2. Bone Disease in CKD
The abnormalities of bone in CKD are defined based on bone histology with histomorphometric analysis. Classically, renal osteodystrophy has been classified in different subtypes encompassing a wide spectrum from high turnover to low turnover. At one end of the spectrum, high PTH levels has been considered a surrogate for high-turnover bone disease, known as hyperparathyroid bone disease or osteitis fibrosa, and characterized by elevated bone turnover, increased number and activity of osteoclasts and osteoblasts, variable alterations in osteoid deposition, usually with a woven pattern, and variable amounts of peritrabecular fibrosis. At the other end of the spectrum, the distinctive pattern is an adynamic bone disease in which typically, bone turnover is decreased with normal mineralization, a paucity of osteoid, bone cells, and a marked decrease in active remodeling sites. Osteomalacia, another lesion with low turnover, is characterized by increased osteoid seam width, increase in the trabecular surface covered with osteoid, and decrease in mineralization as assessed by tetracycline labeling. Mixed uremic osteodystrophy is a complex disorder in which elevated bone turnover coexists with features of osteomalacia
[15,16,17][11][12][13]. The frequency of the two last lesions has decreased consistently in recent decades
[18][14]. More recently, it has been shown that osteoporosis is a frequent feature in patients with CKD-MBD
[19,20,21][15][16][17] that may complicate their outcome. This disorder, characterized by a decreased bone mass, strength and quality predisposing individuals to bone fracture, is frequent in the CKD population and is likely associated with the elevated risk of fracture not only in the aging groups but also in younger strata
[22][18]. The definition of osteoporosis has been based mainly on bone densitometry. The World Health Organization (WHO) defines osteoporosis as a disease characterized by low bone mineral density and microarchitectural deterioration leading to low bone strength and increase in fracture risk
[23][19]. In the normal population, osteoporosis is defined as a DEXA T score ≤2.5 DS below the normal range for the peak obtained in young persons. In the non-CKD population, both cortical and cancellous bone are substantially reduced
[24][20]. However, in CKD patients, histologic changes are more difficult to interpret as osteoporosis has been seen coexisting with other types of renal osteodystrophy
[25][21] with no difference in prevalence, among them.
As a mean to standardize the definition of bone disease in CKD, KDIGO proposed a new classification based on bone turnover (T), mineralization (M) and volume (V) or TMV, which highlights the most significant bone alterations relevant for clinical evaluation and therapeutic implications
[26,27][22][23]. However, the classical definition of renal osteodystrophy described above is still used coexisting with the new classification based on TMV.
Most of the early studies describing the CKD associated bone disease were performed in patients with advanced CKD or ESKD. In many of them, high-turnover bone disease was described as the predominant form of renal osteodystrophy
[17,28,29,30][13][24][25][26]. More recently, low-turnover bone disease has been increasingly reported in patients with ESKD
[31,32][27][28]. In some studies, an important number of patients showed normal bone turnover
[17,33,34][13][29][30]. In two large series on bone biopsy in patient on dialysis, comprising 630 patients from USA and Europe
[34][30] and 492 patients from different countries (Brazil, Portugal, Turkey, and Venezuela)
[35][31], low-turnover bone disease was observed in 58% and 52% of the patients, respectively. Malluche et al.
[34][30] examined the possible role of race in the type of bone lesions occurring in CKD. Biopsies were analyzed using the parameters of TMV as recommended by KDIGO
[2]. As a whole, low turnover was observed in 58% of the cases. This type of lesion was more prevalent in white patients, while in black patients, high turnover was observed in 68% of cases. All patients with high bone turnover were younger. Mineralization defects were seen in only 3% of the cases. Regarding bone volume, biopsies in white patients revealed a similar proportion of normal, high or low bone volume, whereas black patients showed a higher proportion of high volume. No differences were observed regarding diabetes, gender, and treatment with Vitamin D.
In 1976, Malluche et al.
