Risk-stratified management of cervical high-grade squamous intraepithelial lesion based on machine learning.

The concordance rate between conization and colposcopy-directed biopsy (CDB) proven cervical high-grade squamous intraepithelial lesion (HSIL) were 64-85%. We aimed to identify the risk factors associated with pathological upgrading or downgrading after conization in patients with cervical HSIL and to provide risk-stratified management based on a machine learning predictive model. This retrospective study included patients who visited the Obstetrics and Gynecology Hospital of Fudan University from January 1 to December 31, 2019, were diagnosed with cervical HSIL by CDB, and subsequently underwent conization. A wide variety of data were collected from the medical records, including demographic data, laboratory findings, colposcopy descriptions, and pathological results. The patients were categorized into three groups according to their postconization pathological results: low-grade squamous intraepithelial lesion (LSIL) or below (downgrading group), HSIL (HSIL group), and cervical cancer (upgrading group). Univariate and multivariate analyses were performed to identify the independent risk factors for pathological changes in patients with cervical HSIL. Machine learning prediction models were established, evaluated, and subsequently verified using external testing data. In total, 1585 patients were included, of whom 65 (4.1%) were upgraded to cervical cancer after conization, 1147 (72.4%) remained having HSIL, and 373 (23.5%) were downgraded to LSIL or below. Multivariate analysis showed a 2% decrease in the incidence of pathological downgrade for each additional year of age and a 1% increase in lesion size. Patients with cytology > LSIL (odds ratio [OR] = 0.33; 95% confidence interval [CI], 0.21-0.52), human papillomavirus (HPV) infection (OR = 0.33; 95% CI, 0.14-0.81), HPV 33 infection (OR = 0.37; 95% CI, 0.18-0.78), coarse punctate vessels on colposcopy examination (OR = 0.14; 95% CI, 0.06-0.32), HSIL lesions in the endocervical canal (OR = 0.48; 95% CI, 0.30-0.76), and HSIL impression (OR = 0.02; 95% CI, 0.01-0.03) were less likely to experience pathological downgrading after conization than their counterparts. The independent risk factors for pathological upgrading to cervical cancer after conization included the following: age (OR = 1.08; 95% CI, 1.04-1.12), HPV 16 infection (OR = 4.07; 95% CI, 1.70-9.78), the presence of coarse punctate vessels during colposcopy examination (OR = 2.21; 95% CI, 1.08-4.50), atypical vessels (OR = 6.87; 95% CI, 2.81-16.83), and HSIL lesions in the endocervical canal (OR = 2.91; 95% CI, 1.46-5.77). Among the six machine learning prediction models, the back propagation (BP) neural network model demonstrated the highest and most uniform predictive performance in the downgrading, HSIL, and upgrading groups, with areas under the curve (AUCs) of 0.90, 0.84, and 0.69; sensitivities of 0.74, 0.84, and 0.42; specificities of 0.90, 0.71, and 0.95; and accuracies of 0.74, 0.84, and 0.95, respectively. In the external testing set, the BP neural network model showed a higher predictive performance than the logistic regression model, with an overall AUC of 0.91. Therefore, a web-based prediction tool was developed in this study. BP neural network prediction model has excellent predictive performance and can be used for the risk stratification of patients with CDB-diagnosed HSIL.

Clinical analysis of 314 patients with high-grade squamous intraepithelial lesion who underwent total hysterectomy directly: a multi-center, retrospective cohort study.

OBJECTIVE: To identify the risk factors of cervical high-grade squamous intraepithelial lesion(HSIL) complicated with occult cervical cancer and standardize the management of initial treatment for HSIL. METHOD: The clinical data of patients who underwent total hysterectomy directly due to HSIL in the obstetrics and gynecology department of two tertiary hospitals and three secondary hospitals from 2018 to 2023 were collected. Their general characteristics, pathological parameters and survival status were analyzed. Logistic regression model was used to analyze the correlation between clinical parameters and postoperative pathological upgrading. RESULT: 1. Among the 314 patients with HSIL who underwent total hysterectomy directly, 73.2% were from primary hospitals. 2. 25 patients (7.9%) were pathologically upgraded to cervical cancer, all of which were early invasive cancer. 3. Up to now, there was no recurrence or death in the 25 patients with early-stage invasive cancer, and the median follow-up period was 21 months(range 2-59 months). 4. Glandular involvement(OR 3.968; 95%CI 1.244-12.662) and lesion range ≥ 3 quadrants (OR 6.527; 95% CI 1.78-23.931), HPV 16/18 infection (OR 5.382; 95%CI 1.947-14.872), TCT ≥ ASC-H (OR 4.719; 95%CI 1.892-11.766) were independent risk factors that affected the upgrading of postoperative pathology. 5. The area under the curve (AUC) calculated by the Logistic regression model was 0.840, indicating that the predictive value was good. CONCLUSION: There is a risk of occult cervical cancer in patients with HSIL. Glandular involvement, Lesion range ≥ 3 quadrants, HPV 16/18 infection and TCT ≥ ASC-H are independent risk factors for HSIL combined with occult cervical cancer. The prognosis of biopsy-proved HSIL patients who underwent extrafascial hysterectomy and unexpected early invasive cancer was later identified on specimen may be good.

Development of a machine learning-based model for predicting positive margins in high-grade squamous intraepithelial lesion (HSIL) treatment by Cold Knife Conization(CKC): a single-center retrospective study.

