Natural History of Bone Disease following Kidney Transplantation

Significance Statement

The natural history of bone disease in contemporary kidney transplant recipients is not well characterized. Studies are generally limited by small samples sizes or a narrow focus on bone biopsy, imaging, or biochemistry. This prospective, observational study performed extensive bone phenotyping by histomorphometry, densitometry, and biomarkers before and after kidney transplantation in a sizable cohort. The main findings include decreasing bone turnover after kidney transplantation and, in parallel, decreasing levels of bone biomarkers. Delayed bone mineralization occurring in a subset of patients is associated with the duration and severity of hypophosphatemia. Resolution of hyperparathyroidism and high bone turnover are associated with gain in bone density. Thus, optimal control of mineral metabolism and regulation of skeletal remodeling could improve bone health in kidney transplant recipients.

Visual Abstract

Abstract

Background

Knowledge of the effect of kidney transplantation on bone is limited and fragmentary. The aim of this study was to characterize the evolution of bone disease in the first post-transplant year.

Methods

We performed a prospective, observational cohort study in patients referred for kidney transplantation under a steroid-sparing immunosuppressive protocol. Bone phenotyping was done before, or at the time of, kidney transplantation, and repeated at 12 months post-transplant. The phenotyping included bone histomorphometry, bone densitometry by dual-energy x-ray absorptiometry, and biochemical parameters of bone and mineral metabolism.

Results

Paired data were obtained for 97 patients (median age 55 years; 72% male; 21% of patients had diabetes). Bone turnover remained normal or improved in the majority of patients (65%). Bone histomorphometry revealed decreases in bone resorption (eroded perimeter, mean 4.6% pre- to 2.3% post-transplant; P<0.001) and disordered bone formation (fibrosis, 27% pre- versus 2% post-transplant; P<0.001). Whereas bone mineralization was normal in all but one patient pretransplant, delayed mineralization was seen in 15% of patients at 1 year post-transplant. Hypophosphatemia was associated with deterioration in histomorphometric parameters of bone mineralization. Changes in bone mineral density were highly variable, ranging from –18% to +17% per year. Cumulative steroid dose was related to bone loss at the hip, whereas resolution of hyperparathyroidism was related to bone gain at both spine and hip.

Conclusions

Changes in bone turnover, mineralization, and volume post-transplant are related both to steroid exposure and ongoing disturbances of mineral metabolism. Optimal control of mineral metabolism may be key to improving bone quality in kidney transplant recipients.

Clinical Trial registry name and registration number:

Evolution of Bone Histomorphometry and Vascular Calcification Before and After Renal Transplantation, NCT01886950

Fracture risk is increased after kidney transplantation, particularly in the early post-transplant period.^1–3 With the increasing age and longevity of kidney transplant recipients,⁴ improving long-term quality of life is imperative, which includes the prevention of osteoporosis and fractures.⁵

The implementation of steroid-minimization protocols has reduced bone loss and fracture rates post-transplant.^6–8 Consequently, the evolution of bone mineral density (BMD) in contemporary kidney transplant recipients appears overall neutral, with limited bone loss at the central skeleton.^6,9 Still, 20%–30% of kidney transplant recipients have osteoporosis,^10,11 and low BMD at the time of transplantation confers a higher risk of future fracture.¹¹ Recent studies suggest that disturbed mineral metabolism⁶ and changes in skeletal remodeling⁹ modulate bone loss after kidney transplantation. Changes in BMD post-transplant are, to some extent, predicted by decreases in biochemical markers of skeletal remodeling,^9,12 indicating a potential role of such markers for a noninvasive evaluation of bone metabolism. However, longitudinal relationships between these biomarkers and bone histomorphometry have yet to be established.

The transiliac bone biopsy with a subsequent histomorphometric analysis (Figure 1) remains the gold standard to evaluate bone disease in CKD.¹³ Existing literature indicates that kidney transplantation suppresses skeletal remodeling,^14,15 even from the low to normal bone turnover phenotype currently seen in patients with kidney failure.^16,17 Mineralization defects are reported to varying degrees post-transplant,^14,18,19 and have been linked to osteoblast dysfunction¹⁹ and hypophosphatemia.¹⁸ Changes in bone volume by bone biopsy are highly variable and do not correlate well with findings by dual-energy x-ray absorptiometry (DXA).^14,15 However, these previous studies are generally hampered by small sample sizes and highly selected cohorts, limiting generalizability. Further, few performed a complete bone phenotyping including bone histomorphometry, densitometry, mineral metabolism, and bone biomarkers.

Figure 1.

Normal and abnormal bone histology in the context of CKD. (A) Normal bone. (B) Low bone turnover with very few bone cells and bone surfaces without osteoid. (C) High bone turnover with large numbers of osteoblasts (white arrows) and osteoclasts (yellow arrows), zones of bone erosion and fibrosis (green arrow), generous amounts of osteoid (orange). (D) Delayed mineralization with broad osteoid seams (orange), disproportionate to the bone turnover. (E) Low bone volume with reduced amounts of mineralized tissue (blue). (F) High bone volume with generous amounts of well-connected trabecular bone (blue).

The aim of this prospective, observational cohort study was to describe the natural history of bone disease during the first post-transplant year in an unselected cohort of contemporary kidney transplant recipients. We further sought to investigate the relationships between changes in mineral metabolism, bone histomorphometry, and bone densitometry after kidney transplantation.

Methods

Patients and Protocol

Adult patients referred for a single kidney transplant at the University Hospitals Leuven were eligible for inclusion in this prospective, observational study on the natural evolution CKD–mineral and bone disorders after kidney transplantation. Use of antiosteoporotic medications was the only exclusion criteria. Patients with bone biopsies performed before (n=27), or at the time of (n=70) kidney transplantation, with a repeat biopsy at 12 months post-transplant, were included in this analysis. For patients with pretransplant bone biopsies, the median (interquartile range [IQR]) time from biopsy to transplantation was 261 (139–669) days. Biochemical parameters of mineral metabolism were unchanged from time of biopsy to kidney transplantation in these patients (data not shown). All biopsies were performed between September 2010 and March 2016. Out of 126 eligible pairs, 27 were excluded due to insufficient quality of either the baseline or follow-up bone biopsy, and a further two pairs were excluded due to bisphosphonate use. Relevant demographics, comorbidities, and details of medical therapy were extracted from electronic patient files. All clinical and research activities reported here were performed in agreement with the principles of the Declaration of Istanbul on Organ Trafficking and Transplant Tourism. The study adhered to the principles of the Declaration of Helsinki and was approved by the Leuven Research Ethical Committee (study identifier, S52091). All patients provided written informed consent before participation.

