| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the 1Department of Obstetrics and Gynecology, Taichung Veterans General Hospital, Taichung, Taiwan; 2Department of Adult Psychiatry, Taipei City Psychiatric Center, Taipei, Taiwan; 3Institute of Biomedical Sciences, National Chung-Hsing University, Taichung, Taiwan Affiliations, Department of Obstetrics and Gynecology, Taichung Veterans General Hospital, Taichung, Taiwan, Republic of China.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Patients who received conization due to histologic diagnosis of cervical intraepithelial neoplasia III from 1998 to 2000 and who had at least one cytologic/histologic follow-up within one year of conization (n = 449) were enrolled in our study. All available demographic and pathologic parameters were analyzed.
RESULTS: We performed multivariable logistic regression analysis to identify predictive factors for cervical intraepithelial neoplasia III persistence/recurrence. Age (greater than 50 years) was the only preoperative predictor and had an odds ratio equaling 3.070 (95% confidence interval [CI] 1.421–6.630, P = .004). Post–cone endocervical curettage was found to be the most statistically significant factor for predicting persistent disease (odds ratio 7.940, 95% CI 3.428–18.390, P < .001). Positive endocervical curettage was associated with 65.5% (36/55) of persistent disease, whereas negative endocervical curettage was associated with only 7.6% (26/342). Positive endocervical resection margins and multiple-quadrant disease also had predictive values with odds ratios equaling 2.972 (95% CI 1.401–6.281, P = .004) and 2.180 (95% CI 1.014–4.689, P = .046), respectively. The positive predictive values for age (> 50 years), positive endocervical curettage, positive endocervical resection margin, and multiple quadrant disease were 31.7%, 65.5%, 40.0%, and 21.9%, respectively.
CONCLUSION: We found that age is the only preoperative predictive factor. Pathologic parameters, including endocervical curettage, endocervical resection margins, and multiple-quadrant disease are the only postoperative predictive factors for cervical intraepithelial neoplasia persistence or recurrence found in our study. These factors should be considered in patient management before and after therapeutic conization for cervical intraepithelial neoplasia III.
LEVEL OF EVIDENCE: II-3
Conization is the most widely accepted method for treating cervical preinvasive disease. It is not only therapeutic but also diagnostic for occult cervical carcinoma. In most cases, conization is curative for CIN III (encompassing severe dysplasia and carcinoma in situ) and microinvasive cervical cancers. Much is known in predicting persistent/recurrent disease after conization for CIN. However, there has been little focus on CIN III, which has a higher frequency of persistence and progression to invasive cancer than CIN I or CIN II.3 Thus, it is important to identify the risk factors for predicting persistent/recurrent disease after conization for CIN III.
This study comprehensively examines the demographic and pathologic parameters for the preoperative and postoperative prediction of persistent/recurrent disease after conization for CIN III.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The procedures of conization have previously been published.4 All patients had received colposcopic examination for localization of lesions before conization. Conization was then performed with either diathermy loop (BEI Medical Systems Electrosurgical Loop Electrodes, Teterborough, NJ) or fine-needle electrode (CooperSurgical Inc, Trumbull, CT).5 The choice of equipment depended on the size of the lesion with the aim to remove the cervical lesion as a whole. After the excisional procedure, endocervical curettage (ECC) was performed with a Milan uterine curette (Germed USA Inc, Fresh Meadows, NY).
The preparation of the specimens has previously been reported.4 For each specimen, the pathologist reported the size of the sample, the severity of dysplasia on the cervix and ECC, the endocervical and ectocervical marginal status, the endocervical gland involvement, and the numbers of cervical quadrants involved. Unacceptable thermal artifacts or disorientation of specimens were reported separately. Persistent/recurrent disease was defined as histologic confirmation of mild dysplasia or more severe lesions from punch biopsy, conization, or hysterectomy within 12 months of conization. Condyloma was not regarded as positive. Positive resection margins were defined as CIN III presented at the edge of the cone specimen. Positive ECC was defined as the presence of CIN III on the ECC specimen, and multiple-quadrant disease was defined as CIN III present in three or four quadrants of the cone specimen. The depth of conization was measured from the external cervical os to the highest cut end of the cervical canal. An inadequate ECC denoted that neither endocervical nor squamous metaplasia tissues were found on ECC specimens.
