exposure to HPV infection (4). There may, therefore, be factors other than histopathological ones involved in determining the true completeness of a cone biopsy, and these are not apparent in the results obtained in this study. The results obtained here, however, clearly demonstrate the importance of adequate follow-up even in cases of histologically complete excision of CIN lesions. After incomplete excision, this study demonstrated a lower cure rate of 77% comparable to previously reported figures of 70% by Ahlgren et al. (8) and 56% by White et al. (4), and that the likelihood of developing persistent disease may be determined by which margin was incompletely excised. Incomplete excision at both margins is more likely to occur if there is a large area of abnormal epithelium or a multifocal abnormality, and this may partly explain the higher recurrence rate when this occurs. White et al., in analyzing cases of persistent disease after incomplete conization, found a positive correlation with smoking. Unfortunately, in this study, smoking and other patient factors were not able to be evaluated, but may be helpful in future prospective studies in predicting patients who may be more likely to have persistent disease. Seventy-seven percent of patients who had incomplete excision of CIN in this study were cured. Possible explanations for this phenomenon include spontaneous regression and inflammatory responses of healing, which may play a role in destroying areas of residual CIN at the margin of resection. This latter theory may be supported by findings in studies of laser conization, where the surface of the cone bed is unavoidably ablated. These studies describe cure rates of 88 (11) and 94% (12) after incomplete laser conization. Because the time from conization to detection of recurrence is quite variable, regardless of completeness of excision, no conclusion can be drawn as to the optimum duration of colposcopic follow-up after conization.
AID
Gyn 4817
/
6d23$$$143
09-16-97 05:32:17
CONCLUSIONS
Cone biopsy is very likely to cure CIN when the lesion is completely excised. However, even after complete conization, recurrent disease may occur in a small percentage of patients, and adequate follow-up remains an essential part of patient management after conization, regardless of histological findings in the conization specimen. Most cases of incompletely excised CIN will also be cured, especially if the incomplete margin is ectocervical, and cytological and colposcopic follow-up may be an acceptable alternative to hysterectomy in the management of such cases. REFERENCES 1. Gusberg SB, Marshall D: Intraepithelial carcinoma of the cervix: A clinical reappraisal. Obstet Gynecol 19:713–716, 1962 2. Richart RM: Cervical intraepithelial neoplasia. Pathol Annal 8:301– 305, 1973 3. Matseoane S, Williams SB, Navarro C, Hedriana H, Mushayandebvu T: Diagnostic value of conization of the uterine cervix in the management of cervical neoplasia: A review of 756 consecutive patients. Gynecol Oncol 47:287–291, 1992 4. White CD, Cooper WL, Williams RR: Cervical intraepithelial neoplasia extending to the margins of resection in conization of the cervix. J Reprod Med 36:635–638, 1991 5. Grace HF, Pattison NS: Cone biopsy of the cervix: Does surgical technique influence cytological follow-up? Aust N Z J Obstet Gynaecol 27:216–219, 1987 6. Ostor AG: Studies on 200 cases of early squamous cell carcinoma of the cervix. Int J Gynecol Pathol 12:193–207, 1993 7. Bjerre B, Eliasson G, Linell F, Soderberg H, Sjoberg NO: Conization as only treatment of carcinoma in situ of the uterine cervix. Am J Obstet Gynecol 125:143–152, 1976 8. Ahlgren M, Ingemarsson I, Lindberg LG, Nordqvist RB: Conization as treatment of carcinoma in situ of the uterine cervix. Obstet Gynecol 46:135–139, 1975 9. Lopes A, Morgan P, Murdoch J, Piura B, Monaghan JM: The case for conservative management of ‘‘incomplete excision’’ of CIN after laser conization. Gynecol Oncol 49:247–249, 1993 10. 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 70:529–532, 1987 11. Grundsell H, Alm P, Larsson G: Cure rates after laser conization for early cervical neoplasia. Ann Chir Gynaecol 72:218–222, 1983 12. Andersen ES, Nielsen K, Larsen G: Laser conization: Follow-up in patients with cervical intraepithelial neoplasia in the cone margin. Gynecol Oncol 39:328–331, 1990.
goa
AP: GYN