Skinner, Elizabeth N. MD; Gehrig, Paola A. MD; Van Le, Linda MD
From the Department of Obstetrics and Gynecology, Division of Gynecology Oncology, University of North Carolina, Chapel Hill, North Carolina.
Received September 4, 2003. Received in revised form November 5, 2003. Accepted December 4, 2003.
Address reprint requests to: Dr. Linda Van Le, Division of Gynecology Oncology, University of North Carolina, Chapel Hill, CB 7570 Women's Hospital, Chapel Hill, NC 27599; e-mail: firstname.lastname@example.org.
Presented at the South Atlantic Association of Obstetricians and Gynecologists, The Homestead, Hot Springs, Virginia, January 28, 2003.
Each year, approximately 50 million women will undergo Papanicolaou testing (Pap tests) in the United States. Of these women, 3.5 million are diagnosed with cytologic abnormalities that will require additional follow-up and evaluation.1 In the era of managed care and restrictions on health care spending, the need for an efficient and effective treatment algorithm becomes most imperative.
Women diagnosed with high-grade squamous intraepithelial lesions (HSIL) require treatment to prevent the progression to cancer. Currently, the standard treatment for high-grade dysplasia is excision of the transformation zone, which is commonly performed by cold-knife conization or loop excision. One of the benefits of loop excision is that this procedure can be performed in the office under local anesthesia, with minimal blood loss and discomfort. Additionally, loop excision has the added advantage in that it can be therapeutic as well as diagnostic, and treatment outcomes are comparable to those observed with cold-knife conization.2,3
Although loop excision has been shown to successfully treat high-grade dysplasia,4 there are limited data available evaluating the rate and timing of recurrent high-grade dysplasia. Current guidelines from the American College of Obstetricians and Gynecologists recommend that once a woman has been treated for a preinvasive lesion, she should be followed up with cytologic evaluation every 3 to 4 months if feasible for approximately 1 year and then annually thereafter.5 To evaluate whether these guidelines could be further refined, we evaluated the characteristics of disease recurrence to see whether a more efficient surveillance schedule could be defined for following women after treatment for HSIL.
MATERIALS AND METHODS
After obtaining institutional review board approval, a chart review was performed from 1990 to 1999 to identify all women treated with loop excision for high-grade cervical dysplasia. This 10-year study period was selected to evaluate the clinical experience of using loop excision to treat high-grade dysplasia at our institution and to provide a database of women treated with this method who would then have at least 2 years of follow-up surveillance data. International Classification of Diseases, 9th Revision codes for cervical intraepithelial neoplasia (CIN) 2 or 3 were used to identify all women diagnosed and/or treated for CIN 2 or 3. Pathology reports were then reviewed, and those women with the diagnosis of high-grade dysplasia who subsequently underwent loop excision were included in the study. Because gynecologic pathologists at our institution primarily review all cytologic and histologic specimens, the specimens were not rereviewed for purposes of this study.
Using our hospital's centralized computer system, a database was generated containing patient information, including age, race, indication for excision, and pathologic diagnosis with margin status. Follow-up information included all subsequent Pap tests and/or cervical biopsies performed at our institution.
During this time period, women who were treated with loop excision for high-grade dysplasia were followed up for 2 years with conventional cervical cytology obtained from the endocervix and ectocervix performed every 3 to 4 months during the first year and every 4 to 6 months during the second year. If a Pap test returned showing moderate- or high-grade dysplasia or if a patient had 2 consecutive Pap tests showing low-grade dysplasia or atypical squamous cells of undetermined significance, a colposcopic examination was performed. Patients were designated as having “recurrent or persistent dysplasia” when a Pap test, tissue specimen, or both revealed CIN 2 or 3. Women returning for at least 1 Pap test within the first 2 years of their treatment were included in the statistical analysis. Patients with no follow-up within the first 2 years after treatment were excluded from the statistical analysis.
