Low bone mineral density is associated with intracranial posterior circulation atherosclerosis in women

Bone 81 (2015) 669–674

Contents lists available at ScienceDirect

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

Original Full Length Article

Low bone mineral density is associated with intracranial posterior circulation atherosclerosis in women K. Kang ⁎ Department of Neurology, Eulji General Hospital, Eulji University, Seoul, Republic of Korea Department of Neurology, Gyeongsang National University Hospital, Jinju, Republic of Korea

a r t i c l e

i n f o

Article history: Received 26 March 2015 Revised 14 August 2015 Accepted 28 September 2015 Available online 3 October 2015 Keywords: Bone mineral density; Brain magnetic resonance angiography; Intracranial atherosclerosis; Osteoporosis; Phosphate

a b s t r a c t Low bone mineral density (BMD) is associated with carotid atherosclerosis and the incidence of stroke. However, there are no data on the association of BMD with intracranial atherosclerosis. The study population consisted of 357 participants who underwent dual energy x-ray absorptiometric scanning of the lumbar spine and brain 3D time of flight magnetic resonance angiography as part of their voluntary health checks. The basilar, middle cerebral, intracranial internal carotid and intracranial vertebral arteries were evaluated. Low BMD was defined as a T-score of less than −1. All analyses were stratified by sex and intracranial atherosclerosis location. One hundred seventy-six women (53 years; 66.9% postmenopausal; 33.5% low BMD; 60.2% intracranial atherosclerosis in the anterior circulation (AC); 60.2% intracranial atherosclerosis in the posterior circulation (PC)) and 181 men (51 years; 28.7% low BMD; 61.9% intracranial atherosclerosis in the AC; 55.8% intracranial atherosclerosis in the PC) were included. In women, low BMD was significantly associated with intracranial atherosclerosis in the PC with the odds ratio of 2.57 (95% confidence interval 1.11–5.99). However, intracranial atherosclerosis in the AC was not associated with BMD in women. In men, there were no significant associations between BMD and intracranial atherosclerosis. In conclusion, this study shows that low BMD is associated with subclinical intracranial PC atherosclerosis in women but not in men. © 2015 Elsevier Inc. All rights reserved.

1. Introduction The common and frequently coexisting processes of atherosclerosis and osteoporosis were previously viewed as two separate diseases associated with aging. However, mounting evidence raises the possibility of age-independent relationships between bone mineral density (BMD) and atherosclerosis. A recent meta-analysis demonstrated that decreased BMD is a risk factor for stroke in women [1]. Although extracranial carotid atherosclerosis is known to be related to decreased BMD [2–5], there are no data on the association of BMD with intracranial atherosclerosis, a well-known risk factor for ischemic stroke [6]. Intracranial anterior circulation (AC) arteries and intracranial posterior circulation (PC) arteries are different in many ways. Anatomically, the caliber of PC arteries is smaller than that of AC arteries [7]. Therefore, about 700 mL/min flows through two internal carotid arteries, whereas

Abbreviations: AC, anterior circulation; BMD, bone mineral density; MRA, magnetic resonance angiography; PC, posterior circulation. ⁎ Department of Neurology, Eulji General Hospital, Eulji University, 68 Hangeulbiseokro, Nowon-gu, Seoul 01830, Republic of Korea. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.bone.2015.09.016 8756-3282/© 2015 Elsevier Inc. All rights reserved.

100 to 200 mL/min flows through the vertebrobasilar system [8]. Common and internal carotid arteries bifurcate to form smaller distal arteries (internal and external carotid arteries and anterior and middle cerebral arteries), whereas paired vertebral arteries merge into a single larger distal artery (basilar artery) [8,9]. Histologically, the concentration of perivascular sympathetic nerves is lower in the PC than in the AC [10]. In addition, different risk factors are associated with different locations of intracranial atherosclerosis [11]. Risk factors for middle cerebral artery atherosclerosis were older age and hypertension whereas those for intracranial vertebral artery atherosclerosis were hyperlipidemia and coronary heart disease in an asymptomatic Japanese population [11]. The aim of the present study was to assess the relationship between decreased BMD and intracranial atherosclerosis in a stroke-free Korean population spanning a large age range. Because of relatively few publications on the relationship between BMD and atherosclerosis in men [3], these analyses evaluated sex differences in associations of BMD with intracranial atherosclerosis. In addition, intracranial atherosclerosis in the AC (middle cerebral and intracranial internal carotid arteries) was compared and contrasted with that in the PC (basilar and intracranial vertebral arteries) in terms of their associations with BMD, because different locations of intracranial atherosclerosis are associated with different vascular risk factors [11].

