PREVALENCE OF SERRATED LESIONS IN AVERAGE-RISK POPULATION RECEIVING COLONOSCOPY SCREENING: A SYSTEMATIC REVIEW AND METAANALYSIS

Background Serrated lesions are precursors of approximately one-third of colorectal cancers (CRCs). They can be classified into hyperplastic polyps, sessile serrated polyps (SSPs), and traditional serrated adenomas (TSAs). Despite increasing knowledge on the histological categorisation of serrated lesions, the detection rate was highly variable and endoscopist dependent. This study aims to perform a systematic review and meta-analysis to determine the overall prevalence of serrated lesions and their subtypes in average-risk population undergoing CRC screening. Methods MEDLINE and Embase were searched to identify population-based studies that reported the prevalence of serrated lesions. Studies on average-risk populations using colonoscopy as a screening tool were included. The histologic classification of serrated lesions followed the criteria recommended by the World Health Organization. Metaprop was applied to model within-study variability by binomial distribution and Freeman-Tukey double arcsine transformation was adopted to stabilise the variances. The prevalence figures were presented by proportions and their 95% confidence intervals (CIs) using random-effects models. Results This meta-analysis included 17 studies involving 129,001 individuals. The overall prevalence rates of serrated lesions (19.0%, 95% CI=15.3%-23.0%), SSPs (2.5%, 95% CI=1.5%-3.8%) and TSAs (0.3%, 95% CI=0.1%-0.8%) were shown in figure 1 (figure 1). Subgroup analysis indicated their higher prevalence among male (22.0%, 95% CI=16.4%-28.2%) than female (14.0%, 95% CI=8.2%-21.0%), and Caucasian (25.9%, 95% CI=16.4%-28.2%) than Asian (14.6%, 95% CI=11.6%-18.2%). The prevalence of SSPs was also higher among Caucasian (2.9%, 95% CI=1.5%-4.7%) than Asian (0.7%, 95% CI=0.4%-1.2%). However, their prevalence were similar between male (2.9%, 95% CI=1.0%-5.7%) and female (2.4%, 95% CI=0.8%-4.9%). Conclusions This was the first systematic review and meta-analysis to determine the overall prevalence of serrated lesions and their different subtypes. The pooled prevalence estimates can be used as indicators for establishing high-quality CRC screening programmes. Future studies should be performed to evaluate the factors independently associated with the presence of serrated lesions, which could inform clinical practice to enhance their detection rate.