Cardiovascular diseases (CVD) are the most common cause of mortality and morbidity worldwide and lead to a high socio-economic burden on society. Despite the high mortality in CVDs, early diagnosis, careful prevention and treatment can reduce the global burden of CVDs significantly.1

Among the different etiologies of CVDs, heart failure (HF) is associated with very high mortality and morbidity, such as sudden cardiac death. HF is the most frequent cause of hospital admission in patients over 65 years. Considering universal increase in life expectancy, the number of patients diagnosed with HF is growing globally.2

In recent literature, different treatments are suggested for HF and there is agreement on the use of diuretics in many of them. At least small dosage of diuretics is prescribed for several patients with HF. Modern treatments for HF mostly target left ventricular systolic dysfunction and for this goal angiotensin converting enzyme inhibitors and beta-blockers are commonly used.3

Despite major improvements in the treatment of HF over the last three decades, hospitalization, mortality and morbidity rate still remain high. The effectiveness of vitamin D on different prognostic markers of HF has been investigated in studies with different designs. Controversial efficacy is reported for vitamin D in recent articles including primary research, review articles and even meta-analyses.4–7

In different articles, the importance of left ventricular ejection fraction (LVEF) is assessed in patients with HF. Heart failure is a pathological clinical condition defined by impaired pump function of the heart muscle, which is categorized by LVEF. Important prior researches, such as the CHARM trial,8 reported that lower LVEF is associated with higher cardiovascular death, mortality and rehospitalization rate9,10 in patients with HF.

There are several primary studies and a limited number of systematic reviews published regarding the efficacy of vitamin D supplementation in patients with HF and the results are controversial. There is no consensus among these papers and many of them have reported clearly opposing results. We therefore believe that an updated systematic review is required in this field.6,7,11

Considering the helpfulness of a cheap, accessible, and safe treatment such as vitamin D in clinical practice, that clinical trials are the best types of studies for decision-making and the major importance of LVEF index in estimating myocardial function in patients with HF; we focused specifically on the effect of vitamin D supplements on ejection fraction (EF) in a systematic review limited to clinical trials to achieve the greatest accuracy.

This study was a comprehensive and focused systematic review and meta-analysis for assessing the efficacy of vitamin D supplements on LVEF in patients with HF.

Systematic reviews are carried out to sum up results of a large number of published papers for final decision making and their difference from narrative and review articles in that systematic reviews have a narrower title. They seek to solve a real clinical challenge27 and use systematic methods to collect relevant data.

As clinicians, we are not yet certain whether patients with HF benefit from vitamin D supplementations. At the moment, HF is mainly defined as a pathological clinical condition with impaired myocardial muscle function and inability of the heart to pump adequate blood levels to meet the metabolic demands of peripheral tissues.28 Reviewing the latest up-to-date literature, several studies and systematic reviews can be found on different HF markers, such as inflammatory markers,29 molecular pathways and signalling.30 Few articles focus on myocardial function indices, of which EF is an important and frequently used one, and few studies are performed in a clinical trial design.

Based on our knowledge, EF is rarely reported in studies and is less pronounced in systematic reviews as the literature trend is mostly focused on molecular and inflammatory indices. Few systematic reviews were performed that reported the effect of vitamin D on EF in patients with HF. Some reported a positive effect7 while others recorded no significant effect.6,11

The number of studies were really limited and even the published systematic reviews are not in agreement on included studies. This seems to be mainly because of different search strategies and databases searches along with lack of focus on a single outcome of interest.

In this study we observed that vitamin D supplements can be significantly effective in enhancing the most important and the most common myocardial function index, EF (p=0.006) with a raw mean difference of 3.304 (95% CI 0.954, 5.654).

The clinical importance of this 3.3% enhancement of EF in patients with HF gains further value when we look back in the literature for the significance of LVEF in patients with HF. In different validated studies and clinical trials, researchers concluded that enhancement in LVEF is directly associated with lower rehospitalization and cardiovascular mortality.8–10

In this systematic review and meta-analysis, we were unable to differentiate the effect of vitamin D in different categories of HF based on EF (such as HF with reduced EF, HF with mid-range EF, HF with preserved EF) since not all studies separated them accurately or reported the results separately.

While performing the statistical analysis, we aimed to at least differentiate HF categories in those studies that reported it, but our results did not indicate any association between category of HF based on EF and efficacy of vitamin D treatment.

Other analyses were also performed in order to find any possible association between dosage of vitamin D or duration of treatment with vitamin D but we could not find any significant association in such analyses.

