Bisphosphonates are structural analogues of inorganic pyrophosphate, which modulates bone remodeling55 (Table 4). Bisphosphonates have a high affinity for bone because it binds to hydroxyapatite crystals (with a variable half-life in the skeleton that could be up to 10 years).55,56 Bisphosphonates reduce the rate of bone resorption by inhibiting the activity of osteoclasts and, through different mechanisms, induce apoptosis of all bone cells.55,57

Bisphosphonates are drugs of first choice in the current treatment of all types of OP. This is based on years of clinical experience, multiple indications in different types of metabolic bone disorders and their low cost.58–61 The difficulty of absorption in the oral presentations (alendronate, risedronate) and the risk of esophagitis make it necessary to take it on an empty stomach, with water, in an upright position and to wait 30min before any additional oral intake. In the KDIGO 2009 guidelines (Table 1) it was suggested that patients with CKD stages 1–2 with OP and/or high risk of fracture, and CKD stage-3 with PTH in the normal range should be followed the same treatment as in the general population.18 In fact, patients with CKD stage 3 usually have few identifiable abnormalities of mineral metabolism, sharing the same risk factors as the general population.62 Alternatively, in patients with CKD stages 1–3 with biochemical abnormalities of CKD-MBD and decreased BMD and/or fragility fractures, it was suggested that there were other factors to be considered in order to decide treatments8 (Table 1). Finally, in CKD stages 4–5D, it was suggested additional evaluation including bone biopsy before administration of antiresorptive therapy.8

Today, we know that with the deterioration of kidney function there is an increase in the prevalence of OP,63 especially in CKD stages 4–5D. Other factors (ROD, metabolic acidosis, vitamin D deficiency, excess of FGF23, hyperparathyroidism, skeletal resistance to PTH, etc.) complicate the diagnosis of OP based on BMD. Although bone biopsy is the only method to exclude other abnormalities of advanced CKD, there is a growing consensus that basing OP treatments on the need for a bone biopsy could impermissibly limit treatment in patients who have a very high risk of fracture. Bisphosphonates seem to prevent the loss of BMD in CKD64–66 but it has not been proven to be effective in reducing the risk of fractures; this is why it is recommended a prudent and individualized (risk/benefit) use of bisphosphonates.2,62,67

Guidelines for the use of bisphosphonates in the general population (applicable to patients with mild CKD without significant alterations of the mineral metabolism)5,14 propose to initiate treatment with oral bisphosphonates in the situations such as those indicated in Table 5. Also, in the general population it is indicated to re-evaluate the fracture risk by BMD 2 or 3 years after the initiation of treatment. In a recent consensus14 it is being considered FRAX®+BMD together as appropriate objectives. Patients who are receiving bisphosphonates and who are no longer at a high risk of fracture are potential candidates for “drug holiday”.13,14,68 Conversely, in patients with a BMD with T<−2.5 in the femoral neck, patients with history of fractures and with T<−2.0 in the femoral neck, or family history of fragility fractures, treatment with oral bisphosphonates could be extended up to 10 years, since the benefit would probably be greater than the associated risks.14,68 If after this period of time the risk persists, it would be indicated the transition to another drug (sequential therapy).14

The reason for these limitations in the length of treatment are fundamentally linked to two adverse effects: osteonecrosis of the jaw (ONJ) and atypical fractures of the femur (subtrochanteric-diaphyseal). In addition to an individual predisposition, they have been associated with comorbidity (mainly neoplastic), use of intravenous route (i.e., zoledronate) or high doses, prolonged exposure or co-administration with glucocorticoids.68–70 The risk of ONJ is estimated at 1/10,000 and <1/100,000patients-year,14,15,69 and is also associated to poor oral hygiene, history of dental procedures and the use of dental prosthesis.55 The risk of atypical fractures of the femur seems to be more associated with an individual predisposition in patients with OP than with the treatment itself since there is lack of association of atypical fractures with cumulative doses.14,70,71 In the general population, it has been estimated that for every atypical fracture that occurred in a treated patient, there would have been prevented 50–100 hip fractures and hundreds of vertebral and non-vertebral fractures with the consequent reduction in mortality.14,15