Protocols of Immunosuppression and Mineral Metabolism Therapy

The standard immunosuppressive regimen consisted of a calcineurin inhibitor (tacrolimus or cyclosporine), an antimetabolite (mycophenolate mofetil), and glucocorticoids. Methylprednisolone was administered intravenously on the day of transplantation (500 mg) and the first day postoperatively (40 mg), followed by oral prednisolone at a daily dose of 16 mg, tapered gradually to 5 mg by the third month. A protocol kidney graft biopsy at month 3, and the overall clinical course, determined whether glucocorticoids were discontinued or not. Mineral metabolism therapy was discontinued at the time of transplantation and reinstated at the discretion of the treating physician. Levels of 25-hydroxyvitamin D (25[OH]D) were regularly monitored and supplemented to a target of 30 ng/ml.

Biochemical Analyses

Nonfasting blood samples were taken at the time of the bone biopsy. Full-blood samples were stored for <2 hours at 5°C before arrival at the laboratory, and samples were then centrifuged at 3000 rpm for 10 minutes, aliquoted, and stored at –80°C until analysis. Creatinine, total alkaline phosphatase (ALP), total calcium, and phosphate were measured using standard laboratory techniques. The eGFR was calculated by the Chronic Kidney Disease Epidemiology Collaboration equation.²⁰

Serum concentrations of full-length, “biointact” parathyroid hormone (PTH) was determined by an in-house immunoradiometric assay (normal range, 3–40 pg/ml).²¹ Serum 25(OH)D (calcidiol) concentrations were measured using an RIA.²² Bone-specific ALP (BALP; assay range, 1–75 µg/L), trimeric procollagen type I amino-terminal propeptide (intact PINP; assay range, 2–230 µg/L), and tartrate-resistant acid phosphatase isoform 5b (TRAP5b; assay range, 0.9–14.0 U/L) were measured using the ImmunoDiagnostic Systems iSYS instrument (IDS, Boldon, United Kingdom). Values of BALP, intact PINP, and TRAP5b above the assay limit of measurement were determined by dilution. Inter- and intra-assay coefficients of variation were <10% for all assays. Time-averaged levels of PTH, calcidiol, phosphate, calcium, and bicarbonate were calculated using the linear trapezoidal rule.

Bone Biopsy and Histomorphometry

Bone biopsies were performed as an outpatient procedure with light sedation and local anesthesia, or under general anesthesia if performed at the time of transplantation. The sample was retrieved from a site 2-cm posterior and 2-cm inferior to the anterior iliac spine, using a smaller trephine with an external/internal diameter of 4.50/3.55 mm (8G; Biopsybell, Mirandola, Italy). All follow-up biopsies at 12 months post-transplant were preceded by double tetracycline (TC) labeling, which involved administration of oral TC 500 mg twice daily for 3 days, during two separate time-periods, with a TC-free interval of 11 days in between. The bone biopsy was scheduled 4 days after the last intake of TC. Baseline bone biopsies performed pretransplant were with prior TC labeling (n=27), whereas biopsies conducted at the time of deceased donor kidney transplantation were without (n=70). If no TC labels were visible despite the administration of TC, an arbitrary value of one was set for the bone formation rate (BFR).²³ This was the case for one patient at baseline and six patients at follow-up. There were no cases of high turnover with diffuse, unmeasurable TC labels.

After extraction, bone cores were fixed in 70% ethanol and embedded in a methyl methacrylate resin. Static bone parameters were determined from undecalcified, 5 μm–thick sections, stained by the Goldner method. For biopsies with prior TC labeling, unstained 10 μm–thick sections were additionally mounted in 100% glycerol for fluorescence microscopy to determine histodynamic parameters. All bone histomorphometric analyses were performed by an experienced bone pathologist at the Laboratory of Pathophysiology at the University of Antwerp (Antwerp, Belgium). An image analysis software, running a customized program (AxioVision version 4.51, Zeiss Microscopy; Zeiss, Jena, Germany), was used. Bone histomorphometric parameters are given in two dimensions using standardized nomenclature.²³

Patients were categorized by the bone turnover, mineralization, and volume (TMV) classification,¹³ on the basis of quantitative cutoffs of key histomorphometric parameters as detailed below. In addition, the initial, semiquantitative diagnosis, as set by the bone pathologist, is included as Supplemental Figure 1. An overview of histomorphometric variables and their normal values are given in Table 1. Bone turnover was classified on the basis of BFR per total tissue volume (11.5–110 mm³/cm³ per year), as previously determined for an adult reference population.¹⁷ In the absence of TC labeling, the evaluation was made on the basis of the static histomorphometric variables, using cutoffs calculated from a cohort of >200 bone biopsy specimens, as previously published,²⁴ but recalculated to the current BFR reference. Bone turnover was considered low with osteoblast perimeter per bone perimeter (ObPm/BPm) <2.3%, osteoclast perimeter per BPm (OcPm/BPm) <0.6%, and osteoid area per bone area (OAr/BAr) <2.1%, in the absence of fibrosis. Bone turnover was considered high with ObPm/BPm >3.0%, OcPm/BPm >1.3%, and OAr/BAr, >2.4%, with the presence of fibrosis. Mineralization was considered defective if OAr/BAr >12.0% and delayed if OAr/BAr >5.0% with increased mineralization lag time (Mlt >50 days)²⁵ or, in the absence of TC labeling, if OAr/BAr >5.0% without any indications of high bone turnover. Bone volume was evaluated by bone per total tissue area, with a normal range of 16.8%–22.9%.¹⁷

Table 1.

Bone histomorphometric variables for the evaluation of bone TMV

Abbreviation	Description	Unit	Reference Values17	Interpretation
Bone formation
BFR/TV	BFR/total tissue volume	mm3/cm3 per year	11.5–110	↓ Low turnover, ↑ high turnover
MAR	Mineral apposition rate	µm/d	N/A	↓ Low turnover, ↑ high turnover
MPm/BPm	Mineralizing perimeter/bone perimeter	%	N/A	↓ Low turnover, ↑ high turnover
ObPm/BPm	Osteoblast perimeter/bone perimeter	%	N/A	↓ Low turnover, ↑ high turnover
Fibrosis	Marrow fibrosis	None; trace; moderate; severe	None	↑ High turnover
EPm/BPm	Eroded perimeter/bone perimeter	%	0.1–5.69	↓ Low turnover, ↑ high turnover
OcPm/BPm	Osteoclast perimeter/bone perimeter	%	N/A	↓ Low turnover, ↑ high turnover
Mineralization
Mlt	Mineralization lag time	days	13–58	↑ Delayed mineralization
Omt	Osteoid maturation time	days	7–25	↑ Delayed mineralization
OWi	Osteoid width	µm	4.4–11.8	↓ Low turnover, ↑ high turnover, ↑↑ delayed mineralization
OAr/BAr	Osteoid area/bone area	%	0.6–6.0	↓ Low turnover, ↑ high turnover, ↑↑ delayed mineralization
OPm/BPm	Osteoid perimeter/bone perimeter	%	3.5–37.9	↓ Low turnover, ↑ high turnover, ↑↑ delayed mineralization
Volume
BAr/TAr	Bone area/total tissue area	%	16.8–22.9	↓ Low bone volume

N/A, not available.