[45][32] examined bone histology in 50 patients with different stages of kidney disease. It was observed histological evidence of PTH excess, particularly osteoclastic surface resorption, empty osteoclastic lacunae, and woven osteoid in more than 50% of patients with a GFR of 40 mL/1.73 mL/min, whereas endosteal fibrosis was seen when GFR fell below 30 mL/min suggesting that hyperparathyroid bone disease was present since early stages and progressed with advanced CKD. In contrast, in more recent studies, low-turnover bone disease has been increasingly reported predominantly in patients with CKD stages 2 to 4
[46,47,48][33][34][35], with a prevalence that in some cases reached 80 to 100% of the patients. In the study by El-Husseini
[48][35], in patients with a mean eGFR of 44 ± 16 mL/min/1.73 m
2, low turnover was observed in 84% of the patients. Of interest, most of them had vascular calcifications which correlated positively with levels of phosphorus, FGF-23 and activin, and negatively with bone turnover as has been also reported previously, whereas others have found less prevalence of vascular calcification in lower CKD stages. Other observation of relatively recent studies is the finding of osteoporosis. Nevertheless, the results are not uniform as the prevalence of low-turnover bone varies among studies likely reflecting the different population examined, degree of kidney function, age, ethnicity, geographic distribution, medications including corticosteroids, immunosuppressants, Vitamin D, and type of renal disease leading to CKD, which may play a role in the type and degree of turnover alteration. Indeed, in preliminary studies (
Abstract ASN 2011) performed in the laboratory of two of the authors in a group of 46 patients with CKD stages 3 to 5, and patients with ESKD on hemodialysis, low bone turnover was observed in 7 (15.2%) of the patients with CKD stages 3 to 5 not on dialysis, whereas high bone turnover was seen in 20 patients (43.5%). Bone alterations consistent with osteoporosis
[24,25,34][20][21][30] was found in 12 patients (26.1%). Of interest, this finding was more frequently observed in diabetics (33.5%) compared with non-diabetics (27.3%), female gender, and older age. Normal bone histology was seen in 15.2% of the patients. In contrast, in ESKD, 40 (80%) patients had high bone turnover and only 10 patients (20%) had low bone turnover. There is no clear explanation for the different types of bone turnover predominating at CKD stages as well as the apparent increase in the incidence of low bone turnover lesions along the years reported by different authors. It has been considered that medications, particularly those associated with a possible effect in bone metabolism could explain the differences. Vitamin D analogs and cinacalcet have been shown to affect bone metabolism and particularly bone turnover
[49,50,51][36][37][38]. Indeed, in prospective studies, it has been shown that cinacalcet diminishes bone turnover after one year of treatment
[51,52][38][39]. Thus, it has been argued that oversuppression of PTH by these drugs may result in adynamic bone disease.
3. Pathogenesis of Bone Disease in CKD-MBD
Extracellular calcium concentrations are tightly controlled within a narrow physiological range optimal for proper cellular functions. Calcium absorption in the intestine is regulated by calcitriol. Calcium acts directly on the parathyroid cell through its specific receptor, the calcium sensing receptor (CaSR), that detects subtle decreases in extracellular calcium, leading to an immediate release of PTH. Conversely, an increase in extracellular calcium suppresses PTH secretion. Numerous studies assign the pathogenesis of the bone alterations of CKD-MBD to the early changes in the metabolism of phosphorus, calcium, FGF-23, and calcitriol that occur as kidney function declines
[62,63,64,65,66][40][41][42][43][44]. These alterations manifest as an elevation in the levels of FGF-23 and PTH to increase renal phosphate excretion. Studies suggest that FGF-23 increases early in CKD, even prior to a measurable elevation of PTH
[67][45]. It rises phosphate excretion through binding to its klotho co-receptor activating FGFR-1 and FGF-3 receptors leading to a decrease in NaPi2a and NaPi2c cotransporters expression which ultimately will increase phosphate excretion. PTH directly decreases phosphate reabsorption through similar mechanisms. FGF-23 and PTH can maintain phosphate balance until GFR approaches stage 4. Thereafter, neither PTH or FGF-23 are capable of completely maintaining phosphate balance and hyperphosphatemia ensues. The 1,25-(OH)
