Body Mass Index, Coronary Artery Calcification, and Kidney Function Decline in Stage 3 to 5 Chronic Kidney Disease Patients

ORIGINAL RESEARCH

Body Mass Index, Coronary Artery Calcification, and Kidney Function Decline in Stage 3 to 5 Chronic Kidney Disease Patients Jocelyn S. Garland, MD,*† Rachel M. Holden, MD,*† Wilma M. Hopman, MA,‡ Sudeep S. Gill, MD,§ Robert L. Nolan, MD,{ and A. Ross Morton, MD*† Objective: To determine whether body mass index (BMI) and coronary artery calcification (CAC) are risk factors for kidney function decline in predialysis chronic kidney disease (CKD) patients. Design: Prospective cohort study of 125 stage 3 to 5 predialysis CKD patients. Subjects and Setting: CKD patients receiving care in Kingston, Ontario, Canada. Methods: BMI, CAC, and kidney function were measured at baseline. CAC was measured by multislice computed tomography scan. Kidney function was determined by the 4-variable reexpressed Modification of Diet in Renal Disease Study equation. At study end, kidney function decline among patients was compared according to baseline BMI and CAC. Main Outcome: Kidney function decline was defined as a 1-year decline in estimated glomerular filtration rate (eGFR) of $5%. Results: Individuals with a decline in eGFR of $5% at 1 year had higher baseline BMI (33.5 6 8.3 vs. 28.4 6 4.9 kg/ m2; P 5 .0001) and higher baseline median CAC scores (239 vs. 25 Agatston units; P 5 .01) compared with subjects without such a decline. BMI (r 5 0.35; P , .0001) and logarithmically transformed CAC score (r 5 0.22; P 5 .01) correlated with an eGFR decline of $5%. Both crude and adjusted logistic regression analyses showed escalating CAC (with CAC reported in quintiles and CAC score 5 0 Agatston unit as the reference group) was associated with an increased risk of eGFR decline of $5%. Conclusions: CAC and BMI were associated with kidney function decline over 1 year. The risk of kidney function decline was greater in those with increasing burden of CAC, which remained robust in the adjusted analysis accounting for the risk factors for CKD progression. Larger studies will be required for independent validation of the associations of BMI, CAC, and kidney function decline, and to investigate whether obesity and CAC treatment strategies are efficacious in attenuating kidney function decline in predialysis CKD patients. Ó 2013 by the National Kidney Foundation, Inc. All rights reserved.

This article has an online CPE activity available at www.kidney.org/professionals/CRN/ceuMain.cfm

*Division of Nephrology, Department of Medicine, Queen’s University, Kingston, Ontario, Canada. †Queen’s University Vascular Calcification Investigators, Kingston, Ontario, Canada. ‡Clinical Research Center, Kingston General Hospital, and Department of Community Health and Epidemiology, Queen’s University, Kingston, Ontario, Canada. §Division of Geriatric Medicine, Department of Medicine, Queen’s University, Kingston, Ontario, Canada. {Department of Radiology, Queen’s University, Kingston, Ontario, Canada. Financial Disclosure: J.S.G., R.M.H., and R.M. have received paid honoraria for providing continuing medical education and consulting fees. J.S.G. has received paid honoraria for providing continuing

4

medical education for Amgen Canada, and Bristol Myers Squibb Canada. R.M.H. has received paid honoraria from Hoffman LaRoche. R.M. has received paid honoraria from Genzyme Canada Inc., Novartis, and Shire Biochem Inc. Support: This study was funded through an unrestricted educational grant from Pfizer Canada, Inc. Address correspondence to Jocelyn S. Garland, MD, Division of Nephrology, Department of Medicine, Queen’s University, Room 2043, Etherington Hall, Kingston, ON K7L 3N6, Canada.

