Preoperative bone assessment by bone mineral density and bone turnover in patients undergoing total knee arthroplasty

Abstract

Background

The preoperative overall bone quality in patients undergoing total knee arthroplasty (TKA) might influence the development of postoperative bone-derived complications. This study was performed to simultaneously evaluate the baseline bone mineral density (BMD) and bone turnover in these patients.

Methods

We prospectively measured the lumbar spine, total hip, and femoral neck BMD using dual-energy X-ray absorptiometry (DXA) in 214 patients (239 knees; women, 199 knees; men, 40 knees) 1 day preoperatively. Women were stratified into three age groups: <70 years (60s), <80 years (70s), and ≥80 years (80s). We assessed bone turnover using the urinary level of N-telopeptide (NTx) normalized to creatinine (NTx/Cre).

Results

The prevalence of osteoporosis as assessed by DXA (T-score of ≤ −2.5) was 22% in women and 5% in men. A decline in BMD with aging was found in the hip but not in the lumbar spine with significant differences among ages. Women showed a higher NTx/Cre level than men (p = 0.033). An NTx/Cre level suggesting a high fracture risk (>56.5 mmol BCE/mmol Cre) was present in 54% of women and 35% of men (p = 0.037). No significant differences were found in this prevalence among age groups in women. Seventeen percent of women and 0% of men had both osteoporosis and a high fracture risk simultaneously (p = 0.002). The proportion of women in this category was lower in the 60s than 70s age group (p = 0.019).

Conclusions

Preoperative bone health screening before TKA might be necessary for women aged ≥70 years to prevent complications caused by perioperative poor bone condition.

1. Introduction

Osteoporosis prior to total knee arthroplasty (TKA) is an area of increased focus within the field of arthroplasty, although the prevalence depends on sex, race, age, lifestyle, and environment. In fact, various papers have reported the prevalence of osteoporosis in patients undergoing TKA.1, 2, 3, 4, 5, 6, 7 Aseptic loosening, subsidence of the tibial implant, and fractures around implants, which are the main reasons for revision TKA surgery, reportedly occur secondary to poor bone quality (including osteoporosis) around the implants as a patient-related factor.⁸ Therefore, several factors that affect bone mineral density (BMD) around implants have been reported in long-term follow-up study,⁹ and preoperative BMD evaluation has been performed as a baseline for follow-up.

The systemic pathologic condition of osteoporosis contributes to the high incidence of fragility fractures, especially vertebral and hip fractures, in the elderly.^10,11 These post-TKA fractures have negative impacts on the clinical result.¹² Optimization of bone health may help mitigate these complications and is thus an important perioperative consideration. In some studies to date, only BMD was evaluated without the use of other bone assessment tools,^2,3,5,6 only women were evaluated,^2,5,7 or both patients who underwent TKA and those who underwent total hip arthroplasty were evaluated simultaneously.^1,4,6 However, preoperative bone quality and bone turnover in patients undergoing TKA are neither routinely nor simultaneously assessed as a systematic evaluation during pre-TKA planning.

Therefore, the present study was performed to compare the lumbar and hip BMD and bone metabolism markers between women and men as a systemic evaluation of patients with advanced knee osteoarthritis undergoing TKA. In addition, women, who tend to have a higher incidence of osteoporosis than men,¹³ were further analyzed to clarify the impact of age on BMD and bone turnover.

