Association of serum uric acid with bone mineral density and clinical fractures in Chinese type 2 diabetes mellitus patients: A cross-sectional study

Clinica Chimica Acta 486 (2018) 76–85

Contents lists available at ScienceDirect

Clinica Chimica Acta journal homepage: www.elsevier.com/locate/cca

Association of serum uric acid with bone mineral density and clinical fractures in Chinese type 2 diabetes mellitus patients: A cross-sectional study

T

⁎

Pijun Yana, , Zhihong Zhangb, Qin Wana, Jianhua Zhua, Hua Lia, Chenlin Gaoa, Hongyan Maa, Yong Xua a b

Department of Endocrinology, The Affiliated Hospital of Southwest Medical University, Luzhou, Sichuan 646000, China Department of General Practice, The Affiliated Hospital of Southwest Medical University, Luzhou, Sichuan 646000, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Serum uric acid Bone mineral density Clinical fractures Osteoporosis Type 2 diabetes mellitus

Background: We evaluate the associations of serum uric acid (UA) with bone mineral density (BMD) and prevalence of clinical fractures in type 2 diabetes mellitus (T2DM) patients. Methods: 1562 T2DM patients undergoing BMD measurement and clinical fractures assessment were enrolled andserum UA concentrations were measured. Results: T2DM patients with osteoporosis had lower serum UA concentrations compared with those with normal BMD values and osteopenia. Serum UA concentration was significantly correlated with BMD values at the lumbar spine, femoral neck, and total hip in postmenopausal women, and serum UA concentration was positively associated with BMD values at the lumbar spine in men. Moreover, patients with clinical fractures had lower serum UA than those without. Multiple logistic regression analysis showed that serum UA concentrations were significantly and inversely associated with the presence of clinical fractures after adjustment for age, BMI, diabetes duration, fasting blood glucose (FBG), Glycated hemoglobin A1c (HbA1c), alkaline phosphatase (ALP), creatinine (Cr), neutrophil to lymphocyte ratio (NLR), diabetic vascular complications [men: OR = 0.996, 95% CI = 0.993–1.000, P = 0.039; women: OR = 0.996,95% CI = 0.994–0.998, P = 0.001]. The results were not statistically significant when models were further adjusted for BMD values at each site. Conclusions: Lower serum UA concentrations may be associatedwith lower BMD values and higher prevalence of clinical fractures independent of potential confounders except for BMD values at each site. These findings need to be confirmed by further prospective studies.

1. Introduction Osteoporosis is a multi-factorial skeletal disease characterized by low bone mass and altered bone quality, and can put patients at an increased risk for abnormal bone strength and fragility fractures [1]. It is estimated that > 200 million people have osteoporosis worldwide, and osteoporosis has therefore become an alarming health problem [2]. Many clinical studies have reported that osteoporosis is one of the chronic complications associated with diabetes mellitus (DM) [1]. Both type 1 and type 2 diabetes mellitus (T2DM) can affect areal bone mineral density (BMD) and the risk of bone fractures [3]. BMD value measured by dual energy X-ray absorptiometry (DXA) is recognized as a major tool to detect osteoporosis and predict fracture risk [4]. Of interest is that most, but not all, epidemiologic studies have found that despite the normal or increased BMD values, T2DM patients have an increased risk for fracture compared to nondiabetic subjects [5, 6],

⁎

suggesting that increased fracture risk associated with T2DM may be due to impaired bone quality (not revealed from BMD values) and extra-skeletal factors, but the specific mechanisms accounting for diabetes-related bone fracture and the factors for altered BMD values has not been identified clearly, and there are few effective therapies for diabetic osteoporosis. It is therefore an urgent task to seek clinically suitable surrogate markers for diabetic osteoporosis. Substantial evidence indicates the role of oxidative stress and low circulating antioxidants concentrations in the initiation and progression of osteoporosis, especially type 2 diabetic osteoporosis [7–10]. Serum uric acid (UA), an important endogenous antioxidant, effectively can scavenge superoxide, hydroxyl radicals, and singlet oxygen, as wellas block formation of the strong oxidant peroxynitrite [11]. The antioxidant activityof serum UA is much higher than that of other antioxidants, including vitamins and enzymatic antioxidants [12], with serum UA accounting for approximately 50% of the antioxidant activity

Corresponding author. E-mail address: [email protected] (P. Yan).

https://doi.org/10.1016/j.cca.2018.07.033 Received 6 February 2017; Received in revised form 15 July 2018; Accepted 18 July 2018 Available online 19 July 2018 0009-8981/ © 2018 Elsevier B.V. All rights reserved.

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

Fig. 1. Flowchart of the study participants.

hyperuricemia [25]. Although many studies have reported the relationship between serum UA concentrations, BMD values and bone fractures, the conclusions were inconsistent and controversial. Two investigated the association of serum UA concentrations with BMD values in men aged 19–84 y and postmenopausal women with T2DM, and showed inconsistent conclusions [15, 25], and they did not explore the association of serum UA concentrations with the prevalence of clinical fractures.

in human plasma [13]. Moreover, the systemic infusion of serum UA not only increases plasma antioxidant capacity at rest, but also reduces the exercise-associated oxidative stress in healthy subjects [14]. The antioxidant effect of serum UA may potentially protect against osteoporosis. Several studies have demonstrated that serum UA concentrations were positively associated with BMD values at various skeletal sites, T-score and Z-score, and inversely with the prevalence of osteoporosis, vertebral and nonvertebral fractures, and bone resorption markers in peri- and postmenopausal women, and young and middleaged, and elderly men with and without T2DM [10, 15–20], suggesting of a potential protective role of serum UA against osteoporosis. Yet, some recent observational studies have shown the contrary. Sritara et al. and Mehta et al. found that serum UA concentrations were inversely associated with BMD at the femoral neck (FN) after controlling for covariates in females aged 25–54 y, and higherserum UA concentrations were associated with an increased risk of hip fractures in community-dwelling elderly men, respectively [21, 22]. Recent studies reported that serum UA concentration has no relationship with BMD values at various skeletal sites and the onset of new osteoporotic fractures after adjustment for potential confounders in young and middleaged, and elderly men and pre-menopausal women, and post-menopausal women not treated with estrogen or withT2DM [23–26], and confirmed the conclusion in a rodent model of chronic mild

2. Materials and methods 2.1. Study population The population consisted of 5941 patients with T2DM who were initially consecutively enrolled for an education, evaluation, or treatment of diabetes mellitus and osteoporosis at the inpatient clinicof the Endocrinology Department at the Affiliated Hospital of Southwest Medical University during the period between August 2012 and May 2017. The diagnosis of T2DM was based on oral glucose tolerance tests (OGTT) and the 1999 World Health Organization (WHO) criteria. All the patients were being treated with oral medications and insulin injection. Inclusion criteria were: 1) Previously confirmed or newly diagnosed T2DM patients; 2) Postmenopausal women aged ≥45 y who 77

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

Patients who had gout or urinary stones or received treatment for hyperuricemia; 7) Patients who had both a normal BMD values and verified self-reported fractures; and 8) Subjects with incomplete baseline data and without available informations. Some subjects met two or more exclusion criteria. Consequently, 1562 patients (582 men and 980 women) were enrolled in the study. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the human research ethics committee of the Affiliated Hospital of Southwest Medical University in Sichuan province. The requirement of informed consent was waived owing to the retrospective nature of the study.

