The Effects of 3 Years of Calcium Supplementation on Common Carotid Artery Intimal Medial Thickness and Carotid Atherosclerosis in Older Women: An Ancillary Study of the CAIFOS Randomized Controlled Trial

Authors

  • Joshua R Lewis,

    Corresponding author
    1. School of Medicine and Pharmacology, Sir Charles Gairdner Hospital Unit, University of Western Australia, Nedlands, Australia
    2. Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands, Australia
    • Address correspondence to: Joshua Lewis, PhD, School of Medicine and Pharmacology, University of Western Australia, Sir Charles Gairdner Hospital, Hospital Avenue, Nedlands, Perth, WA 6009, Australia. E-mail: joshua.lewis@uwa.edu.au

    Search for more papers by this author
  • Kun Zhu,

    1. School of Medicine and Pharmacology, Sir Charles Gairdner Hospital Unit, University of Western Australia, Nedlands, Australia
    2. Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands, Australia
    Search for more papers by this author
  • Peter L Thompson,

    1. Department of Cardiology, Sir Charles Gairdner Hospital, Nedlands, Australia
    Search for more papers by this author
  • Richard L Prince

    1. School of Medicine and Pharmacology, Sir Charles Gairdner Hospital Unit, University of Western Australia, Nedlands, Australia
    2. Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands, Australia
    Search for more papers by this author

  • For a Commentary on this article, please see Bauer (J Bone Miner Res. 2014;29:531–533. DOI: 10.1002/jbmr.2184).

ABSTRACT

Calcium is an essential nutrient for skeletal health; however, it has been suggested that supplemental calcium may be associated with adverse cardiovascular effects, raising widespread concern about their use. One suggested mechanism is via increasing carotid atherosclerosis, however few randomized controlled trials (RCT) of calcium supplements have assessed these mechanisms. The calcium intake fracture outcome study (CAIFOS) was a 5-year RCT (1998 to 2003) of 1.2 g of elemental calcium in the form of calcium carbonate in 1460 elderly women. An ancillary study of 1103 women assessed common carotid artery intimal medial thickness (CCA-IMT) and carotid atherosclerosis at year 3 (2001). The effects of supplementation were studied in intention-to-treat (ITT) and per-protocol (PP) analyses before and after adjustment for baseline cardiovascular risk factors. The mean age of participants at baseline was 75.2 ± 2.7 years. In ITT analyses, women randomized to calcium supplementation had no difference in multivariable-adjusted mean CCA-IMT (calcium 0.778 ± 0.006 mm, placebo 0.783 ± 0.006 mm, p = 0.491) and maximum CCA-IMT (calcium 0.921 ± 0.007 mm, placebo 0.929 ± 0.006 mm, p = 0.404). Women randomized to calcium did not have increased carotid atherosclerosis (calcium 47.2%, placebo 52.7%, p = 0.066). However, in women taking at least 80% of the supplements, a significant reduction in carotid atherosclerosis was observed in unadjusted but not in multivariate-adjusted models (p = 0.033 and p = 0.064, respectively). Participants in the highest tertile of total calcium (diet and supplements) had reduced carotid atherosclerosis in unadjusted and multivariable-adjusted analyses compared with participants in the lowest tertile (odds ratio [OR] = 0.67 [95% confidence interval (CI) 0.50–0.90], p = 0.008, and OR = 0.70 [95% CI 0.51–0.96], p = 0.028, respectively). In conclusion, these findings do not support the hypothesis that calcium supplementation increases carotid artery intimal medial thickness or carotid atherosclerosis, and high calcium intake may reduce this surrogate cardiovascular risk factor. © 2014 American Society for Bone and Mineral Research.

Introduction

Calcium is a widely used dietary supplement for the prevention of age-related bone loss in the elderly.[1, 2] The recent Institute of Medicine review of the scientific literature concluded that the available scientific evidence supports a key role of calcium and vitamin D in the maintenance of skeletal health and recommends a daily intake of 1200 mg calcium with 800 IU of vitamin D in the elderly.[3] Calcium supplementation has been shown to reduce blood pressure[4] and may improve energy balance and body composition;[5] however, recently the safety of calcium for extraskeletal outcomes, in particular cardiovascular disease outcomes in the elderly, has been questioned. Given the widespread use of calcium supplementation, even small increases in cardiovascular risk factors could have major public health implications.

