DOI: https://doi.org/10.21203/rs.2.14921/v2
Background: The relationship between renal function and bone mineral density (BMD) is controversial. The aim of this study was to determine the relationship of renal function with BMD and osteoporosis risk in healthy postmenopausal Chinese women.
Methods: A cross-sectional study was conducted in 776 healthy postmenopausal Chinese women. Dual-energy X-ray absorptiometry was used to measure BMDs. Clinical, demographic, and biochemical data were obtained at the time of image acquisition. Estimated glomerular filtration rate (eGFR) was calculated using a Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.
Results: Women with eGFR levels of at least 90 ml/min/1.73m2 had a lower prevalence of osteoporosis compared with women with decreased eGFR levels (60 ml/min/1.73 m2 ≤ eGFR < 90.0 ml/min/1.73 m2). BMDs at femoral neck and total hip were significantly lower in the lower eGFR class than the higher class (0.717 ± 0.106 vs 0.744 ± 0.125 g/cm2, P < 0.01; 0.796 ± 0.116 vs 0.823 ± 0.129 g/cm2, P < 0.01, respectively). eGFR was positively correlated with BMDs at femoral neck and total hip in unadjusted analysis (P <0.05). After controlling for age, menopausal duration and body mass index (BMI), decreased eGFR was not associated with osteoporosis risk.
Conclusions: After adjustments for age, menopausal duration and BMI, the decline in renal function was not independently associated with osteoporosis risk in healthy postmenopausal Chinese women.
Renal function declines with age such that close to 95% of women aged 75 years have a mild to moderate reduction of renal function [1]. Currently, we are experiencing an unprecedented rise in the number of older adults [2], and consequently, the prevalence of many age-related diseases will increase, including chronic kidney disease (CKD) and osteoporosis [3, 4]. Studies show that patients with renal dysfunction exhibit reduced bone mineral density (BMD) and increased risk of fracture [5–9]. Renal dysfunction influences bone in various ways, such as secondary hyperparathyroidism, calcium and phosphorus metabolism disorder, and vitamin D deficiency [10, 11]. These effects on mineral homeostasis may be associated with increased bone fragility [12]. However, there are conflicting reports, on whether renal function affects BMD or fracture risk. Elliott et al. [13] showed no association between renal dysfunction and fracture risk. Fujita et al. [14] reported that low renal function is not associated with decreased BMD in community-dwelling elderly men.
Osteoporosis is a serious problem affecting a large number of elderly people. Several studies reported a high prevalence of osteoporosis in postmenopausal women [11, 15, 16]. The prevalence of osteoporosis in China was reported as 6.46% and 29.13% for men and women aged 50 years and older [17]. Due to the increase of proportion of the elderly population, osteoporosis and osteoporotic fractures have become global health issues of major concern [18, 19]. The annual number of osteoporosis-related fractures is predicted to be 5.99 million by 2050, which will cost approximately $25.43 billion annually [20].
Although other studies have investigated the relationship between renal function and BMD, they are mostly in defined kidney disease [5, 21]. The data about the effect of renal function on BMD and osteoporosis in otherwise healthy Chinese postmenopausal women are scare. Therefore, the aim of the present study was to investigate the relationship of renal function (using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) 2009 equations to calculate estimated glomerular filtration rate (eGFR)) with BMD and osteoporosis risk in healthy Chinese postmenopausal women.
The design and operation of this study have been described [22]. Briefly, a total of 1096 urban postmenopausal Chinese women were recruited at random. Subjects with any pathological disorders (such as diabetes mellitus, hyperthyroidism, oligomenorrhoea, malabsorption, rheumatoid arthritis, or hepatic dysfunction) or subjects using medications (such as glucocorticoids, bisphosphonate, oestrogen, thyroid hormones or statins) known to alter bone metabolism were excluded from our study. Of these, 232 subjects had to be excluded. 15 women exited because of lacking of time. 14 subjects had to be excluded because of impaired renal function (eGFR < 60 ml/min/1.73m2) [23]. After excluding 59 participants with abnormal laboratory test results, 776 qualifying healthy postmenopausal women were selected. All subjects agreed to participate in the study and gave informed consent. The study was approved by the ethics committee of the Second Xiangya Hospital of Central South University (Changsha, China) and was in compliance with the Helsinki Declaration.