OBJECTIVES: This study aims to analyze factors associated with positive surgical margins following cold knife conization (CKC) in patients with cervical high-grade squamous intraepithelial lesion (HSIL) and to develop a machine-learning-based risk prediction model. METHOD: We conducted a retrospective analysis of 3,343 patients who underwent CKC for HSIL at our institution. Logistic regression was employed to examine the relationship between demographic and pathological characteristics and the occurrence of positive surgical margins. Various machine learning methods were then applied to construct and evaluate the performance of the risk prediction model. RESULTS: The overall rate of positive surgical margins was 12.9%. Independent risk factors identified included glandular involvement (OR = 1.716, 95% CI: 1.345-2.189), transformation zone III (OR = 2.838, 95% CI: 2.258-3.568), HPV16/18 infection (OR = 2.863, 95% CI: 2.247-3.648), multiple HR-HPV infections (OR = 1.930, 95% CI: 1.537-2.425), TCT ≥ ASC-H (OR = 3.251, 95% CI: 2.584-4.091), and lesions covering ≥ 3 quadrants (OR = 3.264, 95% CI: 2.593-4.110). Logistic regression demonstrated the best prediction performance, with an accuracy of 74.7%, sensitivity of 76.7%, specificity of 74.4%, and AUC of 0.826. CONCLUSION: Independent risk factors for positive margins after CKC include HPV16/18 infection, multiple HR-HPV infections, glandular involvement, extensive lesion coverage, high TCT grades, and involvement of transformation zone III. The logistic regression model provides a robust and clinically valuable tool for predicting the risk of positive margins, guiding clinical decisions and patient management post-CKC.

[Clinical effect of secondary LEEP combined with transcervical resection of endocervical tissue for cervical precancerous lesions with positive internal margin after the first LEEP].

Objective: To preliminarily investigate the clinical outcomes of secondary loop electrosurgical excision procedure (LEEP) combined with transcervical resection of endocervical tissue (i.e., second combined surgeries) in patients with positive endocervical margins following the initial LEEP for high-grade squamous intraepithelial lesion (HSIL) or adenocarcinoma in situ (AIS) of the cervix. Methods: Patients who underwent second combined surgeries due to positive endocervical margins after the initial LEEP for cervical HSIL or AIS at Obstetrics and Gynecology Hospital, Fudan University between August 2015 and September 2023 were included. Postoperative cytological examinations, high-risk human papillomavirus (HR-HPV) testing, colposcopic biopsy results, and cervical canal length were followed up to evaluate the clinical efficacy of second combined surgeries. Results: (1) General clinical data: a total of 67 patients were enrolled, including 34 with cervical HSIL (HSIL group) and 33 with AIS (AIS group). In the HSIL group before the time of initial LEEP, the mean age was (41.3±5.3) years, with all patients positive for HR-HPV preoperatively. Preoperative cytology results revealed ≤low-grade squamous intraepithelial lesion (LSIL) in 13 cases and ≥HSIL in 21 cases. The preoperative cervical canal length was (3.71±0.17) cm. Patients in the AIS group before their the first LEEP were at an average age of (39.1±8.7) years old, with preoperative HR-HPV positive. Among them, 16 cases showed preoperative cytological results of ≤LSIL, while 17 cases showed ≥HSIL. The preoperative cervical canal length was (3.64±0.21) cm. (2) Pathological findings and postoperative follow-up of the HSIL group following second combined surgeries:in the HSIL group, the residual rate of HSIL in the endocervical canal tissue (ECT) was 24% (8/34). Out of the 34 HSIL patients, 10 cases (29%, 10/34) remained with positive endocervical margins post-second combined surgeries. Among these 10 patients, 5 cases (5/10) had no lesion detected in ECT, while the remaining 5 cases (5/10) exhibited HSIL in their ECT. Conversely, 24 patients (71%, 24/34) had negative endocervical margins after second combined surgeries. Of these 24 patients, 3 cases (12%, 3/24) were found to have HSIL in ECT, and 21 cases (88%, 21/24) had no lesion in ECT. During follow-ups conducted at 6 and 12 months post-second combined surgeries, the clearance rates of HR-HPV were 91% (31/34) and 100% (34/34), respectively. Notably, among the 29 patients (85%, 29/34) who were followed up for a period of 2 years or longer, all cases maintained a consistently negative HR-HPV status, highlighting the effectiveness of second combined surgeries in achieving long-term HR-HPV clearance (100%, 29/29). (3) Pathological findings and postoperative follow-up of the AIS group following second combined surgeries: the residual rate of AIS in the ECT following second combined surgeries among AIS patients was 15% (5/33). Out of the 33 AIS patients, 11 cases (33%, 11/33) had positive endocervical margins post-operation, among whom AIS was detected in the ECT of 2 cases (2/11), while 1 case (1/11) was diagnosed with adenocarcinoma in the cervical canal tissue (subsequently underwent radical surgery and was excluded from this study). In contrast, 22 patients (67%, 22/33) had negative endocervical margins post-operation, with AIS found in the ECT of 2 cases (9%, 2/22) and no lesions detected in the remaining 20 cases (91%, 20/22). Follow-up evaluations conducted at 6 and 12 months postoperatively revealed HR-HPV clearance rates of 91% (29/32) and 97% (31/32), respectively. All 32 (100%, 32/32) AIS patients were followed up for a duration of ≥2 years post-second combined surgeries, during which HR-HPV remained consistently negative. (4) Complications and cervical length following second combined surgeries: neither the HSIL group nor the AIS group experienced significant complications such as hemorrhage, infection or cervical canal adhesion. At the 6-month follow-up, the cervical length of both HSIL and AIS patients exceeded 3 cm. By the 12-month follow-up, the cervical length had recovered to 96.5% and 97.5% when compared to the original length, respectively, for the HSIL and AIS groups. Conclusions: For patients with HSIL or AIS who exhibit positive endocervical margins following the initial LEEP procedure, undergoing second combined surgeries presents as an optimal choice. This surgical intervention guarantees thorough excision of the lesion, and subsequent colposcopic follow-up evaluations consistently demonstrate an absence of residual disease or recurrence. Moreover, it augments the rate of sustained HR-HPV negativity, thereby contributing to more favorable clinical outcomes.

