For hypertriglyceridemia, the adjusted odds ratio was 3.10, with a 95 percent confidence interval of 1.18 to 8.16, and for low HDL it was 3.26, with a tighter interval of 2.28 to 4.65. An odds ratio above one indicates higher odds among vapers, but the wide interval for triglycerides reflects substantial statistical uncertainty, and the heterogeneity statistic for that outcome was high at 80.1 percent, meaning the individual studies varied considerably. By contrast, the low-HDL finding was highly consistent across studies, with heterogeneity of just 3.1 percent.

When the researchers examined unadjusted mean lipid levels rather than dyslipidemia diagnoses, the picture was more mixed. The primary quality-effects model showed no statistically significant difference in total cholesterol, with an unadjusted weighted mean difference of +3.66 mg/dL and a confidence interval crossing zero. However, vapers showed higher low-density lipoprotein cholesterol at +12.40 mg/dL, and higher very-low-density lipoprotein at +3.23 mg/dL.

The publication bias statistic for the total cholesterol comparison, an LFK index of 9.44, indicated extreme asymmetry, and the authors assessed publication bias using Doi plots and the LFK index, with the LFK index as the primary quantitative criterion. Crucially, the certainty of the evidence was rated very low for every lipid outcome using the GRADE framework. All seven included studies were observational, meaning participants were not randomly assigned to vape or abstain, so confounding remains a serious concern.

People who vape differ from non-users in many ways, and although some studies adjusted for measured confounders, residual and unmeasured confounding cannot be excluded. The authors also note that planned subgroup analyses by age and sex could not be performed because the included studies did not report stratified results. Measurement problems compound these limitations.

The authors call for well-designed longitudinal studies that use validated smoking-status assessment, detailed vaping exposure measures, fasting lipid panels, repeated lipid measurements and adjustment for key cardiometabolic confounders before firmer conclusions can be drawn. The search covered PubMed, Scopus, EMBASE, CENTRAL, CINAHL, medRxiv and Google Scholar up to May 31, 2025, and identified 4,160 data sources, of which seven met the inclusion criteria. Eligible studies involved adult populations and assessed the association between exclusive e-cigarette use and serum lipids, specifically total cholesterol, HDL, LDL and triglycerides.

Study quality was assessed using the Methodological Standards for Epidemiological Research scale, and pooling used a bias-adjusted inverse-variance heterogeneity method with heterogeneity evaluated by the I2 statistic. The authors stop short of causal claims, and the very low certainty of evidence means these associations could shift or shrink as better studies accumulate. For now, the finding that exclusive e-cigarette users show higher odds of low HDL and elevated triglycerides, alongside higher unadjusted LDL and VLDL levels, adds a lipid dimension to the growing list of cardiovascular signals surrounding vaping.

It also underscores how much rigorous long-term evidence is still missing before the metabolic consequences of vaping can be characterized with confidence.

Subject of Research: Association between exclusive e-cigarette use and dyslipidemia in a systematic review and meta-analysis of observational studies

Article Title: The association between e-cigarette use and dyslipidemia: A systematic review and meta-analysis of observational studies

Article References: Rajha, H. E., Mohsen, S., Alkuwari, H., Naeim, J., Al-Marri, W., Hamdan, A., Chivese, T., & Al Saady, R. M. (2026). The association between e-cigarette use and dyslipidemia: A systematic review and meta-analysis of observational studies. PLOS One, 21(10), e0361200. https://doi.org/10.1371/journal.pone.0361200

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0361200

Keywords: e-cigarettes, vaping, dyslipidemia, triglycerides, HDL cholesterol, LDL cholesterol, meta-analysis, observational studies, cardiovascular risk, GRADE, lipid profile, association