Clinical Outcomes of Cases with Absent Cervical Dysplasia in Cold Knife Conization Specimens

Cold knife conization is a surgical procedure that provides both diagnosis and treatment of cervical lesions at the same session. It is mainly performed for an indication of high-grade cervical intraepithelial neoplasia (CIN) (i.e. CIN2/3) or in case of suspicion for an occult endocervical lesion. It is also used to diagnose an occult invasive cancer at a very early stage (Apgar et al., 2013). Although today it is largely replaced by loop electrosurgical excision procedure (LEEP), cold knife conization is still considered by many gynecologists as a gold standard for treating premalignant conditions of the cervix. Some women who undergo conization for CIN2/3 show no dysplastic lesions in the surgical cone specimen (Koc et al., 2013). This situation not only raises concern for a possible misdiagnosis, but also creates uncertainty for the appropriate steps to take during follow-up. In this study, clinical outcome of women undergoing conization for CIN2/3 that subsequently receive a negative pathological result for cervical dysplasia were investigated at a tertiary referral center.


Materials and Methods
This retrospective cohort study was performed at a tertiary referral hospital in Ankara, Turkey after obtaining approval from the institutional review board. Gynecologic oncology department surgical database was retrospectively reviewed between January 1 st 2008 and August 1 st 2012 for CIN2/3 cases that underwent cold knife conization. All of the conization procedures were performed by experienced gynecologic oncologists. Final pathological examination reports were obtained from the pathology department database. Group 1 consisted of cases in which no dysplastic changes had been reported in conization specimens (cone-negative), and Group 2 comprised cases that had received a diagnosis of CIN or invasive cancer (cone-positive). All cases were retrospectively evaluated for during their 1-year post-conization follow-up.
Study data included patient age, gravidity, parity, menopausal status, preoperative pap-smear result and preoperative high-risk Human papillomavirus (HPV) DNA status. For cases in Group 1, non-dysplastic conditions present in the specimen were also recorded.
Using the outpatient database, 1-year follow-up for each case was evaluated, which included visits every 3 months. During each of these visits, a pelvic examination and a pap smear was performed. An HPV-DNA test was also performed at the 4 th visit (1 year post-conization). Conventional Pap test was used for assessing cervical cytology. Polymerase chain reaction (PCR) method was used to identify high-risk HPV-DNA 18,31,33,35,39,45,51,52,56,58,59,66,68,and 82).
Statistical analyzes were performed using IBM SPSS 20.0 software (IBM, Armonk, NY, USA). Categorical variables were expressed as number and percentage, continuous numeric variables as mean±standard deviation (SD), and discrete variables as median and range. Taking into consideration the sample sizes in each group and data distribution, Student's t, Mann-Whitney U, Chi-square or Fisher's exact tests were performed to compare groups for the aforementioned variables. P values less than 0.05
One year follow-up data were analyzed for all study cases. A scheme of cytological results obtained during this period is presented in Figure 1. Considering both groups, there were a total of 17 histological recurrences. Recurrence rates in Group 1 and Group 2 were 9.1% (2/22) and 9.9% (15/151), respectively (p>0.05). In Group 1 follow-up, one case was diagnosed with CIN1 and case with HSIL was consequently diagnosed with CIN2. In Group 2 follow-up, 12 cases were diagnosed with CIN1, 2 cases with CIN2 and 1 case with CIN3. A total of 35 cases (20.2%) required hysterectomy within one year after conization. Hysterectomy rates were 3/22 (13.6%)