2. Materials and methods

Our local institutional review board (The Research Board of Healthcare Corporation Ashinokai, Gyoda, Saitama, Japan (ID number: 2021–1)) approved the present study. All patients signed a consent form that included a description of the protocol. This study was a retrospective evaluation of a prospective database. In total, 214 primary osteoarthritic patients (239 knees) (women: 179 patients, 199 knees; men: 35 patients, 40 knees) who underwent TKA from October 2009 to March 2021 were analyzed. The median age of women was 73 years (range, 55–89 years), and that of men was 77 years (range, 37–87 years) (p = 0.232). In addition, the women were stratified into three age groups: <70 years (60s), <80 years (70s), and ≥80 years (80s). Twenty-five patients underwent scheduled staged bilateral TKA with an average interval of 21 months (range, 4–78 months) between the first and second surgery. Thirty-four (17%) of 199 knees among the women were treated with medication for osteoporosis. The exclusion criteria were diagnosis of rheumatoid arthritis and the performance of revision surgeries. Each patient completed a questionnaire regarding their demographic data, medical history, and medication history. The patients’ clinical characteristics are summarized in Table 1.

Table 1.

Characteristics of preoperative patients.

Variables (patients/knees)	Women (179/199)	Men (35/40)
Body height (cm)	149 (145, 153)	159.5 (156, 165)
Body weight (kg)	58 (51, 66)	66 (61.3, 73.9)
Body mass index (kg/m2)	26 (24, 28)	26 (24, 28)
Age (years)	73 (69, 78)	77 (65, 82)
Hospital for Special Surgery score14	43 (34, 50)	49 (42, 57)
Osteoarthritis Grade* (Knees)	III 5, IV 194	III 2, IV 38
Medication for osteoporosis (Knees) (%)	34(17)	0 (0)

Values are expressed as median (interquartile range). *Osteoarthritis grade was evaluated using the Kellgren–Lawrence radiographic grading scale.¹⁵

2.1. Bone mineral density

BMD was measured by dual-energy X-ray absorptiometry (DXA) using a Lunar Prodigy Primo densitometer (GE Healthcare, Chicago, IL, USA) the day before surgery. Preoperative scans of the anteroposterior spine, total hip, and femoral neck were performed using the settings and patient positioning recommended by the manufacturer. The scanning precision was calculated as the coefficient of variation between the scan BMD values. In this series, the precision of the measurements was 2.0% in the lumbar spine (L1–L4), total hip, and femoral neck. The results are expressed as the spine, total hip and femoral neck BMD; T-score; and Z-score. The T-score is the number of standard deviations from the mean BMD of the young adult population, and the Z-score is the number of standard deviations from the age-matched mean BMD. Osteoporosis is defined by a T-score of −2.5 or less, and osteopenia is defined by a T-score between −2.5 and −1.0 in the lumbar, total hip and femoral neck region according to the World Health Organization international reference standard.¹⁶

2.2. Urinary N-telopeptide

N-telopeptide (NTx) is a cross-linker for mature type I collagen that is specific to bone and is released into the circulation during the process of bone resorption. It is excreted unmetabolized in urine, making it a sensitive and specific marker of bone resorption during osteoporosis.¹⁷ NTx was measured using an enzyme-linked immunosorbent assay (Osteomark® kit; Alere Inc., Waltham, MA, USA) 1 day before surgery. For normalization, creatinine (Cre) was measured using the second morning voided urine samples as a baseline before surgery. The precision was calculated as the coefficient of variation (3.8%) in this series. According to the guidelines of the Japanese Osteoporosis Society, the reference ranges for Cre-normalized urinary NTx have been established using a commercially available enzyme immunoassay as 9.3–54.3 mmol bone collagen equivalent (BCE)/mmol Cre in premenopausal females, 14.3–89.0 mmol BCE/mmol Cre in postmenopausal females, and 13.0–66.2 mmol BCE/mmol Cre in males.¹⁸ In addition, NTx values can suggest bone diseases (>54.3 mmol BCE/mmol Cre), metastatic bone tumor (>89.0 mmol BCE/mmol Cre), or abnormal calcium metabolism (>66.2 mmol BCE/mmol Cre). Finally, the cutoff level of NTx that indicates a high risk of fracture is 54.3 mmol BCE/mmol Cre.¹⁸

2.3. Statistical analysis

Because the normality assumption was violated, non-parametric tests were conducted. Data are expressed as median (25th, 75th percentile). The distribution of continuous variables was compared between groups using the Wilcoxon rank sum test and the Kruskal–Wallis test. Discrete variables were compared using Fisher's exact test. Multiple comparisons were performed using Bonferroni's method. In all tests, a p-value of <0.05 was considered significant. All statistical analyses were performed using IBM SPSS Statistics version 23 (IBM Japan, Tokyo, Japan).