Table 1 Baseline characteristics of subjects. Men

Postmenopausal women

Number of subjects Diabetes duration (y) Age (y) BMI (kg/m2) FBG (mmol/l) HbA1c (%) Cr (μmol/l) UA (μmol/l) ALP(U/l) Neutrophil (*109/l) Lymphocyte (*109/l) NLR Urinary ACR (mg/g) ABI VPT (v)

582 7.41 ± 6.49 63.45 ± 9.31 24.19 ± 3.28 10.63 ± 4.97 9.44 ± 2.52 77.04 ± 22.14 332.31 ± 98.84 84.71 ± 57.12 4.63 ± 2.30 1.58 ± 0.64 3.55 ± 3.04 148.49 ± 23.97 1.05 ± 0.13 17.31 ± 10.26

980 8.53 ± 6.68 64.06 ± 9.34 24.38 ± 4.13 10.12 ± 5.02 8.93 ± 2.38 61.50 ± 21.08 295.72 ± 99.00 88.81 ± 51.20 4.63 ± 2.37 1.69 ± 0.68 3.42 ± 3.94 230.43 ± 28.75 1.01 ± 0.15 15.71 ± 8.74

0.000 NS NS 0.017 0.000 0.000 0.000 0.008 0.923 0.000 0.003 NS 0.000 0.026

Microvascular complications DPN DR DN

285 (48.97%) 59 (10.14%) 207 (35.57%)

398 (40.61%) 121 (12.35%) 356 (36.33%)

0.001 NS NS

Macrovascular complications PAD Hypertension CHD CI Hyperuricemia

34 (5.84%) 323 (55.50%) 23 (3.95%) 144 (24.74%) 94 (16.15%)

77 (7.86%) 605 (61.73%) 55 (5.61%) 229 (23.37%) 223 (22.76%)

NS 0.015 NS NS 0.002

Hypoglycemic medication Metformin Sulfonylurea Alpha-glucosidase inhibitor Insulin

298 (51.20%) 242 (41.58%) 202 (34.71%) 41 (7.04%)

526 (53.67%) 386 (39.39%) 67(6.84%) 368 (37.55%)

NS NS 0.000 0.000

1.084 ± 0.169 −0.20 ± 1.47 0.860 ± 0.127 −1.29 ± 1.19 0.797 ± 0.139 −1.40 ± 0.80 68 (11.68%)

0.930 ± 0.161 −1.59 ± 1.41 0.768 ± 0.132 −1.38 ± 1.02 0.689 ± 0.144 −2.22 ± 1.17 224 (22.86%)

0.000 0.000 0.000 0NS 0.000 0.000 0.000

135 (23.20%)

544 (55.51%)

0.000

Bone metabolism LS BMD (g/cm2) T score FN BMD (g/cm2) T score TH BMD (g/cm2) T score Diabetic osteoporotic fractures Diabetic osteoporosis

P

2.2. Clinical and biochemical measurements We used a structured interview questionnaire and reviewed medical records of each patient to collect information regarding smoking history, alcohol consumption, menopausal status, family history, diabetes duration, medication use, history of comorbidities and diabetic macrovascular complications. Information on gender, age and long-term residence was collected. Body weight, height, body mass index (BMI) were measured with the use of standard methods, as described previously [27]. Hyperuricemia was defined as ≥420 μmol/l in men and ≥ 360 μmol/l in women [28]. Blood samples were collected following overnight fasting and subsequently analyzed at a certified laboratory of our hospital. Serum UA, creatinine (Cr), albumin, and alkaline phosphatase (ALP), and fasting blood glucose (FBG) were analyzed using a 7060 full-automatic biochemical analyzer (Hitachi). Serum UA and Cr concentrations were measured by the enzymatic method. FBG was measured by the glucose oxidase method. ALP was measured by the AMP buffer concentrate. Glycated hemoglobin A1c (HbA1c) was measured by the anion exchange high performance liquid chromatography (Arkray Eluent 80A). Neutrophil and lymphocyte count were determined using an automated blood cell counter (Mindray BC-6800), and neutrophil to lymphocyte ratio (NLR) was calculated. Urinary microalbumin and creatinine were measured from three fresh morning spot urine sample on three separate occasions within 6 months. Urinary microalbumin was measured with immunoturbidimetric tests. Urinary creatinine was measured enzymatically. The urinary microalbumin to creatinine ratio (ACR; mg/g creatinine) was calculated. Patients were diagnosed as having diabetic nephropathy (DN) if ACR > 30 mg/g in two out of three random voided urine samples. Two-field fundus photography of eyes from each participant were performed using a Canon CR-2 Digital Retinal Camera. The presence of diabetic retinopathy (DR) was assessed by fundus photography and an ophthalmologist. Ankle-brachial index (ABI) measurements were measured by a continuous-wave Doppler ultrasound probe (Vista AVS, Summit Co.). Leg-specific ABI was calculated by dividing the higher SBP in the posterior tibial or dorsalis pedis by the higher of the right or left brachial SBP. Patients were diagnosed as having peripheral arterial disease (PAD) if an ABI value < 0.9 on either limb. Vibrating perception threshold (VPT) values were measured by a neurothesiometer (Bio-Thesiometer; Bio-Medical Instrument Co.). Patients were diagnosed as having diabetic peripheral neuropathy (DPN), as we described previously [29]. The areal BMD values of the lumbar spine (LS), FN and total hip (TH) were measured by DXA using a GE Lunar Prodigy and was expressed as the number of grams of bone mineral per square centimeter (g/cm2). BMD measurements provided absolute values for each anatomic site and were compared with those of healthy young Chinese adults (T-score). All measurements were taken by the same well-trained and qualified operator on the same machine using standardized protocols for participant positioning to ensure machine accuracy of > 98%. The CVs of measurement of BMD at the LS, FN and TH were 0.84, 1.96,

FBG, fasting blood glucose; HbA1c, glycated hemoglobin A1c; Cr, creatinine; UA, uric acid; ALP, alkaline phosphatase; NLR, neutrophil to lymphocyte ratio; ACR, microalbumin to creatinine ratio; ABI, ankle-brachial index; VPT, vibrating perception threshold; DPN, diabetic peripheral neuropathy; DR, diabetic retinopathy; DN, diabetic nephropathy; PAD, peripheral arterial disease; CHD, coronary heart disease; CI, cerebral infarction; BMD, bone mineral density; LS, lumbar spine; FN, femoral neck; TP, total hip. Values were given as means ± SD.

had not menstruated for at least 1 y or men aged ≥50 y; 3) longtermresidence (≥5 y) in China's Sichuan province. Exclusion criteria were: 1). Subjects who had taken any drugs that might influence bone metabolism for morethan 6 months or within the previous 12 months, such as thiazolidinediones (TZDs), calcium supplements, vitamin D, calcitonin, diuretics, systemic glucocorticoids, immunosuppressant, and estrogens; 2) Subjects with fractures caused by cancer, traffic accidents and high-trauma or family history of fractures; 3) Subjects with having diseases known to effect bone metabolism and/or oxidant-antioxidant status like cancer, thyroid diseases, hypothalamic or pituitary diseases, hyperparathyroidism,hypogonadism, Cushing syndrome, inflammatory diseases, congestive heart failure, asthma/chronic obstructive pulmonary disease; 4) Patients who had hepatic or renaldysfunction or nutritional derangements; 5) Subjects with using medications that may affect the oxidant/antioxidant system, including vitamins (A, C, and E) and minerals (zinc and selenium) during the previous 6 months; 6) 78

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

Table 2 Comparison of serum UA levels, and other clinical and biochemical parameters among T2DM patients with normal BMD, osteopenia and osteoporosis. Men

Postmenopausal women

Normal BMD

Osteopenia

Osteoporosis

P

Normal BMD

Osteopenia

Osteoporosis

P

Number of subjects Diabetes duration (y) Age (y) BMI (kg/m2) FBG (mmol/l) HbA1c (%) Cr (μmol/l) UA (μmol/l) ALP(U/l) Neutrophil (*109/l) Lymphocyte (*109/l) NLR Urinary ACR(mg/g) ABI VPT(v)

211 7.24 ± 5.9 61.72 ± 8.94 25.13 ± 3.04 10.38 ± 4.22 9.28 ± 2.38 76.87 ± 21.17 358.08 ± 104.89 81.85 ± 31.01 4.18 ± 1.67 1.61 ± 0.65 3.07 ± 2.28 161.02 ± 616.23 1.06 ± 0.12 15.86 ± 9.01

236 6.91 ± 6.71 62.51 ± 8.52 23.92 ± 3.33 11.11 ± 5.56 9.69 ± 2.59 74.99 ± 20.78 331.02 ± 91.43 84.44 ± 65.89 4.69 ± 2.14 1.65 ± 0.64 3.33 ± 2.33 111.4 ± 317.83 1.05 ± 0.12 17.26 ± 10.51

135 8.55 ± 6.86 67.81 ± 9.88 23.09 ± 3.17 10.18 ± 4.93 9.23 ± 2.58 80.87 ± 25.36 294.28 ± 89.15 89.57 ± 70.31 5.21 ± 3.12 1.40 ± 0.59 4.66 ± 4.53 196.52 ± 612.67 1.01 ± 0.17 20.85 ± 11.63

