Renal function and bone mineral density in community-dwelling elderly Japanese men: The Fujiwara-kyo Osteoporosis Risk in Men (FORMEN) Study

Bone 56 (2013) 61–66

Contents lists available at SciVerse ScienceDirect

Bone journal homepage: www.elsevier.com/locate/bone

Original Full Length Article

Renal function and bone mineral density in community-dwelling elderly Japanese men: The Fujiwara-kyo Osteoporosis Risk in Men (FORMEN) Study Yuki Fujita a, Masayuki Iki a,⁎, Junko Tamaki a, Katsuyasu Kouda a, Akiko Yura a, Eiko Kadowaki a, Yuho Sato b, Jong-Seong Moon c, Kimiko Tomioka d, Nozomi Okamoto d, Norio Kurumatani d a

Department of Public Health, Kinki University Faculty of Medicine, 377-2 Oono-higashi, Osaka-Sayama, Osaka 589-8511, Japan Department of Human Life, Jin-ai University, 3-1-1 Ohdecho, Echizen, Fukui 915-8586, Japan c Faculty of Nursing, Taisei Gakuin University, 1060-1 Hirao, Mihara-ku, Sakai, Osaka 587-8555, Japan d Department of Community Health and Epidemiology, Nara Medical University School of Medicine, 840 Shijocho, Kashihara, Nara 634-8521, Japan b

a r t i c l e

i n f o

Article history: Received 19 December 2012 Revised 22 April 2013 Accepted 9 May 2013 Available online 15 May 2013 Edited by: Toshio Matsumoto Keywords: Bone density Creatinine Cystatin C Men Renal insufficiency

a b s t r a c t End-stage renal failure deteriorates bone mass and increases fracture risk. However, there are conflicting reports in the literature regarding the effects of mild to moderate renal dysfunction on bone mineral density (BMD). We investigated the association between renal function and BMD at the spine and hip and bone metabolism markers in community-dwelling elderly Japanese men. From 2174 male volunteers aged ≥ 65 years, we examined 1477 men after excluding those with diseases or medications known to affect bone metabolism. Renal function was assessed by serum cystatin C and estimated glomerular filtration rate (eGFR) calculated using the Modification of Diet in Renal Disease Study equation. Bone metabolism was evaluated using levels of serum amino-terminal propeptide of type I procollagen (PINP) and tartrate-resistant acid phosphatase isoenzyme 5b (TRACP-5b), which represent bone metabolic status independent of renal function. eGFR was inversely associated with BMD after adjusting for potential confounders (P b 0.01). Cystatin C showed a weaker but significant association with BMD. eGFR was modestly positively associated with PINP levels (P = 0.04), although cystatin C concentrations were neither associated with PINP nor TRACP-5b levels. Since BMD integrates bone metabolism from the past to present, inverse associations between renal function and BMD may be attributed to past factors, such as obesity. Our findings suggest that low renal function does not affect bone metabolism in a population of community-dwelling elderly Japanese men. Longitudinal studies will be necessary to clarify whether low renal function affects bone loss. © 2013 Elsevier Inc. All rights reserved.

Introduction Patients with end-stage kidney disease exhibit reduced bone mineral density (BMD) and increased risk of hip fracture [1,2]. Patients with predialysis renal failure have lower BMD than control subjects matched for gender, age, and body weight [3]. There are conflicting reports, however, on whether mild to moderate impairment of renal function adversely affects BMD and bone metabolism in the general

Abbreviations: BMD, bone mineral density; BMI, body mass index; Cr, creatinine; CTX, cross-linked carboxy-terminal telopeptide of type I collagen; CV, coefficient of variation; DXA, dual X-ray absorptiometry; eCCr, estimated creatinine clearance; eGFR, estimated glomerular filtration rate; FORMEN, Fujiwara-kyo Osteoporosis Risk in Men; Hcy, homocysteine; hs-CRP, high sensitivity C-reactive protein; iOC, intact osteocalcin; iPTH, intact parathyroid hormone; MDRD, Modification of Diet in Renal Disease; METs, metabolic equivalents; P1NP, amino-terminal propeptide of type I procollagen; Q, quintile; SD, standard deviation; SE, standard error; TRACP-5b, tartrate-resistant acid phosphatase isoenzyme 5b. ⁎ Corresponding author. Fax: +81 72 367 8262. E-mail address: [email protected] (M. Iki). 8756-3282/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bone.2013.05.004

population [4–7]. For instance, Jassal et al. reported a significant positive association between hip BMD and estimated creatinine (Cr) clearance (eCCr) using the Cockcroft–Gault formula or estimated glomerular filtration rate (eGFR) using the Modification of Diet in Renal Disease (MDRD) Study equation in community-dwelling elderly people [4]. Kaji et al. found a similar association in postmenopausal Japanese women with mild renal dysfunction [5]. However, a crosssectional study using National Health and Nutrition Examination Survey data showed no association between mild to moderate renal dysfunction assessed using eCCR and femoral neck BMD [6]. These discrepancies may be attributed to the incorporation of serum Cr levels in the equations for calculating eCCr and eGFR. Cr is produced in the muscle [8] and its serum concentration is associated with muscle mass even after adjusting for protein/meat intake and physical activity [9]. Muscle mass also affects bone mass and BMD, and hence may confound associations of interest. Cystatin C is a small 13-kDa cysteine protease inhibitor which is synthesized by all nucleated cells at a constant rate, filtrated freely in the glomerulus, and completely reabsorbed and catabolized by

