Vitamin D in Dialysis is a Dud

#NephJC Chat

Tuesday January 15, 2019 at 9 pm Eastern

Wednesday January 16, 2019 at 9 pm Indian Standard Time (New)

Wednesday January 16, 2019 at 8 pm GMT, 12 noon Pacific

JAMA. 2018;320(22):2325-2334.

Effect of Oral Alfacalcidol on Clinical Outcomes in Patients Without Secondary Hyperparathyroidism Receiving Maintenance Hemodialysis. The J-DAVID Randomized Clinical Trial

J-DAVID Investigators: Shoji T, Inaba M, Fukagawa M, Ando R, Emoto M, Fujii H, Fujimori A, Fukui M, Hase H, Hashimoto T, Hirakata H, Honda H, Hosoya T, Ikari Y, Inaguma D, Inoue T, Isaka Y, Iseki K, Ishimura E, Itami N, Ito C, Kakuta T, Kawai T, Kawanishi H, Kobayashi S, Kumagai J, Maekawa K, Masakane I, Minakuchi J, Mitsuiki K, Mizuguchi T, Morimoto S, Murohara T, Nakatani T, Negi S, Nishi S, Nishikawa M, Ogawa T, Ohta K, Ohtake T, Okamura M, Okuno S, Shigematsu T, Sugimoto T, Suzuki M, Tahara H, Takemoto Y, Tanaka K, Tominaga Y, Tsubakihara Y, Tsujimoto Y, Tsuruya K, Ueda S, Watanabe Y, Yamagata K, Yamakawa T, Yano S, Yokoyama K, Yorioka N, Yoshiyama M, Nishizawa Y.

PMID: 30535217 Full Text at JAMA


Activated vitamin D agents and Vitamin D receptor agonists (VDRAs) are commonly prescribed to chronic dialysis patients and those with late-stage CKD to replace low levels of 1,25 vitamin D. The goal of activated vitamin D in CKD is primarily to prevent progression of secondary hyperparathyroidism and its consequences. The 2009 KDIGO Guidelines (link, PDF) reported that this strategy improved various biochemical endpoints, at the expense of hypercalcemia and other adverse events. A more important question is whether a decrease in serum PTH levels, a surrogate biochemical marker, should be used to justify the widespread use of activated vitamin D therapy.

Clinical and experimental studies report beneficial effects of VDRAs, such as suppression of the renin-angiotensin system; modulation of immune functions; anti-inflammatory effects; anti-atherosclerotic properties; inhibition of cardiac hypertrophy; and increase of proteins potentially protective against arterial calcification, including fetuin-A and klotho. But data on hard clinical outcomes, and on patient centered outcomes are lacking. Thus an events-driven randomized trial of activated vitamin D therapy was critical to understanding the role of active vitamin D in the outcomes of CKD patients.

Two important RCTs (PRIMO and OPERA) looked at non-dialytic CKD patients with moderately increased PTH levels and demonstrated increased risk of hypercalcemia but no beneficial effects on cardiac MRI endpoints. These results were reflected in the updated 2017 KDIGO Guidelines:

“4.2.2: In adult patients with CKD G3a to G5 not on dialysis, we suggest that calcitriol and vitamin D analogues not be routinely used. (Grade 2C recommendation) It is reasonable to reserve the use of calcitriol and vitamin D analogues for patients with CKD G4 to G5 with severe and progressive hyperparathyroidism. (Not graded)”

“4.2.4: In patients with CKD G5D requiring PTH lowering therapy, we suggest calcimimetics, calcitriol, or vitamin D analogues, or a combination of calcimimetics with calcitriol or vitamin D analogues. (Grade 2B recommendation)”

The guideline on use of Calcitriol in CKD G5D with secondary hyperparathyroidism is not based on RCT evidence of the effects of VDRA on mortality or patient centered endpoints, hence it is a 2B recommendation. Moreover, this guideline pertains to patients requiring PTH lowering. There is no evidence on use of these agents in patients without hyperparathyroidism to prevent worsening of hyperparathyroidism, or its consequences.

The Japan Dialysis Active Vitamin D (J-DAVID) trial was performed to test the hypothesis that treatment with VDRAs reduces the risk of cardiovascular events and mortality in patients with CKD undergoing hemodialysis without secondary hyperparathyroidism. Another important aspect should be considered before you read further. The Japanese guidelines in this area are quite different than the rest of the world. The target PTH in Japan while the trial was conceived and started was < 180 pg/mL (< 19 pmol/L) and while trial was ongoing, this was liberalized a bit to < 240 pg/mL (< 25 pmol/L). These target levels are quite different than the rest of the world. The KDIGO 2017 Update Work Group felt that modest increases in PTH may represent an appropriate adaptive response to declining kidney function, due to phosphaturic effects and increasing bone resistance to PTH. Therefore, the Update Work Group revised the 2009 Guideline recommendation to reflect the fact that treatment should not be based on a single elevated PTH value.