2 Vitamin D (calcitriol) synthesis in the kidney is reduced due to the inhibitory effects of elevated FGF-23 and phosphate on 1-alfa- Hydroxylase
[32][28]. Calcitriol deficiency decreases intestinal calcium absorption and diminishes tissue levels of VDR, resulting in resistance to calcitriol-mediated regulation of PTH secretion. The concurrent decrease in the expression of CaSr in the parathyroid cells stimulates PTH secretion
[68,69][46][47]. In addition, elevated serum phosphorus increases PTH secretion by mechanisms that include a direct action on the CaSr
[70][48]. All these factors in concert lead to the development of secondary hyperparathyroidism. Elevated PTH activate osteoblasts and osteoclasts via the receptor activator of nuclear factor-kappaB (RANK-L) and osteoprotegerin signaling pathway lading to an increase in bone turnover resulting in a bone structure with lower strength and increased fragility, which contributes to an elevation of bone fracture risk and alterations in vascular metabolism resulting in vascular and valvular calcifications
[64,65,71,72,73][42][43][49][50][51]. Of interest, heparin, an agent to which ESRD patients on hemodialysis are frequently exposed, has been shown to increase osteoprotegerin intra-and postdialytic levels, thus suggesting that this could be an additional factor in the pathogenesis of bone and vascular disease in hemodialysis patients
[74][52]. Additional factors and mechanisms, including inhibition of the canonical Wnt/B catenin signaling pathway, and accumulation of uremic toxins such as indoxyl sulfate have also been proposed as factors that may lead to disruption of the normal regulation of bone turnover leading to renal osteodystrophy and vascular disease
[59,75,76,77][53][54][55][56].
The direct role of FGF-23 on bone metabolism is not clear. Studies have shown that in CKD, osteocytes exhibit an increased synthesis of FGF-23. A relationship between FGF-23 and bone abnormalities occurs through its effects on phosphate excretion and suppression of calcitriol synthesis
[65,78][43][57]. However, FGF-23 levels have been shown to be elevated even before phosphate levels are increased
[67][45]. In addition, it has been suggested an association between FGF-23 with alterations in bone mineralization
[79][58], and FGF-23 and alfa-klotho, the coreceptor for FGF-23, have been shown to stimulate osteoblastic-like cell proliferation and inhibit mineralization
[47,78][34][57]. Thus, a study in adult dialysis patients found that patients with high bone turnover had higher serum levels of FGF-23 compared with patients with low bone turnover. Likewise, patients with high FGF-23 had normal mineralization, whereas delayed mineralization correlated negatively with FGF-3 levels. Using regression analysis, FGF-23 was the only independent predictor for mineralization lag time
[60][59]. Similarly, in pediatric patients with high bone turnover renal osteodystrophy, it has been shown an association between high serum levels of FGF-23 and improved mineralization, although a correlation between FGF-23 and bone formation rates was not observed
[80][60]. Furthermore, studies examining the expression of proteins in bone tissue of patients with CKD stages 2 to stage 5 on dialysis and healthy individuals have shown that as serum calcium declines, serum alkaline phosphatase, FGF-23, PTH, and osteoprotegerin increase with progression of CKD
[47][34]. These alterations occurred while there was an increase in bone resorption, decreased bone formation and impairment in bone mineralization. Of interest, sclerostin and PTH-receptor-1 expression in the bone was higher in early stages of CKD whereas FGF-23 was elevated in late stages. FGF-23 expression was observed mainly in early osteocytes, whereas sclerostin, which is considered a marker of mature osteocytes, was expressed in cells deeply embedded in the mineralization matrix. These proteins did not co-localize in the same cells. In other studies, high bone turnover was associated with high FGF-23, whereas low bone turnover was observed with lower FGF-23
[81][61]. Thus, FGF-23 seems to play a direct role in bone metabolism and may be a predictor of bone mineralization in patient with CKD on dialysis and a marker to predict alterations in bone metabolism.