Recurrent High-Grade Squamous Intraepithelial Lesion After Loop Excision Procedure Versus Loop Procedure With Top Hat.

OBJECTIVE: To evaluate single-pass loop electrosurgical excision procedure (LEEP-SP) versus LEEP with top hat (LEEP-TH) in terms of treatment failure defined as high-grade squamous intraepithelial lesion (HSIL) cytology within 2 years' follow-up. METHODS: This single-institution cohort study used a prospectively collected cervical dysplasia database including all patients who underwent LEEP-SP or LEEP-TH for biopsy-proven cervical intraepithelial neoplasia between 2005 and 2019. RESULTS: Of 340 patients included, 178 underwent LEEP-SP and 162 LEEP-TH. The LEEP-TH patients were more likely to be older (mean age, 40.4 vs 36.5 years; p < .001) and have a positive preprocedure endocervical sampling (68.5% vs 11.8%; p < .001). Positive margins were found in 23 LEEP-SP (12.9%) and in 25 LEEP-TH (15.4%; p = .507). There was no significant difference in depth of excision between LEEP-SP (13.21 ± 23.19 mm) and LEEP-TH (17.37 ± 28.26 mm; p = .138). At 2 years, there was no difference in the rates of HSIL cytology (5.2% vs 6.3%; p = .698), any positive human papillomavirus test, or HSIL cytology (25% vs 15%; p = .284). The 57 patients undergoing repeat excision were more likely to be older (mean age, 40.95 vs 37.52 years; p = .023), have had a LEEP-TH (26.3% vs 73.7%; p < .001), and have initial cytologic HSIL (64.9% vs 35.0%; p < .001). CONCLUSIONS: In this single-institution study, there is no difference in the rate of recurrent HSIL in patients undergoing LEEP-SP versus LEEP-TH. A LEEP-TH may have limited additional benefit over a LEEP-SP in the treatment of cervical HSIL.

Recurrence in Cervical High-Grade Squamous Intraepithelial Lesion: The Role of the Excised Endocervical Canal Length-Analysis of 2,427 Patients.

OBJECTIVE: This study aimed to evaluate the influence of the excised canal length on relapse rates of cervical high-grade squamous intraepithelial lesion (HSIL) treated by loop electrosurgical excision procedure and to find a cut-off point, above which lower recurrence rates could be observed, with low probability of compromising future obstetric outcome, and the relationship with other individual factors related to HSIL recurrence. METHOD: This was a retrospective cohort study of 2,427 women diagnosed with cervical intraepithelial neoplasia CIN2+ who underwent cervical conization using the high-frequency loop electrosurgical excision procedure surgery technique, to analyze the role of endocervical canal length associated with individual factors in the recurrent disease after CIN2+ treatment and determine a cut-off point for the excised canal length needed to decrease the risk of disease relapse. RESULTS: In 2,427 cases, the relapse rate of HSIL treated was 12%. Compromised margins of conization, HIV+, and endocervical canal length were related directly to relapses ( p < .001). The cut-off point, by receiver operating characteristic curve, to calculate the endocervical canal length related to relapses was 1.25 cm of canal excised. Canal length of less than 1.25 cm increased the recurrence rate 2.5 times. Compromised margins and HIV+ increased recurrence rates by more than 5 times. CONCLUSION: Cervical HSIL recurrence was directly related to the endocervical canal length: excised canal length of 1.25 cm or more decreases recurrence rate; HIV and compromised margins increase the chance of recurrence by more than 5 times.

Benefit and burden in the Dutch cytology-based vs high-risk human papillomavirus-based cervical cancer screening program.