Bone Densitometry

Bone densitometry was performed by DXA using a Hologic densitometer (QDR-4500A or Discovery; Hologic, Marlborough, MA). Scans were performed from 6 months before to 4 weeks after the baseline index date, and from 4 weeks before to 4 weeks after the follow-up index date at 12 months post-transplant. DXA scans at both time points were available for a subset of patients (spine and hip, n=52; forearm, n=27). The main reason for missing DXA scans was logistic constraints. Baseline demographics and mineral metabolism parameters of patients without DXA scans were comparable with those of the overall cohort (data not shown). All DXA scans were analyzed by a single, certified operator. The Hologic Spine Phantom was scanned daily to monitor the device performance and long-term stability. Coefficients of variation for repeated patient scans were 0.58% at lumbar spine, 0.56% for total hip, 1.40% for femoral neck, 1.10% for ultradistal radius, and 0.98% for distal third radius. Results are expressed as areal BMD (g/cm²), with T- and Z-scores calculated using reference data from the third National Health and Nutrition Examination Survey, or as provided by the manufacturer. An increase or decrease of DXA BMD ≥5% was considered the least significant change.²⁶

Statistical Analyses

For continuous variables, summary statistics are given as means±SD if normally distributed, or medians with IQR if skewed. Categoric variables are given as number and proportion (%). Changes from baseline to follow-up were evaluated by the paired t test, or Wilcoxon matched-pairs signed-rank test for normally and non-normally distributed variables, respectively. Changes across categories of bone turnover were evaluated by one-way ANOVA or Kruskal–Wallis equality-of-populations rank test. Univariate correlations between biochemical variables, bone histomorphometry, and bone density were examined by Spearman rank correlation. Multivariable linear regression analyses with a stepwise backward selection of variables with P<0.20, followed by a forward selection of variables with P<0.05, were then applied to adjust for potential confounders (Supplemental Tables 1 and 2). Statistical analyses were performed using STATA IC version 16.1 (StataCorp LP, College Station, TX).

Results

Patient Characteristics

Demographic data and list of medications at baseline and 12 months post-transplant are given in Table 2. The cause of CKD was GN or vasculitis (25%), cystic or hereditary disease (22%), diabetes mellitus (10%), hypertension or atherosclerosis (9%), chronic interstitial nephritis (6%), other (3%), or unknown (25%). The majority of patients had received either hemodialysis (65%) or peritoneal dialysis (32%) before transplantation, with a median (IQR) dialysis vintage of 26 (15–38) months. At 12 months post-transplant, mean±SD eGFR was 49±17 ml/min per 1.73 m²; 11 patients had an eGFR <30 ml/min per 1.73 m², and no patients had an eGFR <15 ml/min per 1.73 m². A total of 21 patients (22%) had suffered an episode of acute kidney graft rejection during the first post-transplant year. The cumulative steroid dose at 12 months post-transplant was 2.43 (IQR, 1.81–2.85) g, and 78% of patients were still receiving oral prednisolone.

Table 2.

Demographic data of participating kidney transplant recipients at baseline and 1 year post-transplant

Characteristic	Baseline (n=97)	Post-transplant (n=97)
Demography
Age (yr)	55±12	56±12
Male sex	70 (72%)	70 (72)
Weight (kg)	75.22±13.29	75.62±15.36
Body mass index (kg/m2)	25.39±4.50	25.34±5.00
Diabetes mellitus	20 (21)	34 (35)
Parathyroidectomy	5 (5)	10 (10)
Dialysis before transplantation
Hemodialysis	63 (65)	63 (65)
Peritoneal dialysis	31 (32)	31 (32)
Dialysis vintage (mo)	26 (15–38)	26 (15–38)
Medications
Calcium-containing phosphate binder	63 (68)	0 (0)
Noncalcium-containing phosphate binder	32 (35)	1 (1)
Vitamin D supplement	66 (68)	41 (42)
Active vitamin D	48 (49)	13 (13)
Cinacalcet hydrochloride	10 (10)	0 (0)
Bicarbonate supplement	27 (28)	36 (37)
Tacrolimus	12 (12)	94 (97)
Mycophenolate mofetil	5 (5)	84 (87)
Prednisolone	14 (14)	76 (78)
Biochemistry
eGFR (ml/min per 1.73 m2)	N/A	47 (38–59)
Biointact PTH (pg/ml)	241 (137–386)	64 (42–117)
Biointact PTH (xUNL)	6.0 (3.4–9.6)	1.6 (1.1–2.9)
Total calcium (mg/dl)	9.41±0.63	9.78±0.66
Magnesium (mg/dl)	2.21±0.33	1.68±0.24
Phosphate (mg/dl)	4.61±1.33	3.09±0.72
Bicarbonate (mmol/L)	24.5±3.2	22.8±3.2
Calcidiol (ng/ml)	44.4±14.9	35.6±13.9
Total ALP (U/L)	93.0 (73.4–129.6)	78.4 (61.7–92.1)
BALP (μg/L)	27.9 (18.2–41.2)	21.9 (14.3–29.3)
Intact PINP (μg/L)	100.8 (57.5–156.6)	59.1 (30.3–87.5)
TRAP5b (U/L)	6.0 (4.2–8.1)	3.6 (2.4–4.8)

Data are mean±SD, median (IQR), or n (%), as appropriate. N/A, not available; xUNL, times upper normal limit of the assay.