Human arteries are comprised of three layers: the intima, the media, and the adventitia. The intima is the innermost layer of the artery and is composed of connective tissue, whereas the media is the wall of the artery and consists of layers of smooth muscle cells. The outermost layer, the adventitia, is highly vascular and comprised of a dense collagenous structure and thus provides much of the nutrition needed for the vessel wall. Atherosclerosis occurs principally in the intima of medium-sized and large arteries such as the coronary, carotid, vertebral, aortic, and iliac arteries. Early lesions typically begin in childhood and early adulthood and primarily consist of lipid-laden macrophages and smooth muscle cells that can be distinguished from normal artery as areas of discoloration on the vessel wall and will either disappear or progress further to become atherosclerotic lesions that worsen with age.[6] Advanced atherosclerotic lesions form fibrous plaques composed of smooth muscle cells, lipid-laden macrophages, and T lymphocytes and appear as raised white areas in the vessel wall that are easily visualized using either invasive or noninvasive imaging modalities.[7]

B-mode carotid ultrasound is a noninvasive, sensitive, and highly reproducible modality that measures the distance from the lumen-intima interface to the media-adventitia interface of the carotid wall, giving a measure of the carotid intimal medial thickness and presence of focal plaques. It is considered a surrogate measure to quantify subclinical carotid calcification and atherosclerosis, and it is considered a cardiovascular disease risk factor.[8-11]

The objective of this study was to use B-mode carotid ultrasound data from an ancillary study of 1103 elderly women enrolled in the CAIFOS randomized controlled trial to investigate whether 3 years of calcium supplementation increased carotid artery intimal medial thickness (CCA-IMT) and carotid atherosclerosis.

Materials and Methods

Participants

The participants involved in this study were recruited in 1998 to a 5-year, randomized, controlled trial of oral calcium supplements to prevent osteoporotic fractures as described previously.[12] Briefly, women were recruited from the Western Australian general population of women older than 70 years by mail using the electoral roll, a requirement of citizenship. More than 99% of Australians of this age are registered on the roll. Of the 5586 women who responded to a letter inviting participation, 1510 women were willing and eligible, and, of these, 1460 women were recruited for the study. Participants were ambulant and did not have any medical conditions likely to influence 5-year survival. They were excluded if they were receiving bone-active agent, including hormone replacement therapy. Participants were similar in terms of disease burden and pharmaceutical consumption to whole populations of this age, but they were more likely to be from higher socioeconomic groups.[13] In the 5 years of the trial, participants received 1.2 g of elemental calcium as calcium carbonate daily or a matched placebo. A preplanned ancillary study at year 3 to investigate epidemiological determinants of CCA-IMT and carotid atherosclerosis by B-mode carotid ultrasound examination was undertaken in 1103 participants of the original cohort. Informed consent was obtained, and the Human Ethics Committee of the University of Western Australia approved the study.

Overview of Calcium Intake Fracture Outcome Study (CAIFOS) randomized controlled trial

Patients received calcium carbonate tablets, 0.6 g twice per day (with morning and evening meals), or identical placebo tablets (Wyeth Consumer Healthcare, Baulkham Hills, Australia). The randomization list was produced by generating 146 blocks of 10 numbers. In each block, 5 positions representing placebo and 5 positions representing calcium treatment were ordered using a letter code according to a random number generator. The numbered blocks were ordered according to randomly generated numbers, and an identification number was assigned to each letter code in the randomized list. The Pharmacy Department of the Sir Charles Gairdner Hospital (Nedlands, Australia) assigned a treatment to the letter code and assigned the appropriate medications to the patient according to this list. The randomization was stratified by allocating patients to blocks according to whether a prevalent nontraumatic fracture had occurred after age 50 years, ensuring that an equal number of patients with and without a prevalent fracture received placebo or calcium. Medication compliance was checked at the completion of the study by counting returned tablets at each 12-month review and was calculated as a percentage of the optimum. Average yearly compliance of less than 80% was classified as noncompliant.