3. Results

In a comparison between women and men, men had significantly higher BMD values in all anatomical sites in this study. In women and men, the BMD of the lumbar spine was 1.00 (0.89, 1.16) and 1.26 (1.07, 1.40), respectively (p < 0.001); the total hip BMD was 0.78 (0.71, 0.85) and 0.83 (0.78, 0.97), respectively (p < 0.001); and the femoral neck BMD was 0.72 (0.65, 0.80) and 0.79 (0.71, 0.94), respectively (p < 0.001) (Table 2). The T-score showed osteopenia (T-score between −2.5 and −1.0) in the total hip and femoral neck in women and in the femoral neck in men (Table 2). Of the 34 patients taking an anti-osteoporosis medication, 9 showed a T-score of ≤ −2.5 and 25 did not. The Z-score was positive in three sites for both women and men, indicating better BMD than those in the age-matched general population (Table 2).

Table 2.

Comparison of preoperative bone mineral density and NTx/Cre between women and men before total knee arthroplasty using Wilcoxon rank sum test.

Median (IQR)	Site	Women (N = 199)	Men (N = 40)	Sig.
Absolute value	Lumbar	1.008 (0.894, 1.162)	1.260 (1.098, 1.395)	<0.001
	Total-Hip	0.777 (0.715, 0.852)	0.829 (0.781, 0.964)	<0.001
	Femoral neck	0.724 (0.648, 0.796)	0.786 (0.711, 0.927)	<0.001
T-score	Lumbar	−0.8 (−1.8, 0.5)	0.7 (−0.83, 1.88)
	Total-Hip	−1.3 (−1.8, −0.6)	−0.9 (−1.3, 0.2)
	Femoral neck	−1.5 (−2.1, −0.9)	−1.3 (−1.8, −0.5)
Z-score	Lumbar	1.0 (0.0, 1.9)	1.9 (0.6, 2.9)
	Total-Hip	0.6 (0.2, 1.2)	0.4 (0.0, 1.5)
	Femoral neck	0.7 (0.2, 1.2)	0.4 (−0.4, 1.2)
NTX/Cre		55.5 (40.2, 69.8)	46.4 (38.7, 60.2)	0.033

IQR: interquartile range, Sig.: significance, NTx/Cre: urinary level of N-telopeptide normalized to creatinine. Boldface p-values are considered statistically significant.

A comparison of women by age group showed that the BMD of the lumbar spine was 1.05 (0.93, 1.20), 0.98 (0.86, 1.13), and 1.02 (0.94, 1.14) in the 60s, 70s, and 80s age groups, respectively (p = 0.015); that of the total hip was 0.82 (0.75, 0.93), 0.78 (0.73, 0.85), and 0.72 (0.65, 0.81), respectively (p < 0.001); and that of the femoral neck was 0.79 (0.69, 0.85), 0.73 (0.64, 0.78), and 0.66 (0.59, 0.74), respectively (p < 0.001) (Table 3). The results of multiple comparisons corrected by Bonferroni's method are shown in Table 3. The lumbar spine showed higher BMD in the 80s than 70s age group and no significant difference compared with the 60s age group. Declining BMD with aging was found in the total hip and femoral neck but not in the lumbar spine (Table 3).

Table 3.

Comparison of preoperative bone mineral density and NTx/Cre among ages in women before total knee arthroplasty using Kruskal–Wallis test with multiple comparisons corrected by Bonferroni's method.

