Systematic review and meta-analysis of individual serum lipids and analysis of lipid ratios in ankylosing spondylitis and healthy control cohorts: significantly lower mean HDL-cholesterol level in ankylosing spondylitis cohorts.

Abstract

Ankylosing spondylitis (AS) is suspected to have increased risk of atherosclerosis and cardiovascular disease (CVD) mortality. This systematic review and meta-analysis aims to critically study serum lipids and lipoprotein ratios in AS compared to healthy control (HC) subjects and determine any significant difference. English-language articles were systematically searched in PubMed, Ovid Medline, Embase (Medline records removed), and Scopus databases from 1970 to 2021. Random-effects model was used to pool results expressed as standardised mean difference (SMD) in the lipid outcomes. Lipid ratios of total ÷ HDL-C and the log10 (TG/HDL-C), i.e. atherogenic index of plasma (AIP), were analysed by histograms of differences in weighted means and weighted SDs between AS and HC exposure cohorts. The meta-analysis included a total of 68 articles, 47 from database search and 21 from reference reviews. Pooled Hedges' g effect size revealed no difference in mean total cholesterol, mean triglycerides, and mean LDL-C between AS and HC subjects. However, mean HDL-C was significantly (p<0.001) lower in AS than HC subjects, with pooled Hedges' g (SE) for HDL-C of -0.484 (0.092), with 95% mean CIs [-0.664, -0.305]. In comparing differencesin AS minus HC weighted means of total HDL-C ratios, 8 values in HC were below the lowest ratio in AS. Highly significantly lower HDL-C levels occurred in AS versus HC subjects. The lower HDL-C levels in AS than HC populations deserve further study and may be attributable to uninvestigated demographic, exercise capacity, or clinical manifestations.