0.015 0.000 0.000 NS NS NS 0.000 NS 0.004 0.001 0.000 0.031 NS 0.002

138 6.97 ± 6.27 56.12 ± 8.62 25.19 ± 4.1 11.02 ± 5.14 9.37 ± 2.29 57.9 ± 21.79 341.18 ± 124.43 85.3 ± 43.21 4.73 ± 2.37 1.71 ± 0.6 3.24 ± 2.66 187.86 ± 654.33 1.04 ± 0.14 13.63 ± 8.23

298 8.71 ± 6.39 61.47 ± 8.42 24.85 ± 4.08 10.53 ± 5.27 9.14 ± 2.32 61.01 ± 21.31 304.97 ± 96.53 88.29 ± 64.03 4.49 ± 2.29 1.85 ± 0.83 2.84 ± 2.35 198.02 ± 667.8 1.03 ± 0.1 14.66 ± 7.22

544 8.83 ± 6.89 67.49 ± 8.27 23.89 ± 4.12 9.66 ± 4.80 8.70 ± 2.41 62.68 ± 20.69 279.12 ± 88.43 89.99 ± 44.81 4.68 ± 2.42 1.59 ± 0.58 3.78 ± 4.80 261.03 ± 837.35 0.98 ± 0.18 17.11 ± 9.58

0.006 0.000 0.000 0.000 0.000 0.004 0.000 NS NS 0.000 0.000 0.016 0.001 0.000

Microvascular complications DPN DR DN

111 (19.04%) 20 (3.43%) 71 (12.18%)

117(20.07%) 26 (4.46%) 84 (14.41%)

57 (9.79%) 13 (2.23%) 52 (8.93%)

NS NS NS

59 (6.01%) 19 (1.94%) 46 (4.69%)

135 (13.76%) 46 (4.69%) 86 (8.77%)

204 (20.80%) 56 (5.71%) 200 (20.39%)

NS NS NS

Macrovascular complications PAD Hypertension CHD CI Hyperuricemia

12 (2.06%) 122 (20.93%) 11 (1.89%) 45 (7.72%) 45 (21.33%)

12 (2.06%) 124 (21.27%) 7 (1.20%) 57 (9.78%) 38 (16.10%)

10 (1.72%) 77 (13.21%) 5 (0.86%) 42 (7.22%) 11 (8.09%)

NS NS NS NS 0.005

8 (0.82%) 75 (7.65%) 3 (0.31%) 21 (2.14%) 47 (34.06%)

15 (1.53%) 185 (18.86%) 15 (1.53%) 64 (6.52%) 73 (24.50%)

54 (5.55%) 345 (35.17%) 37 (3.77%) 144 (14.68%) 103 (18.90%)

0.027 NS NS 0.014 0.001

Hypoglycemic medication Metformin Sulfonylurea Alpha-glucosidase inhibitor Insulin

102 (17.50%) 84 (14.41%) 15 (2.57%) 70 (12.01%)

120 (20.58%) 101 (17.32%) 19 (3.26%) 82 (14.07%)

76 (13.04%) 57 (9.78%) 8 (1.37%) 49(8.42%)

NS NS NS NS

79 56 11 55

165 (16.82%) 121 (12.33%) 17 (1.73%) 118 (12.03%)

282 (28.78%) 209 (21.33%) 29 (2.96%) 195 (19.88%)

NS NS NS NS

Bone metabolism LS BMD (g/cm2) T score FN BMD (g/cm2) T score TP BMD (g/cm2) T score clinical fractures

1.19 ± 0.149 0.78 ± 1.29 0.957 ± 0.098 −0.38 ± 0.93 0.89 ± 0.121 −0.86 ± 0.7 0 (0)

1.04 ± 0.121 −0.59 ± 0.98 0.818 ± 0.076 −1.66 ± 0.69 0.761 ± 0.088 −1.6 ± 0.51 12(5.08%)

0.942 ± 0.165 −1.55 ± 1.41 0.732 ± 0.122 −2.51 ± 1.11 0.667 ± 0.140 −2.15 ± 0.81 56(41.48%)

0.000 0.000 0.000 0.000 0.000 0.000 0.000

1.132 ± 0.123 0.18 ± 1.07 0.925 ± 0.116 −0.21 ± 0.86 0.846 ± 0.132 −0.95 ± 1.07 0 (0)

1 ± 0.106 −1.03 ± 0.77 0.801 ± 0.093 −1.13 ± 0.72 0.728 ± 0.106 −1.9 ± 0.87 17(5.70%)

0.815 ± 0.097 −2.57 ± 1.01 0.692 ± 0.099 −1.95 ± 0.83 0.609 ± 0.113 −2.87 ± 0.92 207(38.05%)

0.000 0.000 0.000 0.000 0.000 0.000 0.000

Prevalent clinical fractures Fresh fractures Old fractures

0 (0) 0 (0)

1(0.42%) 11(4.66%)

3(2.22%) 53(39.26%)

0.042 0.000

0 (0) 0 (0)

2(0.67%) 15(5.03%)

8(1.47%) 199(36.58%)

NS 0.000

(8.05%) (5.71%) (1.12%) (5.61%)

groups. Comparisons among three or more groups were performed using one-way analysis of variance (ANOVA) or the Kruskall-Wallis test followed, when significant, by the Post hoc least significant difference test or the Wilcoxon signed-rank test for pairwise comparisons. Comparisons of categorical variables were made using the χ2 test. Correlations between serum UA and parameters were assessed by the Spearman correlation or Pearson correlation test depending on the distribution of parameters. Multiple stepwise linear regression analysis was carried out to identify the independent variable associated with BMD at the LS, FN and TH. The collinearity diagnostics analysis was also performed to assess whether multiple collinearity exists in these covariates. Multiple logistic regression analysis was performed to ascertain the association between serum UA and the presence of clinical fractures with adjustment for covariates. An enter method was used. Odds ratio (OR) and 95% confidence intervals (CI) for the presence of clinical fractures were estimated. All statistical analyses were performed using the SPSS statistical software (ver 20.0). All reported P values were 2-sided, with a p < 0.05 considered statistically significant.

and < 1.72%, respectively. According to the WHO, osteopenia is diagnosed by a − 2.5 < T-score < −1.0 SD and osteoporosis is diagnosed by a T-score ≤ −2.5 SD at any of sites on the LS, FN and TH [7]. In our present study, the osteoporotic with and without verified selfreported fractures and osteopenic patients with verified self-reported fractures were diagnosed as having diabetic osteoporosis. All participants were asked about the occurrence of clinical fractures, including the time, site and cause of fractures. Self-reported fractures were included in analyses if verified by clinical symptoms, physical examination, medical and hospital records, and radiographs or computed tomography (CT) or magnetic resonance imaging (MRI) or other methods such as whole body bone scan. Diabetic osteoporotic fractures included fractures of the hip, spine, distal forearm or ribs that resulted from minimal or moderate trauma.

2.3. Statistical analysis All data were first analyzed for normality of distribution using the Kolmogorov-Smirnov test of normality. Data are expressed as mean ± SD for continuous variables or percentages (%) for categorical variables, respectively, unless otherwise specified. A Student's t-test or Mann–Whitney U test was used to compare parameters between two 79

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

Table 3 Comparison of serum UA levels, other biochemical and clinical parameters between patients with and without clinical fractures. Men

Postmenopausal women

No clinical fractures

clinical fractures

P

No clinical fractures

clinical fractures

P

Number of subjects Diabetes duration (y) Age (y) BMI (kg/m2) FBG (mmol/l) HbA1c (%) Cr (μmol/l) UA (μmol/l) ALP (U/l) Neutrophil (*109/l) Lymphocyte (*109/l) NLR Urinary ACR(mg/g) ABI VPT (v)

514 7.27 ± 6.5 62.81 ± 9.07 24.35 ± 3.27 10.79 ± 5.03 9.48 ± 2.51 76.53 ± 21.81 336.52 ± 99.85 83.97 ± 51.87 4.57 ± 2.2 1.61 ± 0.64 3.35 ± 2.52 144.95 ± 526.03 1.05 ± 0.13 16.96 ± 10.07