E-mail: [email protected] Ó 2013 by the National Kidney Foundation, Inc. All rights reserved. 1051-2276/$36.00 doi:10.1053/j.jrn.2011.12.008

Journal of Renal Nutrition, Vol 23, No 1 (January), 2013: pp 4-11

BMI, CAC, AND KIDNEY FUNCTION DECLINE

C

ORONARY ARTERY CALCIFICATION (CAC) is a phenomenon described in individuals with chronic kidney disease (CKD), and its presence is associated with adverse cardiovascular events.1,2 We have previously examined the prevalence and risk factors of CAC in a cohort of stage 3 to 5 CKD patients who did not have a history of cardiovascular disease.3 Body mass index (BMI; r 5 0.28, P 5.002) and a composite cardiovascular disease risk factor score based on traditional risk factors (r 5 0.35; P , .001) were also found to be positively correlated with CAC.3 In previous studies of stage 3 to 5 predialysis CKD patients, we have identified that approximately 50% of the patients were obese by BMI criteria (BMI: .30),4 and that most (70%) had detectable CAC.3 To date, most studies evaluating the risk conferred by obesity and CAC presence have focused on cardiovascular mortality or cardiovascular events as the main clinical outcomes of interest. Individuals with CAC scores $10 AU have decreased survival2 and have also been shown to have an increased risk of CAC progression.5,6 However, the potential deleterious impact of obesity and CAC in individuals with CKD on kidney function has not been studied. Currently, an important challenge for the nephrology community is the ability to identify CKD patients who are at highest risk of progressive kidney impairment. In this prospective study of 125 individuals with confirmed stage 3 to 5 CKD, our primary objective was to determine whether CKD patients with baseline evidence of elevated BMI and CAC have an increased risk of kidney function decline at 1 year, while controlling for other known risk factors for CKD progression.

5

Foundation criteria were applied to diagnose CKD.7 For each participant, cardiovascular disease was determined by assessing both current symptoms (through patient interview using Canadian Cardiovascular Society criteria for heart failure and angina)8,9 and history of cardiovascular events (by patient interview and detailed chart review). Weight and height data were collected on each individual to calculate BMI. Overweight status was defined as a BMI $25.0 kg/m2, and obesity was defined as a BMI $30.0 kg/m2.10 Diagnoses of hypertension and diabetes mellitus were made as per the 2006 Canadian Hypertension Education Program Guidelines11 and the Canadian Diabetes Association criteria, respectively.12 Current smokers were defined as patients who smoked at least 1 cigarette per day during the previous 6 months. Random urine samples were obtained to determine urinary albumin to creatinine ratios (UACRs). All patients gave informed consent, and the study protocol was approved by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board.

Kidney Function Measurement Serum creatinine (sCr) was measured by the Roche Creatinine Plus Modular assay (enzymatic; coefficients of variation, ,3% for an sCr level of $1.84 mg/dL [163 mmol/L] and 1% for an sCr level of $6.67 mg/dL [590 mmol/L]) (Roche Diagnostics, Indianapolis, IN). All sCr values were obtained from the same laboratory to minimize interlaboratory variability and misclassification errors. The 4-variable Modification of Diet in Renal Disease Study equation,13 reexpressed for standardized sCr,14 was used to calculate estimated glomerular filtration rate (eGFR).

Subjects and Methods Consecutive patients who consented to participate in a study evaluating the relationship between CAC and decreased kidney function from July 2005 to September 2006 were enrolled from Kingston General Hospital’s CKD clinic. The full methods are described elsewhere.3 In brief, patients were eligible to participate if they (1) were .18 years of age, (2) had stage 3 to 5 CKD (not requiring dialysis, and excluding acute kidney injury), (3) had no documented history of cardiovascular disease, (4) were not prescribed warfarin, and (5) had a BMI $18.5 kg/m2. National Kidney

CAC Measurement CAC scores were evaluated using the Toshiba Aquilion (Nasu, Japan) computed tomography multislice scanner (4 sets of detectors) and VScore analytical software package (Vital images Inc., Minnetonka, MN). The scan thickness was 3 mm 3 4 slices simultaneously over 12 mm per rotation, and the field of coverage was 12 cm. Images were acquired with prospective gating technique using a discrete algorithm.15 A total CAC score was generated as per the method by Agatston et al., which has been described elsewhere.16