Finally, the inhibition of osteoclastic activity induced by bisphosphonates has raised questions about the possibility of generating an “adynamic bone” due to excessive suppression of bone remodeling and, therefore, reduced capacity for microfracture repair and greater skeletal fragility.55,72,73 At least in the general population, it has been shown that the benefit of reducing fractures outweighs the potential risk.68 However, in dogs treated with high doses of bisphosphonates, it has been observed an increase in microfractures,74 and a recent study showed less porosity but more microfractures in the fractured bones of patients treated with bisphosphonates compared with patients with untreated fractures or healthy controls.75 However, iliac crest biopsies performed after 5–10 years of treatment do not seem to show oversuppression of bone remodeling76–78 and, in other studies, an increase in microfractures has not been detected.79,80 Although not constant, the positive (although not constant) effect of anabolic teriparatide in some patients with atypical femoral fractures, previously treated with bisphosphonates, merits consideration. Therefore, the optimal duration of bisphosphonate therapy should be individualized according to the risk factors of the patient, and although the risk seems to be low even in treatments up to 10 years,68,77 the risk/benefit of the extended use remains to be assessed in low-risk patients or patients with CKD.5,81

With regard to the use of bisphosphonates in CKD, it is well known the contraindication of intravenous bisphosphonates with creatinine clearances below 30ml/min since they can cause various types of kidney damage.82,83 However, in a post hoc analysis of clinical trials with oral alendronate and risedronate including patients with decreased GFR (CKD stages 3–4) for a maximum of 3 years, there were no differences in the occurrence of renal adverse effects.64,65 Since bisphosphonates can potentially accumulate in the bone of patients with advanced CKD84,85 (Table 4), the KDIGO 2009 suggested to perform a bone biopsy prior to therapy with bisphosphonates before its use in CKD 4–5 (Table 1).8 As discussed in the KDIGO 2017, this claim was based on a cross-sectional study of 13 patients with CKD stages 2–4 who underwent a bone biopsy after a variable period of treatment with bisphosphonates (4–60 months), all of them being diagnosed of ABD.84 In addition, Coco et al.86 showed that 6/6 patients with functioning renal transplant treated with intravenous pamidronate (with a frequency of administration much higher than usual) had ABD vs 3/8 in the controls, although the pretransplant bone status was not known. Only 14/72 patients were biopsied, so it is difficult to extrapolate these data. Currently, the prevalence of ABD is increasing,52,84,87 and it could even precede the high-turnover disease during the evolution of CKD.88,89 Aging, diabetes, calcium overload, malnutrition, inflammation and relative hypoparathyroidism (low or relatively low PTH levels) are risk factors for ABD that should be taken into consideration before the indication of bisphosphonates.52,84,90 In fact, all of them have warnings or contraindications regarding their use in patients with CKD with creatinine clearances <30–35ml/min).

Since the publication KDIGO 2009 guidelines.8 several studies have shown that BMD does predict the risk of fracture in patients with CKD.91–94 Other post hoc analysis of large randomized trials for treatment of postmenopausal OP with alendronate and risedronate have described that, in CKD patients (most of them women in stages 3–4), these drugs improve BMD and reduce the risk of fractures64,65 (Table 6). Therefore, at least in the absence of significant abnormalities in mineral bone metabolism, the use of bisphosphonates and other drugs approved for the treatment of OP could be appropriate in patients with “normal” creatinine levels and a decreased GFR.2,10,11,64,65 Thus, the update of the KDIGO guidelines in 201710 (Table 1) suggests that the need for a previous bone biopsy before deciding treatment is not mandatory in patients with advanced CKD. It is considered that there is not convincing evidence that bisphosphonates cause ABD, and that the accumulation of data after the use of antiresorptive drugs would potentially allow us their use without a bone biopsy is necessary.10 Therefore, the “difficulty” to obtain histomorphometric diagnosis of the type bone disease does not justify therapeutic nihilism in patients with a high risk of fracture.10 However, beyond the guidelines, we must also recognize that other analyses highlight the fact that antiresorptive treatment does not consistently demonstrate additional beneficial effects beside an improvement in BMD, and there is no absolute evidence on a reduction of fracture risk or vascular calcification67 (Table 6). This last issue is important given the important relationship between bone and vessels and its potential additional beneficial effect.95–97 Experimental work and small clinical studies in dialysis patients, suggest a beneficial effect of some bisphosphonates on decreasing the progression of vascular calcifications and in the treatment of calciphylaxis.98–100

