| Summary: | INTRODUCTION: Worldwide, there is a paradigm shift in the screening for cervical cancer with the use of high-risk human papillomavirus (hrHPV) molecular testing. Before South Africa (SA) adopts this technology in the public sector, health funders will need data on the performance of the current cytology and colposcopy-based programmes. This study was done to establish baseline data on the performance of the cytology and colposcopy based cervical cancer screening programme at the Groote Schuur Hospital (GSH) colposcopy clinic. METHODS: This was a retrospective cohort study of all the women with high grade squamous intraepithelial lesion (HSIL) Pap smears seen at GSH colposcopy clinic between 01 January 2010 and 31 December 2015. The outcome measures were; diagnostic concordance between cytology, colposcopy and histology, large loop excision of the transformation zone (LLETZ) and cone biopsy complication rates, cure rate, treatment failure and invasive cervical cancer rates, median time from treatment to recurrence, follow-up default rates. Data were managed and analysed using IBM SPSS Statistics Version 25 and Microsoft Structured Query Language (SQL) version 2014. Regression methods were used to assess the independent effect of baseline sociodemographic characteristics and clinical covariates on treatment failure and clearance of disease in those who had persistent disease after treatment. Kaplan-Meier curves were used to represent the time from treatment to recurrence and from persistence to cure. Time-to-event methods were applied to determine factors associated with treatment to recurrence and persistence to cure. RESULTS: A total of 7601 women were referred to the GSH colposcopy clinic during the study period. HSIL or worse lesions (≥HSIL) were confirmed histologically in 74.1% (2282/3081) women. At the four-month follow-up visit, 61.2% (742/1213) of the women were considered cured, and 17.0% (206/1213) had persistent/residual disease. In women considered cured at four months, recurrence was very low, and it peaked at ten months at 1.5% (11/740). By 24 months the cumulative recurrence rate was 4.6% (34/742). In women with persistent disease at the four-month follow-up visit, only 0.5% (1/202) developed invasive cervical cancer. The default rate for follow-up was very high, at 81% at 24 months. LLETZ and cone biopsy complication rate was 7.2% (117/1628). Log-rank analysis showed that parity ≥ four was significantly associated with a higher risk of disease recurrence (p=0.0004). In a Coxregression model, taking HAART was the only factor associated with a reduced risk of disease recurrence (p=0.0261). CONCLUSION: LLETZ and cone biopsy are safe procedures. After cure, recurrence rates are low. In women who are treated for HSIL, cervical cancer is very rare. Taking HAART was associated with a reduced risk of disease recurrence. There is a need to mitigate on higher default rates to follow up.
|
|---|
