Vitamin D, whether synthesized in the skin in response to exposure to ultraviolet B rays or ingested in the diet, is hydroxylated to 25-hydroxyvitamin D [25(OH)D] in the liver, and 25(OH)D circulates in up to 1000-fold higher concentrations than the most potent vitamin D metabolite, 1,25-dihydroxyvitamin D [1,25(OH)2D, calcitriol](1). Vitamin D deficiency is frequently observed in the majority of HD patients (2). Since the kidney is the main source of circulating 1,25(OH)2D, which is crucial for calcium and phosphorus metabolism as well as PTH homeostasis, supplementation with 1,25(OH)2D and its analogs is routinely provided to many HD patients, and it is associated with improved survival (3). Low serum 25-OHD levels in patients with CKD have been associated with higher mortality (4) and a more frequent evolution toward ESRD (5). In incident HD patients not treated with calcitriol analogs, vitamin D deficiency has been associated with a higher mortality rate (6,7). An observational study of HD patients has reported that calcitriol analog therapy with alfacalcidol was associated with a better survival rate (8). The aim of the present study was to assess the effect of alfacalcidol therapy on survival in a regional Japanese cohort of prevalent HD patients.
The aim of this study was to determine the relationship between alfacalcidol therapy and the outcomes of chronic hemodialysis (HD) patients. We collected demographic and clinical baseline data from 190 prevalent HD patients in a regional Japanese cohort. A 5-year survival analysis was performed according to whether the patients were receiving calcitriol analog therapy. Alfacalcidol therapy at a mean dose of 5.2 ± 1.8 µg/week was performed in 89 (46.8%) of the 190 patients. We recorded 38 deaths during the follow-up period, including 19 deaths from cardiovascular events. A Kaplan–Meier analysis demonstrated that the alfacalcidol users had a significantly lower rate of all-cause mortality and cardiovascular mortality than the non-users. According to a multivariate Cox proportional hazards model, in addition to the use of alfacalcidol (HR=0.347 [0.155–0.714]; P = 0.0035), serum CRP levels (HR= 1.746 [1.184–2.442]; P = 0.0071) and non-HDL-cholesterol levels (HR=1.012 [1.001–1.022]; P = 0.0267) were identified as independent predictors of all-cause mortality, and the presence of diabetes mellitus (HR=3.720 [1.182–12.398]; P = 0.0246) was identified as an independent predictor of cardiovascular mortality. These findings suggest that low-dose alfacalcidol therapy provides a survival advantage to chronic HD patients.
PATIENTS AND METHODS
A prospective clinical study was conducted between January 2005 and December 2010. A total of 190 HD patients were recruited from among patients who had been routinely treated in the dialysis unit of the Hidaka Hospital for at least 6 months. HD patients, who had septicemia and severe illness, were excluded and transferred to another dialysis unit for intensive care. We examined data collected from the prevalent 190 HD patients having a baseline value of demographic and laboratory results and record of drug administration. The underlying renal diseases were diabetic nephropathy (N = 60) and nephropathy of non-diabetic origin (N = 130). The Institutional Research Ethics Committee approved the study protocol, and all patients signed an informed consent form. HD with bicarbonate dialysate was performed three times weekly (4 h/day). The potassium concentration of the dialysate was 2.0 mEq/L, and its calcium concentration was 3.0 mEq/L.
Based on the medications at entry, the 190 patients were divided into two groups, according to whether or not treatment regimen at entry included alfacalcidol therapy. Medical records were carefully checked for prescriptions of antihypertensive drugs, lipid-lowering agents and weekly doses of calcium carbonate (CaCO3) and alfacalcidol. Alfacalcidol was administered when serum calcium was less than 8.5 mg/dL even after titration with CaCO3 and the doses of alfacalcidol and sevelamer were not changed during the follow-up period. The patients in the two groups were compared according to their baseline characteristics and treatment. None had a history of parathyroidectomy before entry into the study. The primary outcome was all-cause mortality and cardiovascular mortality during a 5-year follow-up period. This study was conducted in compliance with the Declaration of Helsinki.
Blood pressure with a brachial sphygmomanometer was recorded three times after the subject had rested in the supine position for at least 10 min, and the average value of the three measurements was adopted. Blood was collected at entry to measure routine laboratory parameters, including hemoglobin concentration, serum levels of urea nitrogen, creatinine, albumin, calcium, phosphorus, total cholesterol, HDL-cholesterol, triglyceride, CRP, and alkaline phosphatase. The non-HDL-cholesterol was calculated by subtracting the HDL-cholesterol from the total cholesterol. Mean values of three measurements during the 3 months before the start of this study were used for analysis.