“4.2.3: In patients with CKD G5D, we suggest maintaining intact PTH levels in the range of approximately 2 to 9 times the upper normal limit for the assay (2C). We suggest that marked changes in PTH levels in either direction within this range prompt an initiation or change in therapy to avoid progression to levels outside of this range (2C).”

Thus, you could consider the J-DAVID trial not as trying to validate this recommendation, but rather, testing whether even below a PTH less than twice the upper limit of normal, whether addition of alfacacidol helps in cardiovascular event reduction.

The Study


Prospective randomized, open-label, blinded end point (eg PROBE design) clinical trial to examine the effect of alfacalcidol vs no VDRAs on cardiovascular events in patients receiving maintenance hemodialysis without secondary hyperparathyroidism over 48 months.

Study Population

  • Age 20-80 years on maintenance hemodialysis

  • Calcium < 10mg/dL (2.5 mmol/L)

  • Phosphate < 6mg/dL (1.9 mmol/L)

  • iPTH < 180pg/mL (19 pmol/L)

Study Period

  • July 1, 2008, to January 26, 2011


  • Patients screened at 207 dialysis centers in Japan.

Intervention & Follow up

The participants in the intervention group were assigned to receive oral alfacalcidol, starting at a dose of 0.5 μg per day. Other VDRAs were allowed in the experimental group while the control group was asked to avoid them. treatment. Dose reduction, modification or temporary cessation was allowed (Dose range of alfacalcidol 0.25 to 7 mcg/wk). If a control patient received VDRA they were dropped from the assigned treatment (see analysis section).

Primary Endpoint

Composite of

  • fatal and nonfatal cardiovascular events, including acute myocardial

  • infarction, congestive heart failure requiring hospitalization, stroke, aortic dissection/rupture, amputation of ischemic limb, and sudden cardiac death

  • coronary interventions (eg, balloon angioplasty, stenting) or bypass grafting

  • lower limb artery interventions (eg, balloon angioplasty, stenting) or bypass grafting.

Secondary Endpoint

  • All-cause death

Study period – 48 months

Sample Size and Analysis

The initial sample size was estimated at 1600, based on a 32% event rate in control arm, and with 80% power to detect a 20% risk reduction. With difficulty in enrolling, the target recruitment was then reset to 972, assuming a 28% event rate and 30% risk reduction.

They defined three analytic sets: an intention to treat (based on assigned intervention); a per protocol set and a modified per protocol set (see figure below).

eFigure 1 from Shoji et al, JAMA 2018. The difference between per-protocol (groups B and C censored) and modified per-protocol was that in the latter, group C in the intervention arm was not excluded from the analysis.

eFigure 1 from Shoji et al, JAMA 2018. The difference between per-protocol (groups B and C censored) and modified per-protocol was that in the latter, group C in the intervention arm was not excluded from the analysis.


The study was funded by the Kidney Foundation of Japan, which received funding to conduct the trial from Chugai pharmaceuticals, the manufacturer of alfacalcidol.


A total of 976 patients were randomized and 964 were included in analysis (488 in the intervention group and 476 in the control group). There was significant dropout in both the intervention and control groups.

Baseline Characteristics

The mean age of patients was 65 years. Cinacalcet was used in 5.5% patients in the alfacalcidiol group compared to 6% of controls.

FIGURE 2.png

Primary and Secondary Endpoints

The primary outcome occurred in 103 patients (21%) in the intervention group and 85 patients (18%) in the control group (absolute difference, 3% [95% CI, −1.75% to 8.24%]; hazard ratio, 1.25 [95% CI, 0.94-1.67]; P = .13).

There was no significant difference between the intervention and control group in the secondary outcome of all-cause death (18.2% vs 16.8%, respectively; hazard ratio, 1.12 [95% CI, 0.83- 1.52]; P = .46).

FIGURE 3.png

The figure shows the Kaplan-Meier curves for the primary and secondary outcomes from the intention-to-treat analysis. The primary and secondary outcomes were not different between 2 groups, in fact, there was a signal of harm in the alfacalcidol group.