68 8.44 ± 6.33 68.28 ± 9.69 22.98 ± 3.14 9.42 ± 4.31 9.08 ± 2.61 80.84 ± 24.31 300.47 ± 84.90 90.25 ± 87.16 5.03 ± 2.89 1.32 ± 0.59 5.01 ± 5.31 177.09 ± 427.82 1.01 ± 0.16 20.48 ± 11.45

0.047 0.000 0.001 0.023 NS NS 0.020 NS NS 0.000 0.001 NS NS 0.015

756 8.69 ± 6.73 62.9 ± 9.39 24.61 ± 4.23 10.19 ± 5 8.94 ± 2.28 61.53 ± 21.51 303.58 ± 100.02 86.47 ± 49.54 4.57 ± 2.32 1.72 ± 0.7 3.24 ± 3.28 213.12 ± 706.51 1.01 ± 0.14 15.02 ± 8.07

224 8.00 ± 6.49 67.99 ± 8.04 23.57 ± 3.66 9.86 ± 5.08 8.87 ± 2.70 61.37 ± 19.61 269.16 ± 90.80 96.84 ± 55.91 4.82 ± 2.53 1.59 ± 0.60 4.00 ± 5.55 292.53 ± 933.88 0.98 ± 0.17 18.21 ± 10.49

NS 0.000 0.005 NS NS NS 0.000 0.003 NS 0.018 0.002 NS 0.008 0.000

Microvascular complications DPN DR DN

255 (49.61%) 52 (10.12%) 180 (35.02%)

31 (45.59%) 7 (10.29%) 27 (39.71%)

0.533 NS NS

304 (40.21%) 96 (12.70%) 248 (32.80%)

94 (41.96%) 25 (11.16%) 84 (37.50%)

0.639 NS NS

Macrovascular complications PAD Hypertension CHD CI Hyperuricemia

28 (5.45%) 285 (55.45%) 19 (3.70%) 120 (23.35%) 88 (17.12%)

6 (8.82%) 38 (55.88%) 4 (5.88%) 24 (35.29%) 6 (8.82%)

NS NS NS 0.032 NS

51 (6.75%) 472 (62.43%) 40 (5.29%) 172 (22.75%) 186 (24.60%)

26 (11.61%) 133 (59.38%) 15 (6.70%) 57(25.45%) 37 (16.52%)

0.018 NS NS NS 0.011

Hypoglycemic medication Metformin Sulfonylurea Alpha-glucosidase inhibitor Insulin LS BMD (g/cm2) T score FN BMD (g/cm2) T score TP BMD (g/cm2) T score

262 (50.97%) 215 (41.83%) 38 (7.39%) 180 (35.02%) 1.101 ± 0.160 −0.04 ± 1.39 0.877 ± 0.115 −1.13 ± 1.07 0.814 ± 0.127 −1.30 ± 0.73

36 (52.94%) 27 (39.71%) 4 (5.88%) 21 (30.88%) 0.937 ± 0.171 −1.56 ± 1.46 0.716 ± 0.139 −2.67 ± 1.26 0.655 ± 0.159 −2.22 ± 0.92

NS NS NS NS 0.000 0.000 0.000 0.000 0.000 0.000

400 (52.91%) 302 (39.95%) 49 (6.48%) 287 (37.96%) 0.980 ± 0.142 −1.15 ± 1.23 0.797 ± 0.125 −1.16 ± 0.98 0.719 ± 0.136 −1.98 ± 1.1

126 (56.25%) 84 (37.50%) 18 (8.04%) 81 (36.16%) 0.768 ± 0.099 −3.00 ± 0.960 0.674 ± 0.105 −2.11 ± 0.81 0.592 ± 0.127 −3.01 ± 1.03

NS NS NS NS 0.000 0.000 0.000 0.000 0.000 0.000

3. Results

prevalence of hyperuricemia, and BMD values and their corresponding T score at the LS, FN, and TH in both genders (P < 0.01 or P < 0.05). Male patients with diabetic osteoporosis displayed significantly elevated concentrations of neutrophil count and prevalence of fresh fractures compared to those with normal BMD values and osteopenia (P < 0.01 or P < 0.05). Postmenopausal women with diabetic osteoporosis had higher serum Cr concentration and prevalence of PAD and cerebral infarction (CI), and lower concentrations of FBG, HbA1c, and ABI as compared with those with normal BMD values and osteopenia (P < 0.01 or P < 0.05) (Table 2).

3.1. Baseline characteristic of the study population A total of 5941 participants were enrolled. According to the exclusion criteria, 1485 male and 2894 female participants were eliminated. A total of 1562 participants (582 males and 980 females) were finally included in this study (Fig. 1). Table 1 compares the male and postmenopausal female diabetic patients with respect to clinical and biochemical parameters. Serum UA, FBG, HbA1c, Cr, NLR, ABI, VPT, prevalence of DPN, user of alpha-glucosidase inhibitor, and BMD values and their corresponding T score at the LS, FN, and TH were significantly lower in postmenopausal women than in men (P < 0.01 or P < 0.05). On the other hand, ALP, lymphocyte count, prevalence of hypertension and hyperuricemia, user of insulin, and prevalence of diabetic osteoporotic fractures and diabetic osteoporosis were significantly higher, and diabetes duration was significantly longer in postmenopausal women than in men (P < 0.01 or P < 0.05).

3.3. Comparison of serum UA concentrations, other biochemical and clinical parameters between patients with and without clinical fractures Patients with diabetic osteoporotic fractures were significantly older, had lower BMI, serum UA, lymphocyte count, BMD values at the LS, FN, and TH, and higher NLR and VPT values than those without in both genders (P < 0.01 or P < 0.05). Male patients with diabetic osteoporotic fractures had significantly lower concentrations of FBG, longer diabetes duration, and higher prevalence of CI compared to those without (P < 0.01 or P < 0.05). ABI values and prevalence of hyperuricemia were significantly lower, and ALP and prevalence of PAD were significantly higher in postmenopausal women with diabetic osteoporotic fractures than in those without (P < 0.01 or P < 0.05) (Table 3).

3.2. Comparison of serum UA concentrations, and other clinical and biochemical parameters among T2DM patients with normal BMD values, osteopenia and osteoporosis Compared with those with normal BMD values and osteopenia, patients with diabetic osteoporosis had higher age, NLR, urinary ACR, VPT, prevalence of diabetic osteoporotic fractures, and old fractures, longer diabetes duration, and lower BMI, serum UA, lymphocyte count, 80

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

(P < 0.01 or P < 0.05) (Table 4).

Table 4 Bivariate correlation between BMD, serum UA and other biochemical and clinical parameters. LS BMD

FN BMD

r

P

r

P

r

P

0.117 −0.120 0.280 −0.068 −0.154 −0.198 0.212 −0.094 −0.046 −0.002 −0.041 −0.030 −0.005 0.053

0.010 0.008 0.000 NS 0.001 0.000 0.000 0.039 N NS NS NS NS NS

−0.054 −0.242 0.249 −0.006 −0.053 −0.009 0.165 −0.033 −0.127 0.085 −0.136 −0.169 0.032 −0.177

NS 0.000 0.000 NS NS NS 0.000 NS 0.006 NS 0.003 0.001 NS 0.000

−0.042 −0.135 0.266 0.009 −0.031 −0.078 0.175 −0.070 −0.079 0.112 −0.139 −0.135 0.009 −0.125

NS 0.003 0.000 NS NS NS 0.000 NS NS 0.016 0.003 0.007 NS 0.009

Microvascular complications DPN 0.089 DR 0.018 DN 0.007

0.050 NS NS

0.061 −0.069 −0.105

NS NS 0.021

−0.128 −0.042 −0.075

0.005 NS NS

Macrovascular complications PAD 0.067 Hypertension −0.214 CHD −0.098 CI 0.049 Hyperuricemia 0.140

NS 0.000 0.030 NS 0.002

−0.044 0.016 0.014 −0.097 0.087

NS NS NS 0.032 NS

−0.002 0.039 0.041 −0.058 0.100

NS NS NS NS 0.028

Postmenopausal women Diabetes duration 0.040 Age −0.324 BMI 0.209 FBG 0.093 HbA1c 0.055 Cr 0.039 UA 0.230 ALP −0.078 Neutrophil 0.066 Lymphocyte 0.079 NLR −0.019 Urinary ACR −0.023 ABI 0.061 VPT −0.120