6

GARLAND ET AL

Statistical Methods Summary statistics were expressed as means and standard deviations (or as medians and interquartile ranges) counts, and percentages, as appropriate. CAC scores were reported as mean and median total scores, as well as categorically in Agatston units (AU; 0, minimal [score, 1 to 9.9 AU], mild [10 to 99 AU], moderate [100 to 399 AU], and severe [.400 AU]).16 The significance of associations for categorical variables was determined by c2 analysis or by Fisher exact test, as appropriate. Mean change in kidney function at 1 year was determined by paired t test, where the difference between the baseline and 1-year eGFR was calculated for each participant. The mean decline in kidney function was categorized for statistical analysis based on percent mean decline in kidney function at 1 year (.5% decline). Bivariate analysis was performed to evaluate the associations between annual percent decline in kidney function and a priori chosen risk factors (age, sex, hypertension, BMI, hyperlipidemia, diabetes mellitus, UACR, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker use, and CAC). Before analysis, CAC scores and UACR were logarithmically transformed to ensure normality for parametric testing. One AU was added to all CAC scores before logarithmic transformation. Alternatively, CAC scores were analyzed categorically as per Agatston et al.16 A series of logistic regression models were subsequently constructed to identify the risk factors for a kidney function decline of $5% at 1 year. All the variables from the bivariate analysis that showed an association with kidney function decline at the significance level of P , .10 were included in the initial logistic regression models. All statistical analyses were performed using Statistical Package for Social Sciences 17.0 for Windows (SPSS Inc., Chicago, IL). Sample Size Considerations The association between kidney function decline and CAC has not been reported previously in the literature. In designing our study, our aim was to detect a Pearson r of 0.25 or more between CAC score and CKD with 5% significance level and 80% power by recruiting and retaining 125 patients. The calculation was based on the Fisher z transformation of the correlation.17

Results Of the 131 stage 3 to 5 CKD patients enrolled for study participation, 126 had 1-year follow-up eGFR results available for the final analysis. One other patient was excluded owing to a BMI ,18.5 kg/m2, for a final sample size of 125 patients. Table 1 describes the baseline clinical characteristics of the included study participants. The mean age of the participants was 58.7 6 14 years, and 62% were men. Most individuals had a history of hypertension (92%), and 81% had albuminuria (UACR .26.5 mg/g [3 mg/mmol]). Forty-nine percent of the participants were found to be obese (BMI: .30 kg/m2), and 83% were found to be overweight (BMI: .25 kg/m2). Mean baseline eGFR was 27.4 mL/minute/1.73 m2. In terms of CKD stage, 37% (n 5 46) of the individuals had stage 3 CKD, 47% (n 5 59) had stage 4, and 16% (n 5 20) had stage 5. Age, sex, diabetes status, and CAC scores did not differ significantly according to CKD stage. The mean baseline CAC score was 576 6 1,105 AU, and the median score was 115.5 AU (interquartile range, 3 to 661). Fourteen percent of the participants (n 5 18) had no evidence of CAC, 17.0% (n 5 21) had minimal CAC (score, 1 to 9.9 AU), 18.0% (n 5 22) had mild CAC (score, 10 to 99 AU), 18% (n 5 23) had moderate CAC (score, 100 to 399 AU), and 33% (n 5 41) had evidence of severe CAC (score, $400 AU). Patients with diabetes mellitus had higher CAC scores. Sixty-eight percent of diabetic patients had a CAC score .100 (vs. 42% in nondiabetic patients, P 5 .008), and 48% had a CAC score .400 (vs. 25% in nondiabetic patients, P 5.01). Baseline serum phosphorus level (PO4) was 4.06 6 1.02 mg/dL (1.31 6 0.33 mmol/L) and did not differ over 1 year among 120 patients who also had follow-up PO4 level measurements available (3.99 6 1.08 mg/dL; 1.29 6 0.35 mmol/L; mean PO4 difference, 0.02 6 0.35 mmol/L; P 5 .63). At 1 year, the mean change in eGFR for the 125 participants had declined from 27.4 to 25.2 mL/ minute/1.73 m2 (mean difference, 22.2 6 5.4 mL/minute/1.73m2; P ,.0001). In terms of percentage change from baseline eGFR, 58% of the participants (n 5 73) had an eGFR decline of $5%. Baseline CKD stage was not associated with a 1-year eGFR decline of $5%. However, both baseline BMI and CAC scores were significantly higher in patients who had a 1-year eGFR