62

Y. Fujita et al. / Bone 56 (2013) 61–66

proximal tubular cells. Accordingly, cystatin C is used as an endogenous marker of renal function, and its serum levels correlate with inulin clearance more than plasma Cr concentrations [10]. Moreover, unlike Cr, serum cystatin C levels do not correlate with lean mass [9]. Thus, serum cystatin C levels may be a better indicator of renal function than serum Cr levels or Cr-based indices. Despite this, only a few studies have examined the association between renal function evaluated by serum cystatin C levels and BMD [11,12], and no study has examined this aspect in Asian populations. Moreover, there have been no reports on the association between renal function and bone metabolism markers in the general population. The purpose of the present study was to examine associations of low renal function, evaluated using serum cystatin C and eGFR, with BMD and bone metabolism markers in community-dwelling elderly Japanese men. Materials and methods Subjects The original subjects of the present study were participants of the Fujiwara-kyo Osteoporosis Risk in Men (FORMEN) Study. The FORMEN Study was conducted as part of a larger cohort study, the Fujiwara-kyo Study. Details of the FORMEN Study and Fujiwara-kyo Study are described elsewhere [13]. The purpose of the Fujiwara-kyo Study was to provide a scientific basis for strategies used to prevent frailty, extend healthy life expectancy, and maintain quality of life in elderly Japanese men and women. Participants of the Fujiwara-kyo Study were men and women aged ≥65 years at baseline, lived at home in four cities located in Nara Prefecture, and could walk without the assistance of another individual. These individuals were recruited by the Fujiwarakyo Study Administrative Center with cooperation from local resident associations and elderly people clubs organized in each district of the target cities. The baseline study portion of the Fujiwara-kyo Study was conducted in 2007 and 2008 [13]. The FORMEN Study examined all male participants of the Fujiwara-kyo Study cohort. In the present study, we excluded from analysis participants with incomplete clinical, skeletal, or laboratory measurements, including missing serum Cr or cystatin C values, a history of illness or medications known to affect bone metabolism (e.g., type 1 diabetes mellitus, fasting blood glucose ≥ 160 mg/dL or hemoglobin A1c ≥ 8.0%, gastrectomy due to cancer or ulcer, prostate cancer with anti-androgen therapy, oral glucocorticoid therapy ≥ 5 mg/day, connective tissue disease, uncontrolled hyperthyroid diseases, posterior longitudinal ligament ossification, parathyroid diseases, bisphosphonate therapy), and incomplete information on lifestyle variables regarding smoking and alcohol drinking habits or physical activity. The study protocol was approved by the Medical Ethics Committee of Nara Medical University, and the Ethics Committee of Kinki University Faculty of Medicine. The participants provided written informed consent before enrollment in the study. Bone mass measurement BMD was measured by dual-energy X-ray absorptiometry (DXA) at the lumbar spine (L2–4) and the right hip in a posteroanterior projection (QDR4500A; Hologic Inc., Bedford, MA, USA). Subjects with a history of fractures or bone disease in the right hip were scanned on the left side. The short-term precision values measured using the coefficients of variation (CV) of the in vivo lumbar spine, total hip, and femoral neck BMD measurements were 1.2%, 1.2%, and 1.6%, respectively. Laboratory measurements Plasma and serum samples were obtained for conventional biochemical tests (Cr levels). The remaining serum was stored at −80 °C until measurement of bone turnover markers and other biochemical

markers: intact osteocalcin (iOC), amino-terminal propeptide of type I procollagen (P1NP), cross-linked carboxy-terminal telopeptide of type I collagen (CTX), tartrate-resistant acid phosphatase isoenzyme 5b (TRACP-5b), homocysteine (Hcy), high sensitivity C-reactive protein (hs-CRP), intact parathyroid hormone (iPTH), and cystatin C. Serum Cr levels (mg/dL) were measured using an enzymatic method (L-type CRE·M; Wako Pure Chemical Industries, Ltd., Osaka, Japan). To estimate renal function, eGFR was calculated using the MDRD equation modified for Japanese individuals by the Japanese Society of Nephrology as follows: eGFR (mL/min/1.73 m2) = 194 × serum Cr−1.094 × age−0.287 [14]. Serum iOC (ng/mL) was measured using a two-site immunoradiometric assay (BGP IRMA kit; Mitsubishi, Mitsubishi Kagaku Iatron Inc., Tokyo, Japan) that had a sensitivity of 1 ng/mL [15] with intra-assay inter-assay, and overall CVs of 4.9%, 3.7%, and 6.1%, respectively. Serum P1NP (μg/L) was measured using radioimmunoassay (Procollagen Intact P1NP; Espoo, Finland) with intra-assay, interassay, and overall CVs of 7.8%, 5.1%, and 9.4%, respectively. Serum CTX (ng/mL) was measured using enzyme-linked immunosorbent assay (ELISA) (Serum Crosslap ELISA; Immunodiagnostic Systems Ltd, USA) with intra-assay, inter-assay, and overall CVs of 2.5%, 8.4%, and 8.8%, respectively. Serum TRACP-5b was measured using a fragmentsabsorbed immunocapture enzyme assay (Osteolinks-TRAP-5b; Nitto Boseki, Koriyama, Japan) with a sensitivity of 19.2 mU/dL [16], and intra-assay, inter-assay, and overall CVs of 4.9%, 7.3%, and 8.8%, respectively. Serum iPTH (pg/mL) was measured using electrochemiluminescence immunoassay (Elecsys PTH; Roche Diagnostics, Japan). Serum Hcy (nmol/mL) was measured using highperformance liquid chromatography (YMC-UltraHT Pro C18; YMC Co., Kyoto, Japan) [17] with intra-assay, inter-assay, and overall CVs of 3.0%, 2.0%, and 3.6%, respectively. Serum hs-CRP (ng/mL) was measured using the nephelometric method (N-Latex CRP II; Siemens Healthcare Diagnostics, Japan) with intra-assay, inter-assay, and overall CVs of 2.2%, 1.9%, and 2.9%, respectively. Serum cystatin C (mg/L) was measured using the nephelometric method (N-Latex Cystatin C; Siemens Healthcare Diagnostics, Japan). Height and weight measurements Height (cm) and weight (kg) were measured without shoes and with a thin gown over underwear using an automatic scale (Tanita TBF-215; Tanita Inc., Japan), and BMI (kg/m2) was calculated. Medical history and lifestyle factors Detailed interviews were conducted to confirm the information provided on the questionnaire, including 250 items covering past histories of diseases, fractures, and medication, smoking, alcohol drinking habits, and diet. In addition, the energy expenditure index of daily physical activities was estimated using the International Physical Activity Questionnaire [18]. Interviews were conducted by trained nurses or physicians under a friendly atmosphere. Statistical analysis The SAS statistical software (version 9.1; SAS Institute, Cary, NC, USA) was used for all statistical analyses. Physical activity, Hcy, hs-CRP, Cr, cystatin C, iOC, CTX, P1NP, and TRACP-5b levels, which followed a logarithmic normal distribution, are expressed as geometric mean (SD). The chi-square statistic was used to compare the prevalence of lifestyle factors, such as smoking and alcohol drinking. Adjusted mean values of BMD and bone metabolic markers among quintile groups of eGFR or cystatin C were obtained as the least square means from the analysis of covariance (ANCOVA) model, and their pair-wise differences were determined using the Tukey–Kramer test. The association between renal function and BMD was evaluated

