Vascular calcium is well studied in the coronary and peripheral arteries, although there are limited data focusing on calcium deposits specific to renal arteries. The associations among renal artery calcium (RAC), cardiovascular disease risk factors, and indexes of renal function are unknown. We examined 2,699 Framingham Heart Study participants who were part of a multidetector computed tomography substudy from 2008 to 2011. RAC was measured as a calcified plaque of >130 HU and an area of >3 contiguous pixels. Detectable RAC was defined as an Agatston score >0. Chronic kidney disease was defined as an estimated glomerular filtration rate of <60 ml/min/1.73 m 2 . Microalbuminuria was defined as an albumin/creatinine ratio of ≥17 mg/g for men and ≥25 mg/g for women. Multivariable adjusted logistic regression models were used to evaluate the associations between RAC, cardiovascular disease risk factors, and renal function. The associations were secondarily adjusted for coronary artery calcium (CAC) that was used as a marker of nonrenal systemic vascular calcium. The prevalence of RAC was 28.2%; this was similar in women (28.8%) and men (27.5%). Patients with RAC had a higher odds of microalbuminuria (odds ratio [OR] 1.79, 95% confidence interval [CI] 1.22 to 2.61, p = 0.003), hypertension (OR 2.11, 95% CI 1.69 to 2.64, p <0.001), and diabetes (OR 1.60, 95% CI 1.14 to 2.24, p = 0.01) but not chronic kidney disease (OR 0.87, 95% CI 0.58 to 1.32). After adjustment for CAC, the association with microalbuminuria and hypertension persisted, but the association with diabetes became nonsignificant. In conclusion, RAC is common and independently associated with microalbuminuria and hypertension after adjustment for nonrenal vascular calcium. RAC may be uniquely associated with these markers of renal end-organ damage.
Computed tomography (CT) using the multidetector or electron beam method is a noninvasive method that can be effectively applied to quantify arterial wall calcium. Several studies using the electron beam CT method have shown an association between renal artery calcium (RAC) and hypertension. RAC was also shown to be an independent predictor of progression to end-stage renal disease and associated with all-cause mortality. However, these studies were limited by selection bias or small sample size. Thus, the associations between RAC and kidney parameters have not been fully evaluated. The purpose of the present study is to examine the associations between RAC, cardiovascular disease (CVD) risk factors, and renal indexes using a large community dwelling population with rigorous risk factor ascertainment. We hypothesized that the presence of RAC would be uniquely associated with chronic kidney disease (CKD) and microalbuminuria independent of traditional CVD risk factors and nonrenal vascular calcium.
Methods
The Framingham Heart Study is a community-based cohort study that began in 1948. The objectives of the study, selection criteria, and study design have been previously described. Offspring of the original cohort were enrolled in 1971 as well as spouses of the offspring. Data on third generation family members with at least 1 parent in the original cohort were collected starting in 2002 with standardized clinic visits that included an interview conducted by physicians, a physical examination, and laboratory tests. The present study sample consists of 1,333 offspring and 1,431 third generation participants who took part in the second multidetector CT substudy from 2008 to 2011. Participants were drawn from those living in the greater New England area. Men had to be aged ≥35 years and women had to be aged ≥40 years. Women who were pregnant or who had been breast-feeding for <6 months were not eligible. There was a weight limit of ≤450 lb because of scanner limitations. Multiple precautions were taken to limit the risks of radiation exposure including avoidance of the true pelvis, exclusion of young adults, and assessment of pregnancy status for premenopausal women. The study protocol was approved by the Boston University Medical Center and Massachusetts General Hospital institutional review boards. All subjects provided written informed consent. Of the 2,764 participants imaged, 4 had uninterpretable CT measurements for RAC and 61 had missing covariates resulting in a sample size of 2,699 for analysis.
Assessment of RAC was obtained from noncontrast abdominal CT imaging using a Discovery VCT 64-slice positron emission tomography/computed tomography scanner (GE Healthcare, Waukesha, Wisconsin) with fixed 120 kVp and automatic mA adjustment for body mass index (BMI) in a range of 100 to 300 mA with 2.5-mm slice thickness. Images were interpreted on an Aquarius Workstation (TeraRecon, Foster City, California). RAC scoring was consistent with the quantification methods described by Agatston et al. Briefly, regions of interest were identified by a single reader as having a density of >130 HU and an area of >3 contiguous pixels (≥1.0 mm 2 ). Detectable RAC was defined as an Agatston score >0. Final scoring was the sum of both renal arteries and did not include calcified plaque beyond the ostia that extended into the abdominal aorta. Intrareader reproducibility on a sample of 46 scans demonstrated an intraclass correlation coefficient of 0.98. Chest CT images were obtained with fixed 120 kVp, 0.625-mm slice thickness, and 300 mA if the weight was ≤220 lb and 350 mA for weight >220 lb. Coronary artery calcium (CAC) scores were derived using protocols previously described. Intrareader reproducibility demonstrated an intraclass correlation coefficient of 0.99.