BACKGROUND: In 2017, the Dutch cervical cancer screening program had replaced the primary cytology-based screening with primary high-risk human papillomavirus-based screening, including the opportunity to participate through self-sampling. Evaluation and balancing benefit (detection of high-grade cervical intraepithelial neoplasia) and burden of screening (unnecessary referrals, invasive diagnostics, and overtreatment) is needed. OBJECTIVE: This study aimed to compare the referral rates, detection of high-grade cervical intraepithelial neoplasia, overdiagnosis, and overtreatment in the new high-risk human papillomavirus-based screening program, including physician-sampled and self-sampled material, with the previous cytology-based screening program in the Netherlands. STUDY DESIGN: A retrospective cohort study was conducted within the Dutch population-based cervical cancer screening program. Screenees with referrals for colposcopy between 2014 and 2015 (cytology-based screening) and 2017 and 2018 (high-risk human papillomavirus-based screening) were included. Data were retrieved from the Dutch Pathology Registry (PALGA) and compared between the 2 screening programs. The main outcome measures were referral rate, detection of high-grade cervical intraepithelial neoplasia or worse, overdiagnosis (cervical intraepithelial neoplasia grade 1 or less in the histologic specimen), and overtreatment (cervical intraepithelial neoplasia grade 1 or less in the treatment specimen). RESULTS: Of the women included in the study, 19,109 received cytology-based screening, and 26,171 received high-risk human papillomavirus-based screening. Referral rates increased from 2.5% in cytology-based screening to 4.2% in high-risk human papillomavirus-based screening (+70.2%). Detection rates increased to 46.2% for cervical intraepithelial neoplasia grade 2 or worse, 32.2% for cervical intraepithelial neoplasia grade 3 or worse, and 31.0% for cervical cancer, and overdiagnosis increased to 143.4% with high-risk human papillomavirus-based screening. Overtreatment rates were similar in both screening periods. The positive predictive value of referral for detection of cervical intraepithelial neoplasia grade 2 or worse in high-risk human papillomavirus-based screening was 34.6% compared with 40.2% in cytology-based screening. Women screened through self-sampling were at higher risk of cervical intraepithelial neoplasia grade 2 or worse detection (odds ratio, 1.38; 95% confidence interval, 1.20-1.59) and receiving treatment (odds ratio, 1.31; 95% confidence interval, 1.16-1.48) than those screened through physician-sampling. CONCLUSION: Compared with cytology-based screening, high-risk human papillomavirus-based screening increases detection of high-grade cervical intraepithelial neoplasia, with 462 more cervical intraepithelial neoplasia grade 2 or worse cases per 100,000 women but at the expense of 850 more cases per 100,000 women with invasive diagnostics indicating cervical intraepithelial neoplasia grade 1 or less.

Clinical practice variation and overtreatment risk in women with abnormal cervical cytology in the Netherlands: two-step versus see-and-treat approach.

BACKGROUND: Only a few small studies have compared the 2-step method (biopsy followed by treatment) with a see-and-treat (immediate treatment) approach in women both low-grade or high-grade referral cytology. The clinical practice variation in the Netherlands has not been reviewed before. OBJECTIVES: To determine overtreatment rates in the 2-step versus see-and-treat approach in women referred for colposcopy because of abnormal cytology results, and to evaluate clinical practice variation in the Netherlands. MATERIALS AND METHODS: This was a population-based retrospective cohort study including 36,581 women with a histologic result of the cervix identified from the Dutch Pathology Registry (PALGA) between 2016 and 2017. Odds ratios for overtreatment, defined primarily as cervical intraepithelial neoplasia grade 1 or less, were determined for the 2-step and see-and-treat approach in relation to age, high-risk human papillomavirus status, and referral cytology. RESULTS: Of the included women 10,713 women (29.3%) received the 2-step method; 6,851 women (18.7%) underwent see-and-treat; and 19,017 women (52.0%) received conservative management after colposcopy with histologic assessment with cytologic follow-up or another type of treatment. Despite the existence of a national guideline advising see-and-treat only in case of suspected high-grade disease in women who have completed their childbearing, there is a wide practice variation between the 2 strategies in the Netherlands, with 7.0-88.3% of the women receiving see-and-treat per laboratory. The median time between cytology and treatment was 1-2 months (range, 0-12 months) in women receiving see-and-treat and the 2-step method, respectively. A total of 4119 women (23.5%) were overtreated, with older women, high-risk human papillomavirus-negative women, and women with low-grade cytology results being more likely to be overtreated. Women with low-grade cytology results and see-and-treat were associated with a higher overtreatment rate than women receiving the 2-step method (65.0% [1414 of 2174] versus 32.1% [1161 of 3613], respectively; odds ratio, 3.34; 95% confidence interval, 2.92-3.82). However, in women with high-grade cytology results, see-and-treat was inversely associated with overtreatment (11.3% [529 of 4677] versus 14.3% [1015 of 7100], respectively; odds ratio, 0.68; 95% confidence interval, 0.58-0.81). CONCLUSION: A see-and-treat approach is justified only in women with high-grade cytology results who have completed their childbearing. There is a wide practice variation between the 2 strategies in the Netherlands, and gynecologists should adhere to the guideline to prevent overtreatment.

Syringe or mask? Loop electrosurgical excision procedure under local or general anesthesia: a randomized trial.

BACKGROUND: Loop electrosurgical excision procedure may be performed under local anesthesia or general anesthesia, and practice patterns differ worldwide. No randomized head-to-head comparison has been published to confirm or refute either practice. OBJECTIVE: This study aimed to compare loop electrosurgical excision procedure under local anesthesia vs general anesthesia regarding patient satisfaction and procedure-related outcomes such as rates of involved margins, complications, pain, and blood loss. STUDY DESIGN: Consecutive women referred to our colposcopy unit were recruited. Loop electrosurgical excision procedure was performed under local anesthesia with 4 intracervical injections of bupivacaine hydrochloride 0.5% or under general anesthesia with fentanyl, propofol, and a laryngeal mask with sevoflurane maintenance. The primary endpoint was patient satisfaction assessed on the day of surgery and 14 days thereafter using a Likert scale (score 0-100) and a questionnaire. Secondary endpoints included rates of involved margins, procedure-related complications, pain, blood loss, and surgeon preference. Results were compared using nonparametric and chi-square tests. RESULTS: Between July 2018 and February 2020, we randomized 208 women, 108 in the local anesthesia arm and 100 in the general anesthesia arm. In the intention-to-treat analysis, patient satisfaction did not differ between the study groups directly after surgery (Likert scale 100 [90-100] vs 100 [90-100]; P=.077) and 14 days thereafter (Likert scale 100 [80-100] vs 100 [90-100]; P=.079). In the per-protocol analysis, women in the local anesthesia arm had significantly smaller cone volumes (1.11 cm3 [0.70-1.83] vs 1.58 cm3 [1.08-2.69], respectively; P<.001), less intraoperative blood loss (Δhemoglobin, 0.2 g/dL [-0.1 to 0.4] vs 0.5 g/dL [0.2-0.9]; P<.001), and higher satisfaction after 14 days (100 [90-100] vs 100 [80-100]; P=.026), whereas surgeon preference favored general anesthesia (90 [79-100] vs 100 [90-100], respectively; P=.001). All other secondary outcomes did not differ between groups (resection margin status R1, 6.6% vs 2.1% [P=.26]; cone fragmentation, 12.1% vs 6.3% [P=.27]; procedure duration, 151.5 seconds [120-219.5] vs 180 seconds [117-241.5] [P=.34]; time to complete hemostasis, 60 seconds [34-97] vs 70 seconds [48.25-122.25] [P=.08]; complication rate, 3.3% vs 1.1% [P=.59]). In a multivariate analysis, parity (P=.03), type of transformation zone (P=.03), and cone volume (P=.02) and not study group assignment, age, body mass index, and degree of dysplasia independently influenced the primary endpoint. CONCLUSION: Loop electrosurgical excision procedure under local anesthesia is equally well tolerated and offers patient-reported and procedure-related benefits over general anesthesia, supporting the preferred practice in some institutions and refuting the preferred practice in others.