Y. Fujita et al. / Bone 56 (2013) 61–66

using a linear regression model for BMD incorporating eGFR or cystatin C as a continuous independent variable. A regression analysis-based trend test was used to examine trends in mean values of study variables among quintile groups of eGFR or cystatin C.

Results Among the 2174 original subjects, 697 were excluded from analysis. Of these, 156 did not have complete clinical, skeletal, or laboratory measurements, 481 had a history of illness or medications known to affect bone metabolism, and 60 lacked complete information on lifestyle variables. A total of 1477 men were included in the analyses. Excluded subjects were older (mean age, 73.9 years), had lower mean BMD (0.854 g/cm2 at total hip), and had similar mean eGFR (67.4 ml/min/1.73 m2) compared to the original subjects (mean age, 73.3 years; mean hip BMD, 0.878 g/cm2; mean eGFR, 66.9 ml/min/1.73 m2). However, no significant differences were observed in basic characteristics between analyzed and original subjects (mean age, 73.1 and 73.3 years; mean eGFR, 66.7 and 66.9 ml/min/1.73 m2; mean cystatin C, 0.85 and 0.85 mg/L; mean hip BMD, 0.887 and 0.878 g/cm2, respectively). Among analyzed participants, 4.5% had eGFR ≥ 90 ml/min/1.73 m2, 65.9% had 60 ≤ eGFR b 90 ml/min/1.73 m2, 29.1% had 30 ≤ eGFR b 60 ml/min/1.73 m2, and 0.5% had eGFR b 30 ml/min/1.73 m2. Demographics, lifestyle, and clinical characteristics of analyzed participants are presented in Table 1. Lower levels of eGFR and higher concentrations of cystatin C were associated with older age, higher BMI, and lower prevalence of habitual drinking. The proportion of fragility fractures since age 50 was similar across quintile groups of eGFR or cystatin C. Serum biochemical profiles classified by quintile values of eGFR or cystatin C are presented in Table 2. Lower levels of eGFR were associated with higher Hcy and higher hs-CRP, but not with iPTH. Serum cystatin C levels were associated with levels of iPTH, Hcy, and hs-CRP.

63

BMD and bone metabolism marker levels classified by eGFR are presented in Table 3. An inverse association was observed between eGFR and BMD. The result was unchanged after adjusting for age, BMI, physical activity, alcohol intake, smoking, and hs-CRP. This association was also observed in the linear regression analysis for BMD, which incorporated eGFR as a continuous independent variable, after allowing for the same covariates (standardized partial regression coefficient (β) for spine BMD, − 0.108, P b 0.001; β for hip BMD, − 0.098, P b 0.001; β for femoral neck BMD, − 0.083, P = 0.001). Lower eGFR was associated with higher concentrations of iOC and CTX, and with lower concentrations of P1NP. eGFR was not associated with TRACP-5b levels. Serum CTX and P1NP could not be measured in a group of participants due to insufficient volume of serum. However, BMI, physical activity levels, BMD, and serum levels of bone metabolism markers, Cr, and cystatin C of these participants were not significantly different from those of participants whose CTX and P1NP were measured. BMD and bone metabolism marker levels classified by cystatin C are presented in Table 4. Higher levels of cystatin C were associated with higher lumbar spine BMD. A more modest but marginally significant association was observed between cystatin C levels and BMD at the femoral neck. In the regression analysis for BMD using cystatin C as a continuous variable, cystatin C levels were positively associated with spine BMD (β = 0.055, P = 0.049), not associated with hip BMD (β = 0.042, P = 0.107), and marginally associated with femoral neck BMD (β = 0.047, P = 0.080). Cystatin C levels were associated with iOC and CTX levels, but not with P1NP and TRACP-5b concentrations. Discussion This large-scale community-based single-center study first examined the association between renal function, as assessed by serum cystatin C levels, and BMD, as well as biochemical markers of bone

Table 1 Participant characteristics classified according to quintile values of eGFR or cystatin C. Variable

Quintiles of eGFR (mL/min/1.73 m2)

Overall Q1 (23.12–56.41)

Q2 (56.42–63.41)

Q3 (63.42–69.26)

Q4 (69.27–76.60)

Q5 (76.61–124.20)

