Abstract: FR-PO194
Central Blood Pressure, Statins, and LDL-Cholesterol: A Mediation Analysis
Session Information
- Vascular Biology and Dysfunction
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Hypertension
- 1103 Vascular Biology and Dysfunction
Authors
- Lamarche, Florence, Hopital du Sacre-Coeur, Montreal, Quebec, Canada
- Madore, Francois, Hopital du Sacre-Coeur, Montreal, Quebec, Canada
- Agharazii, Mohsen, CHUQ-HDQ, Quebec City, Alberta, Canada
- Goupil, Remi, Hopital du Sacre-Coeur, Montreal, Quebec, Canada
Background
Central blood pressure (CBP) is a better predictor of cardiovascular burden than peripheral blood pressure (BP). While studies have suggested a reduction in peripheral BP with statins, it remains uncertain to what extent statins reduce CBP and whether this reduction is mediated through a decrease in LDL-cholesterol (LDL).
Methods
Of the 20,004 CARTaGENE participants, 17,011 had CBP and LDL measurements (n=3,133 with statins; n=13,439 without). Multivariate regression analyses were used to evaluate the association between CBP, LDL and statin use (after stratification for treatment indication for the latter). The impact of LDL on the association between statin use and CBP was determined by mediation analyses. All analyses were adjusted for age, sex, diabetes, cardiovascular disease, smoking, eGFR, BMI, uric acid, heart rate, anti-hypertensive agents and aspirin.
Results
Lower levels of LDL were associated with lower systolic and diastolic CBP in participants treated with (b=0.098 and 0.125; p<0.001) and without statins (b=0.089 and 0.105; p<0.001). Statin use as primary prevention (per ACC/AHA guidelines; n=8,865) was also associated with lower systolic CBP, diastolic CBP and central pulse pressure (b=-0.091, -0.073 and -0.055; p<0,001). Mediation analyses demonstrated that 15%, 46% and -22% of these effects were achieved through the concomitant changes in LDL (Table 1). In secondary prevention (n=995), statins use was not associated with lower CBP, although the small sample size may lack power.
Conclusion
In this populational cohort, statin use is associated with lower CBP when used as primary prevention. These changes are mediated directly by statin use but also indirectly through the effects on LDL.
Table 1: Mediation analyses
| Path A (total effect) | Path A (direct effect) | Path BC (indirect effect) | Percent mediation | |
| Systolic CBP | -3.0 (-3.8, -2.3) | -2.6 (-3.4, -1.7) | -0.5 (-0.2, -0.0) | 15% |
| Diastolic CBP | -1.7 (-2.2, -1.2) | -1.0 (-1.5, -0.4) | -0.8 (-1.0, -0.5) | 44% |
| Central pulse pressure | -1.3 (-1.8, -0.9) | -1.6 (-2.2, -1.1) | 0.3 (0.0, 0.6) | -22% |
Effect represent change of CBP parameter per 1 standard deviation of LDL (95% CI).
Mediation analysis