670

K. Kang / Bone 81 (2015) 669–674

2. Materials and methods 2.1. Subjects The study population consisted of 378 subjects who consecutively visited a general health promotion center in a university-affiliated hospital from February 2006 through April 2009. Included subjects underwent dual energy x-ray absorptiometry scans of their spine and brain three-dimensional time of flight magnetic resonance angiography (MRA) as part of their voluntary health checks. Medical records were retrospectively analyzed. Twenty-one subjects were excluded from the analysis due to one of the following criteria: (1) the presence of previous stroke (n = 7), (2) the absence of any recorded blood pressure (n = 3), (3) the absence of standard laboratory test results (n = 4), and (4) artifacts in the MRA (n = 3). Therefore, 357 subjects were analyzed. The procedures followed were in accordance with the ethical standards of an independent ethics committee in the hospital (GNUHIRB2010-048). 2.2. Vascular risk factors

atherosclerosis was graded by consensus after reassessment of MRA images. The grade of atherosclerosis of the artery with the greatest severity of atherosclerosis in the anterior circulation (AC) was taken as the representative grade of atherosclerosis in the AC. The same principle was applied to the representative grade of atherosclerosis in the PC. For the purpose of analysis, the representative atherosclerosis grade of the AC and that of the PC were divided into three atherosclerosis grade groups: atherosclerosis grade 0, atherosclerosis grade 1, and atherosclerosis grades 2–4. 2.4. BMD measurements Dual energy x-ray absorptiometry measurements of areal BMD (g/ cm2) at the lumbar spine (L1–L4) were performed using the Lunar DPX NT (GE Healthcare Lunar, Madison, WI, USA). During the quality assurance procedure run on the scanner every other day, the BMD for bone chambers was measured and checked that it was within 0.03 g/ cm2 of its expected value according to the manufacturer's instructions. The T-score was calculated as the number of standard deviations below or above which the patient's BMD differs from peak bone density of an individual of the same gender and ethnicity using the manufacturer's reference population. Low BMD was defined as a Tscore of less than −1. All the measurements were performed on one device and by one technologist.

As a part of health checks, standardized questionnaires were used to obtain information on medical history, including the presence of hypertension, diabetes mellitus, hypercholesterolemia, and cardiac disease, current medications, menopausal status, alcohol consumption, and current smoking habit. Blood pressure, height, and weight were measured. Body mass index was calculated as mass in kilograms divided by height in meters squared. Blood samples were obtained (early in the morning after an overnight fast, with the last meal generally 10 h before the blood draw) to measure serum calcium, phosphate, glucose, total cholesterol, high-density lipoprotein cholesterol, triglycerides, and creatinine. Laboratory measurements were performed by standard autoanalyzer techniques (Modular DP analyzer, Roche Diagnostics, Mannheim, Germany). The low-density lipoprotein cholesterol was estimated by the method of Friedewald et al. [12]. Hypertension was defined as blood pressure of 140/90 mm Hg or more or a history of physician diagnosis of hypertension. Diabetes was defined as a fasting serum glucose concentration ≥126 mg/dl or a history of physician diagnosis of diabetes. Hypercholesterolemia was defined as a prior diagnosis of hypercholesterolemia, low-density lipoprotein cholesterol concentration ≥160 mg/dl, or total cholesterol concentration ≥240 mg/dl. Glomerular filtration rate was estimated by using the 4-variable Modification of Diet in Renal Disease formula [13]. Chronic kidney disease was defined as glomerular filtration rate b 60 ml/min/1.73 m2. Women were defined as postmenopausal if they were aged 60 years or older or reported having a natural menopause or a bilateral oophorectomy [2]. Current drinkers were defined as participants who drank one or more drinks during the past week.

All analyses were stratified by sex. Demographic variables and vascular risk factors were compared between subjects with and without intracranial atherosclerosis in the AC, and then with and without intracranial atherosclerosis in the PC by chi-square or Fisher's exact tests or Student's t-test. Variables that were associated with the outcome of interest in bivariate analysis at a P value of b0.15 were included in bivariate and multivariable logistic regression models. Low BMD was retained in the multivariable model regardless of statistical significance. Odds ratios (ORs) and 95% confidence intervals (CIs) for intracranial atherosclerosis in the AC and for intracranial atherosclerosis in the PC were estimated separately. Kruskal–Wallis tests were used to determine differences among atherosclerosis grade groups and the Jonckheere–Terpstra test was used to assess trends across atherosclerosis grade groups. To exclude participants who might have cerebral vasculitis or Moyamoya disease rather than intracranial atherosclerosis, sensitivity analyses of 281 participants ≥45 years were carried out. In these sensitivity analyses, ORs and 95% CIs for intracranial atherosclerosis in the AC and for intracranial atherosclerosis in the PC were estimated separately. The level of significance was set at P b 0.05 for all statistical analyses. All statistical analyses were performed using PASW Statistics 18 (SPSS Inc., Chicago, Illinois, USA).

