Eight-week-old, male Wistar rats (N = 40) (Charles River Laboratories Inc., Tokyo, Japan) were housed in a controlled environment (temperature 23˚C ± 2˚C, humidity 40% ± 20%) with a 12-h light-dark cycle and free access to water and rat food. The details were described in a previous report and followed for the selection of rat species and sex [6]. Figure 1 shows the protocol of this experiment. Rats were classified into the following groups: a control group comprising 20 rats at two time points (n = 10 each at 22 weeks and 24 weeks of age), and a CKD group comprising 20 rats at two time points (n = 10 each at 22 weeks and 24 weeks of age). Animals were monitored daily for general condition and body weight. Humane endpoints were applied when >20% weight loss over 2 - 3 days or severe debilitation was observed. No animals required humane euthanasia, and no unexpected deaths occurred. However, due to rats dying under anesthesia during surgery by respiratory failure (one rat each in the 22 weeks and 24 weeks control groups) and technical errors (one rat each in the 22 weeks and 24 weeks both of control and CKD groups) that occurred during specimen preparation (e.g., section loss, staining defects), samples were analyzed for 8 rats at 22 weeks and 8 rats at 24 weeks in the control group and 9 rats at 22 weeks and 9 rats at 24 weeks in the CKD group.

Figure 1. Experimental protocol. Cont: non-CKD control rats, CKD: CKD rats.

To generate the CKD model, rats were fed a 0.75% adenine diet (Oriental Yeast Co., Ltd., Tokyo, Japan) for 4 weeks, until 12 weeks of age, followed by a standard rodent chow (CE-7; Clea Japan, Tokyo, Japan). The 4-week adenine diet treatment was selected based on previous studies [6][8], which reported that 4 weeks of feeding an adenine diet induced non-progressive, irreversible renal failure. Control rats were fed a standard diet without adenine throughout the study.

At 20 weeks of age, all rats underwent proximal tibial osteotomy [9]. Figure 2 shows a schematic illustration of the osteotomy procedure. Under general anesthesia with isoflurane (2% - 3%), a longitudinal incision was made on the anterior surface of the hind limb to expose the tibia. The tibia was completely transected using a bone saw. The fragments were secured with non-absorbable sutures, and the incision was closed with sutures. The animals were euthanized at 2 and 4 weeks post-osteotomy for evaluation (n = 8~9 in the CKD and control groups at each timepoint). Euthanasia at both 2 and 4 weeks post-osteotomy was performed under 2% - 3% isoflurane inhalation anesthesia followed by blood collection from vena cava, according to the approved protocol.

Figure 2.Schematic illustration of the surgical procedure of the cancellous bone osteotomy. A complete mid-sagittal osteotomy from the knee joint surface to the tibial outside diaphysis was performed using an electric bone saw (A, B). The fragments were secured with non-absorbable sutures (C).

A post-hoc power analysis was performed for all continuous outcomes. Using the observed effect sizes (Cohen’s d) and actual group sizes parse, the post hoc power analysis (α = 0.05, two-sided) showed that the number of rats for each group should be 8~9 for the micro-CT and hematoxylin and eosin (H&E) staining analyses. All animal experiments adhered to the protocols approved in advance by the Animal Care and Use Committee of Akita University (approval number a-1-0435, approved on September 12, 2022). Furthermore, all subsequent animal experiments were conducted in accordance with the Animal Care and Use Guidelines of our institute.

BW was measured weekly from the beginning (20 weeks of age) until the end of the experiment.

Blood samples were collected from the vena cava at 2 and 4 weeks post-osteotomy. The collected blood was centrifuged at 3,000 rpm for 30 min at 4˚C to separate the serum, which was aliquoted into disposable tubes and stored at −80˚C until analysis. Blood urea nitrogen (BUN), CRE, calcium (Ca), and inorganic phosphorus (IP) levels were measured using a Fujifilm Dri-Chem 3000V (Fujifilm Corp., Tokyo, Japan).