Each participant answered a standardized questionnaire, and underwent a clinical examination. The questionnaire included name, age, menopausal duration, marital status, educational background, smoking or alcohol intake, physical exercise, list of medications and history of fractures. Physical exercise was defined as ≥ 30 min/day. Body mass index (BMI) was calculated as weight (kilograms) divided by the square of the height (metres). BMD values were measured by dual-energy X-ray absorptiometry (DXA) fan-beam bone densitometer (Lunar Prodigy Advance, GE Healthcare, Madison, WI, USA) at the total hip, left femoral neck, and lumbar spine (L1–L4). Subjects were considered to have normal BMD if there T-score was > –2.5 standard deviation (SD) at lumbar spine, femoral neck and total hip, and osteoporosis if there T-score was ≤ –2.5 SD in at least one skeletal site [24, 25].
Blood samples were collected after an overnight fast, and serum samples were stored at –70℃ until assayed. Serum calcium (SCa) levels were measured by calcium kit according to the manufacturer’s instructions (MedicalSystem Biotechnology CO., Ltd, Ningbo, China). Serum creatinine (SCr) levels were measured by enzymatic method (Kanto Chemical, Tokyo, Japan). Blood urea nitrogen (BUN) levels were measured by enzymatic method (Abbott, Chicago, USA). eGFRs were computed according to age, sex, ethnicity, and the SCr level data from each participant, using the CKD-EPI 2009 equations [14]. Participants were staged into categories of renal function based on their eGFR alone as follows: (1) normal eGFR ( 90.0 ml/min/1.73m2), (2) mild decreased eGFR (60 ml/min/1.73m2 ≤ eGFR< 90.0 ml/min/1.73m2) [23].
Log transformation was used for BUN, SCr, eGFR, TG, and HDL-C levels when their results were not normally distributed. Data are presented as mean SD for continuous variables, median (25,75) for skewed variables, and percentages for categorical variables. We used the one-way ANOVA test to compare continuous characteristics, and the chi-squared test or Fisher exact test to compare percentages among groups. The correlations between renal function and independent variables were examined by Spearman’s correlation coefficient. To evaluate the contribution of renal function to the presence of osteoporosis, multivariable logistic analyses were performed. Adjustments were made for age, menopausal duration, and BMI. The accepted level of statistical significance was P < 0.05. All statistical procedures were carried out with SPSS for Windows version 13.0 (SPSS Inc., Chicago, IL, USA).
The characteristics of the study participants are presented in Table 1. A total of 776 relatively healthy postmenopausal Chinese women participated in the present study. The mean age was 62.1 6.1 years. Osteoporosis in at least one site (lumbar spine, femoral neck or total hip) was found in 37.2% of participants. We compared various indices between subjects with and without osteoporosis. As shown in Table 1, participants with normal BMD were younger and had higher BMI and shorter menopausal duration (P < 0.001).
The analysis according to renal function (Table 2) showed that participants with eGFR 90 ml/min/1.73m2 were younger and had shorter menopausal duration (P < 0.001). BMD values at femoral neck and total hip, but not at lumbar spine, were significantly lower in women with declining renal function (P < 0.01). As renal function decreased, the percentage of participants with osteoporosis increased (P <0.05). As shown in Table 3, eGFR was negatively and significantly correlated with age and menopausal duration (P < 0.01). The eGFR was positively and significantly correlated with femoral neck BMD and total hip BMD (P < 0.05). SCr was positively correlated with age (P <0.01), menopausal duration, and TC levels (P < 0.05). BUN was positively correlated with age, menopausal duration (P < 0.01), TC, LDL-C, and SCa levels (P < 0.05). eGFR is computed according to age and SCr concentration. BMD could be influenced by age and menopausal duration. Therefore, we performed subgroup analysis stratified by age and menopausal duration. eGFR did not correlate with BMD at all sites after stratification by age or menopausal duration (Table S1, S2).