Baseline atherosclerotic vascular disease (ASVD) risk assessment

The participants provided their previous medical history and current medications verified by their general practitioner. These data were coded using the International Classification of Primary Care-Plus (ICPC-Plus) method.[14] The coding methodology allows aggregation of different terms for similar pathologic entities as defined by the International Classification of Disease (ICD)-10 coding system. These data were then used to determine the presence of preexisting diabetes (T89001-90009). Cardiovascular medications included beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, HMG-Co A reductase inhibitors, and antiplatelet agents. Smoking status was coded as nonsmoker or ex-smoker/current smoker if they had consumed more than one cigarette per day for more than 3 months at any time in their life. Weight was assessed using digital scales with participants wearing light clothes and no shoes. Height was assessed using a stadiometer, and the body mass index (BMI) was calculated in kg/m2 at baseline. Blood pressure was measured on the right arm with a mercury column manometer using an adult cuff after the participants have been seated in an upright position and had rested for 5 minutes. An average of three blood pressure readings was recorded. Mean arterial pressure (MAP) was calculated using the following equation: [(2 × diastolic blood pressure) + systolic blood pressure]/3.

Dietary calcium intake

A validated semiquantitative food-frequency questionnaire developed by the Cancer Council of Victoria was used to assess baseline dietary intake, including calcium[15] in 1050 of 1103 (95.2%) of participants. The process of collection was identical, whereby a research assistant supervised the completion of the questionnaire in small groups. Food models, cups, spoons, and charts for frequency were provided. Energy and nutrient intakes were estimated based on frequency of consumption and an overall estimate of usual portion size.[16] Total supplemental calcium intake was calculated by multiplying the overall study percentage tablet compliance by 12 to convert to mg/d and adding dietary calcium intake in mg/d to provide a total calcium intake (mg/d).

Biochemical assessment

Blood samples were collected at baseline for assessment. Plasma 25-hydroxy vitamin D3 (25OH-D3) concentrations were determined using liquid chromatography tandem mass spectrometry method (RDDT Laboratories, Melbourne, Australia) using the Agilent 6410 Triple Quadrupole LC-MS/MS. Total cholesterol, high-density lipoprotein cholesterol (HDLC), and triglyceride concentrations were determined using a Hitachi 917 auto analyzer (Roche Diagnostics, Mannheim, Germany). Low-density lipoprotein cholesterol (LDLC) was calculated using Friedewald's method.[17] Baseline creatinine was measured using an isotope dilution mass spectrometry (IDMS) traceable Jaffe kinetic assay on a Hitachi 917 analyzer (Roche Diagnostics). Estimated glomerular filtration rate (eGFR) using creatinine was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation[18] and was added to the multivariable-adjusted model as this has been shown to predict ASVD in this cohort.[19]

Prevalent atherosclerotic vascular disease

Atherosclerotic hospitalizations were retrieved from the Western Australian Data Linkage System (WADLS) for each of the study participants from 1980 to 1998. WADLS provides a complete validated record of every participant's hospitalizations from the coded records of each participant's hospital admissions. Atherosclerotic events were defined using the principal discharge diagnosis codes from the International Classification of Diseases, Injuries and Causes of Death Clinical Modification (ICD-9-CM).[20] These codes included: ischemic heart disease (ICD-9-CM codes 410–414); heart failure (ICD-9-CM code 428); cerebrovascular disease excluding hemorrhage (ICD-9-CM codes 433–438); and peripheral arterial disease (ICD-9-CM codes 440–444).