Discussion
Uterine cervical cancer is a malignancy that most frequently effects underdeveloped and developing countries with poor resources for providing adequate screening opportunities for their citizens (Jemal et al., 2011). Unlike many other organ sites with premalignant lesion of the cervix to progress into invasive cancer (Vink et al., 2013). Histological diagnoses of CIN 2 and CIN 3 are generally considered to have increased risk for progression to cancer (Montz, 2000). For this reason, especially in women with no fertility concerns, gynecologists generally prefer to excise the transformation zone either by cold knife conization or LEEP (Jancar et specimen may infrequently occur. We aimed to outline the clinical outcome of such cases in this study. After analyzing final pathological reports of the surgical specimens, we found that 22 of 173 cases (12.7%) that underwent conization for CIN2/3 was reported as not having any cervical dysplasia (cone-negative). This rate is in compliance with the previously published papers, in which approximately 8-25% of cases that had conization for abnormal cytology or histology were consequently reported as cone-negative (Golbang et al., 1997;Diakomanolis et al., 2003;Murta et al., 2006;Koc et al., 2013). In a study by Koc et al. (2013), among 202 conizations, the authors reanalyzed 25 cone specimens that were previously reported as negative (Koc et al., 2013). They reevaluated preceding punch biopsies that were reported as low-grade and high-grade CIN. After all conization slides were negative for a dysplastic lesion. However, upon reexamining the punch biopsy materials, they found that there were 13 false positive diagnoses in cases reported both as low-grade (n: 9) and high-grade (n: 4) CIN. This study showed that a negative cone biopsy could be a result of a false-positive punch biopsy.
In previous reports, it was reported that conditions such as cervicitis, tubal metaplasia, atrophy, immature squamous metaplasia, and basal cell hyperplasia were most frequently associated with a false-positive punch biopsy report that could eventually lead to negative et al., 2006). In our study, most common non-dysplastic diagnoses in cone specimens were also cervicitis (n: 11, 50%), metaplasia (n: 7, 31.8%) and HPV related changes (n: 3, 13.6%). In some cases, these non-dysplastic lesions might be misdiagnosed as CIN. However, diagnostic accuracy is generally high in CIN2/3, and false-positive results are much less frequent when compared to CIN1 (Koc et al., 2013). In another study by Weigl et al. (2006), 208 consecutive cone specimens were retrospectively evaluated, and 22 cases (10.6%) with benign cervical lesions following a diagnosis of CIN in punch biopsy their frequencies were reported to be very similar to our (18.1%), and HPV infection in 1 (4.5%) and combined lesions in 5 cases (22.7%). Weigl et al. also emphasized be long lasting HPV infections that might result in mild to moderate dysplasia (Weigl et al., 2006). This claim was repeated in a very recent study that investigated factors predicting absence of CIN in the cone specimens (Rodriguez-Manfredi et al., 2013). The authors stated that cases with negative pre-conization HPV test or a low viral load had a high probability of having a negative cone result. In contrast with these reports, we found in our study that cone-negative cases (Group1) had similar high-risk HPV infection rates with cone-positive cases (Group 2).
With regards to outcome of these cases, a study by negative conization (Murta et al., 2006). All patients in this study had undergone conization (cold knife or LEEP) or hysterectomy for a diagnosis of CIN. The frequency of negative conizations in their study was 15.5%. They reported that recurrences were observed less frequently and later cone-negative cases (7.1% vs 11.2%). In our study, we also analyzed 1-year follow-up of all cases, and histological recurrence rates were similar between cone-negative and cone-positive cases (9.1% vs 9.9%) (p>0.05). Therefore, we could not demonstrate an outcome advantage for cone-negative cases in our study. A question still remains to be answered: Could we reduce the discordance rates between punch biopsies and conization specimens? As an answer, the authors of a previously published paper recommended that taking pap-smears after treating atrophy and inflammation, performance of colposcopy by an expert in a cautious cytology-biopsy discordance could reduce the number of negative cone biopsies (Golbang et al., 1997).
In summary, our study suggested that outcomes were similar in terms of histological recurrence and HPV persistence after 1 year of follow-up between cone-negative and cone-positive cases. Therefore, we recommend no additional interventions in cone-negative women. These cases should have follow-up visits similar to cone-positive cases. Treatment of any infectious or atrophic conditions prior to performing a cervical biopsy or conization procedure should be practiced in order to achieve accurate results.