Abstract / Summary
Although multiple prospective studies have examined associations between lipid metabolism disorders and the risk of type 2 diabetes (T2D), a systematic review and data updates are still lacking. Nowadays, some noval lipid metabolism biomarkers have received increasing attention, but who is the most predictive biomarkers of T2D? this study aims to conduct the in-depth exploration. We conducted an umbrella review of meta-analyses of prospective cohort studies by searching PubMed, Web of Science, Cochrane Library, and Embase from inception to 14 February 2025. Methodological quality was assessed using the AMSTAR 2, and evidence strength was evaluated using predefined credibility criteria. The study protocol was registered in PROSPERO (CRD420250649341). A total of 13 meta-analyses, 133 original articles (including 39 newly included articles) and 1, 226, 322 participants were included. The results of meta-analysis showed significant positive correlation between several lipid metabolism parameters and the risk of T2D. The strongest associations were observed for the hypertriglyceridemic waist (HTW) phenotype [Relative risk (RR) = 3.54, 95% CI: 1.92, 6.53]. Significant positive correlations were also confirmed for the lipid accumulation product (LAP) (RR = 2.94, 95% CI: 2.31, 3.73), the triglyceride glucose (TyG) index (RR = 2.51, 95% CI: 2.13, 2.95), the atherogenic index of plasma (AIP) (RR = 1.97, 95% CI: 1.54, 2.52), the visceral adiposity index (VAI) (RR = 1.77, 95% CI: 1.61, 1.94), the triglyceride to high-density lipoprotein cholesterol (TG/HDL-C) ratio (RR = 1.51, 95% CI: 1.36, 1.69), and non-high-density lipoprotein cholesterol (non-HDL-C) (RR = 1.27, 95% CI: 1.07, 1.52). In contrast, lipoprotein(a) [Lp(a)] showed an inverse association (RR = 0.73, 95% CI: 0.56, 0.96). Dose-response meta-analysis suggested that there was a significant linear relationship between the TG/HDL-C ratio and T2D risk (P for nonlinearity = 0.2502), whereas significant nonlinear associations were observed for the VAI, LAP, TyG index, and AIP (P for nonlinearity < 0.05). Furthermore, the analysis results of fatty acids revealed positive associations between palmitic acid (C16:0; RR = 1.19, 95% CI: 1.07, 1.33), palmitoleic acid (C16:1n-7; RR = 1.33, 95% CI: 1.19, 1.49), dihomo-γ-linolenic acid (DGLA, C20:3n-6; RR = 1.36, 95% CI: 1.21, 1.53) and the risk of T2D. In contrast, negative associations were found between pentadecanoic acid (C15:0; RR = 0.82, 95% CI: 0.73, 0.92), arachidic acid (C20:0; RR = 0.78, 95% CI: 0.68, 0.90), linoleic acid (LA, C18:2n-6; RR = 0.80, 95% CI: 0.74, 0.87) and the occurrence of T2D. This systematic review supportsed the predictive value of multiple lipid biomarkers for the risk of T2D, especially some composite indicators such as the HTW, LAP, TyG index, AIP, VAI and TG/HDL-C ratio, but larger-scale and longer-term prospective studies are warranted to validate these associations.
Primary Source
Frontiers in endocrinology
Ask Prognia AI
Have questions about this meta-analysis?
Prognia AI can search this source alongside 35M+ PubMed papers and current ESC, AHA, NICE, and ADA guidelines to give you a fully cited clinical answer.