Patients had a 1-week follow-up after conization. If no further treatment was necessary, follow-up at 3-month intervals with conventional cervical cytology in the initial 2 years was suggested. Patients with cytological findings equal to or more severe than atypical squamous cells were referred to the colposcopic clinic, where the decision to perform punch biopsies was made.
The conditions for study exclusion included vaginal involvement by intraepithelial lesions diagnosed before conization, fragmentation of cone specimens, missing ECC data, the absence of cytologic or histologic follow-up data in the medical records, CIN III lesions presented on preceding punch biopsy specimens but not on the cone specimens, lesions presented on ECC but not on the cone specimens, and adenocarcinoma in situ.
Demographic data and pathologic parameters for all patients with CIN III were obtained for analysis. Pearson 
2 test and Fisher exact test were used for univariable analysis. After excluding the insignificant factors, multivariable logistic regression analysis was used to test for the value of other demographic and pathologic parameters in predicting persistent/recurrent disease. We used Statistical Package for the Social Sciences for Windows 10.0 (SPSS, Chicago, IL) for statistical analyses. A P value of 
0.05 was regarded as statistically significant.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mean age of the study group was 45.7 years with a range from 22 to 77 years. The age of the patients with carcinoma in situ (CIS) was 47.2 ± 11.7 years and of those with severe dysplasia was 42.7 ± 10.3 years. The mean gravidity was 4.3 (0–13), and the mean parity was 3.1 (0–10). Of the 449 cases, CIS was diagnosed in 298 cases (66.4%) and severe dysplasia in 151 cases (33.6%). Endocervical gland involvement occurred in 293 (65.3%) cases. Ectocervical margin involvement was found in 30 cases (6.7%), whereas endocervical margin involvement occurred in 105 cases (23.4%). Positive ECC presented in 55 (12.2%) cases. Cervical intraepithelial neoplasia III lesions involved an average of 2.6 quadrants, while 225 (50.1%) cases involved 3 or 4 quadrants. The average depth of cone specimens was 15.0 mm (3–30 mm).
In total, 64 cases of histologically confirmed persistent/recurrent disease were found during patient follow-up, including 55 cases of CIN III, 5 cases of CIN II, and 4 cases of CIN I. Of the 124 patients who received hysterectomy within 12 months of conization, 53 (42.7%) were positive for CIN lesions, and 71 (57.3%) had no residual disease. However, the indications for hysterectomy included persistent/recurrent CIN or positive resection margins in 107 cases and adenomyosis or uterine leiomyoma in 47 cases (30 patients had both indications). The mean age of the hysterectomy group was 52.6 ± 11.2 years.
In Table 1, we tested whether there was a correlation between various demographic as well as pathologic parameters and persistence/recurrence of disease with 2 analysis. We found that statistically significant predictive factors for persistent disease included: age, gravidity, diagnosis, endocervical gland, endocervical resection margins, ECC, quadrants of disease, and depth of resection.
|
However, when the statistically significant factors in univariable analyses were further tested with multivariable logistic regression analysis, only age, ECC, endocervical margins, and multiple-quadrant disease had predictive values (Table 2). The odds ratio for age, ECC, endocervical margins, and multiple-quadrant disease were 3.070 (persistent/recurrent disease in 40 of 126 patients aged > 50 years and in 24 of 323 patients aged < 50 years), 7.940 (persistent/recurrent disease in 36 of 55 patients with positive ECC and in 26 of 342 patients with negative ECC), 2.972 (persistent/recurrent disease in 42 of 105 patients with positive endocervical resection margins and in 22 of 344 patients with negative margins), and 2.180 (persistent/recurrent disease in 49 of 225 patients with 3 or 4 quadrants involved and in 15 of 224 patients with only 1 or 2 quadrants involved), respectively. We infer from the odds ratio that positive ECC independently predicted persistent/recurrent disease more precisely than any other marker.