2.3. Magnetic resonance angiography

3. Results

All brain three-dimensional time of flight MRA examinations were performed with a 1.5 T magnetic resonance system (Sonata, Siemens, Erlangen, Germany) [14]. Intracranial atherosclerosis eligible for the study had to be located either in the intracranial carotid artery, the horizontal portion of the middle cerebral artery, the basilar artery, or the intracranial vertebral artery. The extent of intracranial atherosclerosis was classified into 5 grades: Grade 0 (normal; signal reduction b25%); Grade 1 (mild stenosis; ≤25% signal reduction b50%); Grade 2 (moderate stenosis; signal reduction ≥50%); Grade 3 (severe stenosis; focal signal loss with the presence of distal artery signal); and Grade 4 (occlusion) [15]. The intracranial atherosclerosis rating scheme used in this study has been validated in a comparative study with conventional angiography [15]. Intracranial atherosclerosis grade was decided by agreement between two neurologists. When there was no agreement, intracranial

The sex-specific characteristics of 176 female and 181 male participants are shown in Table 1. The age ranged from 20 to 78 years. Prevalence of intracranial atherosclerosis in the AC and intracranial atherosclerosis in the PC was 60.2% and 60.2% in women and 61.9% and 55.8% in men, respectively (Supplementary Table 1). Intracranial atherosclerosis in the AC was not associated with low BMD in women or in men (Table 2). Serum phosphate concentrations were significantly higher and T-scores were significantly lower in women with intracranial atherosclerosis in the PC than in women without intracranial atherosclerosis in the PC (Table 3). Hyperlipidemia and low BMD were significantly more common in women with intracranial atherosclerosis in the PC (Table 3). Hypertension was significantly more common in men with intracranial atherosclerosis in the PC than in men without intracranial atherosclerosis in the PC (Table 3). In women, low BMD was

2.5. Statistical methods

K. Kang / Bone 81 (2015) 669–674 Table 1 Baseline characteristics of study population by sex. Variable

Age (years) Serum calcium (mg/dl) Serum phosphate (mg/dl) Body mass index (kg/m2) BMD (g/cm2) T-score Hypertension (%) Diabetes (%) Hyperlipidemia (%) Cardiac disease (%) Chronic kidney disease (%) Current smokers (%) Current drinkers (%) Post-menopausal (%) Medications (%) Antihypertensive Lipid-lowering Antiresorptive Low BMD (%) Osteopenia Osteoporosis

Women

Men

(n = 176)

(n = 181)

53 ± 10 9.0 ± 0.4 3.9 ± 0.6 24.0 ± 3.3 1.077 ± 0.170 −0.3 ± 1.4 75 (42.6) 13 (7.4) 22 (12.5) 3 (1.7) 2 (1.1) 15 (8.5) 32 (18.2) 116 (65.9)

51 ± 11 9.0 ± 0.3 3.5 ± 0.5 24.4 ± 2.9 1.162 ± 0.154 −0.2 ± 1.3 80 (44.2) 18 (9.9) 24 (13.3) 1 (0.6%) 0 (0.0) 63 (34.8) 112 (61.9) –

42 (23.9) 3 (1.7) 4 (2.3) 59 (33.5) 49 (27.8) 10 (5.7)

32 (17.7) 5 (2.8) 0 (0) 52 (28.7) 50 (27.6) 2 (1.1)

Data are given as the number (percentage) or the mean ± standard deviation. BMD: bone mass density.

significantly associated with intracranial atherosclerosis in the PC (Table 3). In men, however, there were no significant associations between low BMD and intracranial atherosclerosis in the PC (Table 3). In a multivariable analysis of women, low BMD, hyperlipidemia, and higher serum phosphate concentrations were related to the presence of intracranial atherosclerosis in the PC; in a multivariable analysis of men, hypertension was associated with the presence of intracranial atherosclerosis in the PC (Table 4). Next, we analyzed the associations between T-scores and the degree of intracranial atherosclerosis. T-score was correlated with the atherosclerosis grade of the AC in neither women (p = 0.36, Kruskal–Wallis test; p for trend = 0.18, Jonckheere–Terpstra test) nor men (p = 0.64,