7

BMI, CAC, AND KIDNEY FUNCTION DECLINE Table 1. Baseline Clinical Characteristics of Study Participants (n 5 125) Variable

Prevalence

Age (years) eGFR (mL/minute) Body mass index (kg/m2) Urinary ACR (mg/g) (n 5 120) Systolic blood pressure Diastolic blood pressure CAC score (AU) Sex (% male) Race (% Caucasian) Diabetes (%) Hypertension (%) Hyperlipidemia (%) Smoker (%) Calcitriol therapy Calcium-based phosphorus binder therapy Kidney disease etiology Hypertension (%) Diabetes (%) Glomerulonephritis (%) Other (%)

Mean/Median

SD/IQR

58.7 27.4 31.5 33.7 136.6 77.5 115.5

14 12 7.3 5.4 to 102.8 15.3 12.1 3 to 661

62% 100% 35% 92% 84% 46% 7% 20% 30% 33% 16% 21%

ACR, albumin to creatinine ratio (to convert ACR in mg/g to mg/mmol, multiply by 0.113); AU, Agatston units; SD, standard deviation; IQR, interquartile range; CAC, coronary artery calcification; eGFR, estimated glomerular filtration rate.

decline of $5% than those who did not (baseline BMI, 33.5 6 8.3 vs. 28.4 6 4.9 kg/m2; P 5 .0001; baseline median CAC score, 239 [interquartile range, 10.5 to 768] vs. 27 [interquartile range, 1 to 403] AU; P 5 .02). Table 2 reports the median 1-year difference of eGFR according to baseline CAC and BMI categories. As the level of BMI and CAC increased, there was a greater median decline in eGFR. Individuals with BMI .30 had the greatest decline in eGFR (median eGFR decline, 22.75 vs. 20.71 Table 2. Median 1-Year Difference of eGFR (mL/ minute/1.73 m2) Decline According to Baseline CAC and BMI Categories Variable CAC 0 CAC .0 CAC ,10 CAC .10 CAC ,100 CAC .100 CAC ,400 CAC .400 BMI ,25 BMI .25 BMI ,30 BMI .30

Median Change in eGFR (6IQR) 20.24 (23.14 to 0.49) 22.32 (24.45 to 20.15) 20.74 (24.07 to 0.21) 22.36 (24.42 to 20.16) 20.84 (24.02 to 0.58) 22.62 (24.96 to 20.45) 21.66 (24.30 to 0.46) 22.36 (25.10 to 20.53) 20.24 (22.68 to 3.70) 22.34 (25.31 to 20.17) 20.71 (23.20 to 1.53) 22.75 (26.90 to 20.81)

BMI, body mass index (kg/m2).

P Value .1 .08 .04 .2 .09 .003

mL/minute/1.73 m2; P 5.003) as did the individuals whose CAC score was .100 AU (median eGFR decline, 22.62 vs. 20.84 mL/minute/ 1.73 m2; P 5 .04). Table 3 reports the findings from the bivariate analysis examining for a priori chosen correlates for a 1-year eGFR decline of $5%. At P ,.05 significance level, BMI most strongly correlated with an eGFR decline of $5% (r 5 0.35, P , .0001), Table 3. Correlations (Unadjusted) Between 1-Year eGFR Decline of $5% and Covariates of Interest (n 5 125) Variables Age (per decade) Body mass index (per 5 point increase) Diabetes status Sex (male vs. female) Angiotensin-converting enzyme inhibitor use Angiotensin receptor blocker use Hypertension Log UACR (n 5 120) Log total CAC CAC .0 CAC .10 CAC .100 CAC .400

r

P

20.14 0.34

.12 ,.0001

0.2 20.03 0.13

.02 .7 .14

0.13

.16

0.23 0.28 0.22 0.21 0.21 0.25 0.14

.01 .002 .01 .02 .02 .005 .12

UACR, urinary albumin to creatinine ratio.