Denosumab (DMab) is a monoclonal antibody that acts by blocking the ligand receptor of nuclear factor kappa-B (RANKL) and as a consequence inhibits osteoclastogenesis101,102 (Table 4). DMab is useful in the prevention of vertebral, non-vertebral and hip fractures in the general population.101,103,104 The metabolism or excretion of DMab does not depend on renal function and it does not appear to alter renal function, so there are neither dosage adjustments needed nor use restrictions in patients with CKD and/or decreased GFR.14,105 DMab has rarely been associated with atypical fractures. The effect of the drug ends upon cessation of its administration.14

In the post hoc analysis of osteoporotic women with CKD stage 3 (n=2817) and 4 (n=73) treated with DMab, it has been shown that after 36 months, as compared with placebo, there was an increase in BMD and reduction of the risk of fracture, independently of renal function and without an effect on serum creatinine and no increase in adverse effects105 (Table 4). However, some studies have shown that DMab may induce hypocalcemia, especially in patients with CKD and in hemodialysis patients with underlying high bone turnover.106–109 This potential effect is important and occasionally severe,110,111 especially during the first and second weeks of treatment, with special attention being paid to patients at risk or who are receiving concomitantly cinacalcet112 (also possible with etelcalcetide). If DMab is considered indicated by the presence of a high risk of fracture, the patient should be informed of the symptoms of hypocalcemia, monitor plasma calcium more frequently and perform a more intense, temporary replacement of calcium and vitamin D (native and/or calcitriol).106,107

Rebound hypercalcemia have also been described after long periods of treatment with DMab; this is attributed to an increase in osteoclastic activity after its suppression, but also to the previous administration of calcium-vitamin-D.108 The administration of DMab does not seem to affect the progression of aortic calcification and cardiovascular events, as compared with placebo, with occasional exceptions.113,114 In hemodialysis patients treated with DMab a relatively safe treatment of hypocalcemia is to modify calcium in the dialysis bath (usual 3mEq/l [1.5mmol/l],110 since each session would produce a positive calcium balance. Despite the concomitant administration of vitamin D, DMab may produce a reversible increase in intact PTH with values greater than 1000pg/ml.110 In any case, after the interruption of the treatment there may be a rapid bone loss so a sequential therapy with another drug should be evaluated to sustain the benefit obtained in the BMD.14,18

Several studies have shown beneficial effects of DMab in dialysis patients. In a prospective pilot study of 12 patients, all of them with PTH >1000pg/ml, T-score <−1.0 and bone pain, not candidates for surgery, Chen et al.115 described the effects of DMab (off-label), calcitriol, phosphate binders and a calcium bath adjusted according to the biochemical data. After 6 months of treatment the BMD increased in the femoral neck and in the lumbar spine and there was a reduction in pain. In the first month, the majority of patients experienced an increase in intact PTH that rapidly decreased at the end of the study after increasing the dose of calcitriol (1702±182 to 519±127pg/ml).115 A longer retrospective experience110 describes 12 hemodialysis patients with osteoporosis who received DMab at usual doses. After 24 months of follow-up, they also observed improvement in bone metabolism parameters and T-score, although the latest was only measured by phalangeal ultrasound.