The excised tibia was secured in a specimen holder. Micro-CT imaging was performed using a Cosmo Scan GX II instrument (Rigaku Corp., Tokyo, Japan) according to the manufacturer’s instructions. The scanning parameters were set at an isotropic voxel size of 36 μm, voltage of 90 kVp, and current of 88 μA. The acquired images were analyzed using TRI/3D BON software (Ratoc System Engineering Co., Ltd., Tokyo, Japan). The cortical bone area/total bone area ratio (Ct.Ar/Tt.Ar [%]) and cortical thickness (Ct.Th [mm]) were determined to evaluate the cortical bone at the proximal tibial osteotomy site. Trabecular bone was evaluated by determining the bone volume (BV/TV [%]), trabecular thickness (Tb.Th [mm]), trabecular number (Tb.N [/mm]), trabecular separation (Tb.Sp, mm), structure model index (SMI), and degree of anisotropy (DA).

After micro-CT imaging, the proximal tibia of each rat was decalcified in 10% neutral ethylenediaminetetraacetic acid for approximately 2 weeks and embedded in paraffin. Mid-frontal slices (thickness: 3 µm) were then sectioned and stained with H&E and Safranin O for cancellous bone histomorphometry. It was difficult to prepare sections appropriate for evaluation for one or two samples in each group or time point. Thus, finally, a total of 8 samples in the control group and 9 samples in the CKD group were examined in micro-CT, bone union, and cartilage formation analyses.

The rate of bone union was defined as the length of continuous trabecular structure, as measured using ImageJ software [10][11] on the central slice of coronal sections from both H&E-stained and micro-CT images, divided by the total length of the osteotomy site in the tibia. Cartilaginous bonding was also considered as an indicator of bone union, whereas fibrous bonding was regarded as indicating non-union. Two authors measured the bone union rate in a blinded fashion. The intraclass correlation coefficient (ICC) measured three times by a single examiner (KT) (ICC 1,3) and ICC measured by two examiner (KT and MW) (ICC 2,2) were as follows: HE staining, 0.999 (ICC 1,3) and 0.980 (ICC 2,2), respectively; and micro-CT, 0.999 (ICC 1,3) and 0.986 (ICC 2,2), respectively.

The width of the osteotomy site was defined as the region within which the proximal tibial growth plate was disrupted. The region of interest (ROI) was set as a 1,200 μm × 1,200 μm area, starting 400 μm distal to the growth plate. The bone formation rate was calculated as the area of continuous trabecular structure, excluding fibrous tissue, divided by the total area of the ROI. This ROI was predefined to fully encompass the osteotomy site while ensuring between-specimen reproducibility based on constant anatomical landmarks.

The rate of cartilage formation was calculated using the same method used to calculate the bone union rate in Safranin O-stained images. Cartilage formation was defined as the length of bone positive for Safranin O staining divided by the length of the osteotomy. The cartilage formation rate was also measured by two authors in a blinded fashion. The ICC (1,3) and ICC (2,2) of cartilage formation were 0.999 and 0.967 respectively.

All data are expressed as the mean ± standard deviation (SD). Data were compared between the control and CKD groups using the Mann-Whitney U test. All statistical analyses were performed using EZR [12], which is a modified version of R Commander designed to add statistical functions frequently used in biostatistics. Differences were considered significant at P < 0.05.

The health status of the animals during the experiment was assessed by measuring the change in BW. At the time of osteotomy (20 weeks of age), the CKD group exhibited significantly lower BW compared to the control group (P < 0.01). At 2 weeks post-osteotomy (22 weeks of age), the BW of rats in the CKD group remained significantly lower than that of rats in the control group (P = 0.035). However, no significant difference in BW was observed between the two groups at 4 weeks post-osteotomy (24 weeks of age).

Serum biochemistry parameters were analyzed at 2 and 4 weeks post-osteotomy. The results are presented in Table 2. At both 2 and 4 weeks post-osteotomy, significant elevations in levels of serum BUN (P< 0.001, and P< 0.001, respectively) and CRE (P< 0.001, and P< 0.01, respectively) were observed in the CKD

Table 1. Body weight.

Mann-Whitney U test; Values are mean ± standard deviation.