8

GARLAND ET AL

followed by albuminuria, hypertension, and diabetes mellitus. Both total CAC score (logarithmically transformed [logCAC]) and CAC categories of .0, .10, and .100 AU correlated with a 1-year eGFR decline of $5%. The result for the highest CAC category (score, .400 AU) did not demonstrate a statistically significant association with a 1-year eGFR decline of $5% in the unadjusted analysis. A series of unadjusted and adjusted logistic regression models were constructed to determine the risk factors for a 1-year eGFR decline of $5% (Fig. 1). For these models, the covariate CAC was categorized and reported in quintiles, with a CAC equal to 0 AU as the reference category. In the multivariable adjusted logistic regression model accounting for age (by decade), hypertension, diabetes, and albuminuria, increasing CAC categories were independent risk factors for a 1-year eGFR decline of $5%. Both moderate (CAC score, .100 AU) and severe CAC (CAC score, .400 AU) were independent risk factors

for a 1-year eGFR decline of $5% (Table 4). Mild CAC (CAC score, 10 to 100 AU) was also associated with an increased risk of eGFR decline of $5% in the adjusted model, although this result missed statistical significance (P 5.09). In a separate series of multivariable logistic regression models, BMI was also included as a covariate to determine its potential role as a risk factor for kidney function decline. There was a statistically significant association of the interaction of CAC (logarithmically transformed) and BMI with 1-year eGFR decline of $5% (P 5.001). Adjusting for age decade (odds ratio [OR], 0.60; 95% confidence interval [95% CI], 0.42 to 0.86; P 5 .006), albuminuria (OR, 4.0; 95% CI, 1.3 to 11.8; P 5 .01), diabetes mellitus (OR, 1.8; 95% CI, 0.68 to 4.9; P 5 .2), and hypertension (OR, 4.4; 95% CI, 0.8 to 24; P 5 .09), the interaction term BMI 3 logCAC score (OR, 1.13; 95% CI, 1.05 to 1.2; P 5.001) was associated with a small, but significant, increased risk for a 1-year eGFR decline of $5%.

Figure 1. Odds ratios for eGFR decline of $5% according to CAC score quintile. Crude and multivariable adjusted odds ratios for eGFR decline are shown according to quintile of CAC score (quintile 1, CAC score 5 0 AU [reference]; quintile 2, 1 to 9.9 AU; quintile 3, 10 to 99 AU; quintile 4, 100 to 399 AU; quintile 5, .400 AU). The multivariable adjusted analysis included age (by decade), hypertension, diabetes mellitus, and albuminuria. Quartile 1 was the reference group in both models. *P , .05, **P , .09. Ref denotes reference group.

BMI, CAC, AND KIDNEY FUNCTION DECLINE Table 4. Multivariable Logistic Regression Risk Factors for an eGFR Decline of $5%

Variable

Odds Ratio

95% Confidence Interval

P

CAC, 0 (reference) CAC, 1 to 9.9 CAC, 10 to 99 CAC, 100 to 399 CAC, 4001 Age (decade) Albuminuria Hypertension Diabetes

1.01 3.7 7.4 8.8 0.56 4.41 5.4 2.35

0.24 to 4.4 0.8 to 17.3 1.45 to 37.9 1.8 to 42.5 0.38 to 0.83 1.48 to 13.3 0.99 to 29.3 0.86 to 6.3

