|Home | About | Journals | Submit | Contact Us | Français|
Recent studies have demonstrated greater risks of cardiovascular events and mortality among persons who have lower 25-hydroxyvitamin D (25-OHD) and higher parathyroid hormone (PTH) levels. We sought to evaluate the association between markers of mineral metabolism and sudden cardiac death (SCD) among the 2,312 participants from the Cardiovascular Health Study who were free of clinical cardiovascular disease at baseline. We estimated associations of baseline 25-OHD and PTH concentrations individually and in combination with SCD using Cox proportional hazards models after adjustment for demographics, cardiovascular risk factors, and kidney function. During a median follow-up of 14 years, there were 73 adjudicated SCD events. The annual incidence of SCD was greater among subjects who had lower 25-OHD concentrations: 2 events per 10,000 for 25-OHD ≥ 20 ng/ml and 4 events per 10,000 for 25-OHD < 20 ng/ml. Similarly, SCD incidence was greater among subjects who had higher PTH concentrations: 2 events per 10,000 for PTH ≤ 65 pg/ml and 4 events per 10,000 for PTH > 65 pg/ml. Multivariate adjustment attenuated associations of 25-OHD and PTH with SCD. Finally, 267 participants (11.7% of the cohort) had high PTH and low 25-OHD concentrations. This combination was associated with a more than 2-fold risk of SCD after adjustment (hazard ratio 2.19, 95% confidence interval 1.17, 4.10, p=0.017) compared to participants with normal levels of PTH and 25-OHD. The combination of lower 25-OHD and higher PTH concentrations appears to be associated independently with SCD risk among older adults without cardiovascular disease.
Disturbances in calcium-phosphorus metabolism, 25-hydroxyvitamin D (25-OHD) and parathyroid hormone (PTH) have been implicated as novel risk factors for cardiovascular events. Community-based, prospective cohort studies suggest that 25-OHD deficiency is associated with increased cardiovascular risk after adjustment for established atherosclerosis risk factors.1–3 Higher plasma PTH levels are also associated with a greater risk for cardiovascular specific death.4 Multiple mechanisms have been proposed to explain these observations. Vitamin D deficiency and hyperparathyroidism each stimulate adverse metabolic changes that could increase the risk of adverse cardiovascular events. In experimental models, disruption of the vitamin D receptor stimulates the renin-angiotensin system5, 6 provoking vasoconstriction,7–10 oxidative stress, and left ventricular hypertrophy.6, 11–14 PTH excess increases intracellular calcium and increases protein synthesis and total cell mass within cardiac myocytes. These mechanisms also contribute to the pathogenesis of sudden cardiac death (SCD) but associations of vitamin D deficiency and PTH with SCD have not been reported.
Vitamin D deficiency and hyperparathyroidism also represent interrelated serologic markers of a unified endocrine pathway. Serum 25-hydroxyvitamin D concentrations indicate substrate vitamin D stores. PTH excess represents, in part, inadequate vitamin D activity, because the parathyroid gland is an established target of 1,25-dihydroxyvitamin D. Thus, lower serum 25-OHD and higher serum PTH concentrations in combination may identify individuals who are at the highest risk of SCD.
We evaluated serum concentrations of 25-OHD and PTH as novel risk markers of adjudicated SCD events in a community based cohort that was free of clinical cardiovascular disease at baseline. We hypothesized that abnormalities in each measure of mineral metabolism would be associated with SCD. We also assessed the vitamin D-PTH axis more thoroughly by evaluating these markers in combination with the hypothesis that together they would identify a subgroup at particularly high SCD risk. In addition, as part of a sensitivity analysis that attempts to evaluate whether markers of mineral metabolism may be unique risk factors for SCD compared to non-sudden cardiac death, we assessed the risk of total and cardiovascular mortality among participants with abnormalities in these metabolic markers.
The Cardiovascular Health Study (CHS)15 is a community-based study of cardiovascular disease risk in ambulatory older adults, sponsored by the National Heart Lung and Blood Institute. This analysis evaluates associations of vitamin D and PTH measurements with the subsequent risk of SCD.
