Rajiv Agarwal and Panagiotis I. Georgianos
Con: Nutritional vitamin D replacement in chronic kidney disease and end-stage renal disease, Nephrol Dial Transplant 2016; doi:
While conducting a meta-analysis on vitamin D supplementation, Joel Gunnarsson, Senior Research Analyst at Quantify Research, Stockholm, Sweden pointed out a typographical error in the meta-analysis of vitamin D supplementation in chronic kidney disease (CKD) 1. The authors confirmed that a minus sign was inadvertently omitted in measuring change from baseline in parathyroid hormone (PTH) values in a 25 patient study 2; they corrected the error and the replacement Figure 2 is reproduced below. The effect size should have been a reduction in PTH of 70.1 pg/mL; instead it was reported as an increase in PTH of 70.1 pg/mL.
As a result of this reanalysis, some changes in the interpretation were necessary. The authors now note that vitamin D supplementation among CKD patients not on dialysis provokes a reduction in PTH (mean difference (MD): -62 pg/mL; 95% CI: -97 – -28)). However, they also note that these data are based on 122 patients in 4 studies and there is substantial heterogeneity among studies (I2 51.8%); the interpretation of the magnitude of heterogeneity is adapted from the Cochrane Handbook for Systematic Reviews of Interventions 3. Similarly, in these patients there is considerable heterogeneity in raising 25 hydroxyvitamin D concentrations in CKD patients not on dialysis (I2 80.6%). Furthermore, substantial heterogeneity in raising 25 hydroxyvitamin D concentrations is also noted among patients on dialysis (I2 65.7%); regardless of the dialysis status, the heterogeneity among studies is substantial (I2 69.3%). In contrast, among 568 patients reported on dialysis, the reduction in PTH was not significant (MD -25 pg/mL; 95% CI -57 - +7) and the heterogeneity among studies might not be important (I2 5.8%). Thus, dialysis patients and those with CKD alone might have different responses to nutritional vitamin D supplementation (heterogeneity between groups p=0.067).
Forest plot depicting the change from baseline in PTH levels in the ‘nutritional vitamin D group’ minus the change from baseline in the ‘placebo group’.
Overall, they believe that a case can be made for nutritional vitamin D supplementation in those with CKD who are not on dialysis to lower PTH concentrations but the data to support this in 2016 came from 4 studies with only 122 patients with substantial heterogeneity in effect sizes. Thus, as is true for many areas in management of CKD, they need better data to support whether nutritional vitamin D is worthwhile; the data to support this notion were not convincing in 2016. The authors regret the error and thank Joel Gunnarsson for graciously pointing out the error in the work.
Comments