Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 May 1.
Published in final edited form as: Diabet Med. 2015 Aug 30;33(5):609–620. doi: 10.1111/dme.12859

Comparison of the heart failure risk stratification performance of the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate in patients with Type 2 diabetes

Y Wang 1, P T Katzmarzyk 1, R Horswell 1, W Zhao 1, J Johnson 2, G Hu 1
PMCID: PMC4723290  NIHMSID: NIHMS710466  PMID: 26202081

Abstract

Aims

To investigate the risk prediction and the risk stratification performances of the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and the Modification of Diet in Renal Disease (MDRD) equation for estimated glomerular filtration rate (eGFRCKD-EPI vs. eGFRMDRD) on heart failure in patients with Type 2 diabetes.

Methods

The study cohort included 12 258 White and 16 886 African American low-income patients with Type 2 diabetes who were 30–90 years old at baseline. Heart failure risk according to different eGFRCKD-EPI and eGFRMDRD categories was prospectively assessed.

Results

During a mean follow-up of 6.5 years, 5043 incident heart failure cases were identified. Multivariable-adjusted hazard ratios (HRs) of heart failure associated with the eGFRCKD-EPI categories [≥ 90 (reference group), 75–89, 60–74, 30–59 and < 30 ml/min/1.73 m2] were 1.00, 1.11, 1.31, 1.75 and 2.93 (Ptrend < 0.001) for African American patients, and 1.00, 1.11, 1.08, 1.59 and 2.92 (Ptrend < 0.001) for White patients, respectively. The model with eGFRCKD-EPI and the other risk factors had significantly higher Harrell’s C than the model with eGFRMDRD and other risk factors. Patients reclassified downward from eGFRMDRD 60–74 to eGFRCKD-EPI 30–59 and from eGFRMDRD 30–59 to eGFRCKD-EPI < 30 ml/min/1.73 m2 showed higher heart failure risk than those who were not reclassified.

Conclusions

Impaired kidney function (i.e. GFR < 60 ml/min/1.73 m2), and even mildly decreased GFR (60–74 ml/min/1.73 m2) estimated by both equations is associated with an increased risk of heart failure. Compared with GFR estimated using the MDRD equation, GFR estimated using the CKD-EPI equation added more predictive power to the model with the other risk factors. Also, eGFRCKD-EPI provided more accurate heart failure risk stratification than eGFRMDRD.

Introduction

Chronic kidney disease (CKD) has emerged as a major health concern worldwide with its high prevalence and heavy economic burdens exerted on society [1,2]. In addition to its risk of progression to end-stage renal disease, CKD is known to be associated with significantly increased risks of cardiovascular disease morbidity and mortality, even at its earliest stage [3,4]. Glomerular filtration rate (GFR) is the best overall index of kidney function and is widely used in the diagnosis, evaluation and management of CKD [5–7]. GFR is most often assessed using estimating equations from serum creatinine measurements [8]. The Modification of Diet in Renal Disease (MDRD) study equation, which was derived from 1628 subjects with CKD, is the most commonly used estimating equation [9]. However, the MDRD study equation, which incorporates age, sex, race and serum creatinine level, has been shown to systematically underestimate GFR in individuals with measured GFR ≥ 60 ml/min/1.73 m2, leading to over-diagnosis of CKD [10]. In 2009, a new estimating equation for GFR based on the same four variables used in the MDRD study equation was proposed by the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) [11]. The CKD-EPI equation, which was developed and internally validated in 10 studies (8254 patients), including the MDRD study, and externally validated in another 16 studies (3896 patients), has been shown to provide more accurate GFR estimates, lower CKD prevalence and better risk predictions [7,12–16]. However, most of these studies were conducted among the general population or among high-risk populations with existing cardiovascular disease and/or CKD [7,13,14,16]. The risk prediction performance of the CKD-EPI equation on cardiovascular disease in patients with diabetes, who are already at high risk of cardiovascular disease compared with people without diabetes [17] is largely unknown [15]. Moreover, no previous study has focused on heart failure as a major outcome. Therefore, this study aims to compare heart failure risk stratification performance of the CKD-EPI equation and the MDRD equation for eGFR in patients with Type 2 diabetes within the Louisiana State University Hospital-Based Longitudinal Study.