|
Age was the only factor that could predict persistent/recurrent disease before ongoing conization. Most of the patients with CIN III were aged between 30 and 50 years (Fig. 1A). When the percentile of persistent/recurrent disease was calculated within each 10-year-age period (Fig. 1B), it was clear that there was a difference between age groups above and below 50 years. There were no statistical differences between the three age groups younger than 50 years (P = .053, Pearson 2 test). At more than 50 years of age, the percentage of persistent/recurrent disease reached a plateau, thus there were no statistical differences between the three age groups above 50 years, either (P = .923, Pearson 2 test).
|
To implement these statistically significant risk factors for clinical utility, we calculated the performance, including sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of these items in predicting persistent/recurrent disease (Table 3). The positive predictive value for age (> 50 years), positive ECC, positive endocervical resection margins, and multiple-quadrant disease were 31.7%, 65.5%, 40.0%, and 21.9%, respectively.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mean age of occurrence of carcinoma in situ (CIS) in western countries is 38 years,11 whereas the mean age of the patients in this study was 47.2 years. Thus, information obtained from data in these studies cannot be applied directly to different cohorts without adjustment. Moore et al8 identified increasing age and severity of CIN disease as the only factors that predict residual dysplasia in hysterectomy specimens after cold knife cone. These studies were consistent with our findings that age was a statistically significant predictive factor for disease persistence/recurrence even after multivariable analysis. It is also the only demographic character that can predict persistent/recurrent disease preoperatively. The explanations for the role of age as a predictive factor include lesions of menopausal women retract deeper into the cervical canal than in younger women, more difficult surgical techniques are required on atrophic cervixes and vaginas found in older women, and the trends toward more severe disease with advanced age.
Positive postcone endocervical curettage was revealed to have the strongest association with persistent/recurrent disease in our study. Few reports emphasize the subject of postcone ECC. Husseinzadeh and Kalogirou both found ECC to be a predictive factor for residual disease in postcone hysterectomy specimens.7,12 Reports that disregarded the value of ECC had fewer cases (94 and 57 cases) in their studies.13,14 Notably, Kobak reported that postcone dysplastic ECC was a predictor of invasive cancer in patients with high-grade squamous intraepithelial lesions.15 Thus, the necessity for routine postcone ECC should again be emphasized.
Positive endocervical margin has long been a debate both for cold-knife cone and loop electrosurgical excision procedure in the prediction of residual dysplasia.6,8,16,17 In the few reports of CIN III treated with loop electrosurgical excision procedure, endocervical margins were consistently regarded as a predictive factor for persistent/recurrent dysplasia. Livasy et al18 recently reported a 39% versus 15% recurrent rate (P = .001) in cases of positive and negative endocervical margins for high-grade dysplasia, respectively. Gardeit et al19 have reported similar results. Our data are consistent with these findings.
In our previous study of patients with CIN III who received postconization hysterectomies,4 we found that multiple-quadrant disease was predictive of persistent disease compared with patients with one or two quadrants of disease (48.4% vs 25.9%, P = .030). Livasy et al18 reported similar findings. However, these results were determined by univariable analysis.
In testing the performance of these risk factors in clinical utility, we found that these risk factors all had high negative predictive values in predicting persistent/recurrent disease (92.4% to 93.6%) (Table 3). Among them, ECC had the highest positive predictive value (65.5%). Positive endocervical resection margin, an age greater than 50 years, and multiple-quadrant disease had positive predictive values of 40.0%, 31.7%, and 21.9%, respectively. These data are invaluable in predicting persistent/recurrent disease before and after conization for CIN III.