Serum creatinine level was measured using the Jaffe assay on a Roche Hitachi 911 (Roche Diagnostics, Indianapolis, Indiana) with an intra-assay coefficient of variation of 3.1%. CKD was defined as an estimated glomerular filtration rate of <60 ml/min/1.73 m 2 using the Chronic Kidney Disease Epidemiology Collaboration equation. This cut-off value is consistent with CKD stage ≥3 set forth in the clinical practice guidelines of the National Kidney Foundation’s Kidney Disease Outcomes Quality Initiative. Urinary albumin and creatinine levels were quantified from spot urine samples collected during participant examinations and stored at −80°C. Urinary albumin level was measured using immunoturbidimetry (Tina-quant albumin assay; Roche Diagnostics, Indianapolis, Indiana; intra-assay coefficient of variation of 1.9%, interassay coefficient of variation of 5.5%), and estimation of urinary creatinine level was made using a modified Jaffe method on a Roche Hitachi 911 with an intra-assay coefficient of variation of 1.2% and interassay coefficient of variation of 2.2%. Microalbuminuria was defined using urinary albumin/creatinine ratios with gender-specific cutoffs of ≥17 mg/g for men and ≥25 mg/g for women to reflect the greater urinary creatinine excretion in men compared with women.
Manual blood pressure measurements were taken in the left upper extremity by a physician using a mercury sphygmomanometer on 2 separate occasions during the same visit following a standardized protocol. Hypertension was defined as a systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or current use of antihypertensive medications. Fasting morning serum samples were obtained for measurement of glucose, total cholesterol, high-density lipoprotein cholesterol, and triglyceride levels. Diabetes was defined as a fasting glucose level of ≥126 mg/dl or use of antidiabetic medication. BMI was calculated by the participant’s weight in kilograms divided by the height in meters squared. Smoking was defined as using a cigarette, pipe, or cigar at least once per day in the year preceding the examination.
Sample means and SDs were calculated for normally distributed continuous values, medians and interquartile ranges were calculated for non-normally distributed variables, and proportions were calculated for categorical data. Spearman correlation coefficients were calculated to study the relation between log(RAC+1) and continuous variables after adjusting for age and gender. The chi-square test was used to measure differences between detectable RAC in each age group and overall stratified by gender. A stepwise logistic regression model with forward selection was used to identify independent correlates for the presence of RAC including age, gender, BMI, hypertension, diabetes, smoking, CKD, microalbuminuria, and total cholesterol, high-density lipoprotein cholesterol, and triglyceride levels.
We then performed a series of multivariable logistic regressions evaluating detectable RAC as a cross-sectional predictor including microalbuminuria, diabetes, hypertension, and CKD. Here, RAC was log transformed using log(RAC+1). For the outcome of microalbuminuria, we adjusted for age, gender, smoking, BMI, hypertension, and diabetes. For the outcomes of diabetes and hypertension, we adjusted for age, gender, smoking, and BMI. For the outcome of CKD, we adjusted for age, gender, smoking, hypertension, and diabetes.
As a secondary analysis, we recreated the stepwise model and all multivariable models to include CAC scores using log(CAC+1) transformation. CAC was used as a proxy for nonrenal vascular calcium to determine if an association was related to more generalized systemic vascular calcification or only to RAC. CAC scores were not available for 112 participants with detectable RAC and 33 participants without RAC. To assess the significance of RAC in the absence of systemic vascular calcium, the chi-square test was used to compare the group of participants with zero RAC and CAC ≥10 versus the group with RAC >0 and CAC <10. This comparison was adjusted for age and gender.
All analyses were performed using SAS, version 9.2 (SAS Institute Inc., Cary, North Carolina). A 1-tailed p value <0.05 was considered statistically significant.
