ORIGINAL ARTICLE: BEHAVIORALAND SOCIAL SCIENCES

Breast arterial calcification andhypertension associated with

vertebral fractureggi_775 330..335

Hiroshi Wada,1 Fuminori Hirano,2 Tatsuhiko Kuroda3 and Masataka Shiraki4

1Wada Women’s Clinic, 2Department of Internal Medicine, Asahikawa Medical College, Asahikawa,3Public health research foundation, Tokyo, and 4Research Institute and Practice for Involutional Diseases,

Nagano, Japan

Aim: Arterial calcification and osteoporosis commonly accompany one another in post-menopausal women. Hypertension is a known contributing factor to arterial calcification.Thus, we aimed to investigate any associations between hypertension, arterial calcificationand vertebral fractures in a cross-sectional study in Japanese postmenopausal women.

Methods: The medical histories of 421 postmenopausal Japanese women diagnosedwith hypertension, diabetes mellitus or hyperlipidemia were investigated. Bodyweight,body height and ultradistal bone mineral density (BMD) were measured. The prevalentvertebral fractures were diagnosed by a semiquantitative method, and the number of breastarterial calcifications (BAC) was investigated by mammography screening.

Results: Patients with vertebral fractures were of a significantly higher age, lower height,lower ultradistal BMD and had a higher number of BAC compared with those withoutvertebral fractures. Furthermore, a significantly higher prevalence of hypertension wasobserved in the patients with vertebral fractures as compared with those without. Amultivariate stepwise regression analysis using these variables for vertebral fracturesshowed that the significant odds ratios (OR) of age (OR 1.76, 95% CI 1.11–2.77,P = 0.016), the prevalence of BAC (OR 2.52, 95% CI 1.62–3.93, P < 0.001) and thepresence of hypertension (OR 1.76, 95% CI 1.11–2.80, P = 0.017) were found as significantindependent risk factors for vertebral fractures.

Conclusion: This is the first report of the relevance of BAC or hypertension to vertebralfractures in Japanese women. The results suggest that hypertension, BAC and osteoporoticfractures share a common metabolic pathway in their pathogenesis. Geriatr Gerontol Int2012; 12: 330–335.

Keywords: breast arterial calcification, hypertension, mammography, vertebral fracture.

Introduction

Because osteoporosis and vascular diseases are commondiseases of the elderly, there might be several etiologicalmechanisms, including lifestyle, in common for thesediseases.1,2 The association of low bone mass with

atherosclerosis has been reported in Caucasians andAsians.3–7 However, the link between hypertension andosteoporosis has not been paid as much attention as therelationship between osteoporosis and atherosclerosis.Previous studies have focused on the relationshipsbetween causal factors of hypertension, such as sodiumor calcium intakes and osteoporosis.8–12 These previousreports showed that high sodium or low calcium intakeinduces high bone turnover or rapid loss of bone mass.However, there are no available data showing the directrelationships between hypertension and osteoporosis orfractures.

Accepted for publication 30 September 2011.

Correspondence: Dr Hiroshi Wada MD, Wada Women’s Clinic,4-15-8 Asahikawa-city, Hokkai-do 070-0034, Japan. Email:wada@potato.hokkai.net

Geriatr Gerontol Int 2012; 12: 330–335

330 � © 2011 Japan Geriatrics Societydoi: 10.1111/j.1447-0594.2011.00775.x

The link between atherosclerosis and osteoporosissuggests that common metabolic pathways might existin bones and vessels in terms of tissue calcification.Among of them, estrogen might play a central role inthe calcification of bone and vessels, because both ofthese organs express estrogen receptors.13 In fact, estro-gen status has been reported to correlate with thecalcium content of coronary atherosclerotic plaques inwomen,14 and estrogen treatment has been reported toreduce coronary artery calcium scores in a subanalysisof a Women’s Health Initiative study,15 in whichestrogen was effective in preventing hip fractures. Fur-thermore, it is well known that long-term estrogentreatment has been shown to increase breast cancerrisks.16 Additionally, higher bone mineral density(BMD) has been described as one of the risks for breastcancer.17,18