Studies have demonstrated a relationship between PTH and FGF-23 in the bone. In osteoblast-like UMR106 cells, PTH increases FGF-23 mRNA levels and inhibits sclerostin mRNA messenger, which is an inhibitor of the Wnt/beta-catenin pathway. These studies directly associate the effects of PTH and FGF-23 likely via stimulation of the Wnt pathway. Sclerostin levels vary with renal function. Thus, the expression of sclerostin in jck mouse, a model of progressive kidney disease occurs at early stages of CKD, even before PTH and FGF-23 increase
[57][62], suggesting that sclerostin may play an early role in the development of renal osteodystrophy. Conversely, sclerostin has been associated with adynamic bone disease and vascular calcification
[82][63]. Increased sclerostin has been observed in early CKD but the mechanisms are not clear. It has been suggested that it may relate to partial calcitonin exposure, disturbed phosphate metabolism and extraskeletal sources
[55,83][64][65].
Serum levels of sclerostin and PTH correlate negatively in patients with CKD stage 5D
[57][62]. In addition, in unadjusted and adjusted analysis, sclerostin correlated negatively with bone turnover, osteoblast number and function, and was superior to PTH for the positive prediction of high bone turnover and osteoblast number. On the other hand, PTH was superior to sclerostin for negative prediction of low bone turnover
[53][66]. These findings agree with studies in mice showing that PTH directly inhibits sclerostin transcription in vivo
[84[67][68],
85], suggesting that sclerostin may be useful as a marker of high bone turnover. This negative correlation is in agreement with the negative regulatory function of sclerostin in the intracellular transduction of the PTH signal described in vitro and in vivo
[84][67]. Moreover, studies in patients with renal osteodystrophy subjected to bone biopsy have shown an inverse correlation of sclerostin and PTH levels, as well as a negative association with bone turnover
[47,57][34][62]. Receiver operator curves analysis showed that PTH and FGF-23 were able to predict high bone turnover, whereas sclerostin was a good predictor of low bone turnover. Sclerostin expression in bone was higher at early stages of CKD and has shown association with lower bone formation rate and greater mineralization defect. These findings favor the notion that sclerostin may play a role in the development of adynamic bone disease in CKD, and hence on fractures
[86][69].
Although elevation of PTH is an early finding in CKD, low bone turnover has been increasingly described in patients with CKD
[31,32][27][28]. The mechanisms behind these apparently incongruent findings are complex and include maladaptation to the pathophysiologic mechanisms described above, inappropriate PTH signaling and hyporesponsiveness of the PTH receptor, repressed Wnt/B-catenin signaling
[55][64], and elevation of sclerostin in early CKD
[47][34]. Other factors may include the use of medication to prevent or control secondary hyperparathyroidism, oxidative modification of PTH
[87][70], nutritional determinants, age, and underlying diseases. Another factor that has been associated with the early events in the pathogenesis of renal osteodystrophy is indoxyl sulfate. This compound increases in early CKD, even before FGF-23, and has been associated with resistance to PTH
[48][35].
In summary, the mechanisms responsible for the development of renal osteodystrophy are complex and multiple. Based on the sequence of pathophysiologic events that occur during the development of CKD-MBD, high bone turnover is expected to be the most frequent histologic alteration observed in early CKD as PTH starts to increase when GFR drops below 60 mL/min/1.73 m
2 [10,67,88,89][10][45][71][72]. Although some studies have confirmed this notion
[90][73], others have shown a high prevalence of adynamic bone disease in the early stages of CKD and even a change of pattern as CKD progresses. Thus, it seems that a single mechanism does not fit all the phenotypes described at the different stages of CKD and that other factors associated with bone disease may also be determinant to the increased bone fragility and fracture in patients with CKD. The interpretation of these findings together is difficult as there is a vast variability in the populations studied, including age, ethnicity, geographic and social background, among others. In evaluating the causes of bone fragility and fracture in CKD, osteoporosis has been increasingly described in patient with CKD and considered to play a major role in bone fracture, particularly given the fact that a high proportion of patients with CKD bone disease have also the usual risk factors for osteoporosis
[91][74].