NS 0.000 0.000 0.009 NS NS 0.000 0.029 NS 0.027 NS NS NS 0.002

−0.161 −0.517 0.220 0.138 0.126 −0.103 0.172 −0.050 0.002 0.138 −0.107 −0.083 0.173 −0.218

0.000 0.000 0.000 0.000 0.000 0.004 0.000 NS NS 0.000 0.003 0.044 0.000 0.000

−0.120 −0.452 0.252 0.108 0.100 −0.059 0.188 −0.050 −0.002 0.136 −0.107 −0.097 0.168 −0.215

0.001 0.000 0.000 0.002 0.005 NS 0.000 NS NS 0.000 0.003 0.019 0.000 0.000

Microvascular complications DPN −0.029 DR 0.053 DN −0.025

NS NS NS

0.031 0.019 −0.086

NS NS 0.016

0.051 −0.006 −0.097

NS NS 0.006

Macrovascular complications PAD −0.041 Hypertension −0.076 CHD −0.001 CI −0.034 Hyperuricemia 0.166

NS 0.032 NS NS 0.000

−0.150 −0.068 −0.058 −0.086 0.127

0.000 NS NS 0.015 0.000

−0.149 −0.015 −0.001 −0.050 0.127

0.000 NS NS NS 0.000

Men Diabetes duration Age BMI FBG HbA1c Cr UA ALP Neutrophil Lymphocyte NLR Urinary ACR ABI VPT

3.5. Multiple stepwise linear regression of independent variables associated with BMD at each site

TP BMD

In men, serum UA showed a significant positive correlation with BMD values at the LS (P < 0.05). Moreover, BMI, urinary ACR, and age and NLR were significantly correlated with BMD values at all sits, the FN and TH, and FN, respectively (P < 0.01 or P < 0.05). On the other hand, in postmenopausal women, serum UA, age and BMI were significantly associated with BMD values at all sits (all P < 0.01). Moreover, diabetes duration, FBG, and ALP were significantly correlated with BMD values at the LS, and FBG and urinary ACR were significantly associated with BMD values at the TH (P < 0.01 or P < 0.05) (Table 5). 3.6. Associations between the presence of clinical fractures and serum UA concentrations in diabetic patients When no adjustment was made, serum UA concentrations were significantly and inversely associated with the presence of clinical fractures in both genders [men: OR = 0.996, 95% CI = 0.993–0.999, P = 0.005; women: OR = 0.996, 95% CI = 0.994–0.998, P = 0.000] (Model 1). The association between serum UA concentration and the presence ofclinical fractures was not affected by adjustment for diabetes duration, age, BMI, FBG, HbA1c, ALP, Cr, NLR, the presence of diabetic vascular complications [men: OR = 0.996, 95% CI = 0.993–1.000, P = 0.039; women: OR = 0.996,95% CI = 0.994–0.998, P = 0.001] (Model 2). The results were not statistically significant when models were further adjusted for BMD values at the LS, FN, andTH (all P > 0.05) (Model 3) (Table 6). 4. Discussion In this cross-sectional study, we examined whether serum UA concentrations were associated with diabetic osteoporosis (lower BMD values and related fractures) in Chinese patients with T2DM. We found that serum UA concentration was positively correlated with BMD values at each site in both genders in Bivariate correlation analysis, and serum UA concentration was significantly correlated with BMD values at each site in postmenopausal women in Multiple linear regression analysis. Moreover, patients with diabetic osteoporotic fractures had lower concentrations of serum UA and BMD values at each site than their counterparts without diabetic osteoporotic fractures in both genders. Last but most important, serum UA concentrations were significantly and inversely associated with the presence of diabetic osteoporotic fractures independently of all risk factors except for BMD values at each site. The present findings suggest that higher serum UA concentrations might be considered as a marker of higher BMD values, and higher serum UA concentrations may be also related to lower prevalence of diabetic osteoporotic fractures dependently of BMD values. Several lines of evidence suggest that oxidative stress plays a pivotal part in the development of osteoporosis [7–10, 17]. Serum UA is the final breakdown product of purine metabolism in humans, and its concentration at 0.15–0.4 mmol/l can. play a beneficial role as a strong endogenous antioxidant through the free radical scavenging capacity during metabolic stress [30]. Therefore, serum UA may be involved in the pathogenesis of osteoporosis. Our study demonstrated that patients with diabetic osteoporosis had lower serum UA concentrations and prevalence of hyperuricemia as compared to those with normal BMD values and osteopenia in both men and postmenopausal women. In addition, Bivariate correlation analysis revealed that serum UA concentrations were positively correlated with BMD values, and Multiple linear regression analysis showed that serum UA concentration was positively correlated with BMD values at each site in postmenopausal women, and serum UA

3.4. Bivariate correlation between BMD values, serum UA and other clinical and biochemical parameters BMD values at the LS, FN, and TH were significantly and positively correlated with BMI and serum UA, and negatively with age in both genders (P < 0.01 or P < 0.05). BMD values at the LS, FN, and TH were positively correlated with FBG, lymphocyte count and prevalence of hyperuricemia, and inversely with VPT in postmenopausal women (P < 0.01or P < 0.05). BMD values at the FN and TH were positively correlated with HbA1c and ABI, and inversely with NLR, urinary ACR, and prevalence of DN and PAD in postmenopausal women (P < 0.01 or P < 0.05). In men, BMD values at the FN and TH were inversely correlated with NLR, urinary ACR, and VPT, and BMD values at the LS and TH were positively correlated with prevalence of hyperuricemia 81

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

Table 5 Multiple stepwise linear regression of independent variables associated with BMD at each site. Men

LS BMD Age UA Diabetes duration FBG ALP BMI FN BMD Age UA FBG BMI urinary ACR NLR TH BMD UA Age BMI urinary ACR

Postmenopausal women

β

Standardized β

P

0.000

0.138

0.013

0.009

0.190

0.001

−0.003

−0.193

0.000

0.009 −0.023 −0.010

0.234 −0.121 −0.147

0.000 0.020 0.005

0.010 −0.023

0.225 −0.112

0.000 0.038

β

Standardized β

P

−0.007 0.000 0.002 0.003 0.000 0.009

−0.403 0.205 0.103 0.112 −0.081 0.238

0.000 0.000 0.017 0.007 0.046 0.000

−0.008 0.000 0.003 0.010 −0.015

−0.535 0.177 0.102 0.289 −0.079

0.000 0.000 0.006 0.000 0.033

0.000 −0.008 0.012

0.147 −0.501 0.328

0.000 0.000 0.000

In multiple stepwise linear stepwise regression analysis, dependent variable was BMD at each site, independent variables included for analysis were diabetes duration, age, BMI, FBG, HbA1c, Cr, UA, urinary ACR, ALP, neutrophil count, lymphocyte count, NLR, ABI, and VPT.

[31] revealed that serum UA concentrations within the physiologic range were significantly associated with an increase in BMD values at the TH and a reduction in incident nonspine fractures even after adjustment for renal disease. Similarly, Nabipour et al. [17] reported that higher serum UA concentrations were significantly correlated with higher BMD values at various skeletal sites after multivariable adjustment in a large population- based study of older men. The same authors also found that higher serum UA concentrations were associated with a lower prevalence of osteoporosis, and vertebral and nonvertebral fractures. Likewise, a cross-sectional study performed by Ishii et al. [20] showed a significant linear association between serum UA concentrations and BMD values at the LS, independent of y after menopause in 615 peri- and postmenopausal women. Also, Makovey et al. [32] confirmed a similar positive relationship between serum UA and BMD values, and demonstrated that higher baseline serum UA concentrations protected against loss of BMD values at the LS, and forearm during their follow-up. Recently, Zhao et al. [15] performed a cross-sectional study and found that higher serum UA concentrations were associated with higher BMD values in 621 male patients with T2DM, and the risk of osteoporosisor osteopenia in patients in the highest tertile of UA significantly decreased. Morerecently, Ahn et al. [18] has suggested that UA treatment in mice decreased osteoclastogenesis in a dose-dependent manner and reduced the production of ROSin osteoclast precursors, leading to a reduction in bone resorption. Taken together, these findings as well as ours suggest that high serum UA concentration may have the potential to increase BMD values and reduce the risk of fractures, and thus protects against osteoporosis because of antioxidant properties of UA. Several lines of evidence suggest that diabetic vascular complications can lead to reduced blood flow to bone, increase risk of falls, and subsequently may contribute to bone loss and fragility [33–35]. It is well known that DPN and DN are the two most common microvascular complications of T2DM, and are associated with lower BMD values and a higher risk of fracture in different population [36, 37, 38, 39–41]. Evidence has suggested that the VPT test appears to be an appropriate and reliable measure for screening DPN in its early stages [29]. Urinary ACR has been considered to be a prognostic predictor for the development and progression of DN. In the present study, we found that, compared with those with normal BMD and osteopenia, T2DM patients with osteoporosis had higher concentrations of urinary ACR and VPT