P-value for trend

N 1477 295 296 295 296 295 Age, years 73.1 (5.1) 75.0 (5.6) 73.5 (5.2) 72.8 (4.7) 72.2 (4.9) 71.9 (4.6) b0.0001 Height, cm 162.8 (5.7) 162.7 (5.4) 162.9 (5.0) 163.1 (6.2) 163.1 (6.1) 162.5 (5.6) 0.8584 Weight, kg 61.2 (8.2) 61.7 (8.1) 61.6 (7.8) 61.7 (8.9) 61.2 (8.3) 60.0 (7.8) 0.0110 BMI, kg/m2 23.1 (2.7) 23.3 (2.6) 23.2 (2.5) 23.2 (2.8) 23.0 (2.7) 22.7 (2.6) 0.0053 Physical activity, METs∙min/day 120.0 (110.2, 130.7) 115.3 (95.6, 139.2) 119.7 (99.9, 143.4) 121.4 (101.5, 145.3) 119.6 (97.4, 146.9) 124.1 (101.0, 152.3) 0.6395 Current smoker, N (%) 250 (16.9) 31 (10.5) 41 (13.9) 53 (18.0) 58 (19.6) 67 (22.7) 0.0008 Ex-smoker, N (%) 893 (60.5) 198 (67.1) 188 (63.5) 165 (55.9) 170 (57.4) 172 (58.3) 0.1046 Smoking, packs/year 31.2 (30.2) 30.3 (30.6) 29.7 (27.8) 30.3 (33.4) 32.1 (29.8) 33.5 (29.0) 0.1155 Habitual drinking, N (%) 730 (49.4) 127 (43.1) 133 (44.9) 151 (51.2) 143 (48.3) 176 (59.7) 0.0443 Ethanol intake, g/day 20.5 (23.9) 17.7 (22.2) 18.0 (21.5) 20.7 (24.0) 21.0 (25.5) 25.2 (25.6) b0.0001 Fragility fractures since age 50, N (%) 71 (4.8) 13 (4.4) 10 (3.4) 19 (6.4) 13 (4.4) 16 (5.4) 0.4860 Variable

Overall

Quintiles of cystatin C concentration (mg/L) Q5 (2.96–0.97)

N Age, years Height, cm Weight, kg BMI, kg/m2 Physical activity, METs∙min/day Current smoker, N (%) Ex-smoker, N (%) Smoking, packs/year Habitual drinking, N (%) Ethanol intake, g/day Fragility fractures since age 50, N (%)

Q4 (0.96–0.87)

Q3 (0.86–0.80)

Q2 (0.79–0.73)

1477 300 310 278 276 73.1 (5.1) 76.0 (5.6) 73.5 (5.1) 72.9 (4.7) 71.9 (4.6) 162.8 (5.7) 162.0 (5.6) 162.7 (5.4) 163.3 (6.2) 163.0 (5.9) 61.2 (8.2) 61.1 (8.1) 61.2 (8.8) 61.7 (8.7) 61.8 (7.9) 23.1 (2.7) 23.3 (2.6) 23.1 (2.8) 23.1 (2.7) 23.2 (2.5) 120.0 (110.2, 130.7) 105.7 (87.2, 128.1) 135.4 (113.8, 161.2) 133.2 (111.3, 159.5) 107.2 (86.3, 133.1) 250 (16.9) 67 (22.3) 60 (19.4) 44 (15.8) 36 (13.0) 893 (60.5) 182 (60.7) 182 (58.7) 159 (57.2) 172 (62.3) 31.2 (30.2) 33.1 (29.9) 31.6 (28.8) 31.0 (31.3) 31.0 (32.1) 730 (49.4) 111 (37.0) 151 (48.7) 130 (46.8) 158 (57.2) 20.5 (23.9) 15.4 (20.6) 20.9 (25.6) 18.9 (22.5) 23.4 (24.2) 71 (4.8) 15 (4.8) 15 (5.4) 14 (5.0) 19 (6.1)

Q1 (0.72–0.47) 313 71.0 (4.1) 163.2 (5.4) 60.4 (7.7) 22.7 (2.6) 121.0 (99.5, 147.1) 43 (13.7) 198 (63.3) 29.2 (29.1) 180 (57.5) 24.0 (25.2) 8 (2.7)

P-value for trend

b0.0001 0.0051 0.5406 0.0225 0.8925 0.0144 0.3310 0.1245 0.0256 b0.0001 0.4651

Data are expressed as mean (SD) or number with proportion (%) in parentheses. Physical activity values represent the geometric mean with mean − SD and mean + SD in parentheses. Q1–Q5, lowest to highest quintile groups; eGFR, estimated glomerular filtration rate; N, number; BMI, body mass index; MET, metabolic equivalent. Habitual drinking refers to alcohol intake of six or more times per week.

64

Y. Fujita et al. / Bone 56 (2013) 61–66

Table 2 Serum biochemical profiles classified according to quintile values of eGFR or cystatin C. Variable

N

eGFR, mL/min/1.73 m2 Cystatin C, mg/L Cr, mg/dL Intact PTH, pg/mL Hcy, nmol/mL hs-CRP, ng/mL

1477 1477 1477 1473 1477 1477

66.7 (12.9) 0.85 (0.84, 0.85) 0.88 (0.87, 0.89) 22.3 (9.9) 12.6 (12.4, 12.8) 664.2 (626.9, 703.8)

Variable

N

Overall

Cystatin C, mg/L eGFR, mL/min/1.73 m2 Cr, mg/dL Intact PTH, pg/mL Hcy, nmol/mL hs-CRP, ng/mL

1477 1477 1477 1473 1477 1477

Quintiles of eGFR (mL/min/1.73 m2)

Overall

0.85 (0.84, 0.85) 66.7 (12.9) 0.88 (0.87, 0.89) 22.3 (9.9) 12.6 (12.4, 12.8) 664.2 (626.9, 703.8)

Q1 (23.12–56.41)

Q2 (56.42–63.41)

Q3 (63.42–69.26)

Q4 (69.27–76.60)

Q5 (76.61–124.20)

49.2 (6.6) 1.04 (1.02, 1.07) 1.14 (1.12, 1.16) 23.6 (11.9) 14.9 (14.4, 15.5) 801.5 (700.3, 917.4)

60.1 (2.1) 0.88 (0.86, 0.89) 0.95 (0.94, 0.95) 21.7 (9.1) 13.0 (12.5, 13.5) 654.5 (574.5, 745.7)