Median (IQR)	Age	60s (N = 56)	70s (N = 104)	80s (N = 39)	Sig.
Absolute Values	Lumbar	1.051 (0.941, 1.204)	0.984 (0.858, 1.125)	1.016 (0.943, 1.126)	0.015(1)
	Total-Hip	0.824 (0.755, 0.926)	0.775 (0.727, 0.848)	0.722 (0.658, 0.806)	<0.001(2)
	Femoral neck	0.789 (0.688, 0.853)	0.726 (0.644, 0.783)	0.663 (0.597, 0.737)	<0.001(3)
T-score	Lumbar	−0.5 (−1.5, 0.8)	−1.1 (−2.1, 0.3)	−0.8 (−1.4, −1.0)
	Total-Hip	−0.9 (−1.5, −0.1)	−1.3 (−1.8, −0.7)	−1.7 (−2.3, −1.0)
	Femoral neck	−1.0 (−1.8, −0.4)	−1.5 (−2.1, −1.0)	−2.0 (−2.6, −1.3)
Z-score	Lumbar	0.9 (−0.2, 1.8)	0.9 (−0.1, 1.7)	1.5 (1.0, 2.4)
	Total-Hip	0.5 (−0.2, 1.0)	0.7 (0.2, 1.3)	0.8 (0.3, 1.2)
	Femoral neck	0.8 (0.2, 1.3)	0.6 (0.2, 1.2)	0.5 (0.0, 1.0)
NTX/Cre		54.7 (44.1, 78.1)	56.8 (39.5, 68.5)	49.1 (35.0, 67.2)	0.297

IQR: interquartile range, Sig.: significance, NTx/Cre: urinary level of N-telopeptide normalized to creatinine. <70 years: 60s, <80 years: 70s, and ≥80 years: 80s. Boldface p-values are considered statistically significant.

The results of multiple comparisons are as follows: ⁽¹⁾p = 0.013 between 60s and 70s; ⁽²⁾p < 0.001 between 60s and 80s, p = 0.019 between 70s and 80s; and ⁽³⁾p = 0.009 between 60s and 70s, p = 0.029 between 70s and 80s, p < 0.001 between 60s and 80s.

The number of knees showing osteoporosis by DXA (T-score of ≤ −2.5) at any anatomical site was 44 (22%) of 199 in women and 2 (5%) of 40 in men. The prevalence of osteoporosis between women and men among the three anatomical sites was 22 (11%) and 2 (5%) in the lumbar spine (p = 0.387), 19 (10%) and 0 (0%) in the total hip (p = 0.05), and 24 (12%) and 1 (3%) in the femoral neck (p = 0.089) (Table 4). Among these women, osteoporosis was found in the lumbar spine in 5 (23%) of 22 patients, in the total hip in 4 (21%) of 19 patients, and in the femoral neck in 4 (17%) of 24 patients despite taking an anti-osteoporosis medication. In a comparison of prevalence of osteoporosis among women of each age group, the 70s group showed the highest prevalence in the lumbar spine. The prevalence of osteoporosis in the total hip and femoral neck increased with age with significant differences between the 60s and 80s groups (p = 0.018 in total hip, p < 0.001 in femoral neck) (Table 5).

Table 4.

Comparison of prevalence of osteoporosis by dual-energy X-ray absorptiometry (T-score of ≤ −2.5) in lumbar spine and hip between women and men before total knee arthroplasty using Fisher's exact test.

Anatomical Site; N (%)	Women (N = 199)	Men (N = 40)	Sig.
Lumbar	22 (11)	2 (5)	0.387
Total-Hip	19 (10)	0 (0)	0.050
Femoral neck	24 (12)	1 (3)	0.089

Sig.: significance.

Table 5.

Comparison of prevalence of osteoporosis by dual-energy X-ray absorptiometry (T-score of ≤ −2.5) in lumbar spine and hip among ages in women before total knee arthroplasty using Fisher's exact test with multiple comparisons corrected by Bonferroni's method.

