Non-invasive peripheral vascular function, incident cardiovascular disease, and mortality in the general population

Abstract Aims  Evidence suggests that peripheral vascular function is related to cardiovascular disease (CVD) and mortality. We evaluated the associations of non-invasive measures of flow-mediated dilatation and peripheral arterial tonometry with incident CVD and mortality. Methods and results  In a post-hoc analysis of the community-based Gutenberg Health Study, median age 55 years (25th/75th percentile 46/65) and 49.5% women, we measured brachial artery flow-mediated dilatation (N=12 599) and fingertip peripheral arterial tonometry (N=11 125). After a follow-up of up to 11.7 years, we observed 595 incident CVD events, 106 cardiac deaths, and 860 deaths in total. Survival curves showed decreased event-free survival with higher mean brachial artery diameter and baseline pulse amplitude and better survival with higher mean flow-mediated dilatation and peripheral arterial tonometry ratio (all Plog rank <0.05). In multivariable-adjusted Cox regression analyses only baseline pulse amplitude was inversely related to mortality [hazard ratio (HR) per standard deviation increase, 0.86, 95% confidence interval (95% CI), 0.79–0.94; P=0.0009]. After exclusion of individuals with prevalent CVD the association was no longer statistically significant in multivariable-adjusted models (HR 0.91, 95% CI 0.81–1.02; P=0.11). None of the vascular variables substantially increased the C-index of a model comprising clinical risk factors. Conclusions  In our cohort, non-invasive measures of peripheral vascular structure and function did not reveal clinically relevant associations with incident CVD or mortality. Whether determination of pulse amplitude by peripheral arterial tonometry improves clinical decision-making in primary prevention needs to be demonstrated.


Supplementary Table 3. C-indices for the clinical model and vascular function measures in intermediate risk individuals based on SCORE Deutschland (N=3,239)
Bootstrap analysis with 1000-fold re-sampling was used to correct for over-optimism.
The basic model comprised age, sex, current smoking, body mass index, systolic blood pressure, heart rate, hypertension treatment, diabetes, LDL/HDL cholesterol, lipid treatment, and prevalent cardiovascular disease.  Figure 1. Survival curves stratified by median vascular function measures for baseline brachial artery diameter, FMD, ln baseline pulse amplitude, and PAT ratio for cardiac mortality (N=106). P-values are for the log rank test. Figure 2. Vascular function measures in relation to incident cardiovascular disease in multivariable-adjusted Cox regression analyses in the subgroup of individuals with a low ten-year risk of fatal cardiovascular disease (318 events). Provided are hazard ratios per standard deviation increase in vascular function measure and 95% confidence intervals. Hazard ratios (HR) are per standard deviation (SD).

Supplementary
Age-and sex-adjusted and cardiovascular risk factors (CVRF)-adjusted models are presented. The latter include age, sex, current smoking, body mass index, systolic blood pressure, heart rate, hypertension treatment, diabetes, LDL/HDL cholesterol, and lipid treatment. FMD stands for flow-mediated dilatation, PAT for peripheral arterial tonometry. Figure 4. Vascular function measures in relation to cardiac mortality in multivariable-adjusted Cox regression analyses in the subgroup of individuals with a low tenyear risk of fatal cardiovascular disease (45 deaths). Provided are hazard ratios (HR) per standard deviation (SD) increase in vascular function measure and 95% confidence intervals.

Supplementary
Age-and sex-adjusted and cardiovascular risk factors (CVRF)-adjusted models are presented. The latter include age, sex, current smoking, body mass index, systolic blood pressure, heart rate, hypertension treatment, diabetes, LDL/HDL cholesterol, lipid treatment, and prevalent cardiovascular disease. FMD stands for flow-mediated dilatation, PAT for peripheral arterial tonometry. Figure 5. Vascular function measures in relation to incident cardiovascular disease in multivariable-adjusted Cox regression analyses in the subgroup of individuals with an intermediate ten-year risk of fatal cardiovascular disease (268 events). Provided are hazard ratios (HR) per standard deviation (SD) increase in vascular function measure and 95% confidence intervals.

Supplementary
Age-and sex-adjusted and cardiovascular risk factors (CVRF)-adjusted models are presented. The latter include age, sex, current smoking, body mass index, systolic blood pressure, heart rate, hypertension treatment, diabetes, LDL/HDL cholesterol, lipid treatment, and prevalent cardiovascular disease. FMD stands for flow-mediated dilatation, PAT for peripheral arterial tonometry. Figure 6. Vascular function measures in relation to mortality in multivariable-adjusted Cox regression analyses in the subgroup of individuals with an intermediate ten-year risk of fatal cardiovascular disease (518 events). Provided are hazard ratios (HR) per standard deviation (SD) increase in vascular function measure and 95% confidence intervals.

Supplementary
Age-and sex-adjusted and cardiovascular risk factors (CVRF)-adjusted models are presented. The latter include age, sex, current smoking, body mass index, systolic blood pressure, heart rate, hypertension treatment, diabetes, LDL/HDL cholesterol, lipid treatment, and prevalent cardiovascular disease. FMD stands for flow-mediated dilatation, PAT for peripheral arterial tonometry. Figure 7. Vascular function measures in relation to cardiac mortality in multivariable-adjusted Cox regression analyses in the subgroup of individuals with an intermediate ten-year risk of fatal cardiovascular disease (45 deaths). Provided are hazard ratios (HR) per standard deviation (SD) increase in vascular function measure and 95% confidence intervals.

Supplementary
Age-and sex-adjusted and cardiovascular risk factors (CVRF)-adjusted models are presented. The latter include age, sex, current smoking, body mass index, systolic blood pressure, heart rate, hypertension treatment, diabetes, LDL/HDL cholesterol, lipid treatment, and prevalent cardiovascular disease. FMD stands for flow-mediated dilatation, PAT for peripheral arterial tonometry.