Table 6 Associations between the presence of clinical fractures and serum UA levels in diabetic patients. Men

Model 1 UA Model 2 UA Age diabetes duration FBG NLR BMI Model 3 Age PAD NLR LS BMD

Postmenopausal women

OR (95% CI)

P

OR (95% CI)

P

0.996 (0.993–0.999)

0.005

0.996 (0.994–0.998)

0.000

0.996 (0.993–1.000) 1.057 (1.020–1.095)

0.039 0.002

0.996 (0.994–0.998) 1.085 (1.060–1.110) 0.965 (0.938–0.992)

0.001 0.000 0.012

0.912 (0.839–0.991) 1.131 (1.037–1.234) 0.881 (0.795–0.975)

0.030 0.006 0.015

0.955 (0.913–1.000)

0.049

1.044 (1.004–1.086) 2.750 (1.041–7.266)

0.032 0.041

0.005(0.000–0.110)

0.001

1.153 (1.016–1.309) 0.011(0.000–0.265)

0.027 0.005

Multiple logistic regression analysis was performed to ascertain the association between serum UA and the presence of clinical fractures with adjustment for covariates. Model 1 -unadjusted; Model 2-adjusted for diabetes duration, age, BMI, FBG, HbA1c, ALP, Cr, NLR, diabetic microvascular complications (DR, DN, and DPN), diabetic macrovascular complications (PAD, Hypertension, CHD, and CI); Model 3- adjusted for factors listed in Model 2 plus BMD at the LS, FN, and TP.

concentration was positively associated with BMD values at the LS in men. Also, Weshowed that diabetic osteoporotic fractures had lower concentrations of serum UA and prevalence of hyperuricemia compared with those without in both genders, and serum UA concentrations were significantly and inversely associated with the presence of diabetic osteoporotic fractures after adjustment for all risk factors except for BMDvalues at each site. These results indicated that lower concentrations of serum UA may beassociated with lower BMD values and related fractures, and lower serum UA concentrations may participated in the development of diabetic osteoporosis. In line with our findings, several studies investigated the relationship between serum UA concentrationsand BMD values and bone fractures in subjects without diabetes. In a large community-based cohort of elderly men, Lane et al. 82

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

fractures in men. These findings further imply that systemic inflammation, as reflected by increased neutrophil count and NLR and decreased lymphocyte count, may be involved in altered bone metabolism in T2DM patients, and may be etiologically relevant to diabetic osteoporosis and related fracture. Several other risk factors, such as advancing age, estrogen deficiency, longer diabetes duration, low body weight, poor glycemic control, hyperinsulinemia, deposition of advanced glycosylation end products in collagen, inappropriate homeostatic response of parathyroid hormone (PTH) secretion, complex alterations of vitamin D regulation, hypercalciuria, lower insulin-like growth factor-1 (IGF-1), and hypoalbuminemia, have been also reported to be associated with low BMD values and high risk of fractures in T2DM patients. As expected, we found that patients with diabetic osteoporosis had higher age, longer diabetes duration, and lower BMI compared with those with normal BMD and osteopenia, and patients with diabetic osteoporotic fractures were significantly older, and had lower BMI in both genders, and had significantly longer diabetes duration only in men than those without fractures. Moreover, BMD values at all sites were significantly and positively correlated with BMI, and negatively with age in both genders. Multiple linear regression analysis revealed that BMI was positively correlated with BMD at all sites in both genders, and age was inversely associated with BMD at the FN in men and at all sites in postmenopausal women, and diabetes duration was significantly correlated with BMD values at the LS in postmenopausal women. Furthermore, BMI and age were significantly associated with the presence of diabetic osteoporotic fractures in both genders, and diabetes duration was significantly and inversely associated with the presence of diabetic osteoporotic fractures in postmenopausal women. The results of our study largely corroborate previous findings [38, 52, 54] and suggest that these factors, especiallyadvancing age, low body weight, and longer diabetes duration, may play an important role in the pathogenesis of diabetic osteoporosis and related fracture. Hyperglycemia per se may cause hypercalciuria, lower concentrations of 25-hydroxyvitaminD [25(OH)D], increase the generation of advanced glycation end products and adversely affect bone mineralization, bone collagen, bone remodeling and/or strength [55]. Our results showed that patients with diabetic osteoporosis in postmenopausal women had lower FBG and HbA1c compared with those with normal BMD values and osteopenia, and patients with diabetic osteoporotic fractures in men had significantly lower concentrations of FBG compared to those without,implying a better glycemic control in patients with diabetic osteoporosis and related fracture. Surprisingly, BMD values at all sites were positively correlated with FBG and HbA1c in postmenopausal women. Additionally, Multiple linear regression analysis revealed that FBG was significantly correlated with BMD. values at the LS and FN in postmenopausal women, and FBG was significantly associated with the presence of diabetic osteoporotic fractures in men. However, these positive associations between FBG and HbA1c and BMD values were proved by other studies [24, 54, 56]. These results suggest that diabetic osteoporosis and releated fracture appear to be associated with the degree of glycometabolic control. A larger-scale epidemiological and prospective study will be needed to explore the mechanisms underlying the association in the future. This study has some limitations. First, the study design was crosssectionaland did not allow us to infer a causal relationship between serum UA concentration, BMD values and clinical fractures. Therefore, prospective studies with large samples and intervention strategies are further needed to verify our results. Second,we analyzed only subjects aged 45–90 who visited our hospital, a tertiary center, for the education, evaluation or treatment of diabetes mellitus and osteoporosis. Therefore, the patients enrolled in this study might have relatively severe states ofthe disorders and might not be representative of standard Chinese diabetic patients,especially younger men and pre- menopausal women. Third, although we considered the effects of potential