Although preoperative and postoperative prediction factors have been identified in this study, no single factor can accurately predict persistent/recurrent disease for conization at this time. Thus, it is critical that pathologists report their findings in detail for additional study. The presence of any of these statistically significant risk factors deserves early and close follow-up schedules or surgical intervention. Additional research on surrogate markers other than demographic or pathologic parameters may be needed for an even better increase in prediction rates.
|
Footnotes |
|---|
Corresponding author: Dr. Esther Shih-Chu Ho, Department of Obstetrics and Gynecology, Taichung Veterans General Hospital, 160 Chung-Kang Road Sec 3, Taichung City, Taiwan 40705, ROC; e-mail: chlu{at}mail.vghtc.gov.tw.
doi:10.1097/01.AOG.0000206777.28541.fc
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Harper DM, Franco EL, Wheeler C, Ferris DG, Jenkins D, Schuind A, et al. Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial. Lancet. 2004;364):1757–65.[Medline]
3. Ostor AG. Natural history of cervical intraepithelial neoplasia: a critical review. Int J Gynecol Pathol 1993;12:186–92.[Medline]
4. Lu CH, Liu FS, Tseng JJ, Ho ES. Predictive factors for residual disease in subsequent hysterectomy following conization for CIN III. Gynecol Oncol 2000;79:284–8.[Medline]
5. Ferenczy A. Electroconization of the cervix with a fine-needle electrode. Obstet Gynecol 1994;84:152–9.
6. Mohamed Noor K, Quinn MA, Tan J. Outcomes after cervical cold knife conization with complete and incomplete excision of abnormal epithelium: a review of 699 cases. Gynecol Oncol 1997;67:34–8.[Medline]
7. Husseinzadeh N, Shbaro I, Wesseler T. Predictive value of cone margins and post-cone endocervical curettage with residual disease in subsequent hysterectomy. Gynecol Oncol 1989;33:198–200.[Medline]
8. Moore BC, Higgins RV, Laurent SL, Marroum MC, Bellitt P. Predictive factors from cold knife conization for residual cervical intraepithelial neoplasia in subsequent hysterectomy [see comments]. Am J Obstet Gynecol 1995;173:361–6.[Medline]
9. Paterson-Brown S, Chappatte OA, Clark SK, Wright A, Maxwell P, Taub NA. The significance of cone biopsy resection margins. Gynecol Oncol 1992;46:182–5.[Medline]
10. Houlard S, Perrotin F, Fourquet F, Marret H, Lansac J, Body G. Risk factors for cervical stenosis after laser cone biopsy. Eur J Obstet Gynecol Reprod Biol 2002;104:144–7.[Medline]
11. Shingleton HM, Orr JW Jr. Cancer of the Cervix: Screening. Philadelphia (PA): JB Lippincott Company; 1995. p 18.
12. Kalogirou D, Antoniou G, Karakitsos P, Botsis D, Kalogirou O, Giannikos L. Predictive factors used to justify hysterectomy after loop conization: increasing age and severity of disease. Eur J Gynaecol Oncol 1997;18:113–6.[Medline]
13. Vierhout ME, de Planque PM. Concomitant endocervical curettage and cervical conization. Acta Obstet Gynecol Scand 1991;70:359–61.[Medline]
14. Felix JC, Muderspach LI, Duggan BD, Roman LD. The significance of positive margins in loop electrosurgical cone biopsies. Obstet Gynecol 1994;84:996–1000.
15. Kobak WH, Roman LD, Felix JC, Muderspach LI, Schlaerth JB, Morrow CP. The role of endocervical curettage at cervical conization for high-grade dysplasia. Obstet Gynecol 1995;85:197–201.[Abstract]
16. Demopoulos RI, Horowitz LF, Vamvakas EC. Endocervical gland involvement by cervical intraepithelial neoplasia grade III: predictive value for residual and/or recurrent disease. Cancer 1991;68:1932–6.[Medline]
17. Narducci F, Occelli B, Boman F, Vinatier D, Loroy JL. Positive margins after conization and risk of persistent lesion. Gynecol Oncol 2000;76:311–4.[Medline]
18. Livasy CA, Maygarden SJ, Rajaratnam CT, Novotny DB. Predictors of recurrent dysplasia after a cervical loop electrocautery excision procedure for CIN-3: a study of margin, endocervical gland, and quadrant involvement. Mod Pathol 1999;12:233–8.[Medline]
19. Gardeil F, Barry-Walsh C, Prendiville W, Clinch J, Turner MJ. Persistent intraepithelial neoplasia after excision for cervical intraepithelial neoplasia grade III. Obstet Gynecol 1997;89:419–22.[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