Mammography (MMG) has been applied widely toJapanese women to screen for breast cancer. Mammo-grams have been shown to allow for a secondary analy-sis of associations between breast arterial calcification(BAC) and other pathologies, such as systemic athero-sclerosis19 and low BMD.20 The data from these studiesshow that there might be a common pathogenesisunderlying osteoporosis and systemic vascular sclerosis.Because hypertension is known to be a prominent riskfor atherosclerosis, it is interesting to investigate therelationships between hypertension, atherosclerosis andosteoporotic fractures. Thus, in the present study, weattempted to investigate the possible associations ofhypertension, vascular sclerosis and osteoporotic frac-tures in postmenopausal women. To achieve this, weevaluated BAC, as it might be a good indicator foratherosclerosis because of its quantitative nature; andvertebral fractures, which are thought to be a typicalphenotype of osteoporosis. Consequently, we investi-gated the relationships among BAC, hypertension andvertebral fractures in Japanese postmenopausal womenusing a cross-sectional study design.

Methods

Subject selection

The present study was a cross-sectional study offemale residents in Hokkaido who voluntarily visitedour clinics for a complete health check, including thepresence or absence of hypertension, diabetes mellitus,hyperlipidemia and osteoporosis, and MMG screening.We enrolled all consecutive patients after exclusion ofthose who met the exclusion criteria, which were use ofsteroids, secondary osteoporosis, renal dysfunction orrheumatism. All the participants were postmenopausalwomen. After giving informed consent, participantswere subjected to BMD and MMG screening. The studyperiod was from April 2008 to May 2009.

Data collection

Patients’ age and years after the onset of menopausewere recorded, as well as body mass index (BMI; kg/m2)calculated from bodyweight and body height. Medicalhistories were investigated by consulting the question-naire and results of the clinical laboratory tests, includ-ing diagnoses of hypertension (systolic blood pressure3140 mmHg or diastolic pressure 390 mmHg), diabetesmellitus (HbA1c 36.5%), hyperlipidemia (low-densitylipoprotein 3140 mg/dL or high-density lipoprotein<40 mg/dL) and vascular events were carefully recordedfrom all the participants. Furthermore, the participantswere asked for the medical histories of the treatments ofthe aforementioned diseases. Serum and urine sampleswere obtained to measure serum levels of calcium (Ca),phosphorus (P), alkaline phosphatase (Al-P) and urinaryexcretion of the N-terminal telopeptide of type I col-lagen (NTX; Osteomark Ostex, Princeton, NJ, USA).

Measurements of BMD and assessment of prevalentvertebral fractures

Ultradistal (UD) BMD was measured by dual energyX-ray absorptiometry (DXA) using a Hologic QDR4500 (Hologic, Waltham, MA, USA) apparatus, andvertebral X-ray films were taken to assess the presenceor absence of vertebral fractures according to themethod of Genant et al.21 The diagnosis of osteoporosiswas made in accordance with the Japanese Criteria ofOsteoporosis.22 Briefly, a UD BMD less than 70% of theYoung Adult Mean (YAM), by measurements of WadaWomen’s Clinic in Hokkaido, Japan, was considered asan osteoporotic diagnosis. Likewise, a BMD of 70–80%of the YAM with the prevalence of a fragility fracturewas also diagnosed as osteoporosis.

Assessment of BAC

The MMG was converted into a photograph usingalpha-RT (Instrumentarium Corporation, Tuusula,Finland). BAC was assessed using images saved by theFuji Film Profect CS+ Synapse system (Fujifilm Medical,Tokyo, Japan). The total number of BAC in both breastsin a mediolateral oblique image of both breasts wasevaluated by the method of Kemmeren et al.23 The sta-bility over time of the intra-rater reliability of the BACmeasurements in our system was evaluated (n = 30).The kappa coefficient value was >0.90, reflecting anexcellent level.

Statistical analysis

The data are expressed as mean 1 SD. Baseline charac-teristics were compared by ANOVA between groups withor without vertebral fractures. The prevalent rate of

Vascular calcification and fracture

© 2011 Japan Geriatrics Society � 331

vertebral fractures and BAC, between the presence andabsence of complications, were tested by a c2-test. Sig-nificant independent predictors for vertebral fracturewere evaluated simultaneously through stepwise mul-tiple regression analyses. A significant difference wasdetermined as P < 0.05 in the two-tailed test. All theanalyses were carried out using JMP software, Version9.0.0 (SAS Institute, Cary, NC, USA).