The CHS recruited participants from Medicare eligibility lists in Forsyth County, North Carolina; Sacramento County, California; Washington County, Maryland; and Allegheny County, Pennsylvania. Recruiting participants from these 4 US communities allowed for a mixture of urban and rural populations and adequate numbers of both men and women. To be eligible, persons had to be at least 65 years of age, not institutionalized, expected to remain in the current community for 3 years or longer, not under active treatment for cancer, and able to provide written informed consent. The initial 5201 participants were enrolled from January 1989 to June 1990; an additional 687 black participants (with race self-reported) were recruited in 1992 to 1993 and enrolled by June 1993.
We evaluated CHS participants at the time of their 1992–1993 examinations. This analysis excluded participants who had prevalent cardiovascular disease defined by a prior history of coronary heart disease, heart failure, stroke, transient ischemic attack, claudication, atrial fibrillation, or the presence of a pacemaker or implantable cardioverter-defibrillator (ICD) at the time of the 1992–1993 exam. Although this analysis excluded those with clinical cardiovascular disease, this cohort included elderly individuals, of whom the majority had hypertension, and a small proportion were smokers or had diabetes. CHS determined prevalent cardiovascular conditions by review of medical records, electrocardiographic findings, participant responses to questionnaires, and interim events that occurred between the 1988–1989 and the 1992–1993 CHS examinations.16 We further excluded 948 participants who had inadequate serum volume (<500 ul) to perform the mineral metabolism measurements. Finally, we excluded participants who were taking lithium, which may alter mineral metabolism, had a history of primary hyperparathyroidism, or had missing values for 25-OHD or PTH. The final study sample consisted of 2,312 participants (figure 1). All participants provided written informed consent, and the institutional review boards at all participating sites approved the study protocol.
Participants underwent a comprehensive examination at baseline, which included a thorough medical history, physical examination, laboratory testing, a 12-lead ECG, and assessment of cardiovascular disease status. Additionally, functional assessment was performed and included an evaluation of gait speed and grip strength. Gait speed was defined as the time to walk 4.6 meters at an individual’s usual pace beginning from a standstill post. Grip strength was assessed as the average force (measured in kilograms) generated by the dominant hand using a Jamar dynamometer.15 The study design, quality-control procedures, laboratory methods, and procedures for blood pressure measurement have been published previously.15, 17
Vitamin D and PTH were measured on fasting sera specimens that had been stored at −70 degree Celsius since the 1992–1993 CHS exam without prior thaw. Total 25-hydroxyvitamin D (25-OHD) was measured using high performance liquid chromatography-tandem mass spectrometry (Waters Quattro Micro mass spectrometer). The interassay coefficient of variation (COV) is < 3.4%. Intact serum PTH was quantified using a two-site immunoassay (Beckman Unicell DxI clinical analyzer). The reference range is 17–66 pg/mL. The interassay COV for PTH is < 4.5% at 37 pg/mL. Investigators who were blinded to the outcome status for each participant conducted measures for both markers.
SCD was defined as a sudden pulseless condition from a cardiac origin in a previously stable individual occurring out of the hospital or in the emergency room. For unwitnessed deaths, the participant must have been seen within 24 hours of the arrest in a stable condition and without evidence of a noncardiac cause of cardiac arrest. Finally, these cases could not have life-threatening, noncardiac comorbidities or be under hospice or nursing home care. These definitions are concordant with the ones proposed by the National Heart Lung and Blood Institute and the Heart Rhythm Society working group on SCD.18
The adjudication process was comprised of multiple steps. A specialized committee in CHS first adjudicated the cause of death. SCD cases were identified by a cardiologist’s (NS) record review of all cardiac deaths. Comprehensive data were gathered on cardiovascular deaths from hospital records, physician interviews, next of kin and/or witnesses, death certificates and autopsy reports when available. Survivors or successfully resuscitated events were not included in the definition of SCD. A second physician conducted blind review of a sample of 70 potential cases with an 88% inter-reviewer agreement and kappa value of 0.74 for SCD.19
We compared baseline characteristics of participants across 4 groups defined by 25-OHD and PTH concentrations using either chi-squared or ANOVA tests. Both measures of mineral metabolism were also evaluated as continuous variables. Vitamin D was analyzed according to previously published categories of < 20 and ≥ 20 ng/ml,1–3 and we analyzed serum PTH concentrations as ≤ 65 versus > 65 pg/l. This PTH concentration represents the upper limit of normal for the assay as determined from the central 95% of values from 43 normal laboratory personnel with normal 25-OHD concentrations in March 2005. Further, previous studies have also used this PTH cutpoint for evaluation in observational cohorts.4 We calculated SCD rates per 1000 person years according to categories of 25-OHD and PTH.