4. Bone Fracture in CKD
Bone fracture is a frequent complication of CKD.
Table 1 summarizes a number of studies examining the incidence of fractures in patients with CKD. Patients with CKD stages 3a to 5D have higher rate of bone fractures compared with the general population
[5,6,7,8,9,10,23,92][5][6][7][8][9][10][19][75]. Most studies show that the incidence of fracture increases as GFR decreases, as well as an association between fracture and age in patients with CKD which is superior to that in patients of similar age in the general population. Likewise, the risk of mortality is superior in CKD patients with fracture compared with the general dialysis population. The causes are diverse. Thus, in addition to the pathophysiologic mechanisms leading to renal osteodystrophy, many other factors that include nutrition, medications, concurrent diseases such as diabetes, cardiovascular disease, sarcopenia, increased propensity to fall, among others, may per se increase the risk of fracture in patients with CKD
[93][76]. The estimated prevalence of CKD varies greatly between countries but may reach between 9% and 13.4% of the global population
[94,95][77][78]. Therefore, the expected prevalence of bone fracture associated with CKD is also high
[96][79]. In most patients, CKD is thought to progress slowly and the bone and the mineral derangement that start in early stages progress relatively silent until it reaches advanced stages in which fracture incidence increases and is more evident. As shown in
Table 1, there is an increase in fractures, particularly of the hip
[6,7,8,97][6][7][8][80] but is also seen in other bones. Although this is frequently associated with older age, evidence shows that in general, the incidence of fractures in patients with CKD is elevated. Indeed, as shown in
Table 1, several studies have revealed that patients with CKD stages 3 to 5, and those undergoing dialysis have an increased incidence of fractures compared with age-matched subjects without CKD
[23][19]. In addition, the study by Klawansky et al.
[98][81], based on the NHANES III population, a strong trend was noted in which lower bone mineral density (BMD) as determined by dual energy X-ray absorptiometry (DEXA) was strongly associated with reduced creatinine clearance as estimate by the Cockcroft-Gault (CCr) equation. The percentage of women with CCr < 35 mL/min, increased from 0.3% for women with BMD in the normal range to 4% for women with osteopenia to 24% for those with osteoporosis. Similar trend was observed in men, albeit to a lesser extent. Likewise, in another population-based study the prevalence of osteoporosis was 31.8% in patients with CKD stages 3–5
[9]. Thus, the combination of CKD, osteoporosis, and age, results in an elevation of fracture risk.
Table 1. Incidence of fractures in CKD.
| Reference |
Cohort Number of pts (N) |
CKD Stage/CCl ml/min or eGFR ml/min/1.73 m | 2 |
Fracture Type and Number |
Fracture Incidence 1000 Person-Years |
Comments |
Alem [5] 2000 |
USRDS data base N: 326,464 person-year |
ESKD |
Hip 6542 |
Men 7.45 Women 13.63 |
Relative risk highest in younger people. Added incidence of fracture increased with age and was greater for women than for men. |
Coco [6] 2000 |
N: 4039 person-year, N: 1272 pts treated |
ESKD HD |
Hip 56 |
Men 11.7 Women 24.1 |
The one-year mortality rate from hip fracture was ~2.5 times higher in dialysis pts compared with general population. |
Jadoul [99] 2006 | Jadoul [82] 2006 |
DOPPS: HD pts. 12,782 |
ESKD HD |
Hip 174 Any 498 |
8.9 for hip 25.6 for any new fracture |
Older age, female sex, prior kidney transplant and low serum albumin were predictive of new fracture. PTH > 900 was associated with risk of new fracture |