Methods

Study population

Between 1997 and 2012, the Louisiana State University Health Care Services Division (LSUHCSD) operated seven public hospitals and affiliated clinics in Louisiana, which provided quality medical care to the residents of Louisiana regardless of their income or insurance coverage [18–24]. Overall, LSUHCSD facilities have served about 1.6 million patients (35% of the Louisiana population) since 1997. Administrative, anthropometric, laboratory and clinical diagnosis data collected at these facilities have been available in electronic form since 1997 for both inpatients and outpatients. Using these data, we have established the Louisiana State University Hospital-Based Longitudinal Study. Since 1997, LSUHCSD’s internal diabetes disease management guidelines have called for physician confirmation of diabetes diagnoses by applying the American Diabetes Association (ADA) criteria: a fasting plasma glucose level ≥ 126 mg/dl; 2-h glucose level ≥ 200 mg/dl after a 75 g 2-h oral glucose tolerance test (OGTT); one or more classic symptoms plus a random plasma glucose level ≥ 200 mg/dl [25]. A cohort of diabetic patients was identified through the Louisiana State University Hospital-Based Longitudinal Study database between 1 January 1999 and 31 December 2009 by using the International Classification of Disease Code (ICD) 250 (ICD-9). The first record of diabetes diagnosis was used to establish the baseline for each patient our analyses due to the design of the cohort study. Before being diagnosed with diabetes, these patients had used our system for an average of 5.0 years. We have validated the diabetes diagnosis in LSUHCSD hospitals. The agreement of diabetes diagnosis was 97%: 20 919 of a sample of 21 566 hospital discharge diagnoses based on ICD codes also had physician-confirmed diabetes by using the ADA diabetes diagnosis criteria [22].

This study included 29 144 patients with newly diagnosed diabetes (12 258 White and 16 886 African American) who were 30–90 years of age without a history of dialysis, heart failure or CHD, and with complete data on all risk factor variables. In these patients with Type 2 diabetes, ~ 78.9% qualify for free care (by virtue of being low income and uninsured – any individual or family unit whose income is at or below 200% of the Federal Poverty Level), ~ 5.0% of patients are self-pay (uninsured, but incomes not low enough to qualify for free care), ~ 5.0% of patients are covered by Medicaid, ~ 8.9% of patients have Medicare, and ~ 2.2% of patients are covered by commercial insurance. The study and analysis plan were approved by the Pennington Biomedical Research Center and the LSU Health Sciences Center Institutional Review Boards, LSU System. We did not obtain informed consent from patients involved in our study because we used anonymized data compiled from electronic medical records.

Baseline measurements

The patient’s characteristics, including age of diabetes diagnosis, sex, race/ethnicity, family income, smoking status, types of health insurance, body weight, height, BMI, blood pressure, total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, HbA1c, creatinine, history and incidence of heart failure, and CHD, and medication (anti-hypertensive drug, cholesterol-lowing drug and antidiabetic drug) within half year after the diabetes diagnosis (baseline) were extracted from the computerized hospitalization records. At each clinical visit, nurses measured height, weight and blood pressure. BMI was calculated by dividing weight in kilograms by the square of height in metres. Plasma total, LDL cholesterol, HDL cholesterol and triglycerides were measured by enzymatic colorimetric methods. Serum glucose was measured by the glucose-oxidase method. HbA1c was measured by immunoassay. Serum creatinine, which was measured using the modified kinetic Jaffe method, was standardized to isotope dilution mass spectrometry. Creatinine concentrations were reduced by 5%, the established calibration factor [26].

GFR estimation

GFR was estimated using the MDRD study equation (eGFRMDRD) [8]: eGFRMDRD = 186 × serum creatinine−1.154 × Age−0.203 × 0.742 (if female) × 1.210 (if African American) and the CKD-EPI equation (eGFRCKD-EPI) [11]: eGFRCKD-EPI = 141 × min (serum creatinine/k, 1)α × max (serum creatinine/k, 1)−1.209 × 0.993Age × 1.018 (if female) × 1.159 (if African American), where k is 0.7 for females and 0.9 for males, α is −0.329 for females and −0.411 for males, min indicates the minimum of serum creatinine/k or 1, and max indicates the maximum of serum creatinine/k or 1.