Biochemical parameters of mineral metabolism pre- and post-transplant are shown in Table 2, and the evolution of biointact PTH, total calcium, phosphate, calcidiol, and bicarbonate levels are shown in Figure 2. Changes in mineral metabolism parameters were consistent with resolution of secondary hyperparathyroidism. Before transplantation, 56% of patients had biointact PTH levels within two to nine times the upper normal limit of the assay; 30% were above and 14% below this target range. Vitamin D insufficiency, defined as 25(OH)D <30 ng/ml, was seen in 11% of patients, and 3% of patients had 25(OH)D <20 ng/ml. At 12 months post-transplant, 56% of patients had persistent hyperparathyroidism (biointact PTH >1.5 times the upper normal limit), 19% had hypercalcemia (>10.3 mg/dl), and 13% had hypophosphatemia (<2.3 mg/dl). The prevalence of 25(OH)D <30 ng/ml and <20 ng/ml had increased to 41% and 12%, respectively. Circulating levels of bone turnover markers decreased during the first post-transplant year: total ALP decreased by 23%, BALP by 23%, intact PINP by 47%, and TRAP5b by 42%.

Figure 2.

Evolution of mineral metabolism in the first post-transplant year. Biointact PTH, total calcium, phosphate, calcidiol, and bicarbonate at baseline, lowest registered value (nadir), time-averaged concentration (TAC) months 0–3, TAC months 3–12, and measurements at 12 months post-transplant, as violin plots with median and IQR (dotted lines).

Effect of Kidney Transplantation on Bone Histomorphometry

Bone TMV at baseline and 12 months post-transplant is shown in Figure 3. Normal or low bone turnover were the most common phenotypes at both time points, particularly at 12 months post-transplant (87%). Bone mineralization was normal for all but one patient at baseline, whereas delayed mineralization was found in 15 patients at 12 months post-transplant. The prevalence of low bone volume by bone biopsy was comparable at the two time points, but with considerable crossover between categories.

Figure 3.

Evolution of TMV classification in the first post-transplant year. Changes in bone TMV by bone histomorphometry from baseline to 1 year after kidney transplantation.

Parameters of the full bone histomorphometric analysis pre- and post-transplant are shown in Table 3. By static histomorphometry, bone resorption (OcPm/BPm, EPm/BPm) decreased, and disordered bone formation (marrow fibrosis) became less prevalent. In the subset of patients with dynamic parameters available at both time points (n=21), BFR decreased and Mlt increased (Figure 4). Bone volume remained unchanged overall.

Table 3.

Bone histomorphometry and BMD at baseline and 12 months post kidney transplantation

Bone Parameters	n	Baseline	Post-transplant	P Value
Turnover
BFR/TV, mm3/cm3 per year	21	107 (52–167)	46 (19–80)	0.004
MAR, µm/d	21	0.89 (0.77–1.25)	0.86 (0.79–0.94)	0.29
MPm/BPm, %	21	14.0 (5.7–18.8)	5.7 (4.4–9.3)	0.003
ObPm/BPm, %	97	2.4 (0.0–5.2)	3.6 (0.5–9.2)	0.05
Marrow fibrosis, n (%)	97	26 (27)	2 (2)	<0.001
OcPm/BPm, %	97	0.9 (0.0–1.7)	0.4 (0.0–0.9)	<0.001
EPm/BPm, %	97	4.6 (2.9–7.2)	2.3 (1.1–3.9)	<0.001
Mineralization
Mlt, d	21	16 (8–27)	31 (19–51)	0.03
Omt, d	21	9 (6–11)	8 (6–10)	0.76
OWi, µm	97	7.4 (6.2–10.0)	7.1 (5.7–10.1)	0.99
OAr/BAr, %	97	2.1 (1.1–3.8)	2.6 (1.2–5.5)	0.14
OPm/BPm, %	97	19.0 (10.5–27.3)	22.2 (11.2–37.4)	0.08
Volume
BAr/TAr, %	97	21.7±8.1	21.0±6.2	0.66
Densitometry
Lumbar spine BMD, g/cm2	54	0.928±0.163	0.942±0.151	0.96
Lumbar spine T-score	54	−1.61±1.46	−1.48±1.37	0.96
Total hip BMD, g/cm2	54	0.809±0.145	0.804±0.142	0.07
Total hip T-score	54	−1.39±1.02	−1.42±0.99	0.08
Femoral neck BMD, g/cm2	54	0.649±0.110	0.661±0.113	0.36
Femora neck T-score	54	−2.01±0.85	−1.91±0.86	0.37
Distal third radius BMD, g/cm2	28	0.722±0.109	0.688±0.104	0.01
Distal third radius T-score	28	−1.07±1.57	−1.72±1.77	0.02
Ultradistal radius BMD, g/cm2	28	0.392±0.072	0.374±0.072	<0.001
Ultradistal radius T-score	28	−2.10±1.04	−2.42±1.23	<0.001

Data are median (IQR), mean±SD, or n (%), with P by paired t test, Wilcoxon matched-pairs signed-rank test, and Pearson chi-squared test, respectively.

Figure 4.

Bone turnover declines and bone mineralization slows in the first post-transplant year. Changes in bone formation rate by total tissue volume (BFR/TV) and mineralization lag time (Mlt) in the subset patients with TC-labeled bone biopsies taken median 261 (IQR, 139–669) days before transplantation and repeated at 12 months post-transplant. P by Wilcoxon matched-pairs signed-rank test.

The cumulative steroid dose was directly correlated to change in Mlt (ρ=0.61, P=0.004), and inversely to change in BFR (ρ=−0.44, P=0.05); indicating a slowing of both bone turnover and mineralization with higher glucocorticoid exposure.

Effect of Kidney Transplantation on Bone Densitometry

Baseline and 12 months post-transplant values of DXA BMD and T-scores are shown in Table 3. Significant bone loss was detected at the distal third radius (–1.7%; 95% CI, –3.0 to –0.4%; P=0.01) and ultradistal radius (–3.6%; 95% CI, –5.7 to –0.8%; P<0.001) only, with no overall changes in BMD at the spine or hip. However, there were notable interindividual variations in the evolution of BMD post-transplant, with changes ranging from –18% to +17% per year (Figure 5).

Figure 5.

Subsets of patients experience bone loss or bone gain in the first post-transplant year. Annualized change in BMD by DXA at lumbar spine (%LS), total hip (%TH), femoral neck (%FN), distal third radius (%13R), and ultradistal radius (%UDR); dashed lines represent least significant change (±5%).

Significant bone loss (change in BMD of less than –5%/yr) was seen in 15% of patients at the lumbar spine, 17% at the total hip, 21% at the femoral neck, 22% at the distal third radius, and 33% at the ultradistal radius. Conversely, bone gain (BMD change of more than +5%/yr) was seen in 19% of patients at lumbar spine, 13% at the total hip, 10% at the femoral neck, 4% at the distal third radius, and 7% at the ultradistal radius. There was little consistency in bone loss or bone gain between different skeletal sites.