671

Kruskal–Wallis test; p for trend = 0.59, Jonckheere–Terpstra test) (Fig. 1A). The median T-score decreased significantly with increasing atherosclerosis grade of the PC in women (p = 0.01, Kruskal–Wallis test; p for trend = 0.007, Jonckheere–Terpstra test) (Fig. 1B); T-score was not correlated with the atherosclerosis grade of the PC in men (p = 0.94, Kruskal–Wallis test; p for trend = 0.74, Jonckheere–Terpstra test) (Fig. 1B). In addition, a sensitivity analysis was conducted to determine whether low BMD was related to the presence of intracranial atherosclerosis among participants ≥ 45 years (Supplementary Tables 2–4). Low BMD was significantly associated with intracranial atherosclerosis in the PC with an OR of 2.49 (1.14–5.44, 95% CI) only in women (Supplementary Tables 3, 4). Intracranial atherosclerosis in the AC was not associated with low BMD in women or men (Supplementary Tables 2, 4). 4. Discussion This study shows that low BMD is associated with subclinical intracranial PC atherosclerosis in women but not in men. Similar relationships were found in sensitivity analyses limited to older participants (≥45 years). In addition, an inverse relationship was identified between BMD and degree of intracranial PC atherosclerosis in women. The results from this study are similar to those of other investigators who have shown associations between BMD and extracranial carotid atherosclerosis and between BMD and incident stroke [1–4]. Many factors can affect disease development in both the bones and arteries. First, an atherosclerosis-lipid-bone model has been suggested to account for associations between low BMD and atherosclerosis [16]. Oxidized low-density lipoprotein-cholesterol promotes osteoblastic differentiation of vascular smooth muscle cells leading to the mineralization in the artery wall [17]. In bone, however, oxidized low-density lipoprotein-cholesterol stimulates bone marrow stromal cells to undergo adipogenic rather than osteoblastic differentiation leading to decreased bone mass [18]. Second, osteoprotegerin, a soluble decoy receptor that inhibits the receptor activator of nuclear factor кB ligand, is recognized as a regulator of both bone metabolism and atherosclerosis [5,19–22]. Third, the importance of inflammation has been investigated. Proinflammatory cytokines such as interleukin-1, tumor

Table 2 Vascular risk factors by the absence or presence of intracranial atherosclerosis in the AC. Variable

Age (years) Serum calcium (mg/dl) Serum phosphate (mg/dl) Body mass index (kg/m2) BMD (g/cm2) T-score Hypertension (%) Diabetes (%) Hyperlipidemia (%) Cardiac disease (%) Chronic kidney disease (%) Current smokers (%) Current drinkers (%) Post-menopausal (%) Medications (%) Antihypertensive Lipid-lowering Antiresorptive Low BMD (%) Osteopenia Osteoporosis

Women

Men

No intracranial atherosclerosis in the AC

Intracranial atherosclerosis in the AC

p

(n = 70)

(n = 106)

52 ± 9 9.0 ± 0.3 3.9 ± 0.5 24.4 ± 3.3 1.097 ± 0.160 −0.2 ± 1.3 26 (37.1) 2 (2.9) 9 (12.9) 1 (1.4) 2 (2.9) 6 (8.6) 16 (22.9) 45 (64.3)

54 ± 9 9.0 ± 0.4 3.9 ± 0.6 23.8 ± 3.2 1.064 ± 0.176 −0.4 ± 1.4 49 (46.2) 11 (10.4) 13 (12.3) 2 (1.9) 0 (0) 9 (8.5) 16 (15.1) 71 (67.0)

0.19 0.38 0.47 0.28 0.21 0.25 0.23 0.06 0.91 N0.99 0.16 0.99 0.19 0.71

16 (22.9) 1 (1.4) 1 (1.4) 21 (30) 19 (27.1) 2 (2.9)

26 (24.5) 2 (1.9) 3 (2.8) 38 (35.8) 30 (28.3) 8 (7.5)

0.8 N0.99 N0.99 0.42

No intracranial atherosclerosis in the AC

Intracranial atherosclerosis in the AC

(n = 69)

(n = 112)

50 ± 12 9.0 ± 0.3 3.5 ± 0.5 24.4 ± 3.1 1.175 ± 0.168 −0.2 ± 1.4 25 (36.2) 10 (14.5) 9 (12.9) 0 (0) 0 (0) 25 (36.2) 46 (66.7) –