Table 2. Serum biochemistry.

Mann-Whitney U test P < 0.05; BUN: blood urea nitrogen, CRE: creatinine, Ca: calcium, IP: inorganic phosphorus.

group compared to the control group, indicating impaired renal function in the CKD rats. No significant differences were observed in serum Ca and serum IP levels between the two groups at 2 and 4 weeks post-osteotomy. The combination of elevated serum creatinine with normal serum calcium and phosphate is characteristic of stage 3 CKD, supporting the validity of the present adenine-induced model.

3.2.1. Cortical Bone Evaluation

No significant differences in Ct.Ar/Tt.Ar were observed between the control and CKD groups at 2 and 4 weeks post-osteotomy. Similarly, there were no significant differences in Ct.Th. between the groups at either time point.

3.2.2. Trabecular Bone Evaluation

No significant differences in parameters of trabecular bone (BV/TV, Tb.Th, Tb.N, Tb, Sp, SMI, and DA) were observed between the control and CKD groups at 2 and 4 weeks post-osteotomy.

Although micro-CT demonstrated a gap in the osteotomy site in both the control (Figure 3(A)) and CKD (Figure 3(B)) groups at 2 weeks, the osteotomy site exhibited bony fusion in the control (Figure 3(C)) and CKD (Figure 3(D)) groups at 4 weeks post-osteotomy. Micro-CT images showed no significant differences between the control and CKD groups at either 2- or 4-week post-osteotomy.

Table 3. Micro-computed tomography at the proximal metaphysis of the tibia.

Mann-Whitney U test P < 0.05; Values are mean ± standard deviation; Ct.Ar/Tt.Ar: cortical area/total area, Ct.Th: cortical thickness, BV/TV: bone volume/tissue volume, Tb.Th: trabecular thickness, Tb.N: trabecular number, Tb.Sp: trabecular separation, SMI: structure model index, DA: degree of anisotropy.

Figure 3. Micro-CT images, central slice of coronal sections, showed no significant differences between the control (A and C, respectively) and CKD (B and D, respectively) groups at both 2 and 4 weeks post-osteotomy.

At 2 weeks post-osteotomy, most areas of the osteotomy site in both the control and CKD groups were filled with fibrous tissue (Figure 4(A) and Figure 4(B)) in H&E-stained sections. By 4 weeks post-osteotomy, both groups demonstrated thick trabecular bone and dense bone union at the osteotomy site (Figure 4(C) and Figure 4(D)).

Figure 4. Histological sections of the osteotomy site stained with hematoxylin and eosin. Arrowheads (σ) indicate sites of interruption of the growth plate. At 2 weeks post-osteotomy, most of the areas at the osteotomy site were filled with fibrous tissues (arrows) in both the control (A) and CKD (B) groups. At 4 weeks post-osteotomy, both the control (C) and CKD (D) groups demonstrated thick trabecular bone and dense bone union at the osteotomy site.

3.4.1. Bone Union Rate at the Osteotomy Site

Bone union rates were assessed using micro-CT and showed no significant differences between the control and CKD groups at either the 2- or 4-week post-osteotomy time points. Bone union rates, as assessed using H&E-stained sections, were significantly lower in the CKD group at 2 weeks post-osteotomy compared to the control group (P = 0.048). However, at 4 weeks post-osteotomy, no significant differences were observed between the control and CKD groups.

Table 4. Bone union/formation rates.

Mann-Whitney U test P < 0.05; Values are mean ± standard deviation. H&E: hematoxylin and eosin.

3.4.2. Bone Formation Rate at the Osteotomy Site

Bone formation rates assessed by H&E staining showed no significant differences between the control and CKD groups at both 2 and 4 weeks post-osteotomy.

Figure 5 shows histological sections of the osteotomy site stained with Safranin O. At 2 weeks post-osteotomy, cartilage formation was observed at the osteotomy site in the control group (Figure 5(A)), but not in the CKD group (Figure 5(B)). After 4 weeks, no cartilage formation was observed in either the control (Figure 5(C)) or CKD (Figure 5(D)) group. Cartilage formation rates assessed by Safranin O staining were significantly lower in the CKD group than in the control group at 2 weeks post-osteotomy (P = 0.017) (Table 5). However, no significant differences were observed between the control and CKD groups at 4 weeks post-osteotomy (Table 5).