Results
The clinical characteristics of the study cohort are presented in Table 1 and stratified by the presence or absence of RAC >0. In general, participants with RAC compared with zero RAC were older (71.0 vs 55.7 years, respectively) and had higher rates of microalbuminuria, hypertension, diabetes, and CKD. Median CAC scores were higher in participants with detectable RAC. Overall, the prevalence of detectable RAC was 28.2%, and this was similar in women (28.8%) and men (27.5%, p = 0.46 for gender difference). The prevalence of RAC increased with each 10-year age group, and there were no significant gender differences across any group ( Figure 1 ). There was a significant positive correlation between RAC and clinical covariates including log(urinary albumin/creatinine ratio) and systolic blood pressure ( Table 2 ). From a stepwise regression analysis, significant variables associated with RAC were age, microalbuminuria, hypertension, smoking, and BMI. When we adjusted for CAC, the odds ratios (ORs) were modestly attenuated, but the risk factor associations with RAC remained statistically significant except for BMI ( Table 3 ).
| Variable | RAC >0 (n = 760) | RAC = 0 (n = 1,939) |
|---|---|---|
| Age (yrs) | 71.0 ± 9.3 | 55.7 ± 9.8 |
| Women | 391 (51.4) | 967 (49.9) |
| BMI (kg/m 2 ) | 30.0 (5.3) | 28.3 (5.4) |
| Hypertension | 504 (66.3) | 508 (29.9) |
| Hypertension treatment | 427 (56.4) | 458 (23.6) |
| Systolic blood pressure (mm Hg) | 130.7 ± 17.1 | 119.5 ± 13.8 |
| Diastolic blood pressure (mm Hg) | 72.9 ± 9.9 | 75.0 ± 8.9 |
| Diabetes | 132 (17.4) | 126 (6.5) |
| Current smoker | 53 (7.0) | 140 (7.2) |
| CKD | 109 (14.3) | 56 (2.9) |
| Estimated GFR (ml/min/1.73 m 2 ) | 77.1 ± 16.1 | 89.6 ± 14.3 |
| Microalbuminuria | 110 (14.5) | 86 (4.4) |
| UACR (mg/g) | 6.1 (3.6–12.7) | 4.1 (2.6–8.0) |
| Total cholesterol (mg/dl) | 184.2 ± 37.2 | 191.2 ± 34.7 |
| HDL cholesterol (mg/dl) | 56.5 ± 17.2 | 58.9 ± 18.2 |
| Triglycerides (mg/dl) | 104.0 (79.0–144.5) | 91.2 (79.9–100.7) |
| Lipid treatment | 401 (52.8) | 481 (24.8) |
| CAC score | 204.8 (29.4–645.7) | 0 (0–34.3) |
| RAC score | 71.7 (22.4–179.1) | 0 (0–0) |
| Variable | r | p |
|---|---|---|
| Age | 0.58 | <0.001 |
| BMI | 0.07 | <0.001 |
| Systolic blood pressure | 0.11 | <0.001 |
| Diastolic blood pressure | −0.04 | 0.02 |
| Log(urinary albumin/creatinine ratio) | 0.11 | <0.001 |
| Estimated glomerular filtration rate | −0.01 | 0.71 |
| Glucose | 0.12 | <0.001 |
| Total cholesterol | −0.07 | <0.001 |
| High-density lipoprotein cholesterol | −0.03 | 0.09 |
| Low-density lipoprotein cholesterol | −0.06 | <0.001 |
| Log(triglycerides) | 0.03 | 0.09 |
| Log(CAC score +1) | 0.29 | <0.001 |
| Variable | OR (95% CI) | p | CAC-Adjusted OR (95% CI) | p |
|---|---|---|---|---|
| Age (10 yrs) | 4.12 (3.65–4.67) | <0.001 | 2.96 (2.58–3.40) | <0.001 |
| Gender (women) | 0.86 (0.69–1.06) | 0.16 | 1.52 (1.17–1.96) | 0.001 |
| BMI (per SD of BMI increase) | 1.14 (1.02–1.27) | 0.03 | 1.07 (0.95–1.21) | 0.25 |
| Current smoker | 2.55 (1.70–3.83) | <0.001 | 2.05 (1.31–3.21) | 0.002 |
| Hypertension | 2.13 (1.70–2.68) | <0.001 | 1.80 (1.41–2.30) | <0.001 |
| Microalbuminuria | 1.81 (1.22–2.70) | 0.003 | 1.80 (1.15–2.82) | 0.01 |
| CAC | — | — | 1.32 (1.25–1.39) | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