We estimated the association between mineral metabolism markers and SCD using Cox proportional hazards regression models. We adjusted multivariate analyses for a set of preselected covariates including age, gender, race, season, clinic, education, physical activity, smoking, body mass index, hypertension, diabetes and cystatin C-based estimated glomerular filtration rate (cystatin C-eGFR). We tested combined associations of 25-OHD and PTH with SCD by calculating the relative excess risk of interaction.20
Because the Cox proportional hazards model treats non-SCD causes of death simply as indications to censor under an assumption of non-competing risks, we repeated our analysis using a competing risk approach in which non-sudden cardiac death and non-cardiovascular death are treated as simultaneous competing risks. 21 Next, we assessed the combined associations of 25-OHD and PTH with other mortality endpoints including total mortality, cardiovascular mortality, and cardiovascular mortality that censored for SCD cases. Finally, the estimates of risk for the individual mortality endpoints were compared using the seemingly unrelated estimation model. The proportional hazards assumption was not violated in any of these analyses.
S-Plus (release 8.0, Insightful Inc, Seattle, Wash) and SPSS statistical software (release 15.0, SPSS Inc, Chicago, Ill) were used for the analyses. A probability value <0.05 was considered statistically significant.
In comparison to participants with 25-OHD and PTH levels in the normal range (PTH < 65 pg/ml and 25-OHD ≥ 20 ng/ml), participants with combined disturbances of these mineral metabolism markers were more likely to be women, black, to have a lower socioeconomic status, less education, and lower physical activity. In addition they were more likely to have hypertension, diabetes and be taking a calcium channel blocker. Less than 1% of participants in the Cardiovascular Health Study were taking vitamin D or calcium supplements in 1992–1993. Hormone replacement therapy use appeared minimal and did not differ significantly between categories. There were no significant differences between the groups in calcium, phosphorus, and potassium levels. In addition, the functional status was not significantly different across vitamin D and PTH groups. Finally, there were no differences in ECG parameters including left ventricular hypertrophy, QT interval, or left bundle branch block (table 1).
Among the 2,312 subjects evaluated, 73 adjudicated SCD events occurred over a median follow-up of 14.0 years (incidence rate 2.3 per 1,000 person-years). The annualized incidence of SCD was greater among participants who had lower vitamin D concentrations: 0.2% per year among participants who had 25-OHD concentrations ≥ 20 ng/ml and 0.4% per year among those with 25-OH D levels < 20 ng/ml. Similarly, the incidence of SCD was greater among participants who had higher PTH concentrations: 0.2% per year in participants with PTH < 65 pg/ml and 0.4% per year among those with PTH ≥ 65 pg/ml. When evaluated as combined markers of risk, low vitamin D concentrations and elevated PTH levels identified a subgroup of participants at the highest SCD risk (figure 2).
When 25-OHD and PTH were evaluated as individual, parallel exposures (table 2), each was associated with increased risk of SCD, which was significant at the alpha=0.05 level in unadjusted but not adjusted models. When 25-OHD and PTH were evaluated in combination (table 3), participants with both a low vitamin D and an elevated PTH had a greater than 2-fold risk of SCD. Although the risk of SCD was significantly elevated when combining the two measures, the relative excess risk of interaction between vitamin D and PTH (figure 1) was not significant (p = 0.150).