Anatomical Site; N (%)	60s (N = 56)	70s (N = 104)	80s (N = 39)	Sig.
Lumbar	3 (5)	18 (17)	1 (3)	0.014(1)
Total-Hip	1 (2)	11 (11)	7 (18)	0.018(2)
Femoral neck	1 (2)	12 (12)	11 (29)	< 0.001(3)

Sig.: significance. <70 years: 60s, <80 years: 70s, and ≥80 years: 80s. Boldface p-values are considered statistically significant.

The results of multiple comparisons are as follows: ⁽¹⁾There was a significant difference in the comparison among the three groups, but no significant combination of the two groups could be detected in the multiple comparison. ⁽²⁾p = 0.023 between 60s and 80s, ⁽³⁾p < 0.001 between 60s and 80s.

Finally, with regard to NTx/Cre, women showed significantly higher values than men (p = 0.033) (Table 2). NTx/Cre exceeded the upper bound of the reference range (>89.0 mmol BCE/mmol Cre in women and >66.2 mmol BCE/mmol Cre in men) in 26 knees in women and 6 knees in men (p = 0.805). An NTx/Cre level consistent with a high risk of fracture (NTx/Cre of >56.5 mmol BCE/mmol Cre) was present in 107 (54%) knees in women and 14 (35%) knees in men (p = 0.037) (Table 6A). No significant differences were found in this prevalence among ages in women (Table 6B). Furthermore, 17% of knees in women (33 of 199 knees) and 0% in men had both osteoporosis at any of the measurement sites and a high fracture risk simultaneously (p = 0.002) (Table 6A). Of the 34 patients taking an anti-osteoporosis medication, 4 showed a T-score of ≤ −2.5 and NTx/Cre level of >56.5 mmol BCE/mmol Cre simultaneously. Thus, without the use of anti-osteoporosis medications, the remaining 30 knees could be added to this category (33 knees), in which case the prevalence reached 63 (32%) of 199. With respect to the age of women who met this category, 3 were in their 60s, 23 were in their 70s, and 7 were in their 80s (p = 0.016 and p = 0.019 between 60s and 70s) (Table 6B).

Table 6.

Comparison of prevalence of NTx/Cre level higher than the upper bound of the reference range (66.2 in men and 89.0 in women), NTx/Cre level indicating a high risk of fracture (>54.3 mmol BCE/mmol Cre), and the presence of both osteoporosis at any measurement site (T-score of ≤ −2.5) and a high fracture risk simultaneously using Fisher's exact test with multiple comparisons corrected by Bonferroni's method.

(A): Comparison between women and men
N(%)	Women (N = 199)	Men (N = 40)	Sig.
higher than the upper bound of the reference range a	26(13)	6(15)	0.805
high fracture risk group b	107(54)	14(35)	0.037
Osteoporosis c & high fracture risk group b	33 (17)	0 (0)	0.002

(B) Comparison among ages in women
	60s (N = 56)	70s (N = 104)	80s (N = 39)	Sig.
higher than the upper bound of the reference range a	8 (14)	14 (13)	4 (10)	0.922
high fracture risk group b	31 (55)	59 (57)	17 (44)	0.377
Osteoporosis c & high fracture risk group b	3 (5)	23 (22)	7 (18)	0.016(1)

Sig.: significance, NTx/Cre: urinary level of N-telopeptide normalized to creatinine. ^aNTx/Cre of >66.2 mmol BCE/mmol Cre in men and >89.0 mmol BCE/mmol Cre in women, ^bNTx/Cre of >54.3 mmol BCE/mmol Cre, ^cT-score of ≤ −2.5. <70 years: 60s, <80 years: 70s, and ≥80 years: 80s. ⁽¹⁾p = 0.019 between 60s and 70s. Boldface p-values are considered statistically significant.