In summary, according to the current knowledge, antiresorptive agents such as bisphosphonates or DMab could be used in patients with different stages of CKD (especially CKD 1–3) at least if they have fragility fractures and/or high risk of fracture2,116 This is similar to what is being done in the general population. Bone biopsy may not be necessary if after the individualized evaluation the presence of an ABD is considered unlikely. Bisphosphonates may be indicated in these patients, especially if they have high bone turnover parameters62 (significant elevation of serum levels of PTH and alkaline phosphatases). It may be convenient to space the doses and/or limit the duration of treatment.62,117 The use of DMab could be extended to more advanced stages of CKD (stages 4 or even 5), with special control of the serum calcium concentration. Clinical trials (i.e., with DMab or alendronate in advanced CKD and dialysis [NCT01464931, NCT02792413, NCT02440581]) could provide relevant information in the upcoming years. Until then, a valid option is to adopt diagnostic-therapeutic algorithms based on fracture risk and not to wait for absolute evidence.4,116,118

Teriparatide (Table 4) is approved as an osteoforming (anabolic) therapy for postmenopausal and steroid induced OP, and in men with high risk of fracture (e.g., two or more vertebral fractures).14,19,119 Teriparatide induces an intermittent exposure to PTH which increases bone formation particularly in trabecular bone. An important limitation is the high cost (the most expensive alternative) and after removing teriparatide there is gradual decrease in bone mass which can be mitigated by continuing the therapy with a bisphosphonate.13,120 Another limitation is that it cannot be administered for periods longer than 24 months (Table 4).

Compared to placebo, teriparatide produces a significant increase in BMD of the lumbar spine and femoral neck and there is no evidence that this increase in BMD is modified by CKD121 (Table 6). Although its use is restricted in moderate CKD and contraindicated in severe CKD, in a recent post hoc analysis in Japanese women with OP, CKD stages 4–5 and high risk of fracture (82% had a previous fracture), it was observed that subcutaneous teriparatide 20μg/day during 24 months appeared to be effective (increasing BMD and biomarkers).122 It should be emphasized that in some cases with ABD the use of teriparatide resulted in an increase in bone turnover in patients with ABD confirmed by bone biopsy.52,123,124 So far there is no experience with abaloparatide or the human recombinant PTH Natpara® in patients with CKD.125 In Japanese patients on hemodialysis with hypoparathyroidism and low BMD the use of 56.5μg of teriparatide once a week resulted in an increase in lumbar BMD, although 10/22 patients had to abandon treatment due to side effects.126

It should be taken into consideration that in patients on dialysis with decreased PTH (<2 times the upper limit of normality, and especially if they have a history of fracture) there is a possibility of decreasing calcium in the dialysis fluid to stimulate endogenous PTH production and improve bone remodeling.4,52 Finally, it should be noted that a clinical trial an ongoing clinical trial in patients with OP on dialysis assigned to the arm of low bone turnover a teriparatide–cinacalcet combination (NCT02440581), perhaps to stimulate the endogenous production of PTH by the hypocalcemia induced by cinacalcet.

Romosozumab (RMab) (Table 4) is a humanized monoclonal antibody against sclerostin, an inhibitor of the Wnt pathway a key signaling system responsible for bone formation. The sclerostin produced by the osteocytes binds to the proteins LRP (4-5-6), the ligand receptors of the Wnt pathway. RMab neutralizes the effect of sclerostin resulting in enhanced Wnt activity which stimulates bone formation and inhibits bone resorption (unlike classical osteoanabolic agents).127,128 Clinical studies have shown that RMab produces a significant improvement of spine and hip BMD.127,129,130 A recent study in postmenopausal women with a very high risk of fracture showed that sequential treatment with monthly RMab during 12 months followed by weekly alendronate resulted in a lower risk of fracture than weekly treatment with alendronate alone.130 However, a greater number of some cardiovascular events were attributed to the RMab group during the first year of treatment (2.5% vs 1.9%, respectively), although there were no differences between RMab and placebo.128–130 For this reason, the FDA has held its commercialization pending new studies and reanalyses. The role of this drug in patients with CKD is not sufficiently known to advise its use.131