Table 4 shows the results of binary logistic regression analyses using osteoporosis as the dependent variable and eGFR groups as the independent variables. In the unadjusted model, lower eGFR class was associated with an increased risk of osteoporosis as compared to the higher eGFR class (odds ratio (OR): 1.38, 95% confidential interval (CI): 1.02–1.86, P < 0.05). However, after adjustment for age, menopausal duration, and BMI, the lower eGFR class was not related to an increase risk of osteoporosis as compared to the higher eGFR class (OR: 0.94, 95%CI: 0.67–1.33, P > 0.05). Subgroup analysis by age and menopausal duration also showed that the decline in renal function was not associated with osteoporosis risk (Table S3, S4).
Several studies have explored the association between renal function and BMD [5–9, 13, 14]. However, the literatures are conflicting, and the relationship between renal function and the risk of osteoporosis in healthy older population is rare. In the current study, we examined the relationship of renal function with BMD and the risk of osteoporosis in 776 relatively healthy postmenopausal Chinese women. Our present study found that participants with worse renal function were associated with lower femoral neck and total hip BMD. However, after adjusting for age, menopausal duration and BMI, the decline in renal function was not associated with increased risk of osteoporosis as compared with normal renal function.
As for the relationship between renal function and BMD in the general population, the results are controversial. Kaji et al. [7] reported a positive relationship between eGFR, calculated using the Modification of Diet in Renal Disease (MDRD) equation for assessing renal function, and BMD in postmenopausal women. Similarly, a retrospective study of 1172 CKD outpatients also demonstrated an association between reduced BMD and impaired renal function (using the CKD-EPI equation) [5]. Ensrud et al. [26] suggested that lower eGFR, calculated using CKD-EPI 2012 equation, was associated with increased risk of hip fracture in older community-dwelling men. However, Hsu et al. [27] demonstrated that a decline in renal function was not associated with decreased BMD. Malmgren et al. [28] also showed that the prevalence of osteoporosis did not differ with renal function. Our findings are consistent with the results of these studies. Although Spearman’s correlation analysis showed a positive relationship between eGFR and BMD values, this relationship was attenuated after adjustment for potential confounders. These discrepancies might be due to the differences in sex, site of BMD observation, the equations used for estimating eGFR, and the higher rate of normal population in subjects of the present study. With MDRD equations, the eGFR tends to be underestimated in people with normal renal function [29]. It is more accurate to use CKD-EPI equations for assessing renal function in older population and healthy individuals, as compared with MDRD equations [30, 31].
In a study of 1,815,943 participants, the association was negative for eGFR and rates of fracture, but after adjusting for additional demographic variables and comorbidities, the relationship disappeared [13]. Similar to their findings, our study showed that decreased eGFR was associated with reduced BMD; however, after adjustment for age, menopausal duration and BMI, we found no evidence to support the hypothesis that the decline in renal function is independently associated with osteoporosis risk. Why was reduced eGFR not associated with increased osteoporosis risk, in contrast to the excess osteoporosis risk observed in patients with end-stage renal disease (ESRD) [32]? In comparison to earlier study, we focused on relatively healthy postmenopausal Chinese women, so we excluded subjects with impaired renal function (eGFR < 60ml/min/1.73m2). This may explain in part why our results differ from previous studies [5], which included 415 CKD outpatients with eGFR < 60 ml/min/1.732. In addition, we found that eGFR was associated with age and menopausal duration, both of which are known risk factors for osteoporosis [33, 34]. Thus, we assume that the association between eGFR and BMD can be explained by known confounding factors, such as age and menopausal duration.
Spearman’s correlation analysis of the data showed that BUN and SCr were not associated with BMD values. Despite its widespread use, BUN and SCr can be affected by many factors that are unrelated to eGFR. Therefore, they are much less sensitive for detection of renal function, particularly in the older population [35, 36]. Our findings are consistent with the previous study, which found no relationships between SCr and BUN with BMD [8].