FIGURE 4.png

The table summarizes the composite and individual cardiovascular events that occurred. The number of SAEs was similar between the groups. The number of cardiovascular SAEs was higher than the number of occurrences of the primary outcome because some participants had more than 2 cardiovascular SAEs.

What happened to calcium, PO4 and PTH during the trial? Since this was an open label a lot of changes were made, as can be seen here:

eFigure 2 from Shoji et al, JAMA 2018, showing changes in Ca, PO4 and PTH level over the trial period.

eFigure 2 from Shoji et al, JAMA 2018, showing changes in Ca, PO4 and PTH level over the trial period.

One might wonder why the biochemical parameters not change despite a drug versus no drug comparison? The answer is seen below:

eFigure 2 (second panel) from Shoji et al, JAMA 2018. (red = alfacalcidol, black = control). Change indrugs during the trial period. There was a higher usage of non-calcium phosphate binders in the intervention group, and a higher usage of cinacalcet and IV VDRAs in the control group.

eFigure 2 (second panel) from Shoji et al, JAMA 2018. (red = alfacalcidol, black = control). Change indrugs during the trial period. There was a higher usage of non-calcium phosphate binders in the intervention group, and a higher usage of cinacalcet and IV VDRAs in the control group.


Despite the lack of evidence from prospective trials, many nephrologists consider the administration of VDRAs to be essential in patients undergoing dialysis.  This trial showed that in patients without secondary hyperparathyroidism receiving maintenance hemodialysis, oral alfacalcidol compared with usual care did not reduce the risk of significant cardiovascular events.  

The study has a signal of potential harm rather than benefit of VDRAs in patients without elevated serum intact PTH levels. In the current trial, the occurrence of congestive heart failure and cardiac sudden death was similar between the groups. The occurrence of coronary events was also similar. The occurrence of stroke was twice as high in the intervention group than the control group (28 vs 14).

Interestingly, the last patient was recruited in 2011, and the last follow up was in April 2015, but the study was only published in December 2018.


This is an underpowered trial. The initial power analysis required 1,600 patients. The investigators only enrolled 964. A positive result with a study population this small would have required:

  • 28% of participants in the control group would experience the primary outcome

  • There would be a risk reduction of 30% in the intervention group

  • 5% of participants would be lost to follow-up

How well did they do with those predictions? Not very good. 18% (85/476) of the control group hit the primary outcome, not 28% (note the very low event rate: ~ 18% over 4 years in hemodialysis patients, not unusual for Japan, but possibly unusual elsewhere). The loss to follow-up, however, was crazy good with investigators only losing 2% of the cohort.

Additional issues include:

  • Unblinded design – and consequently there was little difference between Ca, PO4 and PTh levels in between groups. The results would not exclude an effect of alfacalcidol mediated via PTH. The PTh levels were similar in both groups since patients in the control group were more likely to receive cinacalcet or another intravenous VDRA.

  • Results cannot be generalized to patients with true secondary hyperparathyroidism, ie when levels of PTH are higher than allowed for in this trial.

  • As we have discussed above, the low event rate, other practice patterns in Japan are so different, that one should be cautious in extrapolating even otherwise in non-Japanese population.

This was not a trial of different PTH targets, it was a trial of alfacalcidol versus placebo (most of whom did receive another VDRA) at PTH levels where most of us would not use either active vitamin D or a VDRA.

What does this trial truly demonstrate?

  • VDRAs have been promoted for improving cardiovascular health, immune function and other pleotropic effects. This study does not support this practice.

  • Treatment of hemodialysis patients with normal PTH has been said to promote adynamic bone disease and subsequently increase the risk of vascular calcification and adverse cardiovascular outcomes. No adverse effect on cardiovascular outcomes was seen in this, admittedly underpowered, study.

Points to ponder

Most importantly, does this open the door for a trial of active vitamin D, or a VDRAs (versus placebo) at higher levels of PTH? And do we need a trial of differing PTH targets (using calcimimetics or VDRAs and other drugs to achieve the PTH)? Can we isolate the drug effect from any PTH effect?

Was the tendency of harm, seen in the intervention group surprising, considering increased risks of hypercalcemia? Do you, or will you, treat your CKD-D patients with VDRA in the setting of normal PTH? Let us get together and discuss these and many more pertinent questions raised on use of VDRAs in dialysis population.

For 2019, team NephJC is excited to announce the addition of a third chat focused on people in Asia. This live chat will run from 9 pm to 10 pm IST Wednesday. Here is the introduction video.

Summary by Aakash Shingada
Nephrology Fellow, New Delhi, India
NSMC Faculty Mentor