The combination of 25-OHD deficiency and PTH excess were independently associated with total mortality and cardiovascular death, which included both sudden and non-sudden cardiac deaths, after multivariate adjustment (table 4). The association between vitamin D, PTH and non-sudden cardiac death, which was assessed using the cardiovascular mortality endpoint and censoring for SCD, was not significant in either the unadjusted or adjusted models. Finally, although the risk estimates for total mortality and cardiovascular death were lower than the risk for SCD, they were only marginally different than the risk for SCD (p = 0.599 for the difference between SCD and total mortality risk; p = 0.0806 for the difference between SCD and cardiovascular mortality risk).
In this community-based study of ambulatory older adults without known cardiovascular disease, the combination of low vitamin D and high PTH concentrations identified a group of participants at a 2.5-fold risk for SCD compared to the group with normal levels of both markers. As individual risk factors, neither low vitamin D nor high PTH concentrations were associated with SCD events. These findings thus validate only part of our original hypothesis describing a link between the mineral metabolism pathway and SCD.
Participants with both low vitamin D and high PTH concentrations may have a greater degree of mineral metabolism dysregulation and a greater risk for SCD. Since these hormones are inter-related and represent markers of the same pathway, we had hypothesized that a simultaneous assessment may identify a higher risk subgroup at risk for SCD. This combined abnormality likely identifies a subset of participants with decreased vitamin D substrate and inadequate stimulation of the vitamin D receptor. Physiologically, secondary hyperparathyroidism is expected among people with vitamin D deficiency. Our analysis, however, demonstrates that the extent of PTH elevation may vary in the setting of 25-OHD deficiency. Participants with either a low concentration of vitamin D or high concentration of PTH only did not have an increased SCD risk. Characterization of the mineral metabolism pathway with one marker alone does not appear to identify the extent of subclinical cardiovascular disease that predisposes these participants to lethal arrhythmias including ventricular tachycardia or ventricular fibrillation and subsequent SCD.
Important strengths of this study include its focus on individuals who were free of clinical cardiovascular disease at baseline and the evaluation of adjudicated SCD events during long-term follow-up. Although this cohort was comprised of the elderly, the multivariate analyses adjusted for many traditional cardiovascular risk factors including age, hypertension, smoking and diabetes. The sudden nature of these out-of-hospital events distinguishes them from the other cardiovascular deaths that are typically precipitated by hospitalization for heart failure, myocardial infarction or stroke. Baseline assessments of vitamin D and PTH have identified a group of elderly people without prevalent cardiovascular disease, 12% of our sample, who are at an increased risk of SCD. These metabolic disturbances are likely contributing to subclinical disease including both electrophysiologic and structural remodeling of the heart. Although the QT interval was not significantly different across the various mineral metabolism groups at baseline, this baseline measure does not represent longitudinal changes in the repolarization interval. In addition, the ECG does not capture repolarization reserve and subsequent susceptibility to arrhythmias. Finally, these findings complement recent studies which have reported an independent association between these markers of mineral metabolism and SCD in high risk populations including those with end stage renal disease22 and coronary artery disease.23, 24 These associations serve to generate hypothesis regarding the potential impact of mineral metabolism disturbances on SCD; replication of our findings in other population-based cohorts is necessary to prove their validity.