Results

After the exclusion of patients who met the exclusioncriteria from a total of 490 participants, 421 consecutiveparticipants were enrolled in the present analysis. Themean age (SD) of participants was 66.6 years (8.0).Among those, 229 (54.4%) participants were diagnosedwith osteoporosis, 115 (27.3%) participants with hyper-tension, 216 (51.3%) participants with hyperlipidemiaand 43 (10.2%) participants were found to have diabetesmellitus. Of the participants diagnosed with osteoporo-sis, 203 (88.6%) received anti-resorptive treatment,such as bisphosphonate or selective estrogen receptormodulator. As for the participants with hypertension,six (5.2%) received antihypertensive treatment, suchas angiotensin receptor blocker or calcium channelblocker. When the participants were divided into twogroups according to their vertebral fracture statuses,261 (62.0%) participants were diagnosed without preva-lent fractures, whereas 160 (38.0%) participants werediagnosed with prevalent fractures, respectively. Char-acteristics of the two groups are shown in Table 1. Theparticipants with prevalent vertebral fractures were ofhigher age, BMI and BAC number, and were of shorterstature and UD BMD than the participants withoutvertebral fractures. The correlation coefficient by Spear-man’s test between BAC number and the number ofprevalent vertebral fractures was 0.336 (P < 0.001), andthe UD BMD was -0.216 (P < 0.001). Next, we calcu-lated the prevalence of the accompanying diseases in the

patients with or without vertebral fractures or BAC. Asignificant association of hypertension with vertebralfractures and with BAC was observed (Table 2 andFig. 1). In contrast, there were no significant associa-tions between hyperlipidemia or diabetes mellitus withvertebral fractures or BAC. A stepwise regression analy-sis was carried out to evaluate the contribution ofselected factors on vertebral fractures (Table 3). Theodds ratios (OR) for the vertebral fractures were 1.76 forage over 70 years (P = 0.016), 2.52 for presence of BAC(P < 0.001) and 1.76 for hypertension (P = 0.017). Incontrast, height (BMI) and UD BMD were not signifi-cantly associated with vertebral fractures.

Discussion

Osteoporotic vertebral fractures are a factor connectingdisability and increasing mortality in elderly people.24–28

Pre-existing vertebral fractures are reported to lead tosubsequent incident fractures in not only the vertebrae,but also in the proximal femur.29 Notably, the preva-lence of vertebral fractures in the Japanese is higher thanthat in Caucasians.29 Thus, the elucidation of fracturepathogenesis and the prediction of future vertebralfracture occurrence are important for the managementof osteoporosis, especially in Japanese women. In thepresent study, we investigated whether there were anysignificant associations between vertebral fractures,hypertension and vascular calcification in Japanesepostmenopausal women, using a cross-sectional studydesign.

Vascular calcification was assessed within the breastartery, which can be evaluated by MMG, quan-titatively. Indeed, we observed a close relationshipbetween BAC and prevalent vertebral fractures in thepresent study population (Table 1). In addition, hyper-tension was significantly associated with prevalentvertebral fractures and BAC simultaneously (Table 2).Those relationships were significant, even after

Table 1 Characteristics of participants divided by prevalent vertebral fracture

Prevalent vertebral fracture P (ANOVA)No Yes

Age, years 64.5 1 7.2 (261) 69.9 1 8.1 (160) <0.001Height, cm 152.5 1 5.5 (261) 150.1 1 5.6 (160) <0.001Weight, kg 49.0 1 6.6 (261) 50.0 1 6.9 (160) 0.153BMI, kg/cm2 21.1 1 2.8 (261) 22.2 1 2.8 (160) <0.001Ultradistal BMD, g/cm2 0.31 1 0.05 (261) 0.29 1 0.06 (160) <0.001ALP, U/L 256.2 1 67.3 (246) 256.7 1 78.7 (153) 0.945NTX, nmol BCE/mmol Cr 57.0 1 27.6 (218) 53.3 1 29.0 (132) 0.236BAC number 0.62 1 1.06 (261) 1.48 1 1.54 (160) <0.001

Values are mean 1 SD and number. ALP, alkaline phosphatase; BAC, breast arterial calcification; BCE, bone collagenequivalent; BMD, bone mineral density; BMI, body mass index; Cr, creatinine; NTX, N-terminal telopeptide of type I collagen.