4. Discussion

This study has several important findings. First, all of the preoperative median BMD values were significantly lower in women than men, as reported in the general population.¹³ In addition, the prevalence of osteoporosis at any site was 22% in women and 5% in men. Second, a comparison among age groups in women showed different characteristics between the lumbar spine and hip. That is, the median BMD values of the lumbar spine were higher in the 80s than 70s age group, and the prevalence of osteoporosis was higher in the 70s age group than in the other groups. Additionally, both the total hip and femoral neck BMD declined with aging, and the prevalence rate increased with aging. Third, the median NTx/Cre levels in women were significantly higher than those in men. Additionally, the prevalence of a high risk of fracture was significantly higher in women (>50%) than in men. Finally, 17% of women and 0% of men had both osteoporosis (T-score of ≤ −2.5) at any of the measurement sites and a high risk of fracture (NTx/Cre level of >56.5 mmol BCE/mmol Cre) simultaneously. These patients might readily develop bone-related complaints owing to preoperative poor bone quality at baseline and sequential bone resorption unless appropriate measures are taken.

With regard to the sex-related difference in the preoperative prevalence of osteoporosis, the current study is consistent with previous studies,^3,6 in which women showed a higher prevalence than men. Actually, the prevalence of osteoporosis in any sites in the present Japanese study was 22% in women and 5% in men. Investigations performed in other countries have found the prevalence to range from 3% to 33% in the United States or United Kingdom^1,4,6 and from 31% to 60% in Asia^2,3,7 in women. Because 25 (13%) of 199 knees in women without osteoporosis at any measurement site were being treated with osteoporosis medications in this study, the prevalence might have been underestimated by that amount. With consideration of these patients, the maximum prevalence of osteoporosis in this study seems to be 35% (69/199) in Japanese women, which is comparable with previous studies from other countries.^1,2 Conversely, very low prevalences such as 1% in the lumbar spine, 6% in the forearm,⁶ and 7% in the lumbar spine and proximal femur³ have been reported in men. However, based on positive Z-scores in all measurement sites, the preoperative patients (both women and men) had better spine and hip BMDs than those in the age-matched general population. Further studies are necessary to analyze the reasons for the ethnic differences in BMD in terms of genetics, physical activity levels, diet, and lifestyle factors.

Next, in the evaluation of preoperative BMD as a site-specific feature in women, the BMD of both the total hip and femoral neck declined with age, as previously reported in the general population.^19,20 Meanwhile, although the BMD of the lumbar spine was expected to also decline with age as reported in the general population,²⁰ the BMD was higher in the 80s than 70s age group and not significantly different from that in the 60s age group. We speculate that the degree of degeneration of the lumbar spine might have impacted the results, as previously reported.^21,22 Muraki et al.²² reported that degenerative spinal diseases are associated with increased lumbar spine BMD measurements. The femoral neck BMD may therefore be more appropriate than the lumbar spine BMD in evaluating osteoporosis in elderly women. Liu et al.²¹ reported that diagnosis of osteoporosis and assessment of the osteoporotic fracture risk in the elderly should be based on hip BMD and not on anteroposterior lumbar spine BMD unless spinal osteoarthritis has been excluded. Thus, the degeneration of the lumbar spine in the 80s age group might have been much further advanced than that in the 60s and 70s age groups in the present study, although the degree of lumbar degeneration was not confirmed in this study. Taking these previous studies into account, measurement of hip BMD is likely to be more appropriate for evaluating preoperative osteoporosis before TKA than measurement of lumbar spine BMD.

Few reports to date have focused on bone metabolism markers before TKA. A comparison of sex-related differences in bone resorption markers was performed with a focus on urinary deoxypyridinoline,⁴ showing that women had higher values than men; this is consistent with the NTx measurements in the present study. The prevalence of the patients in the high fracture risk group was higher in women than in men. However, a comparison among various ages in women showed neither an increase in NTx with age nor an increase the prevalence in the high fracture risk group with age. Therefore, considering that physical activity is associated with BMD,¹⁹ disuse osteoporosis before TKA owing to a low level of physical activity related to pain caused by advanced knee osteoarthritis may lead to abnormal bone turnover and increased bone resorption resulting from decreased stimulation of bones. The results of this study suggest that women might be more susceptible to the effects of the amount of physical activity than men and that the effects of aging on bone resorption markers are lower in postmenopausal women.

