Association of aortic pulse wave velocity with cardiovascular outcomes and all-cause mortality in diabetes: A systematic review and meta-analysis

Abstract

Aims

Patients with diabetes have experienced excess cardiovascular (CV) diseases. Accurate prediction of CV risk has been one of the ultimate tasks of contemporary diabetology. Aortic stiffness (AS), a key indicator of vascular health, has been increasingly considered as a valuable biomarker for CV risk and mortality prediction. We sought to investigate and calculate the predictive value of AS measured by pulse wave velocity (PWV) for cardiovascular CV events and/or all-cause mortality for patients with diabetes.

Methods

A comprehensive search was carried out in the PubMed, Embase, and Cochrane Library up to 11 June 2024. We included observational studies where investigators reported the association of PWV with CV events or all-cause mortality in patients with diabetes. Newcastle-Ottawa Quality Assessment Scale was used for quality assessment. Information about study design, participant demographics, baseline characteristics, hazard ratio (HR) and so on were extracted from study documents. Heterogeneity among studies was assessed by using the I² statistic. Acceptable levels of heterogeneity were defined as I² < 75 %. Fixed-effects model (Inverse Variance) was utilized for synthesis when I² < 50 %, and a random-effects model (DerSimonian-Laird) was used when I² > 50 %. Visual inspection of funnel plots, Egger's test and Begg's test were used to evaluate the existence of publication bias. Trim-and-fill method was further used to identify and adjust for publication bias.

Results

We included 12 studies. For each 1 m/s increase in cfPWV in diabetes population, the risk of a CV events increased by 7 % [HR 1.07, 95%CI 1.02–1.12, 3132 subjects, random-effects model], and the risk of all-cause mortality increased by 7 % [HR 1.07, 95%CI 1.03–1.12, 1700 subjects, fixed-effects model]. Diabetic patients with higher cfPWV had a 1.71-fold increased risk of CV events compared with those with lower cfPWV [HR 1.71, 95%CI 1.22–2.40, 2075 subjects, random-effects model]. Since different studies may have used different modeling methods, we performed a quantitative synthesis for each modeling method separately, which consequently led to variations in subjects sizes.

Conclusions

Despite certain limitations of this study, such as heterogeneity, our findings indicate that PWV could be a strong predictor of future CV events and all-cause mortality in patients with diabetes.