A Cross-Sectional Study of the Association between Circulating 25-Hydroxyvitamin D Levels and Predicted Operative Mortality of Patients with Cardiovascular Disease.
J Nutr Sci Vitaminol (Tokyo). 2012;58(5):327-32.
Tsutsumi Y, Sanui M, Shimojima A, Ishioka H, Urashima M.
Division of Molecular Epidemiology, Jikei University School of Medicine.
Recent studies have suggested that low levels of 25-hydroxyvitamin D (25OHD) are associated with cardiovascular risks in medical patients. However, these associations have not been well documented in high risk surgical patients. We hypothesized that serum 25OHD, 1,25-dihydroxyvitamin D (1,25OHD) would be associated with the cardiac operative risk stratification score. The study was conducted with a cross-sectional design at a single academic medical center in Japan. Two hundred five adult patients scheduled for major cardiovascular surgery were included consecutively.
Cardiac operative risk was evaluated with the European System for Cardiac Operative Risk Evaluation (EuroSCORE) scoring system. Correlations between 25OHD and 1,25OHD, and EuroSCORE were assessed using simple and multiple linear regression models. Mean 25OHD and 1,25OHD were 20.1±7.1 ng/mL and 51.2±19.2 pg/mL, respectively.
Half and 88% of the study population showed deficient (<20 ng/mL) and insufficient (<30 ng/mL) 25OHD levels, respectively. In contrast, only 3% showed 1,25OHD levels lower than normal (<20 pg/mL).
Circulating 25OHD levels, but not 1,25OHD levels, were negatively correlated with EuroSCORE (p=0.005) even after adjusted for
- body mass index,
- diabetes mellitus,
- use of statin,
- high sensitive C-reactive protein, and
- intact parathyroid hormone.
These results suggest that serum 25OHD levels are inversely associated with operative risk severity of patients undergoing major cardiovascular surgery.
In this study, we found that circulating 25OHD levels were deficient in half of patients who underwent cardiovascular surgery, and higher levels of 25OHD were significantly associated with a lower EuroSCORE even after multivariate adjustment for possible confounders. In contrast, circulating 1,25OHD levels were not reduced in most patients undergoing surgery and were not associated with EuroSCORE even after multi-variate adjustment for possible confounders (Tables 1 and 2). Lee et al. first noted that lower levels of 25OHD were correlated with predicted hospital mortality evaluated by the Simplified Acute Physiology Score II in 42 patients admitted to the intensive care unit and referred to the Department of Endocrinology (8), which is consistent with our results. Levels of hsCRP and iCa were reported to be associated with severity and prognosis of CVD (18), and Anderson et al. identified the possibility of a relationship between elevated iPTH levels, and incidence of cardiovascular risk factors and mortality (19). Therefore, we used them as possible confounders for EuroSCORE in our multivariate analyses. As a result, however, we did not include iCa as a confounder in multiple linear regression analyses, because EuroSCORE was not significantly correlated with iCa in simple linear regression analysis. Additionally, we included statin therapy as a covariate, because statin therapy has been shown to increase circulating vitamin D levels (20) and may be a confounder behind the negative association between 25OHD as well as 1,25OHD levels, and EuroSCORE (21). In contrast, we did not use underlying diseases (i.e., CAD, VHD and AD) as confounders because EuroSCORE already accounts for these operation-related factors. In our study, 25OHD levels remained a significant risk factor of EuroSCORE in all patients even after adjustment for cardiovascular risk factors and other standard risk factors. Furthermore, in stratified analyses of underlying diseases, a significant association between EuroSCORE and 25OHD still remained in the VHD group, not in the CAD and AD groups. One of the most feasible reasons for these results may be that the number of patients in each group is too small for the differences to be significant in stratified analyses compared with the number of independent variables.
It is unclear why 25OHD levels were so low in our patients. Patients with severe CVD may not be able to go out because of the restricted body activity and may have less sun exposure, which could explain why patients with severe CVD had lower 25OHD levels (22). On the other hand, lower 25OHD levels may be a risk factor for CVD. It is also unclear why predicted operative mortality was associated with 25OHD levels but not 1,25OHD levels. We assumed that
- 1) like macrophages (23), cardiac cells take up circulating 25OHD rather than 1,25OHD, because circulating levels of 25OHD are 1,000 times higher than that of 1,25OHD (1 );
- 2) cardiac cells activate 25OHD into 1,25OHD by 1oOHase in the cells (1);
- 3) the 1,25OHD binds to VDR in the cell;
- 4) the complex of 1,25OHD and VDR moves into the nucleus and binds to vitamin D response elements around promoter regions (24);
- 5) this reaction regulates a variety of gene transcriptions (25, 26);
- 6) as a result, some gene transcriptions protect against progression of atherosclerosis, inflammation, and fibrotic scarring (5).
In contrast to 25OHD, most of the circulating 1,25OHD levels were within normal range and 1,25OHD levels were not associated with predicted operative mortality in this study. These results indicates that circulating 1,25OHD may not be a predictor for operative mortality as the assay for 1,25OHD is less reliable than that for 25OHD because of a short half-life of 1,25OHD (27) and it may be tissue rather than circulating 1,25OHD sufficiency which is more important in cardiac operative risks. Additionally, as circulating 1,25OHD levels are tightly regulated by PTH, Ca and phosphate levels, circulating 1,25OHD levels stay within the normal range until severe deficiency of 1,25OHD levels.
There are several limitations to this study. First, the study design was cross-sectional; thus, no cause and effect relationship can be established. Second, we did not collect data on sun exposure or diet except for vitamin D supplements, so we could not determine the association of these factors with 25OHD levels and the severity of patients' conditions. Third, preoperative heart diseases were heterogeneous among our patients. On the other hand, the study population (n=205) was large enough to compute multivariate analyses for all patients using possible and known CVD risk factors, including BMI (28, 29), albumin (30), hypertension (31), dyslipidemia (28), diabetes mellitus (28), renal dysfunction (32), use of statin (20), hsCRP (33), iCa (18) and iPTH (19). This is a strength of this study.
PDF is attached at the bottom of this page
- Vitamin D levels were tested before the surgery - they can drop a LOT after surgery/trama
- This study compares vitamin D to PREDICTION of death following cardiac surgery - not actual death
- 14X more likely to die after first cardiovascular event if vitamin D deficient – Nov 2012
- Cardiovascular disease 50 percent more likely if low vitamin D – meta-analysis Nov 2012
- Increased heart attacks in the winter may be due to lack of sunshine – Nov 2012
- Off topic: CVD: the primary cause of death in senior women, is increasing – Nov 2012 Note - senior women often have lowever vitamin D levels than senior men
- Low levels of vitamin D associated with all cause mortality – Oct 2012
- Half as many heart deaths for those with high levels of vitamin D – meta-analysis Sept 2012
- Shift workers 23 percent more likely to have cardiovascular events – meta-analysis July 2012 shift workers tend to have low levels of vitamin D
- Coronary Artery Disease predicted by low levels of vitamin D – April 2012
- 2X more complications after heart surgery associated with high level of vitamin D – Jan 2013
- Differences in black and non-black mortality and vitamin D – Oct 2012