In our study, only BMD at femoral neck and total hip but not lumbar spine were significantly lower in the eGFR < 90 ml/min/1.73 m2 group compared with the higher eGFR group (eGFR ≥ 90 ml/min/1.73m2). The overlying aortic calcifications make it difficult to measure BMD at lumbar spine in the elderly [37]. Besides, present BMD does not reflect present bone metabolism alone, but rather integrates bone metabolism from the past to present. Spine BMD remains approximately stable or increases over time, whereas BMD of the total hip and femoral neck declines at an increasing rate in elderly people [38, 39]. Finally, altered mineral metabolism occurs in CKD and includes decreased levels of calcitriol, increased levels of parathyroid hormone (PTH) and abnormal calcium-phosphate metabolism. Collectively, these abnormalities are known as CKD-mineral bone disorder (CKD-MBD). The predominance of bone loss in the cortical over trabecular bone was found in CKD-MBD [40, 41]. The femur is rich in cortical bone and is thought to be more affected by CKD-MBD than the lumbar spine, whereas the lumbar spine is rich in trabecular bone. These above may partly explain the skeletal site-specificity differences. In the present study, we focus on relatively healthy subjects with eGFR 60 ml/min/1.73m2. Thus, further studies are needed to explore the relationship between reduced renal function and BMD in CKD.
The present study has notable strengths including its relatively large sample size, focus on the healthy postmenopausal Chinese women, where data on this population are rare. No participant had taken a drug known to affect bone metabolism and renal function. Besides, important confounding factors were adjusted in the regression analysis. The present study has several potential limitations. Firstly, due to the observational study design, it is impossible to establish a causal relation between renal function and BMD. A longitudinal follow-up study is necessary to ascertain these relationships. Another limitation, which should be mentioned, is that participated in this study might be healthier than average. Thus, they might not represent the entire population of Chinese postmenopausal women. Thirdly, the great majority of studies concluded that formulas based on serum cystatin C are superior to SCr-based eGFR [26, 42]. However, Keddis et al [43] showed that SCr-based CKD-EPI equation was preferred over cystatin C-based eGFR in kidney transplant recipients because they are less biased, more accurate. Moreover, the gold standard for detecting eGFR is inulin clearance, which is rarely done due to issues with inconvenience and time consuming for epidemiologic studies. Finally, urine protein, for example, is lacking for confirming impaired renal function.
In conclusion, our data showed that although participants with declined eGFR have significantly reduced femoral neck and total hip BMD, this association disappeared after adjusting for age, menopausal duration and BMI. Renal function itself is not independently associated with the risk of osteoporosis in relatively healthy postmenopausal Chinese women.
BMD: Bone mineral density; BMI: body mass index; BUN: blood urea nitrogen; CI: confidential interval; CKD: chronic kidney disease; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; DXA: dual-energy X-ray absorptiometry; eGFR: estimated glomerular filtration rate; ESRD: end-stage renal disease; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; MDRD: Modification of Diet in Renal Disease; OR: odds ratio; SCa: serum calcium; SCr: serum creatinine; SDs: standard deviations; TC: total cholesterol; TG: triglyceride (TG)
Not applicable.
This work was supported by the National Natural Science Foundation of China (Grant no. 81770833) and the Natural Science Foundation of Hunan Province (Grant no. 2018JJ3721).
The datasets used and analysed during the current study are available from the corresponding author on reasonable request, once the study has been published.
SL and YL were involved in the conception and design of the research. ZS was involved in the acquisition of data. JZ, YW, YW, JH and WH were involved in the statistical analysis. All authors were involved in drafting the manuscript, approving the final draft, and agree to be accountable for the work. All authors read and approved the final manuscript.
All subjects were provided with informed consent forms separately, and the study was carried out in compliance with the Helsinki Declaration II and the Chinese Standards for Good Clinical Practice. The study protocol was approved by the ethics committee of the Second Xiangya Hospital of Central South University (Changsha, China).
Not applicable.
The authors declare that they have no competing interests.