Figure 5.Histological sections of the osteotomy site stained with Safranin O. Cartilage formation was observed at 2 weeks post-osteotomy in the control (A) group (arrows), and the rate of cartilage formation was lower in the CKD group (B) than in the control group (A). No significant changes between the control (C) and CKD (D) groups were observed at 4 weeks post-osteotomy.

Table 5.Cartilage formation rate.

Mann-Whitney U test P < 0.05; Values are mean ± standard deviation.

In this study, we used a rat model of adenine-induced CKD to evaluate whether prolonged bone healing occurs at the osteotomy site of the proximal tibia. Serological examinations confirmed that the model rats had stage 3 CKD, as serum CRE was elevated and Ca and IP were within the normal range. Micro-CT analysis showed no difference in the bone healing rate of the proximal tibial cancellous osteotomy between the control and CKD rats and no evidence of prolonged bone healing. However, bone tissue specimens showed a significant decrease in the rate of bone healing and cartilage formation in CKD rats at a relatively early stage of 2 weeks post-osteotomy. In contrast, there was no difference in bone healing rate at 4 weeks, which was later after osteotomy. The bone microstructure of the proximal tibial osteotomy showed no deterioration of either cortical or trabecular bone parameters in CKD rats. In conclusion, in the adenine-induced CKD rat model, bone healing of the proximal tibial cancellous bone osteotomy was prolonged in the early-stage post-osteotomy due to histologically delayed endochondral ossification, but histological analyses did not indicate prolonged healing at 4 weeks post-osteotomy.

CKD severely affects bone metabolism, resulting in decreased bone formation, increased bone resorption, and abnormal mineral metabolism [1]. These changes increase the risk of fracture in patients with CKD and are likely to result in prolonged bone healing after fracture. Indeed, in clinical practice, fractures are observed more frequently in CKD patients with a low estimated glomerular filtration rate [13], and an increased risk of hip fracture in these patients has also been reported [14]. In a clinical study of 130 patients with femoral neck fractures, those with CKD had a higher risk of pseudoarthrosis than those with normal renal function [4], and pseudoarthrosis of distal femur fractures in patients with renal osteodystrophy due to CKD has been reported [3]. Therefore, renal impairment may prolong bone healing or increase the risk of pseudoarthrosis. Animal studies have shown that CKD reduces BMD and negatively impacts the bone microstructure and bone strength [6][15][16]. Bone regeneration capacity is reportedly reduced with respect to bone healing in CKD models involving artificially created bone defects [17].

In this study, delayed bone healing was observed in the CKD group, especially in the early period post-osteotomy. Possible mechanisms for this delay include the following. Serum BUN and CRE levels were significantly higher in the CKD group than in the control group in this study, suggesting that the CKD was stage 3. This degree of renal dysfunction may have affected bone formation. Both bone resorption and bone formation are stimulated in adenine-induced CKD model rats, resulting in high bone metabolic turnover [6], which in turn leads to bone loss and bone fragility [6][16][18]. CKD also causes secondary hyperparathyroidism, which increases the number of osteoclasts and osteoblasts [17]. In our previous study, intact PTH levels were significantly higher in CKD rats aged 12 to 20 weeks than in control rats [6] under the same conditions to which the adenine-induced CKD model rats in this study were subjected. Because the same adenine-induced CKD protocol was used in this study, the markedly elevated PTH levels indicate the presence of secondary hyperparathyroidism in the current model as well, strengthening the interpretation that altered hormonal conditions contributed to delayed endochondral ossification. Therefore, secondary hyperparathyroidism may have also played a role in prolonged bone healing. Furthermore, FGF23 levels are elevated in CKD, which reportedly leads to decreased vitamin D receptor (VDR) expression, suggesting that decreased VDR expression may inhibit osteoblast differentiation and reduce osteogenic potential [19].