H Wada et al.

332 � © 2011 Japan Geriatrics Society

adjustment of possible confounding factors (Table 3).Therefore, the present results suggest a possiblemediation of hypertension between BAC and vertebralfractures, because a significantly higher prevalenceof hypertension was shown in the subjects with BACand with vertebral fractures than those without elevatedblood pressure (Fig. 1). The findings of the presentstudy are consistent with previous reports regardingvascular calcification and BAC.14 Osteoporosis orlow BMD is known to be associated with cardio-vascular disease,2 peripheral artery disease3,4 andatherosclerosis.5–7 Therefore, a relationship betweenatherosclerosis and osteoporosis is unsurprising.Furthermore, previous reports have suggested thatthe link between hypertension and osteoporosisis mainly caused by high sodium and low calciumintake.9–12 Only one previous paper reported that hyper-tension was a risk factor for hip and vertebral fracturesand not a risk factor for forearm fractures in a largecase–control study.8 However, little is known regardingwhat type of factor(s) might mediate osteoporosis and

atherosclerosis or hypertension.1 One possible mediatorof low bone mass and atherosclerosis might be lowestrogen levels,13 because these two morbid statesusually occur after menopause, and estrogen repletionis reduced in both atherosclerosis and osteoporoticfractures.14–16 Additionally, hypertension might have aconsiderable role in the occurrence of vertebral frac-tures8 and atherosclerosis, because the OR of hyperten-sion for vertebral fractures was 1.76 (P = 0.017), andhypertension was significantly associated with BAC inthe present population (Table 2). Other reports showedthat a high sodium intake induced high bone turnoverand excessive calcium loss into the urine,10,11 and thatthe Dietary Approaches to Stop Hypertension dietreduced bone turnover,12 suggesting that long-termexistence of a mineral imbalance might play an impor-tant role in fracture occurrence.8

As the present study design was a cross-sectionalstudy, a prospective study will be required to confirmthis relationship. In addition, because the presentpopulation consisted of patients who were undergoing a

Table 2 Complication of participants divided by vertebral fracture and breast arterial calcification

Complication Prevalent vertebral fracture BACYes No Yes% P (c2-test) Yes No Yes% P (c2-test)

HypertensionNo 101 205 33.0% <0.001 117 189 38.2% 0.015Yes 59 56 51.3% 59 56 51.3%

HyperlipidemiaNo 71 134 34.6% 0.165 79 126 38.5% 0.168Yes 89 127 41.2% 97 119 44.9%

Diabetes mellitusNo 140 238 37.0% 0.225 153 225 40.5% 0.101Yes 20 23 46.5% 23 20 53.5%

BAC, breast arterial calcification.

0%

25%

50%

75%

100%

BAC no BAC yes BAC no BAC yes

Vertebral fracture no

Vertebral fracture yes

Hypertension yesHypertension no

Figure 1 Percentages of the patients with vertebralfractures in relation to presence or absence of hypertensionand breast arterial calcification (BAC). The vertebral fracturerates of BAC subjects were significantly higher than thosewithout BAC (c2-test; P < 0.001 in no hypertension subjects,P = 0.012 in hypertension subjects).

Table 3 Multiple regression analysis to vertebralfracture

Item OR 95% CI P

Age, 370 vs <70 years 1.76 1.11 2.77 0.016Height, Q4 vs Q1–3 0.67 0.40 1.11 0.120UD BMD, osteoporosis

vs normal + osteopenia0.72 0.46 1.11 0.141

BAC, yes vs no 2.52 1.62 3.93 <0.001Hypertension, yes vs no 1.76 1.11 2.80 0.017

All the data were categorized as shown in the Table. Q4 forheight meant the highest quartile of height. Ultra-distalbone mineral density (UD BMD) was divided into threecategories of osteoporosis, osteopenia and normal inaccordance with the Japanese Criteria. Age, breast arterialcalcification (BAC) and hypertension showed significantlyhigher odds ratio (OR) for vertebral fractures.

