Abstract 108

Abstract

Background and Objective: A comprehensive review in 2003 reported adrenal incidentaloma (AI) prevalence of 0.6 (0.35-varies from 0.6 to 1.5% on computed tomography (CT)-based studies. However, a few subsequent studies reported higher (4.4-7.3%) prevalence, which is often cited as the prevalence of AI and is attributed to a greater detection rate with modern imaging. Hence, we performed a systematic review and meta-analysis to evaluate the prevalence of AI. Methods: The Pubmed database was searched for literature published between the years 2000 and 2021, without language restriction and selection of articles was performed as per the PRISMA guidelines. The data was analysed via meta-analysis for proportions using MedCalc Statistical Software version 19.2.6. Results: The meta-analysis included 22 studies of which 20 were retrospective. Pooled AI prevalence rates were 1.36% (95% CI: 1.33-1.39) and 2.19% (95% CI: 1.77-2.65) with I2 of 99.1% (99-99.3%) (Fig 1). The pooled (fixed effect model) AI prevalence rates were 3.89% (95% CI: 3.34-4.51) in smaller studies (n: < 1000) and 3.76% (95% CI: 3.35-4.2) in studies with selected population (screening for lung cancer, type 2 diabetes mellitus and living potential kidney donors). In larger studies (n: > 1000) including unselected population, the pooled AI prevalence rates were 1.34% (95% CI: 1.31-1.36) and 1.43% (95% CI: 1.03-1.89) with I2 of 99.5% (99.4-99.6%). Conclusions: Pooled AI prevalence in the meta-analysis is around 1.35%. Studies that are smaller or include selected population may overestimate the AI prevalence. However, larger, prospective studies are required to verify whether it is overestimation in smaller studies or underestimation in larger, retrospective studies.