Previous studies have reported that CKD causes impaired endochondral ossification and growth retardation; in CKD model rats, the structure of the growth chondrocyte plate was disrupted, and the normal arrangement of chondrocytes was impaired. In particular, the height of mature hypertrophic chondrocytes was reduced, and the cell maturation process was disrupted. The proliferation rate of cells was also reduced in the growing chondrocyte version of the CKD rat model. The distribution of osteoclasts was also uneven, and their number was increased, resulting in an impairment of the trabecular bone microstructure. Furthermore, it has been reported that secondary hyperparathyroidism interferes with osteogenesis and calcification processes and inhibits the process of endochondral ossification [20]. The above factors may have delayed endochondral ossification and reduced the rate of early bone healing and cartilage formation. CKD also leads to decreased expression of vascular endothelial growth factor (VEGF), which is essential for angiogenesis and endochondral ossification in fracture healing. However, when VEGF expression is reduced due to CKD, vascular invasion into the fracture site is delayed, leading to pseudostratified bone. However, another study reported that when VEGF is decreased due to CKD, both vascular invasion into the fracture site and ossification of pseudo-bone are delayed [21]. In the present study, CKD may have caused impaired endochondral ossification and prolonged bone healing. In this study, VEGF immunostaining did not yield detectable signals. This was likely due to technical limitations associated with hard-tissue processing, including prolonged decalcification, which can reduce antigenicity. In addition, previous studies have reported decreased VEGF expression in CKD, which may have further contributed to the weak staining observed. The delayed cartilage maturation observed in CKD rats was evident in the Safranin O-stained sections in Figure 5, where persistent cartilaginous tissue and immature matrix were present at the osteotomy site compared with the controls. These findings support the interpretation of delayed endochondral ossification in the CKD group.

Notably, radiographic bone-bridging on micro-CT did not always correspond to histological union. A known limitation of micro-CT is its inability to distinguish mature lamellar bone from newly mineralized woven bone or cartilage. Therefore, micro-CT may overestimate apparent bony continuity compared with histology, which directly reflects tissue maturity. This discrepancy reflects methodological differences: micro-CT primarily captures the continuity of mineralized tissue and is affected by thresholding and spatial resolution, whereas histology reveals residual cartilage and tissue maturity at the osteotomy site. Thus, apparent bridging on micro-CT may precede the completion of endochondral ossification on histology, particularly under CKD conditions. Prior studies have implicated cyclin-dependent kinase 1 (Cdk1) in osteoblast proliferation and bone formation. Loss of Cdk1 impairs bone formation without abolishing the anabolic effects of PTH [22], and disease-related tissue changes can involve altered Cdk1-related pathways [23]. Although we did not assess Cdk1 in this study, the delayed endochondral maturation observed in CKD could be somewhat consistent with cell cycle-linked dysregulation. However, the present study focuses on phenotypic/histological findings, thus molecular-level mechanisms are speculative. Future work should examine Cdk1 expression and signaling in CKD-associated delayed union.

First, in this study, bone union was evaluated at 2 and 4 weeks post-osteotomy, but bone fusion was already complete at 4 weeks. Earlier observations, such as at 1-week post-osteotomy, could be necessary. Second, the detailed analysis of abnormalities in bone metabolism was insufficient because serum PTH and FGF23 levels were not measured. However, the model used in this study was the same as that used in a previous study [6], in which serum PTH was elevated, suggesting secondary hyperparathyroidism. In the future, we plan to investigate the possibility of promoting bone healing in CKD patients by combining drug administration and low-power ultrasound pulses. Third, the study examined a relatively low number of rats, even though the post hoc power analysis indicated the inclusion of 8~9 rats for micro-CT analysis and H&E staining in each group. Thus, it was not possible to completely avoid the risk of type II errors in this exploratory study. Finally, we did not perform a dynamic bone histomorphometry and evaluation of Osteoprotegerin/receptor activator of nuclear factor-kappa B ligand due to limited available resources. An evaluation of bone metabolic markers would be needed in future study.