Abstract
Background:
A growing number of cohort studies revealed an inverse association between cheese intake and cardiovascular diseases, yet the causal relationship is unclear.
Objective:
To assess the causal relationship between cheese intake, and cardiovascular diseases and cardiovascular biomarkers.
Methods:
A two-sample Mendelian randomization (MR) analysis based on publicly available genome-wide association studies was employed to infer the causal relationship. The effect estimates were calculated using the random-effects inverse-variance-weighted method.
Results:
Cheese intake per standard deviation increase causally reduced the risks of type 2 diabetes (odds ratio (OR) = 0.46; 95% confidence interval (CI), 0.34-0.63; p = 1.02 × 10-6), heart failure (OR = 0.62; 95% CI, 0.49-0.79; p = 0.0001), coronary heart disease (OR = 0.65; 95% CI, 0.53-0.79; p = 2.01 × 10-5), hypertension (OR = 0.67; 95% CI, 0.53-0.84; p = 0.001), and ischemic stroke (OR = 0.76; 95% CI, 0.63-0.91; p = 0.003). Suggestive evidence of an inverse association between cheese intake and peripheral artery disease was also observed. No associations were observed for atrial fibrillation, cardiac death, pulmonary embolism, or transient ischemic attack. The better prognosis associated with cheese intake may be explained by lower body mass index (BMI; effect estimate = -0.58; 95% CI, from -0.88 to -0.27; p = 0.0002), waist circumference (effect estimate = -0.49; 95% CI, from -0.76 to -0.23; p = 0.0003), triglycerides (effect estimate = -0.33; 95% CI, from -0.50 to -0.17; p = 4.91 × 10-5), and fasting glucose (effect estimate = -0.20; 95% CI, from -0.33 to -0.07; p = 0.0003). There was suggestive evidence of a positive association between cheese intake and high-density lipoprotein. No influences were observed for blood pressure or inflammation biomarkers.
Conclusions:
This two-sample MR analysis found causally inverse associations between cheese intake and type 2 diabetes, heart failure, coronary heart disease, hypertension, and ischemic stroke.
Full text