66.4 (1.8) 0.83 (0.81, 0.84) 0.87 (0.86, 0.87) 22.4 (9.5) 12.2 (11.8, 12.7) 617.2 (543.3, 701.2)

72.7 (2.1) 0.78 (0.77, 0.80) 0.80 (0.80, 0.80) 21.6 (8.8) 11.8 (11.4, 12.2) 678.8 (595.3, 774.0)

85.1 (7.3) 0.73 (0.72, 0.74) 0.69 (0.69, 0.70) 22.4 (9.6) 11.2 (10.9, 11.5) 588.3 (520.8, 664.7)

Q5 (2.96–0.97)

Q4 (0.96–0.87)

Q3 (0.86–0.80)

Q2 (0.79–0.73)

Q1 (0.72–0.47)

1.12 (1.10, 1.15) 54.4 (11.3) 1.05 (1.03, 1.07) 23.7 (11.9) 15.4 (14.8, 16.0) 928.7 (816.4, 1056.5)

0.91 (0.91, 0.91) 62.6 (9.0) 0.92 (0.91, 0.93) 22.7 (10.3) 13.2 (12.8, 13.7) 738.3 (652.1, 835.9)

0.83 (0.82, 0.83) 67.3 (9.1) 0.86 (0.85, 0.87) 21.8 (9.0) 12.4 (12.0, 12.9) 646.3 (561.5, 744.0)

0.76 (0.76, 0.76) 71.4 (9.6) 0.82 (0.81, 0.83) 21.2 (9.2) 11.6 (11.3, 11.9) 601.0 (527.2, 685.1)

0.67 (0.67, 0.68) 77.8 (11.1) 0.76 (0.75, 0.77) 22.1 (8.4) 10.7 (10.4, 11.0) 485.5 (433.2, 544.1)

Quintiles of serum cystatin C concentration (mg/L)

P-value for trend

b0.0001 b0.0001 0.1904 b0.0001 0.0053 P-value for trend

b0.0001 b0.0001 0.0108 b0.0001 b0.0001

PTH and eGFR are expressed as mean (SD). Hcy, hs-CRP, Cr, and cystatin C values represent geometric means with mean − SD and mean + SD in parentheses. Q1–Q5, lowest to highest quintile groups; Hcy, homocysteine; hs-CRP, high sensitivity C-reactive protein; PTH, parathyroid hormone; eGFR, estimated glomerular filtration rate; Cr, creatinine.

metabolism in elderly Japanese men, and found an inverse association between eGFR and BMD, and a similar but less robust association between cystatin C levels and BMD. Cystatin C levels were not associated with P1NP and TRACP-5b levels. These findings suggest that low renal function, as commonly observed in community-dwelling elderly men, is not associated with decreased BMD and elevated bone turnover. In the present study, a significant inverse association between eGFR and BMD was observed even after adjusting for age and BMI. eGFR is estimated from age and serum Cr concentration. Cr is produced in the muscle [8] and its concentration is affected by muscle mass [9]. Given that all participants in the present study were community-dwelling volunteers, they may have had larger muscle mass than the average population of the same age. This may have resulted in higher serum Cr levels, hence, lower eGFR and higher BMD. Moreover, the MDRD equation was formulated in a clinical trial involving men and women with chronic kidney disease (CKD) and GFR ranging from 13 to 55 mL/min/1.73 m2 [19]. Rule et al. reported that the MDRD equation underestimates GFR by 29% in

healthy individuals [20]. Thus, MDRD-derived eGFR may not be a suitable index of renal function in healthy elderly people, such as the subjects of the present study. Serum cystatin C levels correlate with inulin clearance more closely than plasma Cr concentrations [10]. In the present study, cystatin C levels, but not eGFR, were significantly positively associated with levels of iPTH, an essential mediator of secondary hyperparathyroidism in renal osteodystrophy or CKD-mineral and bone disorder [6,21,22]. Moreover, serum cystatin C levels are independent of muscle mass [9]. Cystatin C presents a more appropriate indicator for renal function when assessing the association between renal function and BMD. In the present study, a significant positive association was observed between serum cystatin C levels and spine BMD, but not between cystatin C levels and hip BMD. A cross-sectional analysis in the Osteoporotic Fractures in Men Study also reported no differences in hip BMD among quartile groups of cystatin C concentration [11]. Another cross-sectional analysis in the Cardiovascular Health Study revealed an inverse association between cystatin C levels and total hip BMD, but the association disappeared after adjusting for race, age, total lean

Table 3 Bone mineral density and bone metabolism markers classified according to eGFR. Variable

N

Overall

Quintiles of eGFR (mL/min/1.73 m2) Q1 (23.12–56.41)

Unadjusted BMD, g/cm2 Lumbar spine Total hip Femoral neck OC, ng/mL P1NP, μg/L CTX, ng/mL TRACP-5b, mU/dL Adjusted BMD, g/cm2 Lumbar spine Total hip Femoral neck OC, ng/mL P1NP, μg/L CTX, ng/mL TRACP-5b, mU/dL

Q2 (56.42–63.41)

Q3 (63.42–69.26)

Q4 (69.27–76.60)