| Danese [100] 2006 | Danese [83] 2006 |
DMMS data base N: 9007 pts |
NA |
Hip and vertebral |
580/1000 vs. 217/1000 in the general dialysis population |
Age and sex-adjusted mortality rate after fracture 2.7 times greater than the dialysis population. Pts with lower PTH were more likely to sustain a hip fracture than those with higher PTH. |
Dukas [101] 2005 | Dukas [84] 2005 |
Cross sectional N: Women 5313 N: Men 3238 |
CrCl ml/min 60.9% < 65 39.1% ≥ 65 |
Not reported |
Not reported |
CCl < 65 increased risk of experiencing falls and risk for hip fracture (OR 1.57, 95%CI 1.18–2.09, | p | = 0.002), and for vertebral fracture |
Lin [8] 2014 |
Taiwan NHIRD N: 51,473 incident dialysis patients |
ESKD Dialysis |
Hip 1903 |
8.92 Men 7.54 Female 10.12 |
HD pts had a 31% higher incidence of hip fracture than PD patients (HR 1.31, 95% CI: 1.01–1.70). Patients ≥65 years old had more than 13 times the risk of a hip fracture than those 18–44 years old (HR: 13.65; 95% CI: 10.12–18.40) |
Tentori [90] 2014 | Tentori [73] 2014 |
International DOPPS N: 34,579 |
ESKD/HD |
NA |
Japan 1 Belgium 45 |
Fracture pts had 3.7-fold higher rates of death compared to DOPPS population. In most countries, mortality rates exceeded 500 per 1000 patient-year |
Naylor [97] 2014 | Naylor [80] 2014 |
Data base from Ontario, Canada N: 679,114 |
eGFR ml/min/1.73 m | 2 | : ≥60; 45–59; 30–44; 15–29; <15 |
Hip Forearm Pelvis Humerus |
Not available in women ≥ 60 |
Fracture rate in women ≥ 65 years old at different eGFR (ml/min/1.73 m | 2 | ): >60: 4.3% 45–59: 43%: 5.8% 30–44: 47.9%: 6.5% 15–29: 54.4%: 7.8% <15.54.2%: 9.6 |
Naylor [102] 2015 | Naylor [85] 2015 |
2107 320 individuals with eGFR < 60 mL/min/1.73m | 2 | 1787 individuals with eGFR ≥ 60 mL/min/m | 2 |
eGFR ml/min/1.73 m | 2 | : ≥60; 45–59; 30–44 15–29; <15 |
64 (3%) over 4.8 years |
Not available |
The 5-year observed major osteoporotic fracture risk was 5.3% in individuals with eGFR < 60 mL/min/1.73m | 2 | was 5.3%, comparable to the FRAX predicted fracture risk. No difference in the AUC values for FRAX in individuals with eGFR < 60 mL/min/1.73 m | 2 | vs. those with eGFR ≥ 60 mL/min/m | 2 |
Yamamoto [103] 2015 | Yamamoto [86] 2015 |
3276 |
|
|
* 1.48 ** 2.33 |
Mortality was lower in pts * using ACEI/ARB than those ** not using ACEI/ARB 13.6% vs. 16.8% |
Hung [7] 2017 |
Taiwan’s NHIRD Total of 61,346 first fragility hip fracture nationwide. 997 dialysis hip fracture patients were matched to 4985 non-dialysis hip fracture subjects |
ESKD Dialysis |
Hip 997 |
Not available |
Higher proportion of femoral neck fractures in the dialysis group compared to the non-dialysis group (51% and 42%, respectively; | p | < 0.001) The mortality rate was significantly higher for patients in the dialysis group, with a mortality rate of 91% compared to 71% for those in the non-dialysis group ( | p | < 0.001). |
Desbiens [9] 2020 |
CARTaGENE data base (CAG) N: 679,114 19,391pts with CKD included |
Non-CKD: 9521 CKD: 2: 9114 CKD 3: 756 |
829 Various type |
Non-CKD: 6.9 CKD 2: 7.6 CKD 3: 11.3 |
Compared with the median eGFR of 90 mL/min/1.73 m | 2 | , eGFRs of 60 mL/min/1.73 m | 2 | were associated with increased fracture incidence [adjusted hazard ratio (HR) ¼ 1.25 (95% confidence interval 1.05–1.49) for 60 mL/min/1.73 m | 2 | ; 1.65 (1.14–2.37) for 45 mL/min/1.73 m | 2 | ]. The effect of decreased eGFR on fracture incidence was higher in younger individuals [HR 2.45 (1.28–4.67) at 45 years; 1.11 (0.73–1.67) at 65 years and in men. |
Osteoporosis is frequent in the CKD population and is likely associated with the elevated risk of fracture not only in the aging groups but also in younger strata