1Department of Geriatrics, Institute of Aging and Geriatrics, The Second Xiangya Hospital, Central South University, Changsha, P. R. China. 2Department of Metabolism and Endocrinology, The Second Xiangya Hospital, Central South University, Changsha, P. R. China.
Table1 Characteristics of participants according to BMD
Variables |
Total (n=776) |
Normal (n=487) |
Osteoporosis (n=289) |
P |
Age (years) |
62.1 ± 6.1 |
61.0 ± 5.7 |
63.9 ± 6.2 |
<0.001 |
Menopausal duration (years) |
12.6 ± 7.1 |
11.1 ± 6.7 |
14.9 ±7.3 |
<0.001 |
Height (cm) |
153.8 ± 5.1 |
154.5 ± 5.0 |
152.7 ± 5.2 |
<0.001 |
Weight (kg) |
55.4 ± 7.8 |
57.0 ± 7.6 |
52.7 ± 7.4 |
<0.001 |
BMI (kg/cm2) |
23.4 ± 3.0 |
23.9 ± 2.9 |
22.6 ± 2.9 |
<0.001 |
BUN (mmol/l) |
5.3 (4.4, 6.2) |
5.3 (4.4, 6.3) |
5.2 (4.4, 6.1) |
0.455 |
SCr (mmol/l) |
58.3 (49.4, 66.1) |
58.7 (50.4, 65.4) |
57.7 (47.2, 67.1) |
0.316 |
eGFR (ml/min/ 1.73m2) |
94.7 (85.3, 101.5) |
95.4(86.6, 101.5) |
93.5 (82.4, 101.8) |
0.216 |
L1-L4 BMD (g/cm2) |
0.933 ± 0.149 |
1.004 ± 0.121 |
0.814 ± 0.110 |
<0.001 |
Femoral neck BMD (g/cm2) |
0.735 ± 0.119 |
0.794 ± 0.098 |
0.634 ± 0.078 |
<0.001 |
Total hip BMD (g/cm2) |
0.814 ± 0.125 |
0.877 ± 0.098 |
0.707 ± 0.087 |
<0.001 |
TG (mmol/l) |
1.4 (1.1, 2.0) |
1.5 (1.1, 2.0) |
1.4 (1.1, 2.0) |
0.342 |
TC (mmol/l) |
5.3 ± 0.9 |
5.3 ± 0.9 |
5.3 ± 0.9 |
0.442 |
HDL-C (mmol/l) |
1.5 (1.3, 1.8) |
1.5 (1.3, 1.7) |
1.6 (1.4, 1.8) |
0.044 |
LDL-C (mmol/l) |
3.1 ± 0.7 |
3.1 ± 0.7 |
3.1 ± 0.7 |
0.444 |
SCa (mmol/l) |
2.4 ± 0.1 |
2.4 ± 0.1 |
2.4 ± 0.1 |
0.140 |
Cigarette smoking, % |
13 (1.7%) |
7 (1.4%) |
6 (2.1%) |
0.567 |
Alcohol consumption, % |
20 (2.6%) |
16 (3.3%) |
4 (1.4%) |
0.106 |
Physical exercise, % |
639 (82.3%) |
405 (83.2%) |
234 (81.0%) |
0.439 |
Data are expressed as mean ± SD, median (25, 75) or number with percentage (%) in parentheses.
*eGFR calculated using the CKD Epidemiology Collaboration 2009 equation.
BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SCa, serum calcium; SCr, serum creatinine; TC, total cholesterol; TG, triglyceride.