A higher cumulative steroid dose was associated with BMD decrease at the total hip (ρ=–0.32, P=0.02), but not at the femoral neck (ρ=–0.22, P=0.11) or lumbar spine (ρ=–0.06, P=0.65).

Effects of Changes in Mineral Metabolism

To investigate relationships between changes in mineral metabolism and evolution of bone phenotype, we performed univariate correlation analyses between changing (Δ) values of biochemistry, bone histomorphometry, and bone densitometry (Table 4). For these analyses, a positive correlation indicates parallel development, whereas a negative correlation signifies change in opposite directions. Resolution of hyperparathyroidism associated with a decrease in bone turnover, as indicated by a positive correlation between ΔBFR and ΔPTH, positive correlation between ΔBFR and time-averaged phosphate, and negative correlation between ΔBFR and time-averaged calcium. Further, a decrease in bone turnover was paralleled by decreasing biochemical bone biomarkers, with positive correlations between ΔBFR and Δintact PINP, and positive correlations between static parameters of skeletal remodeling and Δ values of the bone biomarkers.

Table 4.

Correlations between changes in biochemical markers or time-averaged levels of biochemical markers and bone histomorphometry and BMD by DXA

Δ Values	ΔPTH	ΔTotal ALP	ΔBALP	ΔIntact PINP	ΔTRAP5b	TAC Calcium	TAC Phosphate	TAC Bicarb
Bone turnover
ΔBFR/TV (n=21)	0.581a	0.374	0.422	0.474b	0.399	−0.600a	0.538b	−0.148
ΔMPm/BPm (n=21)	0.527b	0.218	0.325	0.339	0.334	−0.518b	0.394	−0.182
ΔObPm/BPm	0.072	0.125	0.161	0.259b	0.154	0.05	−0.073	−0.227b
ΔOcPm/BPm	0.115	0.282a	0.319a	0.260b	0.428c	0.049	−0.201	−0.059
ΔEPm/BPm	0.186	0.284a	0.338a	0.286a	0.376c	0.061	−0.135	−0.012
Bone mineralization
ΔMlt (n=19)	−0.361	−0.146	−0.149	−0.311	−0.037	0.193	−0.765c	−0.218
ΔOAr/BAr	0.314a	0.343a	0.452c	0.370c	0.427c	0.046	−0.255b	−0.159
ΔOPm/BPm	0.322a	0.290a	0.393c	0.311a	0.395c	0.118	−0.230b	−0.261b
ΔOWi	0.266b	0.317a	0.402c	0.396c	0.330a	−0.075	−0.231b	−0.042
Bone volume
ΔBAr/TAr	0.034	−0.135	−0.053	−0.123	−0.079	−0.165	−0.126	−0.109
BMD
%Δ Lumbar spine	−0.423a	−0.449a	−0.540c	−0.594c	−0.385a	−0.128	0.234	0.030
%Δ Total hip	−0.422a	−0.353a	−0.417a	−0.422a	−0.402a	−0.108	−0.155	0.045
%Δ Femoral neck	−0.370a	−0.339b	−0.377a	−0.310b	−0.305b	0.131	−0.160	0.108
%Δ Distal third radius	−0.345	−0.402b	−0.380	−0.400b	−0.450b	−0.028	0.123	0.363
%Δ Ultradistal radius	−0.303	−0.187	−0.225	−0.362	−0.344	−0.100	0.098	0.125

Shown are Spearman ρ values. TAC, time-averaged concentration 0–12 months; Bicarb, bicarbonate

^aP<0.01, unadjusted.

^bP<0.05, unadjusted.

^cP<0.001, unadjusted.

Lower phosphate levels associated with a slowing of bone mineralization, as seen by the inverse relationships between serum phosphate and the mineralization parameters (Supplemental Figure 2). This was a consistent finding for all measures of phosphate kinetics, including the lowest phosphate level (ΔMlt ρ=−0.78, P<0.001; ΔOAr/BAr ρ=−0.30, P=0.004), time-averaged phosphate months 0–3 (ΔMlt ρ=−0.47, P=0.04; ΔOAr/BAr ρ=−0.20, P=0.05), and time-averaged phosphate months 3–12 (ΔMlt ρ=−0.80, P<0.001; ΔOAr/BAr ρ=−0.25, P=0.01).

Changes in biochemical markers of mineral metabolism were related to changes in BMD. Decreasing levels of biointact PTH and bone turnover markers associated with increases in BMD, particularly at trabecular bone sites (Table 4, Supplemental Figure 3), indicating a beneficial effect on BMD with the resolution of hyperparathyroidism and normalization of bone turnover.

The associations between changes in biochemical markers of mineral metabolism and the evolution of bone histomorphometry and densitometry remained significant after adjusting for age, sex, cumulative steroid dose, and kidney function at 12 months post-transplant (Supplemental Tables 1 and 2).

Effect of Pretransplant Bone Turnover Category

Finally, we investigated the effect of baseline bone turnover category on the evolution of bone phenotype post-transplant (Table 5). At baseline, there were no differences in BMD across bone turnover categories. Patients with a high bone turnover at baseline had greater declines in the biochemical bone turnover markers (BALP and TRAP5b) during the first post-transplant year, and histomorphometry revealed greater decreases in the amounts of osteoid (OAr/BAr) in these patients. Patients with a high turnover tended to experience BMD gain, with a significant difference in the ΔBMD at the total hip (+4.4% versus −1.7%, P=0.02) and femoral neck (+5.3% versus −1.1%, P=0.002) when compared with patients with normal bone turnover at baseline. Conversely, patients with low bone turnover at baseline had slightly increasing levels of bone biomarkers (BALP and PINP) during the first post-transplant year, and tended to experience bone loss, but without any significant differences in ΔBMD compared with patents with normal bone turnover.

Table 5.