[22][18]. The prevalence of osteoporosis increases with age, a condition that parallels CKD which is also more prevalent in the aging population. Thus, it has been argued that the higher rates of fracture in CKD patients may be a consequence of an increase in the prevalence of age-related osteoporosis. In the younger population, the data are more conflicting. In a recent population based prospective study performed in 19,391 individuals 40 to 69 years old from Canada, with CKD stages 2 and 3 followed for 70 months, it was found that, compared with the median eGFR of 90 mL/min/1.73 m
2, those with eGFR of ≤60 had an increased risk of fracture in unadjusted and adjusted models
[9]. This effect was more evident in younger individual. Hence, there is clear evidence of an increase in fracture risk at all stages of CKD. However, the studies are difficult to compare because most of them followed different methodology and the population studied are heterogeneous.
The complex alterations that express as low bone mineral density bone mineral density BMD in DEXA studies constitute important components of the bone abnormalities observed in CKD patients and may manifest as osteoporosis and an increased risk for fracture, particularly in those with advanced CKD
[104,105,106][87][88][89]. However, in reality, the mechanical properties of the bone express a combination of factors that include aspects related to the specific changes of bone quality associated with renal osteodystrophy
[21,107,108][17][90][91] and those due to osteoporosis itself. Thus, low turnover is associated with microstructural abnormalities, whereas bone with high turnover is associated with alteration of material and mechanical property
[104][87]. Some have suggested the combination of the bone alterations described in renal osteodystrophy and the addition of bone fragility osteoporosis be described as “uremic osteoporosis”
[65,109][43][92].
As previously discussed, bone quality in patients with CKD is best characterized by analysis of bone remodeling, mineralizing and volume properties by means of a bone biopsy with histomorphometry. However, this is an invasive method that requires technical and financial resources that are not available for clinical purposes in most centers. In contrast, DEXA, a method widely available, allows the evaluation of bone quantity. Several studies analyzing DEXA in patients with different stages of CKD have provided abundant information indicating an association between CKD-MBD and osteoporosis. In the KDIGO CKD-MBD guideline of 2009
[1], routinely evaluation of BMD testing was not suggested to be performed in patients with CKD, because based on the evidence available at that time, BMD would not predict fracture risk as it does in the general population and would not predict the type of renal osteodystrophy. However, in the updated KDIGO CKD-MBD guideline of 2017
[110][93], after new prospective studies documented that lower DXA BMD predicts incident of fractures in patients with CKD G3a–G5D
[102[85][94][95][96],
111,112,113], it is suggested that in patients with CKD stages G3a–G5D with evidence of CKD-MBD and/or risk factors for osteoporosis, BMD testing to assess fracture risk be performed if results will impact treatment decisions. In one of those studies, BMD by DEXA measured yearly in 485 hemodialysis patients was useful to predict any type of fracture for females with low PTH or to discriminate spine fracture for any patient
[111][94]. Of note, a significant greater risk for fractures was observed with PTH levels either below 150 or above 300 pg-ml. Likewise, bone alkaline phosphatase was a useful surrogate marker for any type of fracture
[111,112][94][95]. Another study by Naylor et al.
