LAUSR

Question In patients with chronic kidney disease (CKD), what are the efficacy and safety of osteoporosis medications? Review scope Included English-language studies compared osteoporosis medications (bisphosphonates, teriparatide, denosumab, and raloxifene) with placebo, usual care, or active control for 6 months in 25 patients who had stage G3 to G5 CKD, were receiving dialysis, or had received a kidney transplant. Outcomes were bone mineral density (BMD), fractures, and adverse events. Review methods This is an update of a 2009 review*. MEDLINE and Cochrane Central Register of Controlled Trials (Dec 2006 to Dec 2016) were searched for randomized controlled trials (RCTs). 13 RCTs (n =9850, mean age 44 to 80 y, duration 8 to 36 mo), including 8 new trials and 5 from the previous review*, met selection criteria. Medications assessed were raloxifene (3 RCTs); pamidronate, alendronate, denosumab, and ibandronate (2 RCTs each); risedronate and teriparatide (1 RCT each). 12 RCTs were placebo-controlled. Patients received kidney transplants (6 RCTs), dialysis (2 RCTs), had G3 to G5 CKD (2 RCTs), or were postmenopausal women with CKD (subgroups from 4 RCTs). 5 RCTs had low risk for bias for randomization, 5 for allocation concealment, and 7 for blinding of outcome assessors. Main results The main results of meta-analyses for vertebral fractures are in the Table. For BMD, bisphosphonates may slow loss of lumbar spine BMD in kidney transplant recipients (moderate strength of evidence), and raloxifene may increase BMD (low strength of evidence) compared with placebo; findings were inconclusive for teriparatide or denosumab vs placebo, and ibandronate vs risedronate. Individual RCTs reported increased adverse events for some drugs, but findings were inconsistent or not reported. Conclusion In patients with chronic kidney disease, osteoporosis medications may improve bone mineral density and fractures in some patient groups, but results of many comparisons are inconclusive. Osteoporosis medications vs placebo for vertebral fractures at 8 to 36 mo in patients with chronic kidney disease (CKD) Osteoporosis medications Patients Number of trials (n) Weighted event rates vs placebo Odds ratio (95% CI) Bisphosphonates Transplant recipients 4 (382) 4.7% vs 8.0% 0.56 (0.23 to 1.35) CKD 2 (231) NA 0.71 (0.30 to 1.65) Raloxifene GFR/dialysis 3 (5033) NA 0.55 (0.44 to 0.70) Teriparatide CKD 1 (581) 4.2% vs 19% 0.19 (0.10 to 0.35) GFR = glomerular filtration rate; NA = not available; CI defined in Glossary. Weighted event rates calculated from placebo event rates and odds ratios in article using a random-effects model. GFR <45 to 59 mL/min/1.73m2 or dialysis and G5 CKD. Very low strength of evidence. Low strength of evidence. Commentary The prevalence of both osteoporosis and CKD increases with age. In subgroup analyses of pivotal osteoporosis treatment trials, patients with CKD did not differ from those without CKD for fractures (1). However, the trials excluded patients with abnormal levels of parathyroid hormone (PTH), alkaline phosphatase, or calcium (1). As CKD worsens (glomerular filtration rate [GFR] <45 mL/min/1.73 m2), the physiologic abnormalities of bone and mineral metabolism become more complex (1, 2). By G4 CKD, patients may have renal osteodystrophy and not simply osteoporosis (1). Abnormalities include increased phosphate, PTH, and fibroblast growth factor 23 levels and decreased klotho, calcitriol, and calcium levels (1, 2). Vascular calcifications can result from the altered physiology, and drugs that target bones could also affect vessels (1, 2). Therefore, results from studies of G3 CKD should not be generalized to patients at G4 or G5. Wilson and colleagues conclude that data are insufficient to recommend osteoporosis medications in patients with CKD. The authors did not separate studies by GFR level, and most of the evidence is from patients with G3 CKD, with almost no information on those with G4 and G5. Although the results had low strength of evidence in patients with G3 CKD, they are in the expected directions and include such hard outcomes as fractures. Based on this limited evidence, osteoporosis medications should be used for usual indications in patients with G3 CKD who have normal calcium, PTH, and alkaline phosphatase levels. In G4 to G5 CKD, however, pathophysiology is different and evidence of effectiveness is absent. Because many patients with G4 to G5 CKD have low-turnover bone disease (1), antiresorptive medications do not make physiologic sense. Although denosumab is not cleared by the kidney, there is no current evidence that it reduces fractures in patients with G4 to G5 CKD, and severe hypocalcemia (e.g., seizures, tetany) is a safety concern (3). In such patients, it would be better to manage other aspects of mineral metabolism (PTH, phosphate, calcium). Focus on exercise to help with strength and balance may be effective, but that also needs further study (4).