Prospective follow-up

Follow-up information was obtained from the Louisiana State University Hospital-Based Longitudinal Study inpatient and outpatient database by using the unique number assigned to every patient who visited the LSUHCSD hospitals. Since 1997, diagnosis of heart failure in the LSUHCSD hospitals has been made by the treating physicians using the Framingham Criteria for Heart Failure diagnosis [27]. After clinical diagnosis of heart failure, an echocardiogram was used for each heart failure patient to support the clinical diagnosis, classify heart failure (ejection fraction ≤ 40% or > 40%), and guide the treatment according to the classification. The diagnosis of heart failure was the primary endpoint of interest of the study, and was defined according to the ICD-9: heart failure (ICD-9 codes 402.01, 402.11, 402.91 and 428). We conducted a validation study among 4380 heart failure patients (including patients with and without diabetes) in LSUHCSD hospitals from 2008: of the 4380 heart failure patients, 2353 had an ejection fraction ≤ 40%, and 2027 had an ejection fraction > 40%; 2246 (95%) of the 2353 heart failure patients were confirmed using both the Framingham Criteria for Heart Failure diagnosis [28] and ejection fraction (≤ 40%), and 1430 (71%) of the 2027 heart failure patients were confirmed by using both the Framingham Criteria for Heart Failure diagnosis and ejection fraction (> 40%). Follow-up of each cohort member continued until the date of the diagnosis of heart failure, the date of the last visit if the subject stopped use of LSUHCSD hospitals, death or 31 May 2012 [22].

Statistical analyses

eGFRCKD-EPI and eGFRMDRD were categorized as ≥ 90, 75–89, 60–74, 30–59 and < 30 ml/min/1.73 m2. Cox proportional hazards regression models were used to estimate the association of eGFR with the risk of heart failure. We cross-tabulated eGFR using the above categories and evaluated the proportion of patients in each category of eGFR by the MDRD study equation that was reclassified by the CKD-EPI equation. The risk of heart failure in patients who were reclassified and patients who were not reclassified were assessed using Cox proportional hazards regression models. The analyses were stratified by race (White vs. African American patients) and age (≥ 60 vs. < 60 years). All of the above analyses were first carried out adjusting for age and sex (age- and sex-adjusted model) and further for smoking, income, type of insurance, BMI, systolic blood pressure, HbA1c, LDL cholesterol, triglycerides, myocardial infarction, use of anti-hypertensive drugs, use of diabetes medications and use of cholesterol-lowering agents (multivariate-adjusted model). We computed the Harrell’s C for the model based on all the covariates listed above (model 1), the model based on a combination of the covariates and eGFRCKD-EPI (model 2), and the model based on a combination of the covariates and eGFRMDRD (model 3). The predictive values of these models were compared by using the Harrell’s C associated with the models [29]. Statistical significance was considered to be P < 0.05. All statistical analyses were performed by using SAS for Windows, v. 9.3 (SAS Institute, Cary, NC, USA) and STATA for Windows, v. 13.1 (StataCorp LP, College Station, TX, USA).

Results

The general characteristics of the study population at baseline are presented by race and eGFRCKD-EPI categories in Table 1. Both African American and White patients who had eGFRCKD-EPI ≤ 60 ml/min/1.73 m2 had lower BMI, a lower proportion of current smokers, higher triglycerides and a higher proportion of cholesterol-lowering medication use, when compared with those who had eGFRCKD-EPI > 60 ml/min/1.73 m2. The interaction of eGFR and race was significant on the risk of incident heart failure (P < 0.001 for eGFRCKD-EPI × race and P = 0.011 for eGFRMDRD × race).

Table 1.

Baseline characteristics of African American and White patients with Type 2 diabetes