B-mode carotid ultrasound

The presence of carotid focal plaques and common carotid artery intimal medial thickness (CCA-IMT) were determined at year 3. Assessments were performed using B-mode carotid ultrasound examination by the same sonographer with an 8.0-mHz linear array transducer fitted to an Acuson Sequoia 512 (Mountain View, CA, USA) ultrasound machine using a standard image acquisition protocol in 2001.[21] The far walls of the distal 2 cm of the left and right common carotid arteries were examined, and images were taken from three different angles (anterolateral, lateral, and posterolateral) to account for the possibility of asymmetrical wall thickening. End-diastolic images were recorded and a semi-automated edge-detection software program was used for image analysis. The same technician performed offline analysis of all of the images. After assessment of CCA-IMT and focal plaque on the right side, the process was repeated on the left side. The CCA-IMT from each of the six images (three on either side) was averaged to give an overall mean CCA-IMT. Once IMT images were recorded, the entire carotid tree (CCA, carotid bulb, internal and external carotid) was examined for the presence of focal plaque, defined as a clearly identified area of focal increased thickness (≥1 mm) of the intima-media layer. A short-term precision study of 20 nontrial subjects with repeat IMT measurements between 0 and 31 days apart (mean 10.3 days) was performed, which yielded a coefficient of variation of 5.98% as described previously.[22]

Statistics

Mean or maximum CCA-IMT was a continuous variable (mm), whereas the presence of a carotid atherosclerotic plaque was used as a dichotomous (yes/no) variable. Baseline characteristics were stratified by treatment groups (calcium and placebo), and were expressed as mean and standard deviation (SD) for continuous variables or as number and percentage for categorical variables. For presence of an atherosclerotic plaque, unadjusted and multivariable-adjusted logistic regression models were used to examine the effects of calcium supplementation (yes/no) on atherosclerotic plaque (yes/no) with results expressed as odds ratios (OR) and 95% confidence intervals (CI). To investigate a “dose” relationship between dietary calcium and calcium supplements, tertiles of total calcium intake were calculated and tested in unadjusted and multivariable-adjusted analyses. The covariates in the multivariable-adjusted models included age, BMI, smoking history, cardiovascular disease medications, diabetes, history of clinical ASVD, and estimated glomerular filtration rate by the CKD-EPI equation. Analyses were undertaken using SPSS (version 18; SPSS Inc., Chicago, IL, USA) and SAS (version 9, SAS Institute Inc., Cary, NC, USA) programs. Any p values <0.05 in two-tailed testing were considered statistically significant.

Results

The study design is shown in Fig. 1. Of 1460 participants, 1103 (76%) had B-mode carotid ultrasound assessment at year 3 of the study and were included. A total of 158 participants were unavailable to participate in the study because of withdrawal or death. These participants had a higher proportion with a history of diabetes (p = 0.003) and atherosclerotic disease (p = 0.018) than those who attended the year 3 clinic visit.

Figure 1.

Overview of the study design.

The baseline characteristics of the patients undergoing B-mode carotid ultrasound in the intention-to-treat analysis are shown in Table 1. The mean calcium intake was 965 mg/d in the calcium group and 972 mg/d in the placebo group. Average tablet compliance was 65.5% in the calcium group versus 64.2% in the placebo group. Women in the calcium group had lower systolic blood pressure than the placebo group, but no other differences at baseline between the groups were detected (Table 1). Age, BMI, smoking history, history of atherosclerotic vascular disease, history of diabetes, cardiovascular medications, and estimated glomerular filtration rate by the Chronic Kidney Disease Epidemiology equation did not vary at baseline by treatment group.