Specimen Fragmentation and Loop Electrosurgical Excision Procedure and Cold Knife Cone Biopsy Outcomes.

OBJECTIVE: Surgical technique for loop electrosurgical excision procedure (LEEP) and cold knife cone (CKC) emphasizes a uniform specimen, but sequelae of specimen fragmentation are not established. We evaluated outcomes between fragmented and unfragmented excisional biopsy specimens. MATERIALS AND METHODS: Loop electrosurgical excision procedure and CKCs from January 2010 to October 2013 were reviewed. Intraepithelial lesion grade, fragmentation, margin, and Endocervical curettage status were analyzed. Adenocarcinoma in situ and cancer were excluded. Repeat procedures during the study period were included in follow-up. Loop electrosurgical excision procedures with top hat with no separate fragments were analyzed independently versus those with fragmented LEEP and/or top hat. Indeterminate margin was defined as inconclusive or unevaluable margin, or intraepithelial lesion in unidentifiable margin or fragment. Outcomes involved residual or recurrent disease and repeat procedures for intraepithelial lesion. χ was used for statistical analysis. RESULTS: Fragmented specimens were more likely to have any positive margin (p = .01), multiple positive margins (p < .001), and indeterminate margin (p < .001) than unfragmented specimens. There was no significant difference in rates of positive, insufficient, or high-grade Endocervical curettage (p = .74, 0.54, 0.92). Patients with fragmented specimens were more likely to have high-grade lesion recurrence in the following 3 years (p = .04) versus patients with index unfragmented specimens, though not compared with those with unfragmented LEEP + top-hat cases. Overall rates of repeat LEEP/CKC or hysterectomy for dysplasia were not different (p = .56). CONCLUSIONS: Fragmentation of LEEP and CKC specimens is associated with higher rates of positive margins, recurrent high-grade intraepithelial lesions, and indeterminate margins. These may cause diagnostic uncertainty, require closer follow-up, and increase cost with more visits and studies.

Intraoperative Human Papillomavirus Test Predicts 24-Month High-Grade Squamous Intraepithelial Lesion Recurrence Saving Costs: A Prospective Cohort Study.

OBJECTIVES: The human papilloma virus (HPV) test is recommended in the posttreatment follow-up of cervical intraepithelial neoplasia. The aim of the study was to assess whether the intraoperative HPV (IOP-HPV) test had a similar diagnostic accuracy that HPV test performed at 6 months to predict high-grade squamous intraepithelial lesion (HSIL) recurrence. MATERIALS AND METHODS: In a prospective cohort study, 304 women diagnosed with HSIL by biopsy and/or endocervical curettage before treatment and/or confirmation in the histological specimen were included. Immediately after surgery, HPV testing was performed. This test was compared with the test at 6 months and other predictors of recurrence. Patients were followed for 24 months. An economic analysis was performed to compare the costs of IOP-HPV and HPV test at 6 months. RESULTS: Recurrence rate of HSIL was 6.2% (19 patients). The diagnostic accuracy of the IOP-HPV test to predict HSIL recurrence at 24 months was similar to the HPV test at 6 months, with comparative sensitivities of 100% versus 86.7%, specificities of 82.0% versus 77.9%, positive predictive values of 27.1% versus 18.1%, and negative predictive values of 100% versus 99.0%. Direct economic saving per high-grade intraepithelial lesion patient was 172.8 &OV0556;. CONCLUSIONS: The HPV test performed after loop electrosurgical resection procedure predicted recurrence of HSIL at 24 months with a similar diagnostic accuracy than the HPV test at 6 months. The use of the IOP-HPV test in the management of HSIL will allow early detection of the risk of recurrent disease and to save costs because of potential suppression of the need of HPV and follow-up controls at 6 months.

Can we predict surgical margin positivity while performing cervical excisional procedures?