More than half of the patients in the high fracture risk group were women, suggesting that administration of bone resorption inhibitors such as bisphosphonate might effectively improve such abnormal bone turnover. Two studies supported this speculation. Teng et al.²³ reported that bisphosphonate significantly reduced the rate of revision TKA surgery. Another study showed that bisphosphonate increases the T-score for hip BMD after TKA.⁵ Conversely, several studies showed no beneficial effects of bisphosphonate treatment on BMD after TKA.^24,25 In addition, 17%–23% of patients in the present study showed osteoporosis at any measurement site despite the administration of therapeutic agents for bone metabolism, including 5 of 20 knees in patients who received bisphosphonates. Therefore, the therapeutic effect of bisphosphonate administration is not yet definitive. In any case, high NTx levels suggest abnormal bone turnover and the need for effective treatment, requiring further evidence-based investigation from various perspectives for osteoporosis-induced complaints.

From a medical economic point of view, it is not efficient to evaluate bone quality and its turnover in all preoperative TKA patients. Thus, it seems important to determine which patients should be evaluated. The present study revealed the prevalence of patients who were regarded as having a higher risk of bone-derived complaints secondary to preoperative poor bone storage (i.e., osteoporosis) and abnormal bone resorption (i.e., high resorption marker levels). Women in their 60s and men were likely to be at lower risk for these problems and to thus have less need for their evaluation. However, if women aged ≥70 years regarded as having higher risk are left unattended, various bone-derived problems might be more likely to occur. Accordingly, it seems important to carefully select women aged ≥70 years who meet the adaptation criteria of the fracture risk assessment tool and/or the National Osteoporosis Foundation guidelines for BMD screening¹³ as reported by Bernatz et al.¹

This study had several limitations. First, the relatively small size of this study limits the reliability of the data with respect to the conclusions and comparisons that can be drawn. Second, this study involved Japanese women aged ≥55 years and men. Thus, the results may not be applicable to other populations, such as younger patients with different ethnicities. Third, the numbers of women and men and the distributions among their ages were uneven. However, female predominance is common,³ and the 70s is commonly reported as the mean age,^2,5 especially in BMD-related studies of Asian patients. Despite these limitations, the main strength of this study is that to the best of our knowledge, it is the first study to simultaneously evaluate preoperative BMD and bone turnover in women and men, focusing only on preoperative TKA patients.

In conclusion, the current study has shown that the prevalence of occult osteoporosis and/or excessive turnover of bone resorption prior to TKA is substantial. We recommend bone health screening in select arthroplasty patients preoperatively, especially for women aged ≥70 years, to prevent complications caused by perioperative poor bone condition. Further research is required to investigate the utility and cost-effectiveness of regular BMD and NTx/Cre screening prior to TKA as well as the clinical benefit of perioperative use of antiosteoporosis medications.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding

None declared.

Authors’ contributions

YI contributed to the study conception and design, drafted the article, and ensured the accuracy of the data and analysis. HN, JS, and IT contributed to the study conception and design and to the analysis and interpretation of the data. HI, RI and KI contributed to the data collection. ST provided statistical expertise and contributed to ensuring the accuracy of the data and analysis. All authors approved the final manuscript.

Ethical review committee statement

The local institutional review board approved this study (The Research Board of Healthcare Corporation Ashinokai, Gyoda, Saitama, Japan (ID number: 2021–1)). All patients provided informed consent.

Declaration of competing interest

Yoshinori Ishi, Hideo Noguchi, Junko Sato, Ikuko Takahashi, Hana Ishii. Ryo Ishii, Kei Ishii, Shin-ichi Toyabe declare that they have no conflict of interest.