[113][96] has shown that the Fracture Risk Assessment Tool with or without BMD measurement is also useful to predict major osteoporotic fracture. Over a period of 5 years, the risk for fracture was not different in individuals with eGFR <60 mL/min/1.73 m
2 compared with those with an eGFR > 60 mL/min/1.73 m
2. The trabecular bone score (TBS) is a DEXA-derived algorithm for the evaluation of bone microarchitecture whose utility in patient with osteoporosis has been largely demonstrated
[48,114,115][35][97][98]. Studies have shown that TBS may also be useful to measure bone quality in patients with CKD-MBD
[114,116][97][99]. Indeed, in CKD patient, TBS was significantly associated with the histologic parameters of trabecular bone volume and trabecular spacing but not with dynamic parameters, suggesting that TBS reflects trabecular microarchitecture and cortical width
[117][100]. However, in another study, there was no significant relationship between TBS and bone histomorphometric parameters. TBS has also been shown to correlate inversely with coronary calcification and aortic calcification. Thus, TBS is an important parameter to consider in the evaluation of CKD bone and cardiovascular disease. The fact that TBS may be obtained with DEXA analysis opens an opportunity for further evaluation of bone disease in CKD patients using non-invasive methods. However, further studies are needed to demonstrate whether TBS may predict clinical fractures in patients with CKD-MBD.
Given the limitations of BMD to ascertain the type of bone disease in patients with CKD and the difficulties in performing a bone biopsy as a method to evaluate bone abnormalities in large number of patients, studies have analyzed the possible benefit of other methods than can provide more details of the bone structure. High resolution peripheral quantitative computed tomography (HR-pQCT) can detect microarchitectural changes in patients with CKD
[44,118][101][102]. A recent study in CKD patients, evaluating bone biomarkers, bone histomorphometry, and HR-pQCT revealed lower BMD, mostly due to trabecular bone impairment compared to controls. It was found that radius BMD and microarchitecture were negatively associated with bone turnover in advanced CKD
[44][101]. HR-pQCT was able to discriminate low bone turnover from non-low bone turnover, whereas there was no difference in DXA BMD between different bone turnover classes. A similar trend has been shown on HR-pQCT and bone turnover as determined by biomarkers in women on dialysis
[119][103].
Although several serum biomarkers have been used to assess bone activity in the general population, their use in patients with advanced CKD and those on treatment with dialysis is limited as many of them are affected by renal function. However, recent studies have shown that some biomarkers, are not affected by the kidney function and thus, can be useful to discriminate the type of bone turnover alteration in patients with CKD. BSAP is produced by osteoblasts during bone formation. Thus, it is considered a marker of bone formation and is associated with fracture and all cause cardiovascular mortality
[120,121][104][105]. Studies have shown a positive predictive value of BSAP for low bone turnover
[35,36,58,122][31][106][107][108]. In the study by Sprague et al.
[35][31] evaluating the accuracy of bone turnover markers to discriminate bone turnover in 492 ESRD subjected to bone biopsy, PTH (iPTH) and bone specific alkaline phosphatase (BSAP) were able to discriminate low from non-low and high from non-high bone turnover, whereas P1NP, another marker of bone formation, that has been shown to be reliable in CKD in some studies, did not improve the diagnostic accuracy. Similar results, for iPTH and BSAP, have been shown by other studies. In contrast, it has also been found that BALP, intact P1NP, TRAP5B, and radius HR-pQCT, but not PTH can discriminate low bone turnover
[44][101]. Other biomarkers such as osteocalcin and FGF-23 albeit promising, require further studies. Therefore, differences still exist as to which biomarkers are useful in the diagnosis of bone turnover. In this regard, Evenepoel et al.
[122][108], in an analysis of several studies on the use of different bone biomarkers, concluded that there is not consistent evidence to replace bone histomorphometry for the diagnosis of bone turnover. It is possible that differences in biomarker assays and the populations evaluated may influence the different results across studies. Meanwhile, the KDIGO CKD-MBD guideline update of 2017 suggests that measurements of serum PTH or BSAP can be used to evaluate bone disease in CKD stages 3–5D because markedly high or low values predict underlying bone turnover stage 3 to 5
[109][92].