Changes in biochemistry, bone histomorphometry, and BMD in the first post-transplant year by bone turnover classification at baseline

Variables	Time-point	Low (n=21)	Normal (n=57)	High (n=19)	P Value
Biointact PTH (pg/ml)	Baseline	150 (48–245)a	248 (140–366)	388 (215–477)a	0.002
	1 Year	51 (31–79)	65 (42–124)	90 (49–130)	0.08
	Change	−89 (−157 to 7)a	−165 (−295 to −83)	−259 (−382 to −47)	0.01b
BALP (μg/L)	Baseline	14.8 (11.4–20.9)a	29.4 (19.7–41.2)	40.8 (28.1–121.5)a	<0.001
	1 Year	14.4 (12.2–24.1)	22.7 (15.5–29.5)	22.8 (15.1–33.5)	0.10
	Change	1.9 (−3.6 to 6.8)a	−7.1 (−13.3 to −0.5)	−14.2 (−101.8 to −4.2)a	<0.001b
Intact PINP (μg/L)	Baseline	48.3 (24.6–85.7)a	106.1 (66.1–138.3)	170.2 (109.6–460.5)a	<0.001
	1 Year	32.7 (25.1–82.0)	61.9 (35.0–83.3)	64.9 (31.4–113.2)	0.30
	Change	6.3 (−30.1 to 20.9)a	−49.4 (−87.2 to −0.9)	−78.7 (−430.2 to −24.6)	0.001b
TRAP5b (U/L)	Baseline	2.8 (1.8–4.8)a	6.5 (4.9–8.9)	7.8 (6.5–10.1)a	<0.001
	1 Year	3.4 (1.9–4.2)	3.6 (2.5–5.0)	4.0 (2.5–5.1)	0.48
	Change	−0.2 (−1.4 to 1.1)a	−2.8 (−4.3 to −1.3)	−4.2 (−5.9 to −3.0)a	<0.001b
ObPm/BPm (%)	Baseline	0.0 (0.0–0.4)a	2.8 (0.9–4.4)	6.8 (4.1–17.1)a	0.001
	1 Year	1.2 (0.0–5.8)	3.5 (0.6–10.3)	7.2 (1.2–13.3)	0.05
	Change	0.6 (0.0–5.8)	0.3 (−2.8 to 6.1)	−0.6 (−10.6 to 7.0)	0.19
OcPm/BPm (%)	Baseline	0.0 (0.0–0.8)a	0.9 (0.1–1.6)	1.7 (1.3–2.6)a	<0.001
	1 Year	0.4 (0.0–0.9)	0.4 (0.0–0.8)	0.7 (0.4–1.5)	0.06
	Change	0.1 (−0.4 to 0.7)a	−0.4 (−0.9 to 0.2)	−0.8 (−2.0 to −0.1)	0.008b
EPm/BPm (%)	Baseline	3.0 (1.0–4.6)a	4.6 (2.9–6.8)	6.9 (4.9–8.6)a	<0.001
	1 Year	2.6 (1.0–4.4)	1.6 (0.9–3.5)	2.7 (2.1–5.3)	0.03
	Change	0.0 (−2.3 to 2.1)a	−2.4 (−5.2 to −0.8)	−3.2 (−6.5 to −1.5)	0.007b
OAr/BAr (%)	Baseline	0.9 (0.3–1.1)a	2.3 (1.6–3.6)	4.1 (3.0–7.9)a	<0.001
	1 Year	2.3 (0.7–5.0)	2.3 (1.2–5.4)	3.3 (2.1–5.8)	0.34
	Change	1.6 (0.2–3.9)a	0.3 (−1.5 to 2.2)	−1.7 (−4.6 to 2.7)a	0.004b
OWi (μm)	Baseline	6.4 (5.6–7.4)a	7.5 (6.4–9.7)	10.1 (7.9–14.1)a	<0.001
	1 Year	7.0 (5.3–13.4)	6.9 (5.7–9.4)	7.6 (6.2–10.6)	0.65
	Change	1.9 (−1.4 to 7.5)a	−0.2 (−1.8 to 1.8)	−2.7 (−6.2 to 1.3)	0.01b
BAr/TAr (%)	Baseline	18.5 (17.1–23.0)	22.5 (15.8–25.9)	22.0 (17.2–29.3)	0.42
	1 Year	18.0 (14.3–20.9)	20.8 (17.4–24.9)	20.9 (18.0–24.9)	0.09
	Change	−0.59 (6.63)	0.09 (8.81)	−2.97 (11.32)	0.44
Lumbar spine BMD (g/cm2)	Baseline	0.96 (0.12)	0.93 (0.17)	0.88 (0.16)	0.48
	1 Year	0.98 (0.14)	0.93 (0.15)	0.93 (0.16)	0.54
	% Change	−2.16 (5.92)	0.77 (5.33)	2.67 (6.07)	0.17
Total hip BMD (g/cm2)	Baseline	0.82 (0.14)	0.81 (0.15)	0.80 (0.15)	0.93
	1 Year	0.85 (0.12)	0.78 (0.14)	0.82 (0.15)	0.27
	% Change	−2.48 (6.14)	−1.73 (5.76)	4.38 (6.71)a	0.04b
Femoral neck BMD (g/cm2)	Baseline	0.66 (0.12)	0.65 (0.11)	0.65 (0.11)	0.93
	1 Year	0.67 (0.11)	0.64 (0.10)	0.71 (0.13)	0.08
	% Change	−2.48 (6.62)	−1.08 (4.40)	5.26 (5.81)a	0.01b

Data are mean±SD or median (IQR), with P values by one-way ANOVA or Kruskal–Wallis equality-of-populations rank test, respectively.

^aP<0.05 compared with “normal.”

^bP<0.05.

Discussion

Main Findings

This prospective, observational study included extensive phenotyping of bone during the first year after kidney transplantation. Our main findings are as follows: the majority of patients have persistently normal, or improved, bone turnover after kidney transplantation. Skeletal remodeling showed an overall decrease, with diminishing signs of hyperparathyroid bone disease. Circulating biochemical markers of bone turnover parallel the decrease in skeletal remodeling. A subset of patients show de novo delayed bone mineralization post-transplant, which relates to the severity and duration of hypophosphatemia. Lastly, changes in BMD are highly variable in the first post-transplant year and are determined both by cumulative corticosteroid exposure and the resolution of hyperparathyroidism and high bone turnover.

Bone Turnover

Although the majority of patients had persistently normal, or improved, bone turnover, bone histomorphometry revealed an overall decrease in skeletal remodeling, with significant declines in both dynamic and static histomorphometric parameters. These findings confirm and extend data from our previous pilot study,¹⁵ establishing that bone turnover declines after kidney transplantation.^14,19,27 Changes in bone histomorphometry were compatible with the resolution of hyperparathyroid bone disease,²⁸ with decreased bone resorption and reduced presence of disturbed bone formation (marrow fibrosis). Remaining parameters of bone formation were stable overall. In contrast, reductions in both osteoblast activity and osteoid amounts were previously reported both in the early¹⁹ and late^14,27 post-transplant period. Case mix, particularly variation in the prevalence of high turnover bone disease at time of transplantation and in the cumulative steroid exposure, may account for these differences.