Experimental evidence suggests that dysregulation in vitamin D and PTH synthesis results in structural and ionic channel remodeling that may increase the susceptibility to cardiac arrhythmias. Specifically, changes in these mineral metabolism measures are known to induce renin expression and subsequent activation of the renin-angiotensin-aldosterone system (RAAS).5 Activation of the mineralocorticoid receptor has been shown to cause an increase in the inward calcium channel current and a decrease in the transient outward potassium current.25 These changes can prolong the repolarization interval and cardiac action potential over time and increase the susceptibility to fatal ventricular arrhythmias.26,27 In addition, vitamin D appears to have a direct effect on cardiac repolarization as its administration in hemodialysis patients is associated with a reduction in QTc dispersion.28 Finally, dysregulation in vitamin D and PTH also induce cardiac hypertrophy6, 11–13 and fibrosis,6 which are known risk factors for SCD.29, 30
In addition, our findings may help provide greater insight for the circannual variation in sudden cardiac death rates with peak incidences occurring during the winter months and the lowest in the summer.31–33 Various mechanisms related to daylight exposure have been implicated as a potential explanation for these observations. Prior studies have demonstrated seasonal variation in autonomic tone that may be induced by daylight exposure.34 Further, circannual variations in vitamin D and PTH levels may also contribute toward these differences in SCD rates. The photosynthesis of vitamin D is dependent on the intensity of ultraviolet radiation with peak values in the summer and lowest values in the winter.35, 36 Parathyroid hormone (PTH) also demonstrates an inverse seasonal relationship with 25-hydroxyvitamin D (25-OHD).37, 38 Although our analyses adjusted for both the site and season in which samples were collected, residual confounding may be present.
Our findings suggest that the combined assessment of vitamin D and PTH dysregulation is a stronger risk factor for SCD than non sudden-cardiac death. Future studies will need to validate these findings in other cohorts with more events and evaluate specific mechanisms that link the mineral metabolism pathways to cardiac arrhythmias.
Several limitations of the current study should be considered. The relatively small number of definitive SCD cases in our analyses reduced the power necessary to detect moderate sized effects; this may have limited our ability to find independent associations between either vitamin D or PTH with SCD in isolation. The limited number of SCD events also raises the possibility of model overfitting. In addition, the lack of interaction between vitamin D and PTH is most likely a consequence of low statistical power. However, the magnitude of the point estimates of the adjusted analyses and the findings combining both vitamin D deficiency and elevated PTH, our apriori hypothesis, suggest that these metabolic measures are likely associated with SCD risk. In addition, CHS consisted solely of elderly persons. The described associations of vitamin D and PTH with SCD cannot necessarily be extrapolated to younger populations. Further, given the low prevalence of cardioprotective medications in this cohort, our study cannot evaluate whether pharmacologic agents such as angiotensin converting enzyme inhibitors, statins, or beta blockers alter SCD risk in participants with mineral metabolism dysregulation. In addition, as with any observational study, there is likely to be residual confounding in the association of vitamin D or PTH with SCD.
The findings from our study provide insight that SCD may be related to vitamin D deficiency and hyperparathyroidism. These results suggest that dysregulation in the mineral metabolism axis may increase the risk of alterations in the electrophysiological properties of the myocardium and thereby elevate SCD risk, without any clinical evidence of overt cardiovascular disease. The next step of this research area should be to confirm our findings in other studies, and perhaps to combine cohorts to allow more precise determinations of these associations. If confirmed by future studies, then an investigation should evaluate whether metabolic markers can be used for SCD risk stratification.
Sources of Funding
This work was supported by the National Institute on Aging at the National Institutes of Health [grant number AG-023629]. CHS was supported by contract numbers N01-HC-80007, N01-HC-85079 through N01-HC-85086, N01-HC-35129, N01 HC-15103, N01 HC-55222, N01-HC-75150, N01-HC-45133, grant number U01 HL080295 from the National Heart, Lung, and Blood Institute at the National Institutes of Health, with additional contribution from the National Institute of Neurological Disorders and Stroke at the National Institutes of Health. Additional support was provided through R01 AG-15928, R01 AG-20098, and AG-027058 from the National Institute on Aging, R01 HL-075366 from the National Heart, Lung and Blood Institute, and the University of Pittsburgh Claude. D. Pepper Older Americans Independence Center P30-AG-024827. RD was supported by K23DK089118. Additional grant support includes National Institutes of Health R01 HL084443 and R01 HL088456.
Conflicts of Interest: None