Last updated on February 2018

Effect of Vitamin D Supplementation in Patients With Heart Failure and Vitamin D Deficiency

Brief description of study

The purpose of this study is to determine the effects of vitamin D supplementation in patients with heart failure and vitamin D deficiency on ventricular function, inflammatory cytokines, brain natriuretic peptide, lipid profile, glucose, serum insulin, serum parathyroid hormone and calcium.

Detailed Study Description

The protocol will be held in three different Hospitals from Monterrey. Those patients with heart failure will be invited to participate; serum vitamin D will be measured, and those with vitamin D deficiency will be participating, once they meet all the inclusion criteria and none of the exclusion criteria, and sign an informed consent.

Patients will be assigned to the intervention group or placebo group according to their characteristics so that both groups are equally balanced in proportion to similar characteristics and number to guarantee homogeneity in both groups at baseline.

The study will have a 12-month time frame for each patient. Patients can be enrolled at different times.

All patients will undergo:

  • A baseline measurement of vitamin D concentration in plasma and then every three months (according to season changes) for a total of five measurements.
  • Myocardial function and structure measured by magnetic resonance with contrast media at baseline and end of intervention.
  • Heart failure clinical follow-up with a 6 minute walk test, arterial pressure and clinical parameters every month.
  • A baseline measurement of serum BNP and then every three months (5 in total).
  • Serum lipid profile, glucose and insulin measurements at baseline, intermediate and at the end of the intervention.
  • Anthropometric measurements (weight, body mass index, waist circumference, body fat percentage) every month.
  • A baseline measurement of inflammatory cytokines in serum and then every three months (5 in total).
  • Serum PTH and calcium measured at baseline and end of intervention

At the end of the intervention, patients in the control group will receive vitamin D supplementation of 5 000 IU per day for a whole year, so they receive the same benefits as the intervention group.


Heart failure defined as patients with ventricular ejection fraction less than 40 percent, serum B natriuretic peptide >300 milligrams/milliliters.

Vitamin D deficiency defined as serum vitamin D levels below 30 nano-grams per milliliter.

According to the New York Heart Association (NYHA) Classification System:

  • NYHA III defined as marked limitation of physical activity
  • NYHA IV defined as symptoms occur even at rest; discomfort with any physical activity.

Statistical analysis plan:

Comparison and p-value measured at baseline and end of intervention will be done with paired t-test for means and McNemar test for proportions.

According to the central limit theorem and the big sample size, there will be no need to evaluate normality for the variables; non-parametric rank test with Wilcoxon sign won't be necessary.

For comparison between independent groups, homogeneity of variance will be evaluated using parametric F test for variables with a normal distribution and with non-parametric Levene test for continuous variables with an abnormal distribution, to compare the means through t-test. Mann-Whitney test won't be necessary for the same reasons listed above.

For comparison of proportions between independent groups will be held with the hypothesis test based on the statistic Z-test, and Fisher test for small samples.

The relation between variables measured in the study with our variable serum vitamin D level, will be done through Pearson's correlation coefficient and a multiple linear regression. That regression will be evaluated through parametric and non-parametric tests with diagnosis of autocorrelation, homoscedasticity and multicollinearity. Regression analysis will be done every time a vitamin D measurement is done (5 in total)

All results will use a significance level of 0.05, equivalent to possibility of type 1 error.

Minitab version 16 software (Minitab Inc, State College, Pennsylvania, USA) will be used for statistical analysis, and Microsoft Excel 2007 (Microsoft Corp., Redmond, Washington, USA) will be used to incorporate the input of date.

Clinical Study Identifier: NCT02087683

Find a site near you

Start Over