There were 364 postmenopausal women who underwent colposcopy due to abnormal cytology and/or HPV test results, with a median age of 68 years (IQR: 66–71) and median parity of 3 (IQR: 2–5). Most of the participants (89.8%) were non-smokers and had one lifetime sexual partner (99.5%). 43.4% were sexually active at the time of evaluation. A history of HPV-related surgical treatment was present in 5.8% of the patients, in the form of LEEP, ECC, and cervical biopsies. Also, 9.1% had a history of gynecologic surgery, the most frequent being total abdominal hysterectomy with bilateral salpingo-oophorectomy. One patient (0.3%) was immunocompromised from previous radiotherapy (Table 1).
Table 1 Demographic and clinical characteristics
Among the 364 patients, high-risk HPV was positive in 58.2% and negative in 41.8%. HPV 16 was detected in 12.9%, HPV 18 in 5.5%, HPV 16 + other types in 3.0%, HPV 18 + other types in 1.1%, and other high-risk types alone in 35.7%. HPV was assessed with the Cobas test in all patients. In 28.3% of cases, HPV status changed from negative to positive over time, with a median duration of infection of 1 year (IQR: 0–2). Initial cytology revealed ASC-US in 44.2%, LSIL in 10.4%, ASC-H in 5.2%, HSIL in 4.9%, AGC in 3.8%, and adenocarcinoma in 0.5%; 30.8% had NILM. Repeat cytology was done in 82.1%, and the results were ASC-US (15.1%), ASC-H (1.3%), HSIL (1.3%), LSIL (4.7%), and NILM (76.9%). Liquid-based cytology was performed in 99.5%.
Colposcopy was performed in all patients and was adequate in 94.8%. In 55.5%, abnormal findings were noted, including acetowhite epithelium (47.5%), atypical vessels (1.9%), punctation (0.5%), mosaicism (0.5%), and some other minor findings. Transformation zone was observed in 91.2%, most frequent type 3 (52.5%).
Biopsy was achieved in 99.7% of the cases. Histopathology revealed chronic cervicitis (33.5%), CIN I (24.5%), CIN II (6.0%), CIN III (3.8%), carcinoma (various types) in 3.5%. Excision was performed in 15.4%, most often disclosing chronic cervicitis (25%), CIN II (21.4%), or CIN III (17.9%). Dual immunostaining was undertaken in 87.6%, with 70.2% positivity (Table 2).
Table 2 HPV test results, cytology results (Pap Smear/Bethesda 2014 Classification), biopsy and histology results, and immunohistochemical staining tests (p16/Ki-67)
Biopsies taken during adequate colposcopy mostly showed chronic cervicitis (33.3%), CIN I (25.2%), and koilocytic changes (13%). CIN II and CIN III were present in 5.8% and 4.1%, respectively. Malignancies like cervical adenocarcinoma (1.4%), squamous cell carcinoma (0.6%), carcinosarcoma (0.3%), and cervical serous carcinoma (0.3%) were also noted. Inadequate colposcopy, chronic cervicitis (36.8%) and koilocytosis (31.6%) were prevalent, and CIN I and CIN II were detected in every 10.5%. In the detection of CIN II+, there was no statistically significant difference observed between satisfactory and unsatisfactory colposcopy (p = 0.321) (Table 3).
Table 3 Biopsy results by colposcopy adequacy
CIN II + lesions were diagnosed in 22 (6.0%) of 364 women. The CIN II + group was a little younger (66.5 vs. 68 years, p = 0.201) and of equal parity (median 3 in both groups, p = 0.740). Median HPV infection duration was significantly longer in CIN II + patients (2 vs. 1 year, p < 0.001). Smoking was more frequent in the CIN II + group (18.2% vs. 9.6%, p = 0.261), and sexual activity was significantly higher (77.3% vs. 41.2%, p < 0.001). Prior HPV-related or gynecologic surgery did not differ significantly. HPV positivity was greater in CIN II + patients (95.5% vs. 55.8%, p < 0.001). HPV 18 was strongly associated with CIN II+ (19.0% vs. 8.4%, p = 0.041), but HPV 16 and other high-risk HPV were not significantly different. Colposcopy was adequate in most, with similar adequacy and transformation zone visibility in both groups. Yet, the abnormal colposcopic findings were far more frequent in the CIN II + group (86.4% versus 53.5%, p = 0.003). Type 3 transformation zone was the most prevalent subtype in both groups. Positive p16/Ki-67 immunostaining was found in all patients with CIN II+ (100% versus 24.6%, p < 0.001) (Table 4).
Table 4 Comparison of clinical and diagnostic features between CIN II + and Non-CIN II + Patients
On univariate logistic regression, a number of factors were significantly associated with CIN II + lesions. Increasing duration of HPV infection was an independent predictor (OR: 1.512, 95% CI: 1.093–2.093, p = 0.013). Sexual activity (OR: 4.847, 95% CI: 1.748–13.443, p = 0.002) and positive HPV test result (OR: 16.602, 95% CI: 2.208–124.828, p = 0.006) were also significantly associated with higher risk. Among all HPV genotypes, HPV 18 had a significant association with CIN II+ (OR: 5.042, 95% CI: 1.137–22.364, p = 0.033), while HPV 16 and combined types were not statistically significant. Abnormal colposcopy results were the other independent predictor (OR: 5.503, 95% CI: 1.599–18.940, p = 0.007). Age, parity, smoking, previous surgeries, colposcopy adequacy, transformation zone visibility and type, and immunostaining had no statistically significant associations (Table 5).
Table 5 Univariate logistic regression analysis for predictors of CIN II+