525 ± 10 9.0 ± 0.3 3.4 ± 0.5 24.4 ± 2.8 1.154 ± 0.145 −0.3 ± 1.2 55 (49.1) 14 (12.5) 13 (12.3) 1 (0.9) 0 (0) 38 (33.9) 66 (58.9) –

0.29 0.57 0.5 0.99 0.36 0.5 0.09 0.7 N0.99 N0.99 N0.99 0.75 0.3 –

9 (13.0) 3 (4.3) 0 (0) 20 (29) 18 (26.1) 2 (2.9)

23 (20.5) 2 (1.8) 0 (0) 32 (28.6) 32 (28.6) 0 (0)

0.2 0.37 N0.99 0.95

Data are given as the number (percentage) of each group or the mean ± standard deviation. AC: anterior circulation; BMD: bone mass density.

p

672

K. Kang / Bone 81 (2015) 669–674

Table 3 Vascular risk factors by the absence or presence of intracranial atherosclerosis in the PC. Variable

Age (years) Serum calcium (mg/dl) Serum phosphate (mg/dl) Body mass index (kg/m2) BMD (g/cm2) T-score Hypertension (%) Diabetes (%) Hyperlipidemia (%) Cardiac disease (%) Chronic kidney disease (%) Current smokers (%) Current drinkers (%) Post-menopausal (%) Medications (%) Antihypertensive Lipid-lowering Antiresorptive Low BMD (%) Osteopenia Osteoporosis

Women

Men

No intracranial atherosclerosis in the PC

Intracranial atherosclerosis in the PC

(n = 70) 52 ± 9 9.0 ± 0.3 3.8 ± 0.5 24.2 ± 3.5 1.110 ± 0.164 −0.1 ± 1.3 26 (37.1) 3 (4.3) 3 (4.3) 0 (0%) 1 (1.4) 7 (10.0) 16 (22.9) 43 (61.4)

(n = 106) 54 ± 10 9.0 ± 0.4 4.0 ± 0.6 23.9 ± 3.1 1.056 ± 0.172 −0.5 ± 1.4 49 (46.2) 10 (9.4) 19 (17.9) 3 (2.8) 1 (0.9) 8 (7.5) 16 (15.1) 73 (68.9)

18 (25.7) 0 (0) 2 (2.9) 14 (20) 10 (14.3) 4 (5.7)

24 (22.6) 3 (2.8) 2 (1.9) 45 (42.5) 39 (36.8) 6 (5.7)

p

0.13 0.66 0.01⁎ 0.56 0.17 0.04⁎ 0.23 0.2 0.007⁎ 0.28 N0.99 0.57 0.19 0.31 0.64 0.28 0.65 0.003⁎

No intracranial atherosclerosis in the PC

Intracranial atherosclerosis in the PC

(n = 80) 50 ± 12 9.9 ± 0.3 3.5 ± 0.5 24.0 ± 2.9 1.158 ± 0.145 −0.3 ± 1.2 24 (30) 8 (10) 13 (16.3) 0 (0) 0 (0) 26 (32.5) 49 (61.3) –

(n = 101) 52 ± 9 8.9 ± 0.3 3.4 ± 0.5 24.8 ± 2.9 1.165 ± 0.162 −0.2 ± 1.3 56 (55.4) 10 (9.9) 11 (10.9) 1 (1) 0 (0) 37 (36.6) 63 (62.4) –

12 (15) 2 (2.5) 0 (0) 22 (27.5) 21 (26.3) 1(1.3)

20 (19.8) 3 (3) 0 (0) 30 (29.7) 29 (28.7) 1 (1)

p

0.25 0.5 0.53 0.09 0.76 0.59 0.001⁎ 0.98 0.29 N0.99 N0.99 0.56 0.88 – 0.40 N0.99 N0.99 0.75

Data are given as the number (percentage) of each group or the mean ± standard deviation. PC: anterior circulation; BMD: bone mass density. ⁎ p b 0.05.

necrosis factor, and interleukin-6 can stimulate osteoclast precursor cells and mature osteoclasts leading to augmenting the effects of the receptor activator of nuclear factor кB ligand on osteoclast formation and bone resorption [19]. The proinflammatory cytokines are also produced in vascular tissues, and their roles in atherosclerosis are well known [23]. Another explanation for the inverse association of BMD with atherosclerosis implicates endothelial function, which has an important role in the prevention of atherosclerosis by the release of vasodilators such as nitric oxide [24]. In this study, low BMD and serum phosphate were associated with intracranial atherosclerosis in the PC but not with that in the AC. It is unclear why low BMD would show much stronger associations with

Table 4 Logistic regression model analyses for the presence of intracranial atherosclerosis. Crude OR (95% CI)