Posttreatment Pap test results were grouped into 3-month time intervals beginning with 0 to 3 months up to 24 months. Analysis of the data was performed with the midpoint of each time interval as the time of disease recurrence, assuming that this time point represented the actual time of recurrence. If the last documented visit for a patient showed no evidence of recurrence, withdrawal of that patient was assumed in the next time interval, with the assumption that withdrawal was independent of the recurrence of high-grade dysplasia.
The probability of disease recurrence was calculated by using Kaplan-Meier analysis. Piece-wise exponential modeling was used to calculate the hazard rate for recurrence of CIN 2 or 3 (SAS Institute, Inc, Cary, NC). Finally, hazard ratios were estimated by using the piecewise exponential modeling, with covariates adjusted for age, race, and margin status.
From 1990 to 1999, 705 loop excisions were performed in patients at our dysplasia clinic for biopsy-proven CIN 2 or 3. Of these patients, 546 (77%) women had at least 1 follow-up Pap test performed in our clinic. Five hundred twenty-six (96.3%) of these women had their first Pap test within 2 years of their primary treatment, whereas 20 (3.7%) women returned for their first follow-up Pap test more than 2 years after their procedure. One hundred fifty-nine (23%) women had no further follow-up for their dysplasia in our clinic.
Of the 526 women included in the study, 456 (86.7%) had no evidence of recurrent dysplasia, and 70 (13.3%) were diagnosed with recurrent CIN 2 or 3. Of those women diagnosed with high-grade disease, 51 (72.9%) had high-grade abnormalities noted on cytology, with 38 of these having histologic confirmation. Recurrence was documented in the remaining 19 (27.1%) women by the presence of high-grade dysplasia seen on histologic specimen alone. Of these, 9 women underwent hysterectomies within 3 months of their original procedure, whereas 2 women had repeat loop excisions within 3 months secondary to positive margins. Three patients had histologic confirmation of high-grade dysplasia that was discordant with the preceding Pap test, and 1 patient had documented recurrence after undergoing only histologic evaluation. Of those women with recurrent dysplasia, 52 (74%) were identified within the first 6 months of follow-up and the remaining recurrences were detected over the last 18 months of follow-up (Figure 1). No patient developed invasive squamous cell carcinoma after treatment with loop excision.
Table 1 shows the abstracted data for the women in the study, comparing age, race, and margin status between those with recurrent high-grade dysplasia and those without. An increase in age was independently associated with a 1.6-fold per decade increase in the risk of disease recurrence (95% confidence interval 1.29, 1.9). A negative margin on the pathology specimen was independently associated with 0.29-fold decrease risk of disease recurrence (95% confidence interval 0.17, 0.51). There was no statistical difference in risk of disease recurrence for those women whose pathology report described indeterminate margins or for those that had positive endocervical or ectocervical margins. Additionally, racial background was not shown to have an influence on risk of disease recurrence (P < .2).
After plotting a Kaplan-Meier curve to evaluate the probability of dysplasia-free survival, the inverse plot of this curve was then used to determine the probability of recurrent CIN 2 or 3 (Figure 2). The probability of having recurrent high-grade dysplasia was greatest during the first 6 months of surveillance. A second, but smaller, increase in the probability of recurrence was noted during the last 3 months of follow-up. The probability of recurrence at both of these time intervals was statistically higher than that seen during months 7 to 21 of surveillance (P < .001). Using the hazard rate to define the relative risk of recurrence, we saw that the risk of recurrence was greatest during the first 6 months of surveillance. This risk dropped significantly over months 7 to 21 of follow-up, with a small increase noted during the last 3 months of surveillance (Figure 3).