Continuous variables are reported as means ± SD, and categorical variables are presented as numbers and percentages. The differences between group means were tested for statistical significance by analysis of variance or the χ2 test (categorical variables), as appropriate. Survival curves were plotted by the Kaplan–Meier method and assessed by the log-rank test. Prognostic variables for survival were identified by univariate and multivariate analyses. Multivariate stepwise methods included variables with P-value less than 0.2 by univariate analyses. Differences with P-values <0.05 were considered statistically significant. The statistical analyses were performed by using the Stat View 5 software program (SAS Institute Inc., Cary, NC, USA) for Windows personal computers.
The clinical characteristics of the users and non-users of alfacalcidol are summarized in Table 1. Alfacalcidol therapy was performed in 89 (46.8%) of the 190 patients, and the mean dose was 5.2 ± 1.8 µg/week. There was no significant difference between the users and non-users in demographic or clinical parameters, including their serum calcium, phosphorus, or intact PTH levels, or in regard to prescription of phosphate binders, anti-hypertensive drugs, or lipid-lowering agents.
|All||Users (N = 89)||Non-users (N = 101)||P-value|
|Age (years)||61.0 ± 11.8||60.9 ± 11.0||61.0 ± 12.5||0.9330|
|Male (%)||117 (61.6)||60 (67.4)||57 (56.4)||0.1364|
|Dialysis vintage (months)||104.9 ± 93.7||108.4 ± 102.7||101.9 ± 85.4||0.6350|
|Body mass index (kg/m2)||21.5 ± 4.0||21.1 ± 2.8||21.8 ± 4.8||0.2369|
|Diabetes (%)||60 (31.6)||27 (30.3)||33 (32.7)||0.7565|
|Systolic blood pressure (mm Hg)||137.1 ± 24.0||138.4 ± 23.1||136.0 ± 24.9||0.4832|
|Diastolic blood pressure (mm Hg)||78.7 ± 12.6||79.7 ± 11.5||77.9 ± 13.4||0.3192|
|Pulse pressure (mm Hg)||58.4 ± 16.4||58.8 ± 16.7||58.1 ± 16.2||0.7911|
|Creatinine (mg/dL)||11.1 ± 3.0||11.1 ± 2.8||11.0 ± 3.2||0.6835|
|Calcium (mg/dL)||8.8 ± 0.9||8.8 ± 0.8||8.8 ± 0.9||0.6854|
|Phosphorous (mg/dL)||6.1 ± 1.2||6.0 ± 1.2||6.1 ± 1.2||0.5524|
|Intact parathyroid hormone (pg/mL)||230.4 ± 142.8||213.8 ± 117.8||245.1 ± 160.8||0.1325|
|Total cholesterol (mg/dL)||152.3 ± 34.5||153.0 ± 34.6||151.7 ± 34.6||0.7975|
|HDL cholesterol (mg/dL)||43.5 ± 12.8||44.2 ± 12.3||42.8 ± 13.3||0.4693|
|Non-HDL cholesterol (mg/dL)||108.9 ± 33.1||108.8 ± 34.7||108.9 ± 31.9||0.9899|
|Triglycerides (mg/dL)||111.5 ± 64.0||105.1 ± 62.8||117.0 ± 64.8||0.2016|
|Albumin (g/dL)||3.8 ± 0.3||3.8 ± 0.3||3.7 ± 0.3||0.1258|
|Alkaline phosphatase (U/L)||267.3 ± 100.7||245.0 ± 80.6||287.1 ± 112.2||0.0038|
|C-reactive protein (mg/dL)||0.4 ± 0.7||0.3 ± 0.8||0.4 ± 0.6||0.7377|
|Hemoglobin (g/dL)||10.2 ± 1.0||10.1 ± 1.0||10.3 ± 1.0||0.0909|
|Anti-hypertensive drugs (%)||131 (68.9)||65 (73.0)||66 (65.4)||0.2532|
|Lipid-lowering agents (%)||40 (21.1)||17 (19.1)||23 (22.8)||0.5949|
|CaCO3 (g/week)||17.9 ± 10.0||17.4 ± 8.7||18.3 ± 11.1||0.5409|
|Alfacalcidol (µg/week)||5.2 ± 1.8||0|