values in both genders, and postmenopausal women with osteoporosis had significantly higher concentration of serum Cr, and slightly higher prevalence of DPN and DN. Moreover, T2DM patients with clinical fractures had higher VPT values than those without in both genders. These rusults together suggested that DPN and DN may be associated with the development of osteoporosis. Additionally, BMD values at the LS, FN, and TH were inversely correlated with with VPT, and BMD values at the FN and TH were inversely correlated with urinary ACR, and prevalence of DN in postmenopausal women. In men, BMD values at the FN and TH were inversely correlated with urinary ACR and VPT. Furthermore, urinary ACR was the independently predictor of BMD at the FN in both genders, and urinary ACR was negatively associated with BMD values at the TH in men. The present findings support the theory that DPN and DN may be linked to lower BMD values, and facilitate the onset of osteoporosis. The ABI is initially a simple noninvasive measure of the severity of lower-extremity PAD, which is representative of diabetic macrovascular complications, and now is also an indicator of atherosclerosis at other vascular sites. It has been suggested that PAD and osteoporosis may be interrelated through a common etiopathogenesis [42]. Many clinical studies have shown that osteoporosis is closely associated with atherosclerosis and cardiovascular disease [43, 44]. In our study, we found that compared with their respective control group, T2DM patients with osteoporosis had higher prevalence of PAD and CI, and lower ABI, and T2DM patients with clinical fractures had significantly lower ABI values, and higher prevalence of PAD in postmenopausal women. Moreover, BMD values at one or two sites were positively correlated with ABI, and inversely with prevalence of hypertension, CI, and PAD in postmenopausal women. Male patients with clinical fractures had slightly lower ABI, and significantly higher prevalence of CI compared to those without, and BMD values at one or two sites were negatively correlated with prevalence of hypertension, CI, and CHD in men. In addition, prevalence of PAD was significantly associated with the presence of clinical fractures in men. Collectively, these results, in line with previous ones [36, 45], demonstrated that diabetic macrovascular complications seem to be associated with loss of BMD values and risk of bone fractures, and play an important role in the development of osteoporosis. Compelling evidence has suggested that chronic inflammation is another major risk factor for systemic bone loss leading to osteoporosis and fracture [46–48]. Some observational studies have also reported associations of several major inflammatory markers, such as C-reactive protein (CRP), interleukin-6 and monocyte chemotactic protein-1, with osteoporosis and/or osteopenia [48]. Neutrophil count has been shown to correlate with high-sensitivity CRP (hs-CRP) concentration better than any other major white cell type in non-diabetic individuals [49]. Lymphocytes are expanded in obese adipose tissue and regulate macrophage production of inflammatory mediators [56]. Lower lymphocyte counts represented a dysregulated inflammatory response and are involved in the progression of atherosclerosis [50, 51]. NLR has emerged as a novel potential inflammation marker in cardiac and noncardiac disorders [49]. In our study, we found that, compared with those with normal BMD and osteopenia, T2DM patients with osteoporosis had higher NLR, and lower lymphocyte count in both genders, and male T2DM patients with osteoporosis displayed significantly elevated concentrations of neutrophil count compared to those without. Also, we showed that T2DM patients with clinical fractures had higher NLR and lower lymphocyte count than those without in both genders. These results, in agreement with previous studies [52, 53], demonstrated that inflammation may play an important role in the development of osteoporosis and related fracture. Furthermore, Bivariatecorrelation analysis revealed that BMD values at the FN and TH were inversely correlated with NLR in both genders, and BMD values at the LS, FN, and TH were positively associated with lymphocyte count in postmenopausal women. In addition, NLR was inversely correlated with BMD values at the FN in Multiple linear regression analysis, and was significantly associated with the presence of diabetic osteoporotic 83

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

confounders on the associations, but the residualconfounding still remains possible due to uncontrolled or unmeasured variables. Fourth, DXA-measured BMD values does not account for bone dimensional changes or allow for separation of the cortical and trabecular bone compartments. We did not measure trabecular bone fractions and bone microarchitecture with a quantitative computed tomography or highresolution peripheral quantitative computed tomography or high-resolution MRI or micro-computed tomography because these measurements are expensive, complicated, and time-consuming, and are not approved for clinical use. Fifth, as we did not measure biochemical markers of bone turnover such as 25(OH)D, PTH and IGF-1 in all participants that were not analyzed in the final results, which might have influenced the results. Finally, the clinical fractures were firstly selfreported and then were confirmed by multiple methods, but not all participants received conventional radiographs, which could lead to misclassification. However, self-reported fractureshave good reliability [57]; frequencies of self-reported fractures are found to be similar to those verified using radiographs. Despite these limitations, the key strengths of this study include the strict inclusion/exclusion criteria, and thorough adjustment for possible confounding variables. Furthermore, the number of subjectswas relatively larger than that of previous study in male T2DM patients, and thatwe measured serum UA concentrations and firstly examined their relationship with BMD values and clinical fractures in separate genders. In conclusion, the present study showed that lower serum UA concentrations may be associated with lower BMD values and higher prevalence of clinical fractures independent of potential confounders except for BMD values at each site, suggesting a potential association between serum UA concentrations and diabetic osteoporosis. Further prospective studies are needed to clarify the associations of serum UA with diabetic osteoporosis and to elucidate the precise underlying mechanism of the association.

[10] M.E. Kuyumcu, Y. Yesil, Z.A. Oztürk, et al., The association between homocysteine (hcy) and serum natural antioxidants in elderly bone mineral densitometry (BMD), Arch. Gerontol. Geriatr. 55 (2012) 739–743. [11] D. Keizman, M. Ish-Shalom, S. Berliner, et al., Low uric acid concentrations in serum of patients with ALS: further evidence for oxidative stress? J. Neurol. Sci. 285 (2009) 95–99. [12] M.A. Hediger, R.J. Johnson, H. Miyazaki, et al., Molecular physiology of urate transport, Physiology (Bethesda) 20 (2005) 125–133. [13] D. Masseoud, K. Rott, R. Liu-Bryan, et al., Overview of hyperuricaemia and gout, Curr. Pharm. Des. 11 (2005) 4117–4124. [14] W.S. Waring, A. Convery, V. Mishra, et al., Uric acid reduces exercise-induced oxidative stress in healthy adults, Clin. Sci. (Lond.) 105 (2003) 425–430. [15] D.D. Zhao, P.L. Jiao, J.J. Yu, et al., Higher serum uric acid is associated with higher bone mineral density in Chinese men with type 2 diabetes mellitus, Int. J. Endocrinol. 2016 (2016) 2528956. [16] T. Muka, E.A. de Jonge, J.C. Kiefte-De Jong, et al., The influence of serum uric acid on bone mineral density, hip geometry, and fracture risk: the Rotterdam study, J. Clin. Endocrinol. Metab. 101 (2016) 1113–1122. [17] I. Nabipour, P.N. Sambrook, F.M. Blyth, et al., Serum uric acid is associated with bone health in older men: a cross-sectional population-based study, J. Bone Miner. Res. 26 (2011) 955–964. [18] S.H. Ahn, S.H. Lee, B.J. Kim, et al., Higher serum uric acid is associated with higher bone mass, lower bone turnover, and lower prevalence of vertebral fracture in healthy postmenopausal women, Osteoporos. Int. 24 (2013) 2961–2970. [19] B.J. Kim, S. Baek, S.H. Ahn, et al., Higher serum uric acid as a protective factor against incident osteoporotic fractures in Korean men: a longitudinal study using the National Claim Registry, Osteoporos. Int. 25 (2014) 1837–1844. [20] S. Ishii, M. Miyao, Y. Mizuno, et al., Association between serum uric acid and lumbar spine bone mineral density in peri- and postmenopausal Japanese women, Osteoporos. Int. 25 (2014) 1099–1105. [21] C. Sritara, B. Ongphiphadhanakul, L. Chailurkit, et al., Serum uric acid concentrations in relation to bone-related phenotypes in men and women, J. Clin. Densitom. 16 (2013) 336–340. [22] T. Mehta, P. Bůžková, M.J. Sarnak, et al., Serum urate concentrations and the risk of hip fractures: data from the cardiovascular health study, Metabolism 64 (2015) 438–446. [23] D. Zhang, I.A. Bobulescu, N.M. Maalouf, et al., Relationship between serum uric acid and bone mineral density in the general population and in rats with experimental hyperuricemia, J. Bone Miner. Res. 30 (2015) 992–999. [24] N. Veronese, F. Bolzetta, M. De Rui, et al., Serum uric acid and incident osteoporotic fractures in old people: The PRO V.A study, Bone 79 (2015) 183–189. [25] L. Zhou, J. Song, S. Yang, et al., Bone mass loss is associated with systolic blood pressure in postmenopausal women with type2 diabetes in Tibet: a retrospective cross-sectional study, Osteoporos. Int. 28 (2017) 1693–1698. [26] N. Dalbeth, R. Topless, T. Flynn, et al., Mendelian randomization analysis to examine for a causal effect of urate on bone mineral density, J. Bone Miner. Res. 30 (2015) 985–991. [27] W. Yang, J. Lu, J. Weng, et al., Prevalence of diabetes among men and women in China, N. Engl. J. Med. 362 (2010) 1090–1101. [28] V. Bartáková, K. Kuricová, L. Pácal, et al., Hyperuricemia contributes to the faster progression of diabetic kidney disease in type 2 diabetes mellitus, J. Diabetes Complicat. 30 (2016) 1300–1307. [29] J. Yang, P.J. Yan, Q. Wan, et al., Association between hemoglobin concentrations and diabetic peripheral neuropathy in patients with type 2 diabetes: a cross- sectional study using electronic health records, J. Diabetes Res. 2017 (2017) 2835981. [30] J. Makovey, M. Macara, J.S. Chen, et al., Serum uric acid plays a protective role for bone loss in peri- and postmenopausal women: a longitudinal study, Bone 52 (2013) 400–406. [31] N.E. Lane, N. Parimi, L.Y. Lui, et al., Association of serum uric acid and incident nonspine fractures in elderly men: the osteoporotic fractures in men (MrOS) study, J. Bone Miner. Res. 29 (2014) 1701–1707. [32] J. Makovey, M. Macara, J.S. Chen, et al., Serum uric acid plays a protective role for bone loss in peri- and postmenopausal women: a longitudinal study, Bone 52 (2013) 400–406. [33] P. Jackuliak, J. Payer, Osteoporosis, fractures, and diabetes, Int. J. Endocrinol. 2014 (2014) 820615. [34] S. Wientroub, D. Eisenberg, R. Tardiman, et al., Is diabetic osteoporosis due to microangiopathy? Lancet 2 (1980) 983. [35] M.T. Vogt, J.A. Cauley, L.H. Kuller, et al., Bone mineral density and blood flow to the lower extremities: the study of osteoporotic fractures, J. Bone Miner. Res. 12 (1997) 283–289. [36] J.H. Kim, M.H. Jung, J.M. Lee, et al., Diabetic peripheral neuropathy is highly associated with nontraumatic fractures in Korean patients with type 2 diabetes mellitus, Clin. Endocrinol. 77 (2012) 51–55. [37] A.V. Schwartz, E. Vittinghoff, D.E. Sellmeyer, et al., Diabetes-related complications, glycemic control, and falls in older adults, Diabetes Care 31 (2008) 391–396. [38] I. Anaforoglu, A. Nar-Demirer, N. Bascil-Tutuncu, et al., Prevalence of osteoporosis and factors affecting bone mineral density among postmenopausal Turkish women with type 2 diabetes, J. Diabetes Complicat. 23 (2009) 12–17. [39] Y. Lim, S. Chun, J.H. Lee, et al., Association of bone mineral density and diabetic retinopathy in diabetic subjects: the 2008-2011 Korea National Health and nutrition examination survey, Osteoporos. Int. 27 (2016) 2249–2257. [40] M. Rix, H. Andreassen, P. Eskildsen, Impact of peripheral neuropathy on bone density in patients with type 1 diabetes, Diabetes Care 22 (1999) 827–831. [41] M. Viégas, C. Costa, A. Lopes, et al., Prevalence of osteoporosis and vertebral fractures in postmenopausal women with type 2 diabetes mellitus and their