Vascular calcification and fracture

© 2011 Japan Geriatrics Society � 333

voluntary heath check program, the participants had ahigher awareness of their health than the general popu-lation, and this might have caused a selection bias.However, the prevalence rates of hypertension, diabetesand hyperlipidemia were in good agreement with thoseof the general population of Japanese postmenopausalwomen,30 suggesting that the characteristics of thepresent population were appropriate for the presentstudy. In addition, if the present population hada tendency toward a healthier lifestyle than the generalpopulation, the relationships among hypertension,atherosclerosis and bone fractures might result inan underestimate. However, against expectations, thepresent population showed a significant relationshipafter adjustments for confounding factors. These find-ings might provide proof for the generalization of thepresent results.

In the present study, the ultradistal BMD was notassociated with vertebral fractures. This no-correlationmight be explained by the higher prevalence rate ofmedication history for osteoporosis. However, thiswas not a surprising phenomenon, because significantdecreases in the BMD are usually observed at specificsites in patients with fractures. Thus, if the lumbar spineBMD had been available in the present study, it mighthave been linked to the vertebral fractures. This aspectmust be confirmed in a future study.

In conclusion, BAC in a MMG study showed a sig-nificant association between vertebral fractures andhypertension. In addition, hypertension was linkedto the presence of prevalent vertebral fractures. Thesefindings might show that the pathogenesis of vertebralfracture, BAC and hypertension share common meta-bolic abnormalities. A prospective study will be requiredto confirm this potential relationship.

Disclosure statement

There are no potential conflicts of interest related to thiswork and manuscript, including specific financial inter-ests relevant to the subject of the manuscript.

References

1 McFarlane SI, Muniyappa R, Shin JJ, Bahtiyar G, SowersJR. Osteoporosis and cardiovascular disease: brittle bonesand boned arteries, is there a link? Endocrine 2004; 23:1–10.

2 Tanko LB, Christiansen C, Cox DA, Geiger MJ, McNabbMA, Cummings SR. Relationship between osteoporosisand cardiovascular disease in postmenopausal women.J Bone Miner Res 2005; 20: 1912–1920.

3 Mangiafico RA, Russo E, Riccobene S et al. Increasedprevalence of peripheral arterial disease in osteoporoticpostmenopausal women. J Bone Miner Metab 2006; 24:125–131.

4 van der Klift M, Pols HA, Hak AE, Witteman JC, HofmanA, de Laet CE. Bone mineral density and the risk of

peripheral arterial disease: the Rotterdarm Study. CalcifTissue Int 2002; 70: 443–449.

5 Tamaki J, Iki M, Hirano Y et al. Low bone mass is associ-ated with carotid atherosclerosis in postmenopausalwomen: the Japanese population-based osteoporosis(JOPS) cohort study. Osteoporosis Int 2009; 20: 53–60.

6 Kim SH, Kim YM, Cho MA et al. Echogenic carotid arteryplaques are associated with vertebral fractures in post-menopausal women with low bone mass. Calcif Tissue Int2008; 82: 411–417.

7 Sumino H, Ichikawa S, Kasama S et al. Relationshipbetween carotid atherosclerosis and lumbar spine bonemineral density in postmenopausal women. Hypertens Res2008; 31: 1191–1197.

8 Vestergaard P, Rejnmark L, Mosekilde L. Hypertension isa risk for fracture. Calcif Tissue Int 2009; 84: 103–111.

9 Woo J, Kwok T. Dietary intake, blood pressure andosteoporosis. J Hum Hypertens 2009; 23: 451–455.

10 Cappuccio FP, Kalaitzidis R, Duneclift S, Eastwood JB.Unravelling the links between calcium excretion, saltintake, hypertention, kidney stones and bone metabolism.J Nephrol 2000; 13: 169–177.

11 Caudarella R, Vascini F, Rizzoli E, Francucci CM. Saltintake, hypertension, and osteoporosis. J Endocrinol Invest2009; 32 (4): 15–20.

12 Lin PH, Ginty F, Appel LJ et al. The DASH diet andsodium reduction improve markers of bone turnover andcalcium metabolism in adults. J Nutr 2003; 133: 3130–3136.

13 Smiley DA, Khalil RA. Estrogenic compounds, estrogenreceptors and vascular cell signaling in the aging bloodvessels. Curr Med Chem 2009; 16: 1863–1887.