The treatment and management of women diagnosed with high-grade dysplasia continues to be an important issue for health care providers. Although loop excision has been shown to be highly effective in the treatment of CIN 2 and 3, little information exists on whether our posttreatment management provides appropriate care for women completing this treatment. Previous studies have examined whether surveillance schedules could be modified based on factors that might alter a woman's risk of recurrence. For example, correlation between margin status and risk of recurrence has been studied. Although some studies report a higher risk of recurrence in patients who had incomplete excision of the dysplasia, other studies suggest that the correlation between margin status and risk of recurrence is low.6–13 No studies have looked at the timing of disease recurrence as a means for defining posttreatment surveillance, and we propose that this approach may provide more universal guidelines with which to follow women treated for high-grade dysplasia.
Loop excision is extremely effective in the treatment of high-grade dysplasia, with the incidence of recurrence ranging from 5% to 25%.6–9,14 In our study, the overall rate of recurrent high-grade dysplasia during the first 2 years of follow-up was 13.3%. Of those women with recurrent CIN 2 or 3, 74% were detected during the first 6 months of follow-up, a finding similar to what was reported by Dietrich et al.6 The remainder of recurrences were detected over the subsequent months of follow-up, with a small, but significant increase noted in the last 3 months of surveillance. It is possible that the increased rate of high-grade dysplasia noted in the first 6 months of follow-up represents persistent dysplasia, rather than recurrent dysplasia, and that the true incidence of recurrent high-grade dysplasia is reflected by the small increase noted in the later months of follow-up.
Based on our findings, we feel that that current recommendations for postloop surveillance may be too stringent when considering the rate and timing of recurrent high-grade dysplasia. If the majority of women with abnormal Pap test results are identified within the first 6 months of follow-up, it is reasonable that posttreatment surveillance could begin 6 months after the initial treatment, with a subsequent Pap test performed at 12 months after treatment. If these Pap test results are normal, the women could then return to annual screening. If this Pap test shows abnormal cytology, the patient should undergo further testing with repeat cytology and colposcopy. ThinPrep liquid-based cytology (Cytyc Corp, Boxborough, MA) may enhance the sensitivity of surveillance because this method has been proven to have increased sensitivity and specificity when compared with conventional cytology.15 Additionally, use of human papillomavirus typing in equivocal cases may aid in earlier detection of high-grade dysplasia. By changing the surveillance schedule, clinicians should be able to provide screening for women at a time when they are most likely to have a recurrence, thus reducing the number of clinic visits and unnecessary Pap tests performed.
In addition to evaluating the timing of recurrent high-grade dysplasia, we analyzed the relative risk of recurrence based on margin status. In our study, a positive endocervical or a positive ectocervical margin was not independently associated with a statistically significant increase risk of disease recurrence. The literature contains conflicting reports on whether or not margin status alone can be used to predict the risk of recurrent disease.6–13 Given this information and our findings that the recurrence rate is greatest during the first 6 months of follow-up for all patients, risk factors, although important when counseling our patients, become less important when defining universal posttreatment surveillance strategies.
Twenty-three percent of women treated with loop excision had no further follow-up at our institution, a number similar to that which has been reported by others.6,9 No difference was noted in age or ethnic background between those who returned for follow-up and those who did not. Many of the patients in our study are referred to our university dysplasia clinic from the state health departments. Although these women failed to return for follow-up care at our institution, it is likely that many of these women returned to their referring clinics for continued medical care. If recurrent disease had been identified, these women would have been referred back to our dysplasia clinic for further management of their dysplasia. Additionally, none of these women returned to our hospital with the diagnosis of invasive cancer. Thus, although a subset of patients did not return for follow-up, some assumptions can be made about their care given the referral patterns in our state.
As our specialty continues to learn about and understand the nature of cervical dysplasia, adjustments in the management schema for dysplasia can be implemented. An example of this is the new guideline issued by the American Cancer Society, which provides evidence-based recommendations for cervical cancer screening.16 Along these lines, it seems appropriate to revise posttreatment surveillance for women treated for high-grade dysplasia to reflect the timing of disease recurrence. Nasiell et al17 observed that the time of progression of CIN 2 to carcinoma in situ or cancer ranged from 41 to 80 months depending on the age group studied. Given the slow rate of disease progression, alterations and abbreviations in the surveillance schedules should not compromise the welfare of patients with dysplasia.
