Objectives: To quantify the association between risk factors and cardiovascular disease (CVD) in heterozygous familial hypercholesterolemia (FH).
Design: Systematic review, meta-analysis, and meta-regression.
Data sources: MEDLINE, EMBASE, Global Health, the Cochrane Library, and PubMed for full-text articles published in English from 1990 - 2018.
Methods: Studies reporting adjusted associations between: (1) traditional cardiovascular risk factors; (2) personal health practices; or (3) clinical characteristics and CVD with ≥ 100 participants (50 participants with and 50 without CVD) were included. A sole reviewer extracted data on study characteristics, risk factors, and outcomes of interest, with independent verification by a second reviewer. Odds ratios with 95% confidence intervals (CIs) were calculated for selected risk factors with random effects meta-analysis, from which values for population-attributable fraction and case impact number were derived. We also performed heterogeneity and publication bias investigations, five sensitivity analyses (by study size, design, adjustment, country and year of publication), and meta-regression.
Results: We identified 25 studies representing 52,034 participants and 9,751 CVD events. Age [OR: 1.06; 95% CI: 1.05, 1.10], male sex [OR: 1.79; 95% CI: 1.47, 2.18], hypertension [OR: 1.76; 95% CI: 1.41, 2.21], diabetes [OR: 1.60; 95% CI: 1.18, 2.16], body mass index [OR: 1.04; 95% CI: 1.03, 1.05], smoking [OR: 1.69; 95% CI: 1.45, 1.98], lipoprotein a [OR: 1.78; 95% CI: 1.22, 2.60], high-density lipoprotein cholesterol [OR: 1.39; 95% CI: 1.20, 1.61], low-density lipoprotein (p = 0.045) and total cholesterol (p < 0.001) were associated with increased CVD. Smoking, hypertension, and diabetes were associated with 1 in 8, 1 in 10, and 1 in 38 cases of CVD, respectively. However, they accounted for a lower proportion of CVD risk in FH individuals than reported in the general population. Results were broadly consistent in sensitivity analyses.
Conclusions: Several clinical risk factors are significantly and independently associated with CVD risk in patients with FH and should be targeted for risk factor modification. These data can also inform the selection of variables for prediction models to aid in risk stratifying patients.