.99 .09 .02 .007 .003 .008 .053 .10

Discussion The primary objective of this study was to determine whether CKD patients with baseline evidence of elevated BMI and CAC have an increased risk of kidney function decline. In addition to known risk factors for CKD progression, our results suggest that individuals who have increased BMI or CAC, either separately or in combination, have an increased risk of kidney function decline. CAC is a form of vascular calcification commonly described in CKD patients. CAC is an important clinical finding, as it is progressive and is a risk factor for adverse outcomes, including hospitalizations and death.1,2 In this prospective study of stage 3 to 5 CKD patients, we report a significant and robust association between CAC score and decline in eGFR over 1 year, controlling for known risk factors for CKD progression (younger age, diabetes mellitus, male sex, hypertension, and proteinuria).18 In the multivariable adjusted regression analysis with CAC reported in quintiles and a CAC equal 0 AU as the reference category, the risk of a 1-year eGFR decline of $5% increased with CAC score categories of 100 and 400 AU, with the CAC score category of $10 AU just missing statistical significance. In the bivariate analysis, the finding that CAC scores .10 AU were associated with kidney function decline is intriguing, and is consistent with the findings of other studies that have reported adverse clinical outcomes for CKD patients having minimally elevated CAC scores (CAC score, 10 to 100 AU). Watanabe et al.2 studied 117 predialysis CKD patients, and demonstrated a CAC score $10 AU was associated with an increased risk of hospitalizations, as well as decreased survival. Therefore, we believe our observation that a CAC score

9

.10 AU is associated with an increased risk of kidney function decline is an important addition to current knowledge. The mechanisms of vascular calcification in CKD remain incompletely understood; however, the extent of vascular calcification reflects disequilibrium between processes that inhibit and promote calcification.19 It is known that vascular calcification inhibitors are reduced (for example fetuin-A)20 or defective (for example reduced matrix gla protein secondary to vitamin K deficiency)4 in CKD. In terms of vascular calcification promoters in CKD patients, disrupted phosphorus homeostasis is thought to be central to the induction of vascular calcification.21 Similarly, decreased kidney function is associated with an increased risk of CAC. One large study involving 1,908 CKD patients demonstrated a graded relationship between CAC and CKD severity, in particular, for patients with eGFR ,30 mL/minute/1.73 m2.22 In addition, we,3 and others,22,23 have demonstrated that the conventional risk factors of cardiovascular disease, such as age, diabetes mellitus, and increased BMI, are also important in the etiology of CAC in CKD patients. We recognize that individuals who have CKD, increased BMI, and CAC share similar risk factors (e.g., diabetes, hypertension); thus, the associations we report may simply reflect this phenomenon. However, we have demonstrated previously that approximately 30% of stage 3 to 5 CKD patients have minimal CAC (CAC score, 0 to 9.9 AU) despite possessing the risk factors for CAC occurrence.3 Moreover, CKD patients protected from CAC appear to remain protected over the longer term.24 Therefore, we hypothesized that a CKD patient who also has CAC may have an increased risk of accelerated vascular disease with associated kidney function decline, and our results appear to support this hypothesis. We also examined the potential role of BMI with respect to kidney function decline in this cohort. Patients who were overweight (BMI: $25 kg/ m2) or obese (BMI: $30 kg/m2) had a greater median decline in 1-year eGFR compared with those who were not overweight or obese, respectively. In the multivariable matrix logistic regression analysis, the interaction between the logarithmically transformed CAC score and BMI was modestly, but statistically, significantly associated with eGFR decline. This suggests that the largest eGFR decline was observed in those who had the highest levels

10

GARLAND ET AL

of both CAC scores and BMI. Further studies are warranted to explore this potential relationship. This study has strengths and limitations. Strengths include the 96% follow-up rate and the prospective design. Diagnosis of CKD was determined based on the National Kidney Foundation’s Kidney Disease Outcomes Quality Initiative criteria,7 and patients were well characterized at baseline for important prognostic risk factors for progressive CKD, including albuminuria. In terms of limitations, although we adjusted for the known risk factors for CKD progression, it is possible that other confounders remained unmeasured in our analysis. Inaccuracies in determining eGFR and CAC could have introduced misclassification errors. Kidney function decline was assessed considering only 2 points in time over 1 year. Several eGFR values would have been preferable to have provided a more precise determination of kidney function decline. Small sample size and the single-center design are also potential factors to consider in accounting for the lack of statistically significant associations between 1-year eGFR decline of $5% and all CAC categories. Second, we were unable to determine whether increased CAC directly impairs kidney function or whether CAC presence is a marker of the toxicity of other factors affecting CAC. We were also unable to quantify the impact of other potentially important biomarkers on the vascular calcification process (e.g., fetuin A,20 vitamin K,4 or fibroblast growth factor-23).25 Additional studies are required to examine these questions. Third, elevated BMI, which was used to identify overweight and obese patients in this study, does not differentiate between increased muscle mass versus excess adipose tissues.26 The gold standard for assessing obesity in stage 3 to 5 CKD patients is unknown, and conflicting results have been demonstrated in the medical literature. Waist-hip ratio and abdominal circumference, which were not measured in this study, have also been implicated as potential risk factors for the development of incident CKD,27,28 and cardiovascular endpoints29 in the stage 3 to 5 CKD population. Further studies are required to determine the optimal measurement of obesity in CKD patients and whether a particular measure of obesity is superior in predicting risk of kidney function decline in CKD patients. In summary, in this prospective cohort study of 125 stage 3 to 5 CKD patients, increasing levels of