Table 1. Baseline Characteristics of Participants With Carotid Ultrasound Measured at 3 Years
 PlaceboCalciump Value
  1. SD = standard deviation; BMI = body mass index; SBP = systolic blood pressure; DBP = diastolic blood pressure; HDLC = high-density lipoprotein cholesterol; LDLC = low-density lipoprotein cholesterol; ASVD = atherosclerotic vascular disease; CVD = cardiovascular disease.
  2. Data are expressed as mean ± SD, median (interquartile range) or number (percentage). The p values were calculated by Student's t test or chi-square test where appropriate.
  3. aAdjusted for season of blood collection.
  4. bCorrected calcium level based on the serum albumin level.
No. of participants550553 
≥80% tablet compliance (per protocol), no. (%)353 (64.2)362 (65.5)0.656
Age (years), mean (SD)75.0 (2.7)75.1 (2.6)0.514
Energy intake (kJ [000's]), mean (SD)7.3 (2.4)7.2 (2.3)0.773
BMI (kg/m2), mean (SD)27.3 (4.4)26.8 (4.5)0.091
SBP (mm Hg), mean (SD)138.5 (18.7)136.2 (17.3)0.035
DBP (mm Hg), mean (SD)73.4 (11.1)72.7 (10.9)0.326
Mean arterial pressure (mm Hg)95.1 (12.0)93.9 (11.5)0.090
CKD-EPI eGFR (mL/min/1.73m2), mean (SD)66.1 (13.6)66.6 (12.8)0.342
Calcium intake (mg/d), mean (SD)972 (345)965 (360)0.742
Serum 25OHDa (nmol/L), mean (SD)67.6 (1.3)68.1 (1.3)0.755
Corrected serum calciumb (mmol/L), mean (SD)2.3 (0.1)2.3 (0.1)0.445
Serum phosphate (mmol/L), mean (SD)1.2 (0.1)1.2 (0.1)0.879
Total cholesterol (mmol/L), mean (SD)5.9 (1.1)5.8 (1.1)0.807
Triglycerides (mmol/L), mean (SD)1.6 (0.7)1.6 (0.7)0.510
HDLC (mmol/L), mean (SD)1.5 (0.4)1.4 (0.4)0.700
LDLC (mmol/L), mean (SD)3.7 (1.0)3.7 (1.0)0.813
Smoked ever, no. (%)185 (33.6)198 (35.8)0.410
History of diabetes, no. (%)30 (5.5)27 (4.9)0.668
History of ASVD, no. (%)67 (12.2)56 (10.1)0.278
CVD medication use, no. (%)344 (62.5)323 (58.4)0.160

Carotid ultrasound was not performed in 73 (10.0%) of the participants receiving calcium and 75 (11.6%) of participants receiving placebo because of death or withdrawal, whereas 104 (14.2%) of the participants receiving calcium and 88 (12.0%) of participants receiving placebo were not able to attend because of scheduling difficulties (Fig. 1). There were no differences in the baseline cardiovascular risk factors of either participants who withdrew or died or those unable to attend by treatment group (data not shown). A post hoc power calculation determined that the study had 80% power to detect a 16% increase in the relative risk of carotid atherosclerosis, 0.02 mm difference in mean CCA-IMT, and 0.03 mm difference in maximum CCA-IMT.

Common carotid artery intimal medial thickness (Table 2)

The mean CCA-IMT of the cohort was 0.780 ± 0.127 mm; the maximum CCA-IMT was 0.924 ± 0.151 mm. Calcium supplementation did not increase mean CCA-IMT in either the intention-to-treat analysis or the per protocol analysis compared with placebo (per protocol unadjusted model mean CCA-IMT difference calcium group compared with placebo –0.008 mm, p = 0.386, and intention-to-treat mean CCA-IMT difference –0.001 mm, p = 0.869) or in models adjusted for age, BMI, smoking history, history of atherosclerotic vascular disease, history of diabetes, cardiovascular medications, and estimated glomerular filtration rate by the Chronic Kidney Disease Epidemiology equation (Table 2). Similarly, calcium supplementation did not increase maximum CCA-IMT in participants receiving calcium compared with placebo (mean difference calcium group compared with placebo –0.011 mm, p = 0.295, or intention-to-treat mean difference calcium group compared with placebo –0.003 mm, p = 0.729) or after adjustment (Table 2). There was no significant difference in mean CCA-IMT in unadjusted analyses by tertile of total dietary and supplemental calcium (tertile 1, 0.781 ± 0.147 mm; tertile 2, 0.786 ± 0.122 mm; or tertile 3, 0.771 ± 0.114 mm, p = 0.281) or maximum CCA-IMT (tertile 1, 0.927 ± 0.172 mm; tertile 2, 0.931 ± 0.147 mm; or tertile 3, 0.914 ± 0.135 mm, p = 0.297). This was unchanged after adjusting for baseline cardiovascular risk factors (Table 3).