Follow-up was completed in December, 2010, and during the follow-up period, 38 deaths were recorded including 19 deaths from cardiovascular events. The dates and causes of death were obtained by reviewing the hospital record. The 19 deaths from cardiovascular events were attributable to acute myocardial infarction (N = 7), congestive heart failure (N = 6), stroke (N = 3), and peripheral artery disease (N = 3). Non-cardiovascular causes of death were sepsis in five patients, pneumonia in three patients, cancer in two patients, peritonitis in one patient, pulmonary hypertension in one patient, malnutrition in one patient and unknown origin in six patients. The mean follow-up period was 58 ± 22 months. According to the Kaplan–Meier analysis, the alfacalcidol users had a significant lower rate of all-cause mortality (Fig. 1) and cardiovascular mortality (Fig. 2) than the non-users. The mean daily dose of alfacalcidol was 0.57 µg and was not changed during the follow-up period. During the 3 months before the start of this study, four patients were administered 1.0 µg of alfacalcidol for the treatment of secondary hyperparathyroidism. The death rate in non-users was significantly higher than those in the low-dose group (less than 6 µg/week) (P = 0.0052). In addition, there was no significant difference in death rate between the non-user group and the high-dose group (more than 6 µg/week) (P = 0.1629).
The univariate Cox proportional hazards model was used to identify predictors of all-cause mortality (Table 2) and cardiovascular mortality (Table 3), and the results showed a significant association between alfacalcidol therapy and all-cause mortality and cardiovascular mortality. Other significant univariate predictors of all-cause mortality were age, dialysis vintage, pulse pressure, and the serum creatinine, non-HDL-cholesterol, triglycerides, albumin, alkaline phosphatase, and CRP levels. Age, presence of diabetes, systolic blood pressure, pulse pressure, and serum creatinine levels were identified as univarite predictors of cardiovascular mortality.
|Candidate variables||HR||95% CI||P-value|
|Age (/1 year)||1.037||1.008–1.068||0.0124|
|Male (vs. female)||0.903||0.477–1.749||0.7570|
|Dialysis vintage (/1 months)||0.996||0.992–1.000||0.0387|
|Body mass index (/1 kg/m2)||1.019||0.937–1.087||0.6426|
|Diabetic (vs. Non-diabetic)||1.773||0.915–3.358||0.0882|
|Systolic blood pressure (/1 mm Hg)||1.012||0.999–1.026||0.0717|
|Diastolic blood pressure (/1 mm Hg)||0.996||0.971–1.022||0.7522|
|Pulse pressure (/1 mm Hg)||1.028||1.009–1.047||0.0045|
|Creatinine (/1 mg/dL)||0.861||0.777–0.956||0.0052|
|Calcium (/1 mg/dL)||0.844||0.585–1.225||0.3704|
|Phosphorous (/1 mg/dL)||0.851||0.636–1.121||0.2572|
|Intact-parathyroid hormone (/1 pg/mL)||1.001||0.999–1.003||0.3463|
|Total cholesterol (/1 mg/dL)||1.007||0.998–1.016||0.1049|
|HDL-cholesterol (/1 mg/dL)||0.986||0.958–1.012||0.2965|
|Non-HDL-cholesterol (/1 mg/dL)||1.010||1.001–1.018||0.0379|
|Triglycerides (/1 mg/dL)||1.005||1.000–1.009||0.0666|
|Albumin (/1 g/dL)||0.278||0.107–0.754||0.0124|
|Alkaline phosphatase (/1 U/L)||1.003||1.001–1.006||0.0145|
|C-reactive protein (/1 mg/dL)||1.635||1.169–2.130||0.0066|
|Hemoglobin (/1 g/dL)||0.838||0.605–1.158||0.2854|
|Anti-hypertensive drugs (vs. non users)||1.284||0.645–2.779||0.4898|
|Lipid-lowering agents (vs. non users)||1.895||0.922–3.674||0.0801|
|CaCO3 (vs. non users)||1.001||0.361–4.154||0.4560|
|Alfacalcidol (vs. non users)||0.361||0.167–0.720||0.0033|