14 Christian RC, Harrington S, Edwards WD, Oberg AL,Fitzpatric LA. Estrogen status correlates with the calciumcontent of coronary atherosclerotic plaques in women.J Clin Endocr Metab 2002; 87: 1062–1067.

15 Manson JE, Allison MA, Rossouw JE et al. WHI and WHI-CACS Investigators. Estrogen therapy and coronary-arterycalcification. N Engl J Med 2007; 356: 2591–2602.

16 Writing group for the Women’s Health Initiative Investi-gators. Risks and benefits of estrogen plus progestin inhealthy postmenopausal women. Principal results from theWomen’s Health Initiative randomized controlled trial.JAMA 2002; 288: 321–333.

17 Cauley JA, Song J, Dowsett SA, Mershon JL, CummingsSR. Risk factors for breast cancer in older women: therelative contribution of bone mineral density and otherestablished risk factors. Breast Cancer Res Treat 2007; 102:181–182.

18 Van der Klift M, de Laet CE, Coebergh JW, Hofman A,Pols HA. Rotterdam Study. Bone mineral density and therisk of breast cancer: the Rotterdam Study. Bone 2003; 32:211–216.

19 Van Noord PA, Beijerinck D, Kemmeren JM, van der GraafY. Mammograms may convey more than breast cancerrisk: breast arterial calcification and arterio-scleroticrelated diseases in women of the DOM cohort. Eur J CancerPrev 1996; 5: 483–487.

20 Reddy J, Bilezikian JP, Smith SJ, Mosca L. Reduced bonemineral density is associated with breast arterial calcifica-tion. J Clin Endocrinol Metab 2008; 93: 208–211.

21 Genant HK, Jergas M, Palermo L et al. Comparison ofsemiquantitative visual and quantitative morphometricassessment of prevalent and incident vertebral fractures inosteoporosis: The study of osteoporotic fractures researchgroup. J Bone Miner Res 1996; 11: 984–996.

H Wada et al.

334 � © 2011 Japan Geriatrics Society

22 Orimo H, Hayashi Y, Fukunaga M et al. OsteoporosisDiagnostic Criteria Review Committee: Japanese Societyfor Bone and Mineral Research. Osteoporosis diagnosticcriteria for primary osteoporosis: year 2000 revision. J BoneMiner Metab 2001; 19: 331–337.

23 Kemmeren JM, Beijerinck D, van Noord PA et al. Breastarterial calcifications: association with diabetes mellitusand cardiovascular mortality. Work in progress. Radiology1996; 201: 75–78.

24 Kanis JA, McCloskey EV. Epidemiology of vertebralosteoporosis. Bone 1992; 13: S1–S10.

25 Oleksik A, Lips P, Dawson A et al. Health-related quality oflife in postmenopausal women with low BMD with orwithout prevalent vertebral fractures. J Bone Miner Res 2000;15: 1384–1392.

26 Silverman SL, Minshall ME, Shen W, Harper KD, Xie S,Health-Related Quality of Life Subgroup of the MultipleOutcomes of Raloxifene Evaluation Study. The relation-ship of health-related quality of life to prevalent and

incident vertebral fractures in postmenopausal womenwith osteoporosis: results from the Multiple Outcomes ofRaloxifene Evaluation Study. Arthritis Rheum 2001; 44:2611–2619.

27 Kuroda T, Shiraki M, Tanaka S, Ohta H. Contributions of25-hydroxyvitamin D, co-morbidities and bone mass tomortality in Japanese post-menopausal women. Bone 2009;44: 168–172.

28 Kado DM, Browner WS, Palermo L, Nevitt MC, GenantHK, Cummings SR. Vertebral fractures and mortality inolder women: a prospective study. Study of OsteoporoticFractures Research Group. Arch Intern Med 1999; 159:1215–1220.

29 Fujiwara S, Kasagi F, Masunari N, Naito K, Suzuki G,Fukunaga M. Fracture prediction from bone mineraldensity in Japanese men and women. J Bone Miner Res 2003;18: 1547–1553.

30 National Health and Nutrition Survey. 2006. JapaneseMinistry of Health, Labour and Welfare.

Vascular calcification and fracture

© 2011 Japan Geriatrics Society � 335