Circulating levels of bone turnover markers paralleled the changes in bone histomorphometry. Because these biomarkers are released from bone during the skeletal remodeling process, they have the potential to deliver highly specific information on current bone turnover. BALP and TRAP5b are enzymes released from osteoblasts and osteoclasts, respectively, with BALP considered to reflect the activity of osteoblasts, whereas TRAP5b likely represents the overall number of osteoclasts. PINP is a fragment of procollagen released as bone matrix is laid down.²⁹ None of these three markers are excreted by the kidney, but fragments of PINP accumulate in kidney failure, so assays measuring the intact, trimeric form should be preferred in CKD.³⁰ Similar to what we report, changes in BALP and TRAP5b correlated with changes in histomorphometric parameters in a trial of zoledronic acid after kidney transplantation.²⁷ These findings highlight the potential of biochemical bone turnover markers as surrogates for evaluating changes in skeletal remodeling in CKD. Such noninvasive alternatives are urgently needed, particularly in the context of considering response to treatment, because bone biopsies are not well suited for longitudinal monitoring. Studies on the diagnostic accuracy of biomarkers have, so far, been disappointing,¹⁶ although high negative predictive values for diagnosing both high and low bone turnover have been demonstrated for BALP, intact PINP, and TRAP5b.^31,32 At present, there is insufficient data to support the use of biochemical bone turnover markers as a substitute for the bone biopsy, particularly because no biomarker has yet been found to identify mineralization defects.³³

Bone Mineralization

In line with what has been demonstrated in recent cohorts of patients with kidney failure,³⁴ mineralization defects were rare pretransplant, with just a single case (1%) in this cohort. A substantial subset of patients (15%) were, however, found to have de novo delayed mineralization 1 year after kidney transplantation. Previous studies report mineralization defects in 22%–50% of patients post-transplant^14,15,18,27; this variability may, at least partly, reflect the lack of consensus on diagnostic cutoffs. Generally, however, kidney transplantation seems to associate with a slowing of bone mineralization.^14,27

Time-averaged levels of phosphate, and the phosphate nadir, correlated inversely with changes in the histomorphometric parameters of mineralization. Thus, both the severity and duration of hypophosphatemia may negatively affect bone mineralization, which is consistent with our understanding of phosphate as a necessary component in biomineralization.³⁵ Lower phosphate levels were associated with mineralization defects in previous cross-sectional studies,^18,36 but this is, to our knowledge, the first report to confirm this longitudinally. In our previous study, we found that patients with delayed bone mineralization post-transplant had increased urinary fractional excretion of phosphate.³⁶ Ongoing hyperparathyroidism and hyperphosphatoninism lead to urinary phosphate wasting in the post-transplant period,^37–39 although immunosuppressives, particularly glucocorticoids, may also contribute.⁴⁰ Serum phosphate levels reach a nadir around 1 month post-transplant and improve gradually during months 6–12.^41,42 However, around 10% of patients have persistently low phosphate levels at 1 year post-transplant and in the following years.^9,41,43 In a previous trial of cinacalcet hydrochloride, both a reduction in serum calcium and an increase in serum phosphate were achieved with treatment, whereas no significant benefit was found on BMD by DXA.⁴⁴ The potential effect on bone quality of correcting phosphate levels in the early post-transplant period warrants further investigation.

Bone Volume

Overall, only minor changes occurred in bone volume during the first post-transplant year. Significant bone loss by DXA BMD was detected at the forearm, with a similar trend at the hip, but with no change at the lumbar spine. These findings concur with recent reports of limited bone loss at the central skeleton after kidney transplantation^6,9; a change from the findings of earlier studies^45–47 which has been attributed to reductions in steroid exposure.⁴⁸ Although patients were managed with a steroid-minimization protocol in this study, the majority (78%) remained on low-dose oral prednisolone at 12 months post-transplant. Cumulative steroid dose associated with BMD loss at the total hip, and also with a decrease in bone turnover and a slowing of bone mineralization. These findings are in line with our understanding of the effect of systemic glucocorticoids on bone, with an early and sustained inhibition of osteoblast function leading to low turnover bone loss.⁴⁹ Thus, even under current steroid-minimization protocols, glucocorticoid exposure exerts negative effects on the skeleton post-transplant.

The variation in the evolution of BMD in the first post-transplant year was impressive, with changes in BMD ranging from –18% to +17% per year on the individual level. On the basis of a least significant change of 5% per year, BMD loss at the central skeleton was seen in 15%–21% of patients, whereas a similar proportion (10%–19%) showed BMD gain at either spine or hip. We conclude that different subsets of patients experience substantial bone loss or bone gain post-transplant. Future efforts should be directed at identifying which patients are at risk of substantial bone loss in the early post-transplant period to target fracture-preventive interventions.

Interplay between Mineral Metabolism, Bone Phenotype, and Bone Structure

Normalization of skeletal remodeling associated with a favorable development of BMD. Reduced bone turnover by biochemical markers associated with BMD gain, as did a decreasing amount of unmineralized bone (osteoid). We hypothesize that the decrease in skeletal remodeling rate post-transplant allows for a “catch-up” mineralization of previously formed unmineralized bone, which tends to accumulate in the high turnover form of renal osteodystrophy.²⁸ This would result in a rapid increase in the amount of fully mineralized bone, which is captured by DXA BMD. Our findings can be compared with what is seen after parathyroidectomy, where impressive gains in BMD are reported in the months after parathyroid surgery,^50–52 whereas, simultaneously, bone histomorphometry demonstrates not only a marked reduction in bone resorption,⁵³ but also widespread bone mineralization.⁵⁴ These findings highlight the transitional nature of the skeleton after kidney transplantation and may suggest that skeletal remodeling should be evaluated before initiating therapy aimed at reducing bone loss in kidney transplant recipients.⁵⁵

Patients with low bone turnover at baseline experienced a slight increase, or improvement, in skeletal remodeling as evaluated by biomarkers, but, even so, tended to experience loss of BMD. Similarly, a study investigating post-transplant bone loss by repeat DXA scans found that patients with deterioration in BMD during the first post-transplant year had lower levels of intact PTH pretransplant.⁵⁶ A recent review highlights the association between low bone turnover and comorbid conditions, such as older age, diabetes, and malnutrition, which may confound the relationships between this phenotype and clinical outcomes.⁵⁷ Although of a substantial size for a study including histomorphometry, this cohort was too small for adjustments to be attempted in subgroup analyses. Concerted action from centers performing bone biopsies in renal osteodystrophy will likely be necessary to achieve studies of sufficient power to establish the clinical consequences of bone phenotypes post-transplant.⁵⁸