|Candidate variables||HR||95% CI||P-value|
|Age (/1 year)||1.045||1.003–1.091||0.0331|
|Male (vs. female)||0.906||0.367–2.342||0.8330|
|Dialysis vintage (/1 months)||0.317||0.997–0.991||0.3167|
|Body mass index (/1 kg/m2)||1.029||0.915–1.119||0.5957|
|Diabetic (vs. Non-diabetic)||4.188||1.684–11.258||0.0021|
|Systolic blood pressure (/1 mm Hg)||1.025||1.006–1.046||0.0107|
|Diastolic blood pressure (/1 mm Hg)||1.002||0.966–1.039||0.9281|
|Pulse pressure (/1 mm Hg)||1.049||1.022–1.078||0.0004|
|Creatinine (/1 mg/dL)||0.815||0.702–0.944||0.0064|
|Calcium (/1 mg/dL)||0.989||0.587–1.691||0.9680|
|Phosphorous (/1 mg/dL)||0.817||0.536–1.206||0.3176|
|Intact-parathyroid hormone (/1 pg/mL)||1.000||0.997–1.003||0.9138|
|Total cholesterol (/1 mg/dL)||1.011||0.998–1.023||0.0888|
|HDL-cholesterol (/1 mg/dL)||0.999||0.962–1.033||0.9656|
|Non-HDL-cholesterol (/1 mg/dL)||1.012||0.999–1.024||0.0719|
|Triglycerides (/1 mg/dL)||1.002||0.994–1.009||0.6348|
|Albumin (/1 g/dL)||0.330||0.086–1.393||0.1293|
|Alkaline phosphatase (/1 U/L)||1.003||0.999–1.006||0.1496|
|C-reactive protein (/1 mg/dL)||1.519||0.865–2.233||0.1261|
|Hemoglobin (/1 g/dL)||0.966||0.610–1.513||0.8805|
|Anti-hypertensive drugs (vs. non users)||2.450||0.816–10.531||0.1170|
|Lipid-lowering agents (vs. non users)||2.405||0.895–5.981||0.0795|
|CaCO3 (vs. non users)||0.728||0.209–4.590||0.6835|
|Alfacalcidol (vs. non users)||0.361||0.117–0.945||0.0375|
The multivariate Cox proportional hazards model was used to identify predictors of all-cause mortality (Table 4) and cardiovascular mortality (Table 5). In addition to alfacalcidol therapy (HR=0.347 [0.155–0.714]; P = 0.0035), serum CRP levels (HR= 1.746 [1.184–2.442]; P = 0.0071) and non-HDL-cholesterol levels (HR=1.012 [1.001–1.022]; P = 0.0267) were identified as independent predictors of all-cause mortality, whereas the presence of diabetes mellitus (HR=3.720 [1.182–12.398]; P = 0.0246) was identified as an independent predictor of cardiovascular mortality. The beneficial effect of alfacalcidol therapy on all-cause mortality and cardiovascular mortality remained significant when the model included systolic blood pressure, pulse pressure, or non-HDL cholesterol.
|Alfacalcidol (vs. non users)||0.347||0.155–0.714||0.0035|
|C-reactive protein (/1 mg/dL)||1.746||1.184–2.442||0.0071|
|Non-HDL-cholesterol (/1 mg/dL)||1.012||1.001–1.022||0.0267|
|Alkaline phosphatase (/1 U/L)||1.003||1.000–1.005||0.0695|
|Age (/1 year)||1.014||0.983–1.046||0.3850|
|Dialysis vintage (/1 months)||0.998||0.992–1.002||0.3085|
|Pulse pressure (/1 mm Hg)||1.015||0.993–1.039||0.1897|
|Creatinine (/1 mg/dL)||1.004||0.864–1.168||0.9599|
|Albumin (/1 g/dL)||0.653||0.169–2.597||0.5417|
|Diabetic (vs. Non-diabetic)||3.720||1.182–12.398||0.0246|
|Alfacalcidol (vs. non users)||0.317||0.101–0.839||0.0198|
|Age (/1 year)||1.030||0.976–1.091||0.2961|
|Systolic blood pressure (/1 mm Hg)||0.980||0.935–1.029||0.4186|
|Pulse pressure (/1 mm Hg)||1.050||0.977–1.127||0.1818|
|Creatinine (/1 mg/dL)||0.950||0.802–1.129||0.5518|
The results of this study showed that alfacalcidol therapy was associated with a significantly lower risk of all-cause mortality and cardiovascular mortality in our cohort of chronic HD patients.