Acknowledgments This study is supported by the grants from the Ministry Science and Technology of China (2016YFC0901200, 2016YFC0901205), research grants from Health and Family Planning Commission of Sichuan Province (16129), Sichuan Province Science (Z1448) and Technology Department, and Luzhou Science and Technology Bureau (2013-S-48 (22/30)).The authors would like to thank all the colleagues in linical laboratory center and endocrine laboratory, and all the nurses in our department for their hard work and valuable assistance with this study. References [1] S.A. Abdulameer, S.A. Sulaiman, M.A. Hassali, et al., Osteoporosis and type 2 diabetes mellitus: what do we know, and what we can do? Patient Prefer Adherence 6 (2012) 435–448. [2] C.E. Lampropoulos, I. Papaioannou, D.P. D'Cruz, Osteoporosis–a risk factor for cardiovascular disease? Nat. Rev. Rheumatol. 8 (2012) 587–598. [3] P. Vestergaard, Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2diabetes–a meta-analysis, Osteoporos. Int. 18 (2007) 427–444. [4] Z. Liu, H. Gao, X. Bai, et al., Evaluation of Singh index and osteoporosis self- assessment tool for Asians as riskassessment tools of hip fracture in patients with type 2 diabetes mellitus, J. Orthop. Surg. Res. 12 (2017) 37. [5] L.K. Billings, Y.H. Hsu, R.J. Ackerman, et al., Impact of common variation in bonerelated genes on type 2 diabetes and related traits, Diabetes 61 (2012) 2176–2186. [6] K.K. Nicodemus, A.R. Folsom, Iowa Women's Health Study, Type 1 and type 2 diabetes and incident hip fractures in postmenopausal women, Diabetes Care 24 (2001) 1192–1197. [7] X. Lin, C. Zhao, A. Qin, et al., Association between serum uric acid and bone health in general population: a large and multicentre study, Oncotarget 6 (2015) 35395–35403. [8] D. Maggio, M. Barabani, M. Pierandrei, et al., Marked decrease in plasma antioxidants in aged osteoporotic women: results of a cross-sectional study, J. Clin. Endocrinol. Metab. 88 (2003) 1523–1527. [9] Y. Hamada, H. Fujii, R. Kitazawa, et al., Thioredoxin-1 overexpression in transgenic mice attenuates streptozotocin-induced diabeticosteopenia:a novel role of oxidative stress and therapeutic implications, Bone 44 (2009) 936–941.

84

Clinica Chimica Acta 486 (2018) 76–85

P. Yan et al.

[42]

[43] [44]

[45]

[46] [47] [48] [49]

[50]

relationship with duration of the disease and chronic complications, J. Diabetes Complicat. 25 (2011) 216–221. P. Pennisi, S.S. Signorelli, S. Riccobene, et al., Low bone density and abnormal bone turnover in patients with atherosclerosis of peripheralvessels, Osteoporos. Int. 15 (2004) 389–395. S.I. McFarlane, R. Muniyappa, J.J. Shin, et al., Osteoporosis and cardiovascular disease: brittle bones and boned arteries, is there a link? Endocrine 23 (2004) 1–10. P. von der Recke, M.A. Hansen, C. Hassager, The association between low bone mass at the menopause and cardiovascular mortality, Am. J. Med. 106 (1999) 273–278. T. Adragao, P. Branco, R. Birne, et al., Bone mineral density, vascular calcifications, and arterial stiffness in peritoneal dialysis patients, Perit. Dial. Int. 28 (2008) 668–672. L. Bessueille, D. Magne, Inflammation: a culprit for vascular calcification in atherosclerosis and diabetes, Cell. Mol. Life Sci. 72 (2015) 2475–2489. M.L. Isidro, B. Ruano, Bone disease in diabetes, Curr. Diabetes Rev. 6 (2010) 144–155. J.V. Huang, C.M. Schooling, Inflammation and bone mineral density: a Mendelian randomization study, Sci. Rep. 7 (2017) 8666. C. Lorenzo, A.J. Hanley, S.M. Haffner, Differential white cell count and incident type 2 diabetes: the insulin resistance atherosclerosis study, Diabetologia 57 (2014) 83–92. B. Azab, V. Chainani, N. Shah, et al., Neutrophil-lymphocyte ratio as a predictor of

[51] [52]

[53]

[54]

[55]

[56]

[57]

85

major adverse cardiac events among diabetic population: a 4-year follow-up study, Angiology 64 (2013) 456–465. A.S. Major, S. Fazio, M.F. Linton, B-lymphocyte deficiency increases atherosclerosis in LDL receptor-null mice, Arterioscler. Thromb. Vasc. Biol. 22 (2002) 1892–1898. Z.A. Öztürk, Y. Yesil, M.E. Kuyumcu, et al., Inverse relationship between neutrophil lymphocyte ratio (NLR) and bone mineral density (BMD) in elderly people, Arch. Gerontol. Geriatr. 57 (2013) 81–85. X.Y. Yu, X.S. Li, Y. Li, et al., Neutrophil-lymphocyte ratio is associated with arterial stiffness in postmenopausal women with osteoporosis, Arch. Gerontol. Geriatr. 61 (2015) 76–80. T. Yamaguchi, I. Kanazawa, M. Yamamoto, et al., Associations between components of the metabolic syndrome versus bone mineral density and vertebral fractures in patients with type 2 diabetes, Bone 45 (2009) 174–179. Z. Berberoglu, A.C. Yazici, N.G. Demirag, Effects of rosiglitazone on bone mineral density and remodelling parameters in postmenopausal diabetic women: a 2-year follow-up study, Clin. Endocrinol. 73 (2010) 305–312. L. Ma, L. Oei, L. Jiang, et al., Association between bone mineral density and type 2 diabetes mellitus: a meta-analysis of observational studies, Eur. J. Epidemiol. 27 (2012) 319–332. R.G. Munger, J.R. Cerhan, B.C. Chiu, Prospective study of dietary protein intake and risk of hip fracture in postmenopausal women, Am. J. Clin. Nutr. 69 (1999) 147–152.