CAC and BMI were each identified as independent risk factors for kidney function decline. The association between CAC and eGFR and between the interaction of BMI and CAC both remained robust to adjustment for established risk factors for CKD progression. Given the increased cardiovascular morbidity and mortality attributed to CAC in the CKD population,1,2 as well as the increased mortality rate30 and economic considerations31 associated with the initiation of renal replacement therapies for CKD patients, our findings may have clinical implications. Larger studies will be required for independent validation of these results and to determine whether strategies aimed at reducing BMI and CAC also prevent kidney function decline in CKD patients.

Practical Application In this prospective cohort study of 125 stage 3 to 5 CKD patients, increasing levels of CAC and BMI were identified as independent risk factors for kidney function decline. Results remained robust to adjustment for established risk factors for CKD progression. Obesity is a prevalent problem in predialysis CKD patients, and is identifiable in up to 50% of patients. Therefore, our findings have potential clinical implications for health care providers, in particular, renal dietitians and nephrologists, as obesity and coronary calcification may represent modifiable risk factors in attenuating kidney function decline.

Acknowledgments The authors thank Research Assistant Frances MacLeod for her work with data collection.

References 1. Matsuoka M, Iseki K, Tamashiro M, et al. Impact of high coronary artery calcification score (CACS) on survival in patients on chronic hemodialysis. Clin Exp Nephrol. 2004;8:54-58. 2. Watanabe R, Lemos MM, Manfredi SR, Draibe SA, Canziani ME. Impact of cardiovascular calcification in nondialyzed patients after 24 months of follow-up. Clin J Am Soc Nephrol. 2010;5:189-194. 3. Garland JS, Holden RM, Groome PA, et al. Prevalence and associations of coronary artery calcification in patients with stages 3 to 5 CKD without cardiovascular disease. Am J Kidney Dis. 2008;52:849-858. 4. Holden RM, Morton AR, Garland JS, Pavlov A, Day AG, Booth SL. Vitamins K and D status in stages 3-5 chronic kidney disease. Clin J Am Soc Nephrol. 2010;5:590-597.

BMI, CAC, AND KIDNEY FUNCTION DECLINE 5. Russo D, Miranda I, Ruocco C, et al. The progression of coronary artery calcification in predialysis patients on calcium carbonate or sevelamer. Kidney Int. 2007;72:1255-1261. 6. Block GA, Spiegel DM, Ehrlich J, et al. Effects of sevelamer and calcium on coronary artery calcification in patients new to hemodialysis. Kidney Int. 2005;68:1815-1824. 7. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis. 2004;39(suppl 1):S1-S266. 8. Arnold JM, Liu P, Demers C, et al.; Canadian Cardiovascular Society. Canadian Cardiovascular Society consensus conference recommendations on heart failure 2006: diagnosis and management. Can J Cardiol. 2006;22:23-45. 9. Hemingway H, Fitzpatrick NK, Gnani S, et al. Prospective validity of measuring angina severity with Canadian Cardiovascular Society class: the ACRE study. Can J Cardiol. 2004;20:305-309. 10. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: executive summary: expert panel on the identification, evaluation, and treatment of overweight in adults. Am J Clin Nutr. 1998;68:899-917. 11. Hemmelgarn BR, McAlister FA, Grover S, et al. Canadian Hypertension Education Program. The 2006 Canadian Hypertension Education Program recommendations for the management of hypertension: part I—blood pressure measurement, diagnosis and assessment of risk. Can J Cardiol. 2006;22:573-581. 12. Canadian Diabetes Association 2003 clinical practice guidelines for the prevention and management of diabetes in Canada. Can J Diabetes. 2003;27(suppl 2):S1-S152. 13. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of diet in renal disease study group. Ann Intern Med. 1999; 130:461-470. 14. Levey AS, Coresh J, Greene T, et al.; Chronic Kidney Disease Epidemiology Collaboration. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med. 2006;15(145):247-254. 15. Wexler L, Brundage B, Crouse J, et al.; Writing Group. Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications. A statement for health professionals from the American Heart Association. Circulation. 1996;94:1175-1192. 16. Agatston AS, Janowitz WR, Hildner F. Quantification of coronary artery calcium using ultra fast computed tomography. J Am Coll Cardiol. 1990;15:827-832. 17. Rosner B. Fundamentals of Biostatistics. 6th ed. Pacific Grove, CA: Duxbury Press; 2006.