We designed this study to evaluate any factors associated with positive surgical margin in conisation specimens and to determine the optimal cone size. The medical records of patients who had undergone a loop electrosurgical excision procedure (LEEP), cold-knife conisation (CKC) and needle excision of the transformation zone (NETZ) procedure were reviewed retrospectively. Two hundred and sixty eight women fulfilled the inclusion criteria. Univariate analyses showed that 'postmenopause', 'HSIL on smear', 'previous colposcopic examination revealing HSIL in endocervical curettage (ECC) material and in two or more ectocervical quadrants' and 'managing with LEEP' were significant predictors of surgical margin positivity. Nulliparous patients showed significantly lower rate of surgical margin positivity. 'Postmenopause', 'previous colposcopic examination revealing HSIL in ECC material and in two or more ectocervical quadrants' and 'HSIL on smear' were identified as independent predictors of surgical margin positivity according to multivariate analyses.IMPACT STATEMENTWhat is already known on this subject? Previous studies demonstrated 'menopause', 'Age ≥50', 'managing with LEEP', 'disease involving >2/3 of cervix at visual inspection', 'training level of the surgeon', 'cytology squamous cell carcinoma' and 'mean cone height' as factors associated with positive surgical margin in conisation specimens.What do the results of this study add? In our study, univariate analyses showed that 'postmenopause', 'HSIL on smear', 'previous colposcopic examination revealing HSIL in endocervical curettage material and in two or more ectocervical quadrants' and 'managing with LEEP' were associated with surgical margin positivity. On the other hand, nulliparous women showed significantly lower rate of surgical margin positivity compared with parous women. Multivariate analyses showed that 'postmenopause', 'previous colposcopic examination revealing HSIL in endocervical curettage material and in two or more ectocervical quadrants' and 'HSIL on smear' were independent predictors of surgical margin positivity in conisation specimens.What are the implications of these findings for clinical practice and/or further research? We can predict high-risk patients with regard to surgical margin positivity. Prediction of high-risk patients and management with a tailored approach may help minimise surgical margin positivity rates.

Value of top-hat procedure in management of squamous intraepithelial lesion.

AIM: To investigate the features of skip lesions and evaluate value of top-hat procedure in management of squamous intraepithelial lesion. METHODS: We reviewed the records of patients who underwent loop electrosurgical excision procedure (LEEP) in Peking University First Hospital between 2011 and 2016. Patients were confirmed to have CIN1-3. The term 'skip lesion' refers to lesion lying deep in cervical canal discontiguous with other lesions in transformation zone and was confirmed by top-hat. We compared their lesion grade in patients with or without skip lesion using logistic regression. We further reviewed patients who underwent subsequent hysterectomy within 6 months following LEEP and evaluated if top-hat procedure led to less residual lesions or was able to predict residual lesions. RESULTS: A total of 2260 patients were included and 595 underwent top-hat procedure. Thirty-nine out of 595 patients had skip lesions (6.5%), among whom two patients had CIN1 (5.1%), eight had CIN2 (20.5%) and 29 had CIN3 (74.4%). Logistical regression showed CIN3 was associated with higher risk of skip lesions compared to CIN1 (OR = 4.433, 95%CI: 1.036-18.964), while CIN2 was not (OR = 1.762, 95%CI: 0.366-8.471). Sixty-two patients underwent hysterectomy within 6 months following LEEP (CIN1-3), 24 underwent top-hat. Analysis revealed top-hat procedure did not result in less residual lesions. Colposcopy impression or prior HPV test was unable to predict skip lesions. CONCLUSION: About 9.4% patients with CIN3 had skip lesions in the study, which is associated with elevated risk for residual lesion. Top-hat procedure is able to detect skip lesions, but should not be performed on routinely because its prognostic value is not proved.

Factors Related to Overtreatment in the See-and-Treat Approach: A Retrospective Multicentric Observational Study.

OBJECTIVE: The aim of the study was to evaluate the association between colposcopic features, age, menopausal status, and overtreatment in women subjected to "see-and-treat" approach, to identify subgroups of patients in which this approach could be more appropriate. MATERIALS AND METHODS: Retrospective multicentric cohort study conducted on women older than 25 years, with a high-grade squamous cytological report and a visible squamocolumnar junction, in which colposcopy and the excisional procedure were performed at the same time without a previous cervical biopsy (see and treat). Overtreatment was defined as histopathological finding of cervical intraepithelial lesion grade 1 or normal tissue. RESULTS: Among the 254 included patients, the overall overtreatment rate was 12.6%, whereas in women with a grade 2 colposcopy, it was 3.2% and, in women with grade 1 colposcopy, it was 22.0%. Among the considered factors (age, menopause, and grade 1 colposcopy), only a positive association with overtreatment and grade 1 colposcopy emerged (odds ratio = 8.70, 95% CI = 2.95-25.62, p < .001). CONCLUSIONS: See and treat may be appropriate in women older than 25 years with a visible squamocolumnar junction and a high-grade squamous cervical cytology. Patients need to be informed about the higher risk of overtreatment in case of a grade 1 colposcopic impression, which however may still be considered acceptable. Patient's age and menopausal status should not influence the decision to propose a see-and-treat approach.

Risk Factors of Post-Large Loop Excision of the Transformation Zone Recurrent High-Grade Cervical Intraepithelial Lesion: A Prospective Cohort Study.