Table 2 Characteristics of participants according to renal function, measured by eGFR
Variables |
Category of eGFR* (ml/min/1.73m2) |
p-value |
|
<90 (n=273) |
³90 (n=503) |
||
Age (years) |
65.1 ± 6.0 |
60.4 ± 5.5 |
<0.001 |
Menopausal duration (years) |
15.4 ± 7.0 |
11.0 ± 6.7 |
<0.001 |
Height (cm) |
153.8 ± 4.9 |
153.8 ± 5.3 |
0.858 |
Weight (kg) |
55.3 ± 7.3 |
55.5 ± 8.1 |
0.914 |
BMI (kg/cm2) |
23.4 ± 2.8 |
23.4 ± 3.1 |
0.965 |
BUN (mmol/l) |
5.7 (4.8, 6.6) |
5.0 (4.2, 5.9) |
<0.001 |
SCr (mmol/l) |
69.1 (65.0, 73.7) |
53.3 (44.0, 58.6) |
<0.001 |
eGFR (ml/min/1.73m2) |
80.3 (74.1, 86.2) |
99.3 (95.3, 104.9) |
<0.001 |
L1-L4 BMD (g/cm2) |
0.924 ± 0.142 |
0.938 ± 0.152 |
0.195 |
Femoral neck BMD (g/cm2) |
0.717 ± 0.106 |
0.744 ± 0.125 |
0.004 |
Total hip BMD (g/cm2) |
0.796 ± 0.116 |
0.823 ± 0.129 |
0.004 |
TG (mmol/l) |
1.5 (1.1, 2.1) |
1.4 (1.1, 2.0) |
0.991 |
TC (mmol/l) |
5.4 ± 0.9 |
5.3 ± 0.9 |
0.169 |
HDL-C (mmol/l) |
1.5 (1.3, 1.7) |
1.6 (1.3, 1.8) |
0.350 |
LDL-C (mmol/l) |
3.2 ± 0.7 |
3.1 ± 0.7 |
0.287 |
SCa (mmol/l) |
2.4 ± 0.1 |
2.4 ± 0.1 |
0.849 |
Cigarette smoking, % |
8 (2.9%) |
5 (1.0%) |
0.074 |
Alcohol consumption, % |
6 (2.2%) |
14 (2.8%) |
0.623 |
Physical exercise, % |
231 (84.6%) |
408 (81.1%) |
0.222 |
Osteoporosis,% |
115 (42.1%) |
174 (34.6%) |
0.038 |
Data are expressed as mean ± SD, median (25, 75) or number with percentage (%) in parentheses.
*eGFR calculated using the CKD Epidemiology Collaboration 2009 equation.
BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SCa, serum calcium; SCr, serum creatinine; TC, total cholesterol; TG, triglyceride.
Table 3. Correlation analysis of renal function indicators and other variables
Variables |
eGFR* |
BUN |
Crea |
Age (years) |
-0.426a |
0.136a |
0.107a |
Menopausal duration (years) |
-0.360a |
0.133a |
0.086b |
BMI (kg/cm2) |
-0.012 |
0.014 |
0.010 |
L1-L4 BMD (g/cm2) |
0.057 |
0.014 |
0.022 |
Femoral neck BMD (g/cm2) |
0.086b |
-0.025 |
0.027 |
Total hip BMD (g/cm2) |
0.071b |
-0.022 |
0.016 |
TG (mmol/l) |
0.018 |
-0.064 |
-0.022 |
TC (mmol/l) |
-0.045 |
0.084b |
0.074b |
HDL-C (mmol/l) |
0.038 |
0.061 |
-0.052 |
LDL-C (mmol/l) |
-0.005 |
0.079b |
0.025 |
SCa (mmol/l) |
0.000 |
0.075b |
0.028 |
*eGFR calculated using the CKD Epidemiology Collaboration 2009 equation. r, correlation coefficient.
BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SCa, serum calcium; SCr, serum creatinine; TC, total cholesterol; TG, triglyceride.
a P < 0.01
b P < 0.05
Table 4. Logistic regression of osteoporosis
|
Unadjusted |
Adjusted# |
||||
|
P value |
OR |
95%CI |
P value |
OR |
95%CI |
eGFR (ml/min/1.73m2) |
||||||
³90 |
0.039 |
1.00 (ref) |
|
0.738 |
1.00 (ref) |
|
<90 |
|
1.38 |
1.02-1.86 |
|
0.94 |
0.67-1.33 |
*eGFR calculated using the CKD Epidemiology Collaboration 2009 equation.
CI, confidence interval; eGFR, estimated glomerular filtration rate; OR, odds ratio.
#Adjustment for age, menopausal duration and BMI.