Thomas GN, Hartaigh BO, Bosch JA, et al. Vitamin D levels predict all-cause and cardiovascular disease mortality in subjects with metabolic syndrome: The Ludwigshafen Risk and Cardiovascular health (LURIC) study. Diabetes Care. 2012;35:1158-1164.
Prospective cohort study of 1,801 white people in southwest Germany with metabolic syndrome who underwent elective coronary angiography between 1997 and 2000. Mortality was tracked for a median of 7.7 years (12,514 person-years of follow-up).
Levels of 25-OH-vitamin D were obtained the morning of the angiographic procedure and measured by radioimmunoassay (DiaSorin). Classification of death was performed using reports on death certificates. Participants were categorized based on the following definitions: optimal (≥ 75 nmol/L , ≥30 ng/ml ), insufficient (50–74.99 nmol/L, 20–29.99 ng/ml), moderate deficiency (25–49.99 nmol/L, 10–19.99 ng/ml), or severe deficiency (<25 nmol/L, <10 ng/ml). Metabolic syndrome was defined per the 2009 international consensus statement.
Of all 1,801 participants, only 8% had levels greater than 30 ng/ml. Equally surprising, 42.8% had levels between 10 and 20 ng/ml, and 22.2% had levels less than 10 ng/ml. In this white, German (medically insured) population, only 2% reported taking any vitamin supplements at all.
The results demonstrated a serum level–dependent reduction in all-cause mortality (P<0.001 for trend) with a 75% reduction in the hazard ratio in the participants with optimal vitamin D levels relative to those with severe deficiency after full adjustment (HR: 0.25; 95% CI: 0.13–0.46). The results also demonstrated a serum level–dependent reduction in cardiovascular mortality [P<0.001 for trend; HR: 0.33 (0.16–0.66)]. Risk reduction included that by sudden death (P<0.004 for trend, HR: 0.15; 95% CI: 0.04–0.63) and congestive heart failure [HR: 0.24 (0.06–1.04)].
The results did not change markedly after removing participants with type II diabetes or glomerular filtration rates <60 ml/min/1.73 m2. The results did not change with seasonality (summer/winter) nor with self-reported activity levels. The results did not change significantly when eliminating from the analysis any participant who died within 2 years of follow-up.
This study’s robust findings confirm the strong association of low vitamin D status with increased risk of both all-cause and cardiovascular mortality reported in multiple previous studies from other large databases, including those of the Centers for Disease Control and Prevention CDC (National Health and Nutrition Examination Survey) and of Intermountain Health (Utah).
The strengths of this study include the large numbers, the biological plausibility of low vitamin D for adverse health consequences, and the strength of the association even when considering significant confounders. The population tested (white, insured, undergoing elective angiography) is unique for existing large databases. One would expect this population to have much higher vitamin D levels.
More than 65% of the participants had vitamin D levels below the 2010 Institute of Medicine (IOM) low standard “normal” of 20 ng/ml. Please note that in 2011 the Endocrine Society recommended achievement of a serum level of 40–60 ng/ml.1 Only 8% of the 1,801 participants achieved a level greater than 30 ng/ml. This means that a rather miniscule number of participants achieved the recommended level of 40 ng/ml.
A key point for all future studies: Dosing does not matter. What counts is the serum level at the start and at the end of the intervention.
Perhaps most clinically important fact for readers is that vitamin D is a low-cost, low-toxicity intervention that complements other medical interventions. Even in the absence of randomized, controlled trials demonstrating causality in all-cause and cardiac mortality, per recent authoritative recommendations by the IOM and the US Preventive Services Task Force, clinicians should advocate for their patients’ good bone health and reduced risk of falls by ensuring normal vitamin D levels. This study strongly suggests that additional health benefits are likely to follow.
This study draws conclusions based on just one vitamin D level obtained at the start in a population at increased risk for all-cause and cardiovascular mortality. Readers cannot draw a causal connection. As such, this study affirms the urgent need for prospective, randomized, controlled trials of vitamin D for reduction of all-cause and cardiovascular mortality.
A key point for all future studies: Dosing does not matter. What counts is the serum level at the start and at the end of the intervention. Any study that fails to assess and report these values has limited clinical significance for meeting the needs for an evidence-based medicine recommendation and therefore a population-wide vitamin D recommendation.