1417 1.033 (0.191) 1.067 (0.205) 1.044 (0.195) 1.036 (0.192) 1.028 (0.180) 1473 0.887 (0.124) 0.894 (0.132) 0.900 (0.126) 0.890 (0.122) 0.879 (0.122) 1473 0.747 (0.114) 0.753 (0.117) 0.756 (0.114) 0.745 (0.115) 0.740 (0.118) 1457 4.9 (4.8, 5.0) 5.4 (5.1, 5.7) 5.0 (4.8, 5.2) 4.9 (4.7, 5.2) 4.7 (4.5, 4.9) 1112 34.5 (33.7, 35.3) 34.3 (32.6, 36.0) 32.7 (31.1, 34.4) 34.3 (32.6, 36.2) 35.1 (33.3, 37.0) 1230 0.201 (0.195, 0.206) 0.214 (0.199, 0.231) 0.194 (0.182, 0.206) 0.203 (0.192, 0.215) 0.201 (0.190, 0.213) 1473 208.1 (202.4, 214.0) 215.3 (202.2, 229.3) 201.9 (190.0, 214.5) 206.6 (194.1, 220.0) 203.1 (190.2, 216.9) 1.057 (0.011) 1.043 (0.011) 1.031 (0.011) 0.898 (0.007) 0.901 (0.007) 0.884 (0.007) 0.758 (0.006) 0.758 (0.006) 0.739 (0.006) 5.4 (5.3, 5.5) 5.0 (4.8, 5.1) 4.9 (4.8, 5.0) 34.4 (33.5, 35.3) 32.6 (31.8, 33.4) 34.3 (33.5, 35.2) 0.216 (0.209, 0.223) 0.195 (0.188, 0.201) 0.202 (0.195, 0.208) 214.0 (207.3, 221.0) 202.5 (196.2, 208.9) 207.1 (200.7, 213.7)

1.032 0.879 0.740 4.8 35.1 0.202 207.6

Q5 (76.61–124.20)

P-value for trend

0.992 (0.176) b0.0001 0.869 (0.117) 0.0020 0.739 (0.107) 0.0428 4.6 (4.4, 4.8) b0.0001 36.2 (34.5, 38.1) 0.0288 0.190 (0.179, 0.201) 0.0426 214.0 (201.6, 227.2) 0.9556

(0.011) 1.000 (0.011) 0.0005 (0.007) 0.872 (0.007) 0.0008 (0.006) 0.740 (0.006) 0.0073 (4.7, 4.9) 4.7 (4.6, 4.8) b0.0001 (34.2, 36.0) 36.1 (35.2, 37.1) 0.0411 (0.196, 0.208) 0.188 (0.182, 0.195) 0.0221 (201.2, 214.3) 213.3 (206.6, 220.1) 0.8446

Adjusted for age, BMI, physical activity (METs∙min/day), ethanol intake (g/day), smoking (packs/year), and hs-CRP. Unadjusted values represent mean (SD). OC, P1NP, CTX, and TRACP-5b values represent geometric means with values for mean − SD and mean + SD in parentheses. Adjusted values represent mean (SE). OC, P1NP, CTX, and TRACP-5b values represent geometric means with mean − SE and mean + SE in parentheses. Q1–Q5, lowest to highest quintile groups; eGFR, estimated glomerular filtration rate; BMD, bone mineral density; OC, osteocalcin; P1NP, amino-terminal propeptide of type I procollagen; CTX, crosslinked carboxy-terminal telopeptide of type I collagen; TRACP-5b, tartrate-resistant acid phosphatase isoenzyme-5b; hs-CRP, high sensitivity C-reactive protein.

Y. Fujita et al. / Bone 56 (2013) 61–66

65

Table 4 Bone mineral density and bone metabolism markers classified according to serum cystatin C concentration. Variable

Unadjusted BMD, g/cm2 Lumbar spine Total hip Femoral neck OC, ng/mL P1NP, μg/L CTX, ng/mL TRACP-5b, mU/dL Adjusted BMD, g/cm2 Lumbar spine Total hip Femoral neck OC, ng/mL P1NP, μg/L CTX, ng/mL TRACP-5b, mU/dL

N

1417 1473 1473 1457 1112 1230 1473

Overall

1.033 (0.191) 0.887 (0.124) 0.747 (0.114) 4.9 (4.8, 5.0) 34.5 (33.7, 35.3) 0.201 (0.195, 0.206) 208.1 (202.4, 214.0)

Quintiles of cystatin C concentration (mg/L)

P-value for trend

Q5 (2.96–0.97)

Q4 (0.96–0.87)

Q3 (0.86–0.80)

Q2 (0.79–0.73)

Q1 (0.72–0.47)

1.059 (0.206) 0.893 (0.136) 0.753 (0.123) 5.4 (5.1, 5.7) 34.4 (32.7, 36.2) 0.224 (0.210, 0.239) 216.3 (203.4, 230.1)

1.057 0.884 0.744 5.1 34.8 0.201 201.4

(0.206) (0.122) (0.110) (4.8, 5.3) (33.1, 36.7) (0.189, 0.214) (189.6, 213.9)

1.025 (0.185) 0.884 (0.118) 0.744 (0.105) 4.7 (4.5, 5.0) 33.5 (31.9, 35.3) 0.187 (0.176, 0.199) 196.5 (184.0, 209.8)

1.013 0.884 0.745 4.8 35.8 0.205 207.0

(0.173) (0.120) (0.115) (4.6, 5.0) (34.0, 37.8) (0.192, 0.219) (193.5, 221.6)

1.012 0.887 0.747 4.7 33.9 0.187 219.0

(0.179) 0.0001 (0.124) 0.6494 (0.119) 0.6568 (4.4, 4.9) b0.0001 (32.3, 35.7) 0.9636 (0.177, 0.199) 0.0005 (206.9, 231.8) 0.5590

1.049 (0.012) 0.901 (0.007) 0.762 (0.007) 5.3 (5.2, 5.5) 34.5 (33.6, 35.5) 0.232 (0.224, 0.239) 211.9 (205.1, 218.9)

1.056 0.884 0.744 5.1 35.1 0.201 203.7

(0.011) (0.007) (0.006) (4.9, 5.2) (34.2, 36.0) (0.195, 0.208) (197.5, 210.0)

1.028 (0.011) 0.884 (0.007) 0.743 (0.007) 4.7 (4.6, 4.9) 33.4 (32.5, 34.3) 0.188 (0.182, 0.194) 197.2 (190.9, 203.7)

1.010 0.878 0.739 4.9 35.9 0.203 211.6

(0.011) (0.007) (0.007) (4.7, 5.0) (34.9, 36.8) (0.197, 0.210) (204.8, 218.5)