OBJECTIVE: The aim of the study was to identify the risk factors of post-large loop excision of the transformation zone (LLETZ) recurrent disease and the impact of colposcopic guidance at the time of LLETZ on that risk. MATERIALS AND METHODS: From December 2013 to July 2014, 204 patients who had undergone LLETZ for the treatment of high-grade intraepithelial lesion with fully visible cervical squamocolumnar junction were included. The use of colposcopy during each procedure was systematically documented. The dimensions and volume of LLETZ specimens were measured at the time of the procedure before formaldehyde fixation. All participants were invited for a follow-up. The primary endpoint was the diagnosis of post-LLETZ recurrent disease defined as the histologic diagnosis of a high-grade cervical intraepithelial lesion. RESULTS: The median duration of post-LLETZ follow-up was 25.8 months. Recurrent disease was diagnosed in 8 (3.6%) patients. Older than 38 years (adjusted hazard ratio [aHR] = 11.9, 95% CI = 1.6-86.0), history of excisional therapy (aHR = 21.6, 95% CI = 3.5-135.3), and the absence of colposcopy for the guidance of LLETZ (aHR = 6.4, 95% CI = 1.1-37.7) were found to significantly increase the risk of post-LLETZ recurrent disease. The dimensions and volume of the specimen were not found to have any impact. Only positive endocervical margins were identified to significantly increase the risk of post-LLETZ recurrent disease (aHR = 14.4, 95% CI = 2.0-101.1). CONCLUSIONS: Risk factors of post-LLETZ recurrent disease are older than 38 years, history of excisional therapy, positive endocervical margins, and lack of colposcopic guidance at the time of LLETZ.

Risk Factor Analysis of Persistent High-Grade Squamous Intraepithelial Lesion After Loop Electrosurgical Excision Procedure Conization.

OBJECTIVE: The aim of the study was to analyze the clinical outcomes of high-grade squamous intraepithelial lesion (HSIL) 6 months after loop electrosurgical excision procedure (LEEP). We explored the risk factors of persistent cervical HSIL after LEEP and evaluated the methods of follow-up. PATIENTS AND METHODS: This retrospective study included women who underwent a LEEP and had a diagnosis of HSIL in their LEEP specimen during 2011 to 2015. The purpose was to determine the risk factors among these women for having persistent HSIL disease at their 6-month follow-up visit. At their follow-up visit, each woman underwent cervical cytology and high-risk human papilloma virus (hrHPV) testing, colposcopy-directed punch biopsy, and/or endocervical curettage. RESULTS: A total of 3582 women were enrolled. There were 9 cases invasive cervical cancer found and 101 women had persistent HSIL. The persistence rate was higher in women 50 years or older. The circumference, length, and width of LEEP specimens did not differ statistically between the persistent and nonpersistent group. The persistence rate among women with positive LEEP specimen margins was higher than among women with negative margins. Positive endocervical margins were associated with a higher rate of persistence than positive ectocervical margins. Multivariate logistic analysis showed that age, positive margins, abnormal cytology, and positive hrHPV during follow-up were all independent risk factors for persistent HSIL lesions. CONCLUSIONS: Being 50 years or older, positive margins, particularly endocervical margins, and having abnormal cytology and positive hrHPV testing during follow-up were risk factors for persistent HSIL lesions after LEEP conization. Colposcopy plays an indispensable role in the diagnosis of persistent HSIL and progression.

[Relationship between various histological status of margins of LEEP and residual HSIL or worse at hysterectomy following conization].

Objective: To investigate the relationship between various loop electrosurgical excision procedure (LEEP) margin status and residual high grade squamous intraepithelial lesion (HSIL) or worse at hysterectomy following conization. Methods: The relevant clinicopathological data were collected in the Obstetrics and Gynecology Hospital, Fudan University from Jan. 2014 to Dec. 2015, including 947 cases who underwent hysterectomy within 6 months of LEEP. The residual HSIL or worse at hysterectomy were analyzed among the groups. (1) Clear margins, involved margins, and without 1 mm negative margins. (2) Only one positive margin, two positive margins and three positive margins. (3) A positive margin of internal ostium of cervix, of external ostium of cervix and of the basement. Results: (1) The histological evaluation of the uterine specimens showed residual HSIL or worse in 234 cases (24.7%, 234/947). The proportion of residual lesions was 7.3% (21/286) in population with clear margins, 33.2% (211/635) with involved margins, 7.7% (2/26) without 1 mm negative margins, respectively. The positive margins group had significant difference at the aspect of residual rate in contrast to the negative margins group and the without 1 mm negative margins group (P<0.01). Further studies conclusively showed that the proportion of residual lesions was very similar between the negative margins group and the without 1 mm negative margins group (P>0.05). (2) The involved margins were interpretable in 621 cases. This was detected in 25.3% (111/438) patients with only one positive margin, 47.4% (74/156) with two positive margins and 77.8% (21/27) among three positive margins, respectively (P<0.01). (3) Furthermore, there were 418 cases only one positive margin was definite, and the proportion of residual lesions was 31.0% (62/200) in population with a positive margin of internal ostium of cervix, 18.2% (31/170) of external ostium of cervix and 33.3% (16/48) of the basement. The residual rates were higher in the endocervical and basal margin groups than that in the ectocervical margin group, and the differences were significant (P<0.05). Conclusions: The risk of residual HSIL or worse is significantly greater with involved margins at hysterectomy following LEEP. Both the positive endocervical and basal margin are excellent predictors of residual diseases, while the without 1 mm negative margin may be not. Clinicians should avoid treating it as positive margin and prevent overtreatment.

[Analysis of recurrence and its influencing factors in patients with cervical HSIL within 24 months after LEEP].