11

18. Levin A, Djurdjev O, Beaulieu M, Er L. Variability and risk factors for kidney disease progression and death following attainment of stage 4 CKD in a referred cohort. Am J Kidney Dis. 2008; 52:661-671. 19. Covic A, Kanbay M, Voroneanu L, et al. Vascular calcification in chronic kidney disease. Clin Sci (Lond). 2010;119:111-121. 20. Ketteler M, Bongartz P, Westenfeld R, et al. Association of low fetuin-A (AHSG) concentrations in serum with cardiovascular mortality in patients on dialysis: a cross-sectional study. Lancet. 2003;361:827-833. 21. Reynolds JL, Joannides AJ, Skepper JN, et al. Human vascular smooth muscle cells undergo vesicle-mediated calcification in response to changes in extracellular calcium and phosphate concentrations: a potential mechanism for accelerated vascular calcification in ESRD. J Am Soc Nephrol. 2004;15:2857-2867. 22. Budoff MJ, Rader DJ, Reilly MP, et al. Relationship of estimated GFR and coronary artery calcification in the CRIC (Chronic Renal Insufficiency Cohort) study. Am J Kidney Dis. 2011;58:519-526. 23. Tomiyama C, Higa A, Dalboni MA, et al. The impact of traditional and non-traditional risk factors on coronary calcification in pre-dialysis patients. Nephrol Dial Transplant. 2006;21: 2464-2471. 24. Bellasi A, Kooienga L, Block GA, Veledar E, Spiegel DM, Raggi P. How long is the warranty period for nil or low coronary artery calcium in patients new to.hemodialysis? J Nephrol. 2009; 22:255-262. 25. Gutierrez OM, Januzzi JL, Isakova T, et al. Fibroblast growth factor 23 and left ventricular hypertrophy in chronic kidney disease. Circulation. 2009;119:2545-2552. 26. Beddhu S. The body mass index paradox and an obesity, inflammation, and atherosclerosis syndrome in chronic kidney disease. Semin Dial. 2004;17:229-232. 27. Noori N, Hosseinpanah F, Nasiri AA, Azizi F. Comparison of overall obesity and abdominal adiposity in predicting chronic kidney disease incidence among adults. J Ren Nutr. 2009;19: 228-237. 28. Elsayed EF, Sarnak MJ, Tighiouart H, et al. Waist-to-hip ratio, body mass index, and subsequent kidney disease and death. Am J Kidney Dis. 2008;52:29-38. 29. Elsayed EF, Tighiouart H, Weiner DE, et al. Waist-to-hip ratio and body mass index as risk factors for cardiovascular events in CKD. Am J Kidney Dis. 2008;52:49-57. 30. Collins AJ, Foley RN, Gilbertson DT, Chen SC. The state of chronic kidney disease, ESRD, and morbidity and mortality in the first year of dialysis. Clin J Am Soc Nephrol. 2009;4(suppl 1): S5-S11. 31. Klarenbach S, Manns B. Economic evaluation of dialysis therapies. Semin Nephrol. 2009;29:524-532.