Objective To investigate the correlation between serum vitamin D level and heart failure based on cross-sectional analysis.

Methods A total of 10, 411 subjects included in the National Health and Nutrition Examination Survey (NHANES) database from 2013 to 2016 were selected as the study population. The serum 25-hydroxyvitamin D 25(OH)D levels of all subjects were measured. According to the quartiles of serum vitamin D levels, the subjects were divided into four groups. Multivariate logistic regression analysis and restricted cubic spline regression analysis were employed to explore the correlation between serum vitamin D level and heart failure.

Results There were statistically significant differences among the four groups in terms of age, gender, educational attainment, race, diabetes, hypertension, alcohol consumption, physical activity, and body mass index (P < 0.05). The incidence rates of heart failure in the four vitamin D quartile groups (from low to high) were 3.9%, 2.7%, 2.9% and 4.1% respectively, with a statistically significant difference (P < 0.05). The results of multivariate logistic regression analysis revealed that, compared with the subjects in the first quartile group of vitamin D, the odds ratios (OR) and 95% confidence intervals (CI) of heart failure in the second, third, and fourth quartile groups of vitamin D were 0.64 (0.46 to 0.89), 0.53 (0.38 to 0.73) and 0.48 (0.35 to 0.66) respectively, and the trend test showed P was less than 0.001. The results of restricted cubic spline regression analysis indicated that serum vitamin D level was non-linearly negatively correlated with heart failure (non-linear test P=0.005). At the initial stage, the risk of heart failure decreased with the increase in vitamin D level; when the vitamin D level reached a certain level, the risk of heart failure tended to stabilize or even increase slightly.

Conclusion Serum vitamin D level is non-linearly negatively correlated with the risk of heart failure.