Objective: To analyze clinical outcome of high grade squamous intraepithelial lesion (HSIL) within 24 months after loop electrosurgical excision procedure (LEEP), and to explore risk factors of recurrent cervical HSIL, the risk of progress into cervical cancer and methods of follow-up. Methods: This retrospective study was carried out on 1 005 patients who underwent LEEP, diagnosing with HSIL after LEEP from January 2011 to December 2013 at Obstetrics and Gynecology Hospital Affiliated to Fudan University to confer the difference between non-recurrent group and recurrent group 24 months after the LEEP conization. Patients were followed with ThinPrep cytologic test (TCT), high risk HPV (HR-HPV) test, colposcopy guided biopsy. Results: A total of 1 005 cases were enrolled in this study with HSIL in the LEEP specimen, no residual HSIL in the 6-month follow up, and have follow up records in 24 months after LEEP. HSIL recurred in 5 cases, microinvasive carcinoma in 1 case, low grade squamous intraepithelial lesion (LSIL) in 17 cases at 12 months follow-up. HSIL recurred in 8 cases, LSIL in 11 cases, adenocarcinoma in situ in 1 case, and invasive cervical carcinoma in 1 case in Ⅰb1 stage at 24 months after LEEP. The recurrence rate was 1.3% (13/1 005) , and the progression rate was 0.3% (3/1 005) . There was no significant difference in age, length, circumference and width of LEEP between recurrent and non-recurrent patients (P>0.05) . The recurrence rate was highest in the endocervical positive margin group with 3/16, which was higher than ectocervical positive margin and negative margin (P<0.01, P=0.040, respectively). The recurrence rate of endocervical positive margin group and fibrous interstitial positive margin group showed no significant difference (P=0.320) . There was no significant difference between ectocervical positive margin and negative margin [2.8% (2/72) vs 0.7% (6/882), P=0.117]. Postoperative cytological examination combined with HR-HPV detection has a high sensitive and high negative predictor value of HSIL recurrence with both 100.0%. Multivariate logistic regression analysis showed that positive endocervical positive margin, abnormal follow-up cytological examination and positive HR-HPV after LEEP were independent factors affecting recurrence of HSIL patients after LEEP (P<0.05). Conclusions: Age, length, circumference and width of LEEP have no effect on recurrence within 24 months after HSIL. The high risk factors for HSIL recurrence within 24 months after LEEP in HSIL patients include: positive HPV, abnormal cytology, and positive endocervical positive margin. Colposcopy biopsy and endocervical curettage have important role in diagnosing HSIL recurrence and progression.

Differences in the Immune Microenvironment of Anal Cancer Precursors by HIV Status and Association With Ablation Outcomes.

Background: Anal high-grade squamous intraepithelial lesions (HSILs) are the precursors to anal cancer and frequently persist or recur following electrocautery ablation (EA). Impaired mucosal immunity may facilitate anal carcinogenesis. We characterized the immune microenvironment of anal HSILs in correlation with human immunodeficiency virus (HIV) serostatus and ablation outcomes. Methods: Using immunohistochemistry, mucosa-infiltrating CD4+ and CD8+ lymphocytes were quantified in HSILs and benign mucosa from 70 HIV+ and 45 HIV- patients. Clinicopathological parameters were compared. Results: Anal HSILs harbored more T lymphocytes than benign mucosa regardless of HIV status (P ≤ .03). Total T lymphocyte count and CD8+ subset were significantly higher in HIV+ HSILs versus HIV- HSILs (median cell count, 71 vs 47; 47 vs 22/high power field [HPF]; P < .001), whereas the CD4+ subset was comparable between groups (median, 24 vs. 25; P = .40). Post EA, HSILs persisted in 41% of HIV+ and 19% of HIV- patients (P = .04). Unadjusted analysis showed trends toward EA failures associated with HIV seropositivity (incidence rate ratio [IRR], 2.0; 95% CI, .8-4.9) and increased CD8+ cells (IRR, 2.3; 95% CI, .9-5.3). Conclusions: Human immunodeficiency virus is associated with alterations of the immune microenvironment of anal HSILs manifested by increased local lymphocytic infiltrates, predominately CD8+. Human immunodeficiency virus seropositivity and excess mucosa-infiltrating CD8+ cells may be associated with ablation resistance.

Factors affecting residual/recurrent cervical intraepithelial neoplasia after cervical conization with negative margins.

To identify factors for predicting residual or recurrent cervical intraepithelial neoplasia (CIN) after cervical conization with negative margins. A total of 172 patients with histologically verified high-grade squamous intraepithelial lesions who underwent conization with negative margins were recruited at the General Hospital of Tianjin Medical University from December 2006 to January 2016. Follow-up comprised clinical examination, a liquid-based cytology test, a human papillomavirus (HPV) DNA genotyping test, colposcopy assessment, and if indicated, colposcopy-directed punch biopsy. The Kaplan-Meier method was used to analyze the median recurrent time, whereas log-rank tests and Cox regression models were used to determine the predictors of residual/recurrent CIN. Fourteen residual/recurrent cases (8.1%) were identified in 172 patients. In univariate analysis, cytologic abnormalities on follow-up (P = .000), conization method (P = .017), HPV positivity at any visit (P = .000), persistent HPV infection postconization (P = .000), persistent infection with the same HPV genotype (P = .000), and HPV positivity at 18 months after conization (P = .000) were predictive factors of residual/recurrent CIN. The results of multivariate analysis further revealed that persistent HPV infection postconization (P = .035), HPV positivity at 18 months after conization (P = .017), and cytologic abnormalities on follow-up (P = .000) had an increased risk of residual/recurrent CIN. During follow-up, patients with persistent HPV infection or cytologic abnormalities were at high risk of residual/recurrent CIN and should be identified for close surveillance and monitoring. Meanwhile, patients with HPV who became negative within 18 months after treatment had a low risk of recurrence.