Adjusted OR (95% CI)

Diabetes Current drinkers Low BMD

Women Intracranial atherosclerosis in the AC 3.94 (0.85–18.34) 3.69 (0.78–17.39) 0.6 (0.28–1.3) 0.7 (0.31–1.54) 1.3 (0.68–2.49) 1.24 (0.64–2.42)

Age (per 1 year increase) Serum phosphate (per 1 mg/dl increase) Hyperlipidemia Low BMD

Intracranial atherosclerosis in the PC 1.03 (0.99–1.06) 0.99 (0.96–1.03) 2.05 (1.16–3.63) 1.84 (1.02–3.32) 4.88 (1.39–17.17) 4.43 (1.21–16.25) 2.95 (1.46–5.95) 2.57 (1.11–5.99)

Hypertension Antihypertensive medication Low BMD

Men Intracranial atherosclerosis in the AC 1.7 (0.92–3.14) 1.55 (0.75–3.17) 1.72 (0.75–3.98) 1.28 (0.48–3.4) 0.98 (0.51–1.9) 0.96 (0.49–1.88)

Body mass index (per 1 kg/m2 increase) Hypertension Low BMD

Intracranial atherosclerosis in the PC 1.09 (0.99–1.21) 1.06 (0.95–1.18) 2.9 (1.56–5.39) 2.72 (1.45–5.1) 1.11 (0.58–2.14) 1.14 (0.58–2.26)

OR: odds ratio; CI: confidence interval; AC: anterior circulation; BMD: bone mass density; PC: posterior circulation.

intracranial atherosclerosis in the PC than would that in the AC. Different conventional vascular risk factors are related to different locations of intracranial atherosclerosis [11]. Like other risk factors, BMD could exert differential effects, depending on the location and type of the vascular bed. In the Multi-Ethnic Study of Atherosclerosis, for example, lower BMD was associated with lower ankle-brachial index and greater extracranial internal carotid artery intimal-medial thickness but not with common carotid artery intimal-medial thickness in men [3]. The same study showed a stronger association of BMD with coronary artery calcium score compared with abdominal aortic calcium score assessed by computed tomography in women [4]. It is not clear why an association between BMD and intracranial atherosclerosis was observed in women and not in men in this study. In the aforementioned Multi-Ethnic Study of Atherosclerosis, lower BMD was associated with coronary calcification in women but not in men [4]. In the Framingham Heart Study, metacarpal bone loss was associated with the progression of aortic calcification in women but not in men [25]. In accordance with the results from this study, a recent meta-analysis revealed that decreased BMD is associated with increased risk of incident stroke in women but not in men [1]. Considering that intracranial atherosclerosis is an important risk factor for stroke, hormonal factors unique to women may be involved in the underlying pathophysiology linking BMD and intracranial atherosclerosis. Estrogen deficiency is a well-known risk factor for both osteoporosis and atherosclerosis in women [19,26]. After natural or surgical menopause, the incidence of cardiovascular disease in women increases [26]. Estrogen use reduces atheromatous plaque formation in animal models and incidence of cardiovascular disease in young women [26], while inhibiting bone resorption and increasing bone mass in women [19]. Some randomized clinical trials have shown antiatherogenic effects of estrogen in young, healthy postmenopausal women who do not have pre-existing atherosclerosis [27,28]. The beneficial effects of estrogen on early atherosclerotic lesions are mediated through its influence on the plasma lipid profile, endothelial nitric oxide synthase, proinflammatory cytokines, and osteoprotegerin [21,26]. In addition, estrogen has phosphaturic action [4]. Higher serum phosphate concentrations have been associated with extracranial carotid atherosclerosis and the risk of stroke [14, 29]. Elevated extracellular phosphate concentrations affect multiple

K. Kang / Bone 81 (2015) 669–674

673

rates, as opposed to the static approach by BMD measurements [30]. Therefore, intracranial atherosclerosis might be more closely associated with changes in bone turnover markers than BMD. Fifth, most of the participants were stroke-free adults who were so concerned about their health status to pay considerable money for a voluntary health check that included a brain MRA. Healthy adults might have been more interested in undergoing an MRA if they had headache, dizziness, or a family history of stroke, raising the possibility of selection bias. Last, only a small fraction of participants had osteoporosis or high grade intracranial atherosclerosis in this study population. Results may differ in other populations with a higher prevalence of osteoporosis or severe intracranial atherosclerosis. In conclusion, this investigation demonstrated significant associations between lumbar BMD and intracranial PC atherosclerosis, as assessed by brain MRA, in women but not men. Author's disclosure Kyusik Kang declares that he has no conflict of interest. Acknowledgments This study was supported by a grant of the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare, Republic of Korea (HI10C2020). The funding agency had no role in study design; the collection, analysis, and interpretation of data; the writing of the report; and the decision to submit the article for publication. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bone.2015.09.016. References