Table 2. Intention-to-Treat and Per-Protocol ANCOVA for Common Carotid Artery Intimal Medial Thickness and Atherosclerosis (2001), According to Randomization for Calcium Supplementation at Baseline (1998)a
Intention-to-treat analysisPlacebo (n = 550)Calcium (n = 553)p Value
Mean CCA-IMT (mm)
Unadjusted0.780 (0.770–0.791)0.779 (0.768–0.790)0.869
Multivariable adjusteda0.783 (0.772–0.794)0.778 (0.766–0.789)0.491
Maximum CCA-IMT (mm)
Unadjusted0.925 (0.913–0.938)0.922 (0.910–0.935)0.729
Multivariable adjusteda0.929 (0.916–0.942)0.921 (0.908–0.942)0.404
Presence of carotid atherosclerosis (odds ratio)
Unadjusted1.000.80 (0.63–1.02)0.066
Multivariable adjusteda1.000.80 (0.62–1.04)0.095
Per-protocol analysisPlacebo (n = 353)Calcium (n = 362)p Value
  1. All values are mean or odds ratio and (95% CI).
  2. aMultivariable-adjusted model was adjusted for age, BMI, smoking history, history of atherosclerotic vascular disease, history of diabetes, cardiovascular medications, and estimated glomerular filtration rate by the Chronic Kidney Disease Epidemiology equation.
Mean CCA-IMT (mm)
Unadjusted0.785 (0.772–0.798)0.778 (0.764–0.790)0.386
Multivariable adjusteda0.788 (0.774–0.801)0.776 (0.762–0.789)0.214
Maximum CCA-IMT (mm)
Unadjusted0.932 (0.917–0.948)0.921 (0.905–0.936)0.295
Multivariable adjusteda0.935 (0.919–0.951)0.919 (0.903–0.935)0.172
Presence of carotid atherosclerosis (odds ratio)
Unadjusted1.000.73 (0.54–0.97)0.033
Multivariable adjusteda1.000.74 (0.54–1.02)0.064
Table 3. Multivariable-Adjusted ANCOVA or Logistic Regression for Mean and Maximum Common Carotid Artery Intimal Medial Thickness and Carotid Atherosclerosis (2001), According to Tertiles (n = 350 per Tertile) of Total Calcium Intake (Dietary and Supplemental) in the CAIFOS Cohorta
Tertiles of total calcium intake (diet and supplemental)Multivariable-adjusted values
Mean CCA-IMT (MM)
Tertile 1 (<1010 mg/d)0.782 (0.768–0.796)
Tertile 2 (1010–1795 mg/d)0.790 (0.776–0.804)
Tertile 3 (≥1795 mg/d)0.769 (0.754–0.783)
p trend0.107
Maximum CCA-IMT (mm)
Tertile 1 (<1010 mg/d)0.927 (0.910–0.944)
Tertile 2 (1010–1795 mg/d)0.936 (0.919–0.953)
Tertile 3 (≥1795 mg/d)0.911 (0.895–0.928)
p trend0.124
Presence of carotid atherosclerosis (odds ratio)No. (% with carotid atherosclerosis)Odds ratio
  1. aAll values are mean or odds ratio (95% confidence interval). All models were adjusted for age, BMI, smoking history, history of atherosclerotic vascular disease, history of diabetes, cardiovascular medications, and estimated glomerular filtration rate by the Chronic Kidney Disease Epidemiology equation.
Tertile 1 (<1010 mg/d)187 (53.4)1.00 (referent)
Tertile 2 (1010–1795 mg/d)181 (51.7)0.97 (0.71–1.34)
Tertile 3 (≥1795 mg/d)152 (43.4)0.70 (0.51–0.96)
p trend0.008 