Strengths and Limitations

The main strength of this study is the large number of paired bone biopsy specimens, which enabled us to evaluate the effect of kidney transplantation on bone at the tissue level. Bone biopsies were performed in unselected kidney transplant recipients, at protocol-specified time points, which should increase generalizability. Further, the cohort was contemporary, and the results should, therefore, reflect renal osteodystrophy as it presents under the current treatment regimes, both with regards to mineral metabolism and immunosuppression. Considering our limitations, this was an observational study, although it could be argued that we studied the effects of kidney transplantation as an intervention. Although this was an unselected cohort, some of the changes reported may represent regression toward the mean, particularly in the subgroup analysis looking at evolution of parameters depending on bone phenotype at baseline. We cannot exclude that other factors related to improved kidney function post-transplant contributed to the changes reported and, particularly, because DXA scans were only available for a subset of patients, we had limited statistical power for the prediction of bone loss. The subset of bone biopsies with TC labeling was performed before kidney transplantation, and we cannot exclude that changes in bone metabolism occurred during the time until transplantation. However, biochemical parameters of mineral metabolism and bone turnover markers remained stable during this time, which argues against major shifts, and we do believe this subgroup offers valuable insight into changes in dynamic parameters from late-stage CKD to the post-transplant period. We did not include an evaluation of cortical bone, which is a disadvantage considering hyperparathyroidism may preferentially affect this compartment.⁶ All participants were White and, therefore, results may not be fully applicable to populations of different racial and ethnic distributions. An important limitation to the generalizability of our findings, and those of other histomorphometric studies, is the lack of consensus on diagnostic cutoffs for renal osteodystrophy, which severely hampers aggregation of published data. Collaboration and data harmonization between centers performing bone biopsies should be a future priority.⁵⁸

Clinical Applications

In addition to steroid minimization, optimal control of mineral metabolism is likely key to improving the bone health of contemporary kidney transplant recipients. Efforts should be made to determine treatment targets of PTH both in the short and long term after kidney transplantation, and future trials could focus on the effects of normalization of PTH, phosphate, and calcium on bone and vascular phenotypes. Such trials should preferably include the gold-standard bone histomorphometry, at least in a subset of participants. Because changes in skeletal remodeling determine the evolution of BMD post-transplant, it may be prudent to evaluate the current status of bone turnover before initiating therapy aimed at reducing bone loss. Nonkidney–retained bone biomarkers show promise for a noninvasive evaluation of bone turnover, but further studies are needed to determine their clinical utility, both in the pre- and post-transplant setting.

Disclosures

B. Bammens reports receiving research funding from the Amgen, Astellas, Novartis, Otsuka Pharma, and Roche; receiving honoraria via speaker’s fees from Baxter; and having consultancy agreements with Baxter and Otsuka Pharmaceutical. K. Claes reports serving as a scientific advisor for, or member of, Alexion and Astellas; receiving support from Alexion, Astellas, AstraZeneca, and Sanofi; and having other interests in/relationships with Fresenius Medical Care and Menarini (via speaker’s fee). E. Cavalier reports serving as a consultant for DiaSorin, Fujirebio, IDS, and Nittobo. P. D’Haese reports receiving research funding from Inositec, Rockwell Medical, Shire Pharmaceuticals, and Vifor Pharma. P. Evenepoel reports receiving honoraria from Amgen and Vifor-FMC; serving on speaker’s bureaus for Amgen and Vifor-FMC; having consultancy agreements with Amgen and Vifor Pharma; serving on the editorial boards of Kidney International and Nephrology Dialysis Transplantation; and receiving research funding from Sanofi and Vifor-FMC. H.S. Jørgensen reports having a long-term fellowship from the European Renal Association (ERA) for work within their CKD–mineral and bone disorder (CKD-MBD) working group and received further financial support from the Augustinus Foundation and Kornings Fund. D.R.J. Kuypers reports receiving research funding from, and serving on a speaker’s bureau for, Astellas Inc.; having consultancy agreements with, and receiving honoraria from, Astellas Company, CSL Behring, GlaxoSmithKline, Hansa, and UCB; and serving as an editorial board member of Current Clinical Pharmacology, Therapeutic Drug Monitoring, and Transplantation Reviews, and as an associate editor of Transplantation. B. Meijers reports having consultancy agreements with AstraZeneca, Baxter, Bayer Fresenius, Nipro, Novartis, and Vifor Fresenius; serving on a speaker’s bureau for Baxter; receiving honoraria from Baxter, Fresenius, Nipro, and Vifor; receiving research funding from Bayer, Ionis, and Nipro; and serving on the editorial boards of BMC Nephrology and Toxins. B. Sprangers reports serving as an expert ad hoc for the European Medicines Agency. B. Sprangers and M. Naesens are senior clinical investigators of The Research Foundation Flanders (1842919N and 1844019N, respectively). All remaining authors have nothing to disclose.

Funding

None.

Author Contributions

B. Bammens, G. Behets, E. Cavalier, K. Claes, P. D’Haese, P. Evenepoel, D.R.J. Kuypers, B. Meijers, M. Naesens, and B. Sprangers were responsible for investigation; G. Behets, E. Cavalier, P. D’Haese, P. Evenepoel, and D.R.J. Kuypers were responsible for resources; E. Cavalier, P. D’Haese, and P. Evenepoel conceptualized the study; P. Evenepoel provided supervision and was responsible for funding acquisition; P. Evenepoel and H.S. Jørgensen were responsible for methodology; P. Evenepoel and D.R.J. Kuypers were responsible for project administration; H.S. Jørgensen wrote the original draft and was responsible for formal analysis and visualization; and all authors reviewed and edited the manuscript.

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2021081081/-/DCSupplemental.

Supplemental Table 1. Association between changes in, or time-averaged values of, biochemical parameters of mineral metabolism and changes in bone histomorphometry from baseline to 1 year post kidney transplantation.

Supplemental Table 2. Association between changes in, or time-averaged values of, biochemical parameters of mineral metabolism and changes in bone mineral density (BMD) from baseline to 1 year post kidney transplantation.

Supplemental Figure 1. The semi-quantitative diagnosis of bone turnover, mineralization, and volume at baseline (pre-transplant or time of transplantation) and 12 months post-transplant.

Supplemental Figure 2. Linear relationships between time-averaged concentration (TAC) of phosphate and absolute changes in bone histomorphometric variables of mineralization.

Supplemental Figure 3. Changes in circulating levels of biointact parathyroid hormone (PTH), intact pro-collagen type I N-terminal pro-peptide (PINP) and tartrate resistant acid phosphatase isoform 5b (TRAP5b) and annualized change in bone mineral density (BMD) at lumbar spine (LS) and total hip (TH).