Fig. 1. Changes in T-score with atherosclerosis grade of the AC (A) and that of the PC (B). Box plots demonstrate median, 25th, and 75th percentile values for T-score. As shown, there was a linear decrease in T-scores as the atherosclerosis grade of the PC increased only in women (B). All p values were calculated using the Jonckheere–Terpstra test.

signaling pathways that promote vascular calcification, including decreased calcification inhibitors, increased extracellular matrix degradation, osteogenic differentiation and apoptosis of vascular smooth muscle cells, and vesicle release [14]. Interestingly, this study shows that the inverse associations between serum phosphate and intracranial atherosclerosis and between BMD and intracranial atherosclerosis are all limited to women and to the intracranial atherosclerosis in the PC. Abnormal regulation of mineral metabolism may explain the inverse association of BMD with intracranial atherosclerosis and sex differences in that association. The study has limitations. First, measures of BMD and intracranial atherosclerosis were taken at one point in time, which limits conclusions on causality. Second, BMD was assessed from a single bone site. In elderly participants, spinal BMD should be interpreted with caution because of errors caused by osteoarthritic changes [19]. Third, estrogen deficiency in women was indirectly estimated via age and menopausal status. Fourth, adequate prescribing and adherence data were lacking on the dose, duration of treatment, or type of estrogen, progesterone, or glucocorticoid. In addition, additional markers of osteoporosis and skeletal health (parathyroid and thyroid hormones, vitamin D, and bone turnover markers) were not available. Bone turnover markers provide a dynamic picture of the actual bone resorption and formation

[1] P.K. Myint, A.B. Clark, C.S. Kwok, et al., Bone mineral density and incidence of stroke: European prospective investigation into cancer-Norfolk population-based study, systematic review, and meta-analysis, Stroke 45 (2014) 373–382. [2] L. Jorgensen, O. Joakimsen, G.K. Rosvold Berntsen, I. Heuch, B.K. Jacobsen, Low bone mineral density is related to echogenic carotid artery plaques: a population-based study, Am. J. Epidemiol. 160 (2004) 549–556. [3] J.A. Hyder, M.A. Allison, E. Barrett-Connor, et al., Bone mineral density and atherosclerosis: the multi-ethnic study of atherosclerosis, abdominal aortic calcium study, Atherosclerosis 209 (2010) 283–289. [4] N.E. Jensky, J.A. Hyder, M.A. Allison, et al., The association of bone density and calcified atherosclerosis is stronger in women without dyslipidemia: the multi-ethnic study of atherosclerosis, J. Bone Miner. Res. 26 (2011) 2702–2709. [5] J.R. Shaffer, C.M. Kammerer, D.L. Rainwater, et al., Decreased bone mineral density is correlated with increased subclinical atherosclerosis in older, but not younger, Mexican American women and men: the San Antonio family osteoporosis study, Calcif. Tissue Int. 81 (2007) 430–441. [6] O.Y. Bang, Intracranial atherosclerosis: current understanding and perspectives, J. Stroke 16 (2014) 27–35. [7] M. Misra, M.S. Alp, D. Hier, J.I. Ausman, Multidisciplinary treatment of posterior circulation ischemia, Neurol. Res. 26 (2004) 67–73. [8] M.S.V. Elkind, R.L. Sacco, Pathophysiology, classification, and epidemiology, in: L.P. Rowland, T.A. Pedley (Eds.), Merritt's Neurology, Lippincott Williams & Wilkins, Philadelphia 2010, pp. 250–263. [9] P. Amarenco, L.R. Caplan, M.S. Pessin, Vertebrobasilar occlusive disease, in: H.J.M. Barnett, J.P. Mohr, B.M. Stein, F.M. Yatsu (Eds.), Stroke : Pathophysiology, Diagnosis, and Management, W.B. Saunders, Philadelphia 1998, pp. 513–598. [10] L. Edvinsson, C. Owman, N.O. Sjoberg, Autonomic nerves, mast cells, and amine receptors in human brain vessels. A histochemical and pharmacological study, Brain Res. 115 (1976) 377–393. [11] T. Uehara, M. Tabuchi, E. Mori, Risk factors for occlusive lesions of intracranial arteries in stroke-free Japanese, Eur. J. Neurol. 12 (2005) 218–222. [12] W.T. Friedewald, R.I. Levy, D.S. Fredrickson, Estimation of the concentration of lowdensity lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin. Chem. 18 (1972) 499–502. [13] A. Shankar, K.E. Lee, B.E. Klein, et al., Estimating glomerular filtration rate in a population-based study, Vasc. Health Risk Manag. 6 (2010) 619–627. [14] K. Kang, Serum calcium and phosphate concentrations and intracranial atherosclerosis, Atherosclerosis 232 (2014) 249–253. [15] T. Uehara, E. Mori, M. Tabuchi, Y. Ohsumi, A. Yamadori, Detection of occlusive lesions in intracranial arteries by three-dimensional time-of-flight magnetic resonance angiography, Cerebrovasc. Dis. 4 (1994) 365–370.