With restrictions on healthcare spending, the need for efficient and cost-effective follow-up becomes increasingly important. Creating surveillance protocols that reflect the timing of disease recurrence could lead to the development of more cost-effective guidelines that would benefit both the healthcare provider, pertinent to healthcare dollars and clinical hours expended, and the patient, in regards to time spent away from work and family, without compromising the ability to detect and treat recurrent disease. A clinically and financially optimal surveillance schedule for women treated with loop excision for high-grade dysplasia would be to obtain Pap tests every 6months for 1 year and then return to annual screening. Our study indicates that such an abbreviated surveillance schedule is feasible.
1. Wright TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ; ASCCP-Sponsored Consensus Conference. 2001 Consensus guidelines for the management of women with cervical cytologic abnormalities. JAMA 2002;287:2120–9.
2. Huang LW, Hwang JL. A comparison between loop electrocautery excision procedure and cold knife conization for treatment of cervical dysplasia: residual disease in a subsequent hysterectomy specimen. Gynecol Oncol 1999;73:12–5.
3. Giacalone PL, Laffargue F, Aligier N, Roger P, Combecal J, Daures JP. Randomized study comparing two techniques of conization: cold knife versus loop excision. Gynecol Oncol 1999;75:356–60.
4. Ferenczy A. Management of patients with high-grade squamous intraepithelial lesions. Cancer 1995;76(suppl 20):1928–33.
5. Cervical cytology: evaluation and management of abnormalities. 2000 Compendium of Selected Publications. 2000. Technical Bulletin #183:330–6.
6. Dietrich CS 3rd, Yancey MK, Miyazawa K, Williams DL, Farley J. Risk factors for early cytologic abnormalities after loop electrocautery excision procedure. Obstet Gynecol 2002;99:188–92.
7. Flannelly G, Bolger B, Fawzi H, De Lopes AB, Monaghan JM. Follow up after large loop excision of the transformation zone: could schedules be modified according to risk of recurrence? Br J Obstet Gynaecol 2001;108:1025–30.
8. Gonzalez DI, Zahn CM, Retzloff MG, Moore WF, Kost ER, Snyder RR. Recurrence of dysplasia after loop electrosurgical excision procedures with long-term follow up. Am J Obstet Gynecol 2001;184:315–21.
9. Hanau CA, Bibbo M. The case for cytologic follow up after LEEP. Acta Cytol 1997;41:731–6.
10. 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 Path 1999;12:233–8.
11. Zaitoun AM, McKee G, Coppen MJ, Thomas SM, Wilson PO. Completeness of excision and follow up cytology in patients treated with loop excision biopsy. J Clin Path 2000;53:191–6.
12. Gardeil F, Barry Walsh C, Prendiville W, Clinch J, Turner M. Persistent intraepithelial neoplasia after excision for cervical intraepithelial neoplasia grade III. Obstet Gynecol 1997;89:419–22.
13. Felix JC, Muderspach LI, Duggan BD, Roman LD. The significance of positive margins in loop electrosurgery cone biopsy. Obstet Gynecol 1994;84:996–1000.
14. Biggig 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.
15. Abulafia O, Pezzullo JC, Sherer DM. Performance of ThinPrep liquid-based cervical cytology in comparison with conventionally prepared Papanicolaou smears: a quantitative survey. Gynecol Oncol 2003;90:137–44.
16. Saslow D, Runowicz CD, Solomon D, Moscicki AB, Smith RA, Eyre HJ, et al. American Cancer Society guideline for the early detection of cervical neoplasia and cancer [review]. CA Cancer J Clin 2002;52:342–62.
17. Nasiell K, Nasiell M, Vaclavinkova V. Behavior of moderate cervical dysplasia during long-term follow-up. Obstet Gynecol 1983;61:609–14.