Carotid atherosclerosis

In the per-protocol analysis, there were 169 of 362 (46.7%) of participants in the calcium-treated group with carotid atherosclerosis at 3 years compared with 193 of 353 (54.7%) of participants in the placebo group, p = 0.033. The per-protocol analysis participants randomized to calcium had a reduced risk of the presence of carotid atherosclerosis in the unadjusted analysis that became nonsignificant after adjustment for other baseline cardiovascular risk factors (Table 2). In the intention-to-treat analysis, 261 of 553 (47.2%) participants had carotid atherosclerosis at 3 years in the calcium-treated group compared with 290 of 550 (52.7%) of the placebo group, p = 0.066. The intention-to-treat analysis participants randomized to calcium supplementation did not have an increased risk of the presence of carotid atherosclerosis in either unadjusted analysis or multivariable-adjusted analysis (Table 2). Dividing the cohort by tertiles of total calcium intake (diet and supplemental), 20 of 350 (5.7%) participants were randomized to calcium supplements, 164 of 350 (46.9%) in the second tertile, and 343 of 350 (98%) in the highest tertile of calcium intake. Compared with the lowest tertile, participants in the highest tertile of total calcium intake (≥1,795 mg/d) had a reduced presence of carotid atherosclerosis in both the unadjusted (tertile 1, 1.00 [referent]; tertile 2, OR = 0.93 [CI 0.69–1.26]; and tertile 3, OR = 0.67 [CI 0.50–0.90], p = 0.019), which remained significant in the multivariable-adjusted analyses (Table 3).

Discussion

To our knowledge, this is the only randomized controlled trial (RCT) of calcium supplementation in elderly women to have assessed carotid ultrasound measures of subclinical calcification and atherosclerosis and compared the long-term effects of calcium supplementation on this outcome. In this study of 1103 elderly women, we did not observe any adverse effects of calcium supplementation on increased carotid artery intimal medial thickness or atherosclerosis after 3 years of calcium supplementation before or after adjusting for baseline cardiovascular risk factors. Indeed, there was a reduction in the presence of carotid atherosclerosis in the per-protocol analysis of participants randomized to calcium supplementation. These findings are consistent with our previously reported null effect of calcium supplementation on clinical atherosclerotic vascular disease[23] or myocardial infarctions.[24, 25] Therefore, in this cohort of elderly women, calcium supplementation did not increase either subclinical or clinical atherosclerosis.

The suggested mechanism of calcium supplementation increasing cardiovascular risk is via increased serum calcium increasing carotid plaque thickness and arterial calcification.[26] The observed lack of any effect in this study is similar to the findings from others, including a cross-sectional substudy analysis of the Women's Health Initiative that found no effect of 7 years of 1 g of calcium and 400 IU of vitamin D3 supplementation daily with coronary artery calcified plaque burden.[27] Similarly, Wang and colleagues[28] reported no effect of calcium supplementation on coronary artery calcification in an RCT of elderly men or abdominal aortic calcification in an RCT of elderly postmenopausal women. More recently, an observational study from the Framingham cohort[29] reported that calcium intake from either dietary or supplementary sources was not associated with increased coronary artery calcification in either men or women. Therefore, to date there is little evidence to support the suggestion that calcium supplements may increase vascular calcification, common carotid artery intimal medial thickness, or carotid atherosclerosis.

Although others have suggested that the myocardial infarction event rate in this study was lower than other studies of this age range,[30] the previously published event rate[24] is similar to the placebo group in the Auckland study.[25] Western Australia is fortunate in having a system that provides a comprehensive, population-based linkage system for nearly 40 years of data from more than 30 health-related data sets[31] independent of self-report with the associated problems of loss to follow-up and inaccurate patient reporting. This data set has been verified by more than 250 publications[31] with cardiovascular endpoints verified against both self-reported events and medical records.[32, 33] Therefore, we reiterate from this randomized controlled trial of calcium supplementation in elderly women that there is little evidence to support the hypothesis that these supplements increase either subclinical or clinical atherosclerotic vascular disease.