Across the included studies, mean EF was <50% and given the low rate of adverse effects from vitamin D at the suggested doses or even at higher doses31 and the significance of 3.3% of LVEF in a patients with EF <50%, we believe that vitamin D treatment should be always considered as a treatment for patients with HF.

The most recent systematic review was performed by Wang et al.6 and was published in May 2019. The researchers reported that vitamin D supplementation does not affect LVEF significantly 2.56, (95% CI -2.18, 7.3) which was confirmed in another systematic review performed by Jiang et al.11 in 2016. In this study, the researchers reported that with a mean difference of 4.11 (95% CI -0.91, 9.12), the effect of vitamin D supplements on EF was not significant.

Wang et al.6 published their article on 2019 with a search limited up to October 2017; this is while in another valuable and reliable study conducted by (published in 2018 in BMJ journal and received by journal on November 2017), the researchers claimed a more comprehensive database search which unexpectedly led to a lower number of primary papers. This means that emphasizing on other outcomes rather than EF (as primary outcome) may result in missing papers in which EF is reported as a secondary outcome. In our study, EF was the only outcome of interest, thus we included more studies after a more comprehensive and sensitive search without any limitation.

In Zhao et al.’s study, a significantly positive effect of vitamin D supplements on EF 4.18 (0.36, 7.99) was reported, whereas Wang et al.’s study revealed the opposite. We believe that the controversial results can be due to different primary outcomes in those studies. We are of the opinion that a more focused meta-analysis on EF as a primary outcome, such as our research, can yield better search results.

In this systematic review, we performed a very comprehensive and wide search and all studies were carefully screened by two authors. This accuracy along with help of previous systematic reviews (as we obtained some of our data from them) lead to a total of 11 papers and 12 studies in quantitative data synthesis.

Based on our knowledge, our study is the only systematic review which includes Zittermann et al.’s results after one year and three years23 of follow up. Their study included the largest number of subjects (311 and 242 at one year and three years, respectively) and the results indicate a lack of any total significant effect of vitamin D on EF which contradicts our pooled estimate results.

Our study, like many others, has its own limitations of which we are aware. This study contains a very high inter-study heterogeneity, which can be explained in part by consciously including heterogeneous studies, from patients with a mean age of <1 year to patients <70, a wider range of doses of vitamin D and duration which varied from a few months to three years. Other less significant differences such as gender, genetics and race were also ignored and may also have influenced our results.

As a systematic review and meta-analysis, results of this study were pooled out based on reported outcomes of studies, instead of individual patient data and this means that validated primary researcher and clinical trials are always valuable and required to confirm a scientific hypothesis in clinic.

Another limitation in this paper is that some primary studies included patients with vitamin D deficiency and some enrolled patients with normal levels of vitamin D. On the other hand, many of the primary clinical trials did not even differentiate this important factor and included both patients with and without vitamin D deficiency, which can subsequently reduce the reliability of our results.

Primary trials are not all clear even how EF should be assessed whether it should be performed via echocardiography or cardiac magnetic resonance imaging. Some of them even used a combination and actually measured patients’ EF using different modalities at different time points for the same patient or at the same time point for different patients.

Different categories of HF based on EF such as HF with preserved EF were not discussed and reported. Patient functional class and clinical status were not reported in detail in all studies. In primary trials, the effect of vitamin D was not measured in different patients’ categories according to EF subgroups.

We performed a series of various subgroup analyses and used several meta regression models but unfortunately none of them were problem solving and the heterogeneity remained too high still. We even tried removing the very different studies from all different analyses, such as the Shedeed et al. study,32 but this did not help.

However, returning to the previously mentioned meta analyses, it is notable that they also had a very high heterogeneity and considering all these together, it seems that heterogeneity is somehow inevitable due to the nature of the included studies. The only solution seems to be conducting more high quality clinical trials in this field to enable better meta-analyses in future.

Another reason for performing more primary clinical trials is that among those performed, not all of them were high quality and accurate. These 11 studies which seems to be all available studies, globally does not have an acceptable quality such as clinical trials performed in other fields. We believe that rather than retrospective and descriptive studies, we need additional clinical trials, with enough power to better clarify the role of vitamin D in the treatment of HF.

Based on our knowledge this meta-analysis is currently the most up-to-date one with the most comprehensive data about EF. The results indicate that prescribing vitamin D supplements for patients with HF, especially those with EF <50% can enhance myocardial EF.

We conclude that before publication of more primary clinical trials with higher quality and further meta-analysis, it is logical to bear vitamin D supplementation in mind for patients with HF, especially those with vitamin D deficiency.

This study is self-funded and authors did not receive any financial support from any organization for performing this research.

The authors have no conflicts of interest to declare.