1.020 0.888 0.744 4.7 33.6 0.182 219.5

(0.011) 0.0064 (0.007) 0.1847 (0.006) 0.0602 (4.6, 4.8) 0.0002 (32.8, 34.5) 0.7007 (0.176, 0.188) b0.0001 (212.7, 226.5) 0.2670

Adjusted for age, BMI, physical activity (METs∙min/day), ethanol intake (g/day), smoking (packs/year), and hs-CRP. Unadjusted values represent mean (SD). OC, P1NP, CTX, and TRACP-5b values represent geometric means with values for mean − SD and mean + SD in parentheses. Adjusted values represent mean (SE). OC, P1NP, CTX, and TRACP-5b values represent geometric means with mean − SE and mean + SE in parentheses. Q1–Q5, lowest to highest quintile groups; BMD, bone mineral density; OC, osteocalcin; P1NP, amino-terminal propeptide of type I procollagen; CTX, crosslinked carboxy-terminal telopeptide of type I collagen; TRACP-5b, tartrate-resistant acid phosphatase isoenzyme-5b, high sensitivity C-reactive protein.

mass, and percent fat [12]. Thus, the hip BMD results of the present study are consistent with these previous studies, although the association between cystatin C levels and spine BMD has not been examined to date. Spine BMD values in the group with higher cystatin C levels may have been overestimated due to acceleration of calcifications in the aorta [23] and other tissues [24], which are more frequently observed in patients with renal dysfunction. Moreover, present BMD does not reflect present bone metabolism alone, but rather integrates bone metabolism from the past to present. Accordingly, the observed inverse associations between renal function and BMD may have resulted from past factors, such as obesity. It is widely accepted that obesity, or higher BMI, is associated with increased BMD [25]. Obesity is also well known to increase the risk of diabetes and hypertension, which lead to decreased renal function [26,27]. Thus, obesity in the past may have contributed to low renal function and increased BMD at present. Further studies examining longitudinal changes in renal function and BMD are warranted. Serum cystatin C levels were significantly positively associated with OC and CTX levels, but not with PINP or TRACP-5b levels in the present study. OC and CTX accumulate in the serum when their filtration in kidneys is reduced [28–30]. On the other hand, TRACP-5b and P1NP levels are independent of renal function [28–31]. Thus, the effects of renal function on bone metabolism should be assessed using bone turnover markers that are independent of renal function, such as PINP and TRACP-5b. Our results suggest that low renal function, as observed in community-dwelling elderly men, is not associated with elevated bone turnover. The present study has several strengths. This study is a populationbased study with a sufficiently large sample size, which may reflect the health status of elderly Japanese men. This study was conducted as a part of an ongoing cohort study with a planned 20-year followup, and changes in BMD and incident cases of fractures will be confirmed. We measured serum cystatin C levels in addition to Cr-based indices for accurate assessment of renal function. A single-center study design has an advantage over a multi-center design with respect to quality control of study performance given the lack of inter-center variation. The present study also has potential limitations. Given the cross sectional nature of the study, causality could not be addressed. Direct

measurement of GFR, such as inulin or radioisotope clearance, was not included and may not be feasible in large cohorts. The participants of the FORMEN Study were not randomly selected from the population but are community-dwelling volunteers, and thus caution should be exercised when generalizing the results. In conclusion, our findings suggest that low renal function, as commonly observed in community-dwelling elderly men, is not associated with decreased BMD and elevated bone turnover.

Acknowledgments The Fujiwara-kyo Study Group (chaired by Norio Kurumatani with Nozomi Okamoto as a secretary general) comprising Nobuko Amano, Yuki Fujita, Akihiro Harano, Kan Hazaki, Masayuki Iki, Junko Iwamoto, Akira Minematsu, Masayuki Morikawa, Keigo Saeki, Noriyuki Tanaka, Kimiko Tomioka, and Motokazu Yanagi performed most non-skeletal measurements in the present study and provided the data to the FORMEN Study. The authors acknowledge Toyukai Medical Corporation (Tokyo, Japan) and Toyo Medic Inc., (Osaka, Japan) for DXA scanning and SRL Inc. (Tokyo, Japan) for their technical assistance with laboratory measurements. The FORMEN Study was supported by Grants-in-Aid for Scientific Research (No. 21390210: 2009–2011, No. 20590661: 2008–2010) from the Japanese Society for the Promotion of Science; a Grant-inAid for Young Scientists (No. 20790451: 2008–2010) from the Japanese Ministry of Education, Culture, Sports, Science and Technology; a Grant-in-Aid for Study on Milk Nutrition (2008) from the Japan Dairy Association; a grant (2007) from the Foundation for Total Health Promotion; a St. Luke's Life Science Institute Grant-in-Aid for Epidemiological Research (2008); and a grant (2008) from the Physical Fitness Research Institute, MEIJIYASUDA Life Foundation of Health and Welfare. Disclosure statement: the authors have nothing to disclose.

References [1] Nickolas TL, Leonard MB, Shane E. Chronic kidney disease and bone fracture: a growing concern. Kidney Int 2008;74:721–31. [2] Alem AM, Sherrard DJ, Gillen DL, Weiss NS, Beresford SA, Heckbert SR, et al. Increased risk of hip fracture among patients with end-stage renal disease. Kidney Int 2000;58:396–9.