Support for the hypothesis that calcium supplementation may increase vascular disease is derived from RCT evidence showing that end-stage renal disease (ESRD) patients treated with calcium-based phosphate binders have more rapid progression of vascular calcification compared with patients treated with the phosphate binder Sevelamer.[34] However, because it is considered unethical not to provide phosphate-binding treatment in patients with ESRD, all studies comparing calcium-based phosphate binders to Sevelamer fail to account for the “normal” rate of vascular calcification progression. Therefore, it remains unclear whether these findings are the result of a beneficial Sevelamer effect or an adverse calcium effect. To our knowledge, only one study by Russo and colleagues[35] of predialysis patients had a “normal” untreated comparison group. Participants were randomized to either a low-phosphate diet, a low-phosphate diet plus 2 g/d of calcium carbonate, or a low-phosphate diet plus Sevelamer with 2 years of treatment. The authors found Sevelamer treatment attenuated the progression of coronary artery calcification compared with calcium-treated patients and those with a low-phosphate diet alone; however, there was little difference in the rate of progression of calcification between the low-phosphate diet and those with the low-phosphate diet plus calcium. Therefore, this study supports the concept that calcium per se does not increase vascular calcification in ESRD, but rather Sevelamer attenuates the progression of vascular calcification.

Interestingly, we observed a beneficial effect of calcium supplements in participants compliant with the intervention before but not after adjusting for baseline cardiovascular risk factors and in participants in the highest tertile of total calcium intake (diet and supplemental) both before and after adjustment for risk factors. This may be because of the suggested beneficial effects of calcium supplementation on blood pressure[4] or reduced arterial wave reflection.[36] Burt and colleagues[36] recently undertook a short-term RCT and found calcium supplementation reduced arterial wave reflection and markers of myocardial perfusion that would be expected to reduce, not increase, cardiovascular risk. Similarly, these findings may explain the beneficial associations of calcium supplements with cardiovascular disease and deaths reported by others.[37-39]

Strengths of the study include the comprehensive assessment of cardiovascular risk factors in this cohort in conjunction with hospital records for the previous 18 years (1980 to 1998), detailed medication history, cardiovascular risk factors assessment, and detailed assessment of both dietary calcium intake and calcium tablet compliance in a large cohort of elderly women randomized to calcium or placebo. This comprehensive assessment of baseline risk factors in conjunction with B-mode carotid ultrasound, which is a highly sensitive measure of the carotid intimal medial thickness, atherosclerosis, and focal plaques, in a large cohort of elderly women gave the study sufficient power to detect small differences in CCA-IMT and carotid atherosclerosis owing to calcium supplementation. However, there are several limitations of our study. First, because no baseline carotid ultrasounds were performed, residual confounding by unmeasured differences in carotid wall characteristics cannot be excluded; however, this is unlikely given the nature of the RCT design that aims to account for baseline confounding factors. Second, because this was an ancillary study in 1103 of the original 1460 women recruited at baseline (76%), there is the possibility we may have introduced bias into the study. However, again this is unlikely because there were no significant differences between treatment groups in the baseline cardiovascular risk factors of participants who did not have the year 3 carotid ultrasound and the majority (76%) of the cohort had carotid ultrasound measured. Finally, because this cohort was elderly women with an average age of 78 years at carotid ultrasound measurement, the findings are not generalizable to elderly men or to younger cohorts.

In conclusion, our study does not support the hypothesis that calcium supplementation increases either carotid artery intimal medial thickness or carotid atherosclerosis. In addition, calcium supplementation and high total calcium intake may indeed reduce carotid atherosclerosis.

Disclosures

All authors state that they have no conflicts of interest.

Acknowledgments

The study was supported by a research grant from Healthway Health Promotion Foundation of Western Australia and by the project grants 254627, 303169, and 572604 from the National Health and Medical Research Council of Australia. The salary of JRL is supported by a Raine Medical Research Foundation Priming Grant. None of the funding agencies had any role in the conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.

The authors thank the staff at the Data Linkage Branch, Hospital Morbidity Data Collection, and Registry of Births, Deaths, and Marriages for their work on providing the data for this study.

Authors' roles: Data collection: RLP, KZ, JRL, and PLT. Data analysis: JRL, KZ, and RLP. Data interpretation: JRL, KZ, and RLP. Drafting the manuscript: JRL, KZ, PLT, and RLP. Revising the manuscript content: JRL, KZ, PLT, and RLP. Approving the final version of the manuscript: JRL, KZ, PLT, and RLP. JRL takes responsibility for the integrity of the data analysis.

Ancillary