674

K. Kang / Bone 81 (2015) 669–674

[16] S.I. McFarlane, R. Muniyappa, J.J. Shin, G. Bahtiyar, J.R. Sowers, Osteoporosis and cardiovascular disease: brittle bones and boned arteries, is there a link? Endocrine 23 (2004) 1–10. [17] F. Parhami, A.D. Morrow, J. Balucan, et al., Lipid oxidation products have opposite effects on calcifying vascular cell and bone cell differentiation. A possible explanation for the paradox of arterial calcification in osteoporotic patients, Arterioscler. Thromb. Vasc. Biol. 17 (1997) 680–687. [18] M.E. Nuttall, A.J. Patton, D.L. Olivera, D.P. Nadeau, M. Gowen, Human trabecular bone cells are able to express both osteoblastic and adipocytic phenotype: implications for osteopenic disorders, J. Bone Miner. Res. 13 (1998) 371–382. [19] J.A. Lorenzo, E. Canalis, L.G. Raisz, Metabolic bone disease, in: S. Melmed, K.S. Polonsky, P.R. Larsen, H.M. Kronenberg (Eds.), Williams Textbook of Endocrinology, Elsevier/Saunders, Philadelphia 2011, pp. 1305–1349. [20] M. Schoppet, J.R. Schaefer, L.C. Hofbauer, Low serum levels of soluble RANK ligand are associated with the presence of coronary artery disease in men. Circulation 107 (2003) e76 (author reply e.). [21] M.L. Frost, R. Grella, S.C. Millasseau, et al., Relationship of calcification of atherosclerotic plaque and arterial stiffness to bone mineral density and osteoprotegerin in postmenopausal women referred for osteoporosis screening, Calcif. Tissue Int. 83 (2008) 112–120. [22] F. Biscetti, G. Straface, S. Giovannini, et al., Association between TNFRSF11B gene polymorphisms and history of ischemic stroke in Italian diabetic patients, Hum. Genet. 132 (2013) 49–55.

[23] J. Pfeilschifter, R. Koditz, M. Pfohl, H. Schatz, Changes in proinflammatory cytokine activity after menopause, Endocr. Rev. 23 (2002) 90–119. [24] Hamilton CJ, Reid LS, Jamal SA. Organic nitrates for osteoporosis: an update. Bonekey Rep. 2013;2:259. [25] Kiel DP, Kauppila LI, Cupples LA, Hannan MT, O'Donnell CJ, Wilson PW. Bone loss and the progression of abdominal aortic calcification over a 25 year period: the Framingham heart study. Calcif. Tissue Int. 2001;68:271-6. [26] P. Ouyang, E.D. Michos, R.H. Karas, Hormone replacement therapy and the cardiovascular system lessons learned and unanswered questions, J. Am. Coll. Cardiol. 47 (2006) 1741–1753. [27] H.N. Hodis, W.J. Mack, R.A. Lobo, et al., Estrogen in the prevention of atherosclerosis. A randomized, double-blind, placebo-controlled trial, Ann. Intern. Med. 135 (2001) 939–953. [28] J.E. Manson, M.A. Allison, J.E. Rossouw, et al., Estrogen therapy and coronary-artery calcification, N. Engl. J. Med. 356 (2007) 2591–2602. [29] L. Ruan, W. Chen, S.R. Srinivasan, J. Xu, A. Toprak, G.S. Berenson, Relation of serum phosphorus levels to carotid intima-media thickness in asymptomatic young adults (from the Bogalusa heart study), Am. J. Cardiol. 106 (2010) 793–797. [30] O. Bruyere, J. Collette, P. Delmas, et al., Interest of biochemical markers of bone turnover for long-term prediction of new vertebral fracture in postmenopausal osteoporotic women, Maturitas 44 (2003) 259–265.