66

Y. Fujita et al. / Bone 56 (2013) 61–66

[3] Rix M, Andreassen H, Eskildsen P, Langdahl B, Olgaard K. Bone mineral density and biochemical markers of bone turnover in patients with predialysis chronic renal failure. Kidney Int 1999;56:1084–93. [4] Jassal SK, von Muhlen D, Barrett-Connor E. Measures of renal function, BMD, bone loss, and osteoporotic fracture in older adults: the Rancho Bernardo Study. J Bone Miner Res 2007;22:203–10. [5] Kaji H, Yamauchi M, Yamaguchi T, Shigematsu T, Sugimoto T. Mild renal dysfunction is a risk factor for a decrease in bone mineral density and vertebral fractures in Japanese postmenopausal women. J Clin Endocrinol Metab 2010;95:4635–42. [6] Hsu CY, Cummings SR, McCulloch CE, Chertow GM. Bone mineral density is not diminished by mild to moderate chronic renal insufficiency. Kidney Int 2002;61: 1814–20. [7] Kim HL, Park IY, Choi JM, Hwang SM, Kim HS, Lim JS, et al. A decline in renal function is associated with loss of bone mass in Korean postmenopausal women with mild renal dysfunction. J Korean Med Sci 2011;26:392–8. [8] Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 1992;38:1933–53. [9] Baxmann AC, Ahmed MS, Marques NC, Menon VB, Pereira AB, Kirsztajn GM, et al. Influence of muscle mass and physical activity on serum and urinary creatinine and serum cystatin C. Clin J Am Soc Nephrol 2008;3:348–54. [10] Fliser D, Ritz E. Serum cystatin C concentration as a marker of renal dysfunction in the elderly. Am J Kidney Dis 2001;37:79–83. [11] Ishani A, Paudel M, Taylor BC, Barrett-Connor E, Jamal S, Canales M, et al. Renal function and rate of hip bone loss in older men: the Osteoporotic Fractures in Men Study. Osteoporos Int 2008;19:1549–56. [12] Fried LF, Shlipak MG, Stehman-Breen C, Mittalhenkle A, Seliger S, Sarnak M, et al. Kidney function predicts the rate of bone loss in older individuals: the Cardiovascular Health Study. J Gerontol A Biol Sci Med Sci 2006;61:743–8. [13] Iki M, Fujita Y, Tamaki J, Kouda K, Yura A, Kadowaki E, et al. Design and baseline characteristics of a prospective cohort study for determinants of osteoporotic fracture in community-dwelling elderly Japanese men: the Fujiwara-kyo Osteoporosis Risk in Men (FORMEN) Study. BMC Musculoskelet Disord 2009;10:165. [14] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53:982–92. [15] Kawaguchi H, Matsumoto T, Kurokawa T, Orimo H, Mizunashi K, Takuwa Y, et al., Serum levels of BGP determined by two-site immunoradiometric assay (IRMA) using monoclonal antibodies, Hormone to Rinsho. 38 (1990) 1291–1296. [16] Ishizawa Y, Inaba M, Ishii K, Yamashita H, Miki T, Goto H, et al., Evaluation of newly developed kit for measurement of bonespecific tartrate-resistant acid phosphatase in blood, Igaku to Yakugaku. 54 (2005) 709–717. [17] Araki A, Sako Y. Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. J Chromatogr 1987;422:43–52.

[18] Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35:1381–95. [19] Beck GJ, Berg RL, Coggins CH, Gassman JJ, Hunsicker LG, Schluchter MD, et al. Design and statistical issues of the Modification of Diet in Renal Disease Trial. The Modification of Diet in Renal Disease Study Group, Control. Clin Trials 1991;12:566–86. [20] Rule AD, Larson TS, Bergstralh EJ, Slezak JM, Jacobsen SJ, Cosio FG. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004;141:929–37. [21] Vassalotti JA, Uribarri J, Chen SC, Li S, Wang C, Collins AJ, et al. Trends in mineral metabolism: Kidney Early Evaluation Program (KEEP) and the National Health and Nutrition Examination Survey (NHANES) 1999–2004. Am J Kidney Dis 2008;51: S56–68. [22] Patel S, Barron JL, Mirzazedeh M, Gallagher H, Hyer S, Cantor T, et al. Changes in bone mineral parameters, vitamin D metabolites, and PTH measurements with varying chronic kidney disease stages. J Bone Miner Metab 2010;29:71–9. [23] Qunibi WY. Cardiovascular calcification in nondialyzed patients with chronic kidney disease. Semin Dial 2007;20:134–8. [24] Duhn V, D'Orsi ET, Johnson S, D'Orsi CJ, Adams AL, O'Neill WC. Breast arterial calcification: a marker of medial vascular calcification in chronic kidney disease. Clin J Am Soc Nephrol 2011;6:377–82. [25] Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effects of weight and body mass index on bone mineral density in men and women: the Framingham Study. J Bone Miner Res 1993;8:567–73. [26] Wang Y, Chen X, Song Y, Caballero B, Cheskin LJ. Association between obesity and kidney disease: a systematic review and meta-analysis. Kidney Int 2008;73:19–33. [27] Brancati FL, Whelton PK, Randall BL, Neaton JD, Stamler J, Klag MJ. Risk of end-stage renal disease in diabetes mellitus: a prospective cohort study of men screened for MRFIT. Multiple Risk Factor Intervention Trial. JAMA 1997;278:2069–74. [28] Yamada S, Inaba M, Kurajoh M, Shidara K, Imanishi Y, Ishimura E, et al. Utility of serum tartrate-resistant acid phosphatase (TRACP5b) as a bone resorption marker in patients with chronic kidney disease: independence from renal dysfunction. Clin Endocrinol (Oxf) 2008;69:189–96. [29] Swolin-Eide D, Hansson S, Larsson L, Magnusson P. The novel bone alkaline phosphatase B1x isoform in children with kidney disease. Pediatr Nephrol 2006;21: 1723–9. [30] Haarhaus M, Fernstrom A, Magnusson M, Magnusson P. Clinical significance of bone alkaline phosphatase isoforms, including the novel B1x isoform, in mild to moderate chronic kidney disease. Nephrol Dial Transplant 2009;24:3382–9. [31] Chu P, Chao TY, Lin YF, Janckila AJ, Yam LT. Correlation between histomorphometric parameters of bone resorption and serum type 5b tartrate-resistant acid phosphatase in uremic patients on maintenance hemodialysis. Am J Kidney Dis 2003;41: 1052–9.