Whether active vitamin D treatment has a beneficial or detrimental effect on the cardiovascular system and risk of mortality of HD patients has been a matter of controversy. The concern about a harmful effect of vitamin D is mainly based on the fact that high doses of vitamin D have been shown to increase vascular calcification in experimental animal models (9,10). On the other hand, low doses and more physiological doses of active vitamin D have been found to have a cardioprotective effect (11,12). We previously reported finding that low-dose oral vitamin D therapy protects HD patients from developing vascular calcification (13).
Observational studies have demonstrated the advantages of treatment of HD patients with active vitamin D in regard to cardiovascular mortality (8) and all-cause mortality (14–16). Although oral active vitamin D therapy is widely prescribed, there have been few reports of studies that investigated the survival benefit of treatment with oral active vitamin D in HD patients (8,17). In a study of 242 HD patients, Shoji et al. (8) reported that patients treated with alfacalcidol were at reduced risk of cardiovascular death according to an adjusted Cox model. Naves-Diaz et al. (17) showed a significant survival advantage of oral active vitamin D therapy that appeared to be independent of other potential risk factors and confounders, and the results of the present study are consistent with their findings. A French group recently reported a similar efficacy of vitamin D therapy with either daily calcifediol (10–50 µg/day) (18) or monthly cholecalciferol (100 000 U) (19). They reported no evident toxicity, a small decrease in serum PTH levels, and correction of vitamin D deficiency in more than 80% of the cases.
The positive results in regard to all-cause mortality suggest that the beneficial effect of active vitamin D is greater than its effect on calcium-phosphorus metabolism. Several experimental studies have recently demonstrated an important role of active vitamin D in suppressing cell growth and regulating immune response (20,21). Moreover, nutritional and epidemiological evidence has linked abnormalities in the vitamin D metabolism to increased susceptibility to infection (22). The present study identified serum CRP and non-HDL-cholesterol levels as well as alfacalcidol therapy as independent predictors of all-cause mortality. Inflammation and elevated serum non-HDL-cholesterol levels have been found to be risk factors of atherosclerosis in HD patients. It is likely that active vitamin D affects arterial cells and plays a role against infectious status in HD patients (23).
The benefit of oral vitamin D therapy in regard to the cardiovascular system remains unclear. It is well known that high-doses of active vitamin D increase serum calcium and phosphorus levels and suppresses serum PTH levels. On the other hand, physiological doses of active vitamin D have been shown to have protective effects by reducing the inflammatory response to cardiovascular injury, myocardial cell hypertrophy and proliferation, and renin-angiotensin system activation (24). Some of the paradoxical effects of vitamin D can be explained by the differences between the doses of active vitamin D prescribed. High pharmacological doses may precipitate vascular calcification, whereas low doses may have a protective effect.
The results of the present study showed that mean daily oral alfacalcidol doses below 1 µg had a significant benefit in reducing all-cause mortality and cardiovascular mortality in chronic HD patients, and these findings are consistent with the results of a previous study (17). Naves-Diaz et al. (17) reported that a mean daily oral vitamin D dose below 0.25 µg was able to reduce the mortality rate by 53% in HD patients whose serum PTH levels were below 150 pg/mL, independently of their serum calcium or phosphorus levels.
Alfacalcidol is the most frequently used calcitriol analog in Japan, probably because of the different dosage forms that are available. A once daily dose of an oral preparation is the most common prescription. Daily administration of 0.25 µg is most frequently prescribed. In view of the beneficial effect of alfacalcidol therapy on HD patients who are deficient in vitamin D, the dose prescribed seems to be important. However, measurement of serum 25(OH)D levels has not been regularly performed in Japan. Whether alfacalcidol has a beneficial effect on mortality rate of HD patients is likely to depend on their serum 25(OH)D levels.
Our study had the following limitations. First, the sample size was small. However, the prospective cohort design enabled prediction of mortality risk without changes in the alfacalcidol doses during a 5-year follow-up period. Second, only baseline parameters were recorded, and time-integrated parameters were not. Third, we did not measure serum 25(OH)D levels, and they may affect the beneficial effect of alfacalcidol on mortality risk.
The results of the present study demonstrated that hemodialysis patients receiving low oral doses of active vitamin D had a survival advantage, and this beneficial effect was seen even after adjustment for several confounders. However, only a randomized controlled trial can provide definitive proof of the benefit of vitamin D therapy with an active vitamin D analog.