Characteristics eGFRCKD-EPI categories at baseline (ml/min/1.73 m2)
P
≥ 90 75–89 60–74 30–59 < 30
African American patients
No. patients 9 271 3 473 2 258 1 595 289
Age, mean (SD), years 47.3 (8.8) 52.1 (9.2) 55.2 (9.5) 57.8 (10.5) 48.1 (9.0) < 0.001
Income, mean (SD), $/family 11 971 (10 810) 12 151 (11 593) 12 181 (11 370) 11 663 (9 560) 12 160 (13 047) < 0.001
BMI, mean (SD), kg/m2 33.9 (8.7) 34.1 (8.2) 33.7 (7.9) 32.8 (7.7) 31.6 (8.2) < 0.001
Blood pressure, mean (SD), mmHg
 Systolic 145 (24) 147 (25) 148 (25) 150 (27) 154 (32) < 0.001
 Diastolic 83 (13) 82 (16) 81 (14) 80 (15) 82 (18) 0.367
Total cholesterol, mean (SD), mmol/l 4.9 (1.3) 4.9 (1.2) 4.9 (1.2) 4.9 (1.3) 4.7 (1.5) 0.183
HDL cholesterol, mean (SD), mmol/l 1.2 (0.4) 1.2 (0.4) 1.2 (0.4) 1.2 (0.4) 1.1 (0.4) < 0.001
LDL cholesterol, mean (SD), mmol/l 3.0 (1.0) 3.0 (1.0) 3.0 (1.1) 2.9 (1.1) 2.8 (1.3) 0.134
Triglycerides, mean (SD), mmol/l 1.4 (0.9) 1.4 (0.8) 1.4 (0.8) 1.5 (0.9) 1.6 (0.9) < 0.001
HbA1c, mean (SD), mmol/mol 68 (31) 62 (27) 61 (27) 6.1 (27) 58 (26) < 0.001
HbA1c, mean (SD), % 8.4 (2.8) 7.8 (2.5) 7.7 (2.5) 7.7 (2.5) 7.5 (2.4) < 0.001
Obesity status, % < 0.001
 Normal weight (< 25) 13.9 11.1 11.3 13.5 18.7
 Overweight (25–29.9) 23.5 22.0 23.9 26.3 30.5
 Obesity class I (30.0–34.9) 23.5 26.9 27.1 26.3 24.9
 Obesity class II (≥ 35.0) 39.1 40.1 37.8 33.9 26.0
Current smoker (%) 35.3 32.9 27.5 25.9 23.2 < 0.001
Medication use, %
 Blood pressure 92.0 94.6 96.5 97.1 96.4 < 0.001
 Diabetes 87.2 83.3 84.0 82.8 80.4 0.005
 Cholesterol 67.0 71.7 74.3 78.4 75.0 < 0.001
White patients
No. patients 5 311 2 887 2 306 1 569 185
Age, mean (SD), years 48.1 (9.0) 53.0 (9.2) 56.3 (9.1) 59.7 (9.6) 58.2 (11.7) < 0.001
Income, mean (SD), $/family 13 457 (11 890) 13 656 (12 569) 13 051 (10 263) 13 515 (11 576) 14 752 (16 286) < 0.001
BMI, mean (SD), kg/m2 35.1 (9.0) 35.1 (8.8) 34.7 (8.2) 34.4 (8.6) 33.1 (8.9) < 0.001
Blood pressure, mean (SD), mmHg
 Systolic 141 (21) 141 (21) 143 (22) 144 (24) 141 (26) 0.116
 Diastolic 79 (12) 78 (12) 77 (13) 75 (14) 73 (15) < 0.001
Total cholesterol, mean (SD), mmol/l 5.1 (1.4) 5.0 (1.3) 5.0 (1.3) 4.9 (1.4) 1.5 (57) < 0.001
HDL cholesterol, mean (SD), mmol/l 1.1 (0.3) 1.1 (0.3) 1.1 (0.3) 1.1 (0.3) 1.0 (0.3) < 0.001
LDL cholesterol, mean (SD), mmol/l 2.9 (1.1) 2.9 (1.0) 2.9 (1.0) 2.8 (1.1) 2.4 (1.2) < 0.001
Triglycerides, mean (SD), mmol/l 1.9 (1.1) 2.0 (1.1) 2.0 (1.1) 2.0 (1.0) 2.1 (1.1) < 0.001
HbA1c, mean (SD), mmol/mol 62 (25) 54 (22) 54 (22) 54 (21) 54 (21) < 0.001
HbA1c, mean (SD), % 7.8 (2.3) 7.1 (2.0) 7.1 (2.0) 7.1 (1.9) 7.1 (1.9) < 0.001
Obesity status, % < 0.001
 Normal weight (< 25) 11.2 9.9 8.7 9.9 15.1
 Overweight (25–29.9) 20.0 21.0 21.8 22.8 24.3
 Obesity class I (30.0–34.9) 23.5 24.9 26.1 27.9 21.1
 Obesity class II (≥ 35.0) 45.3 44.1 43.5 39.5 39.5
Current smoker (%) 42.2 36.9 33.7 25.0 24.6 < 0.001
Medication use, %
 Blood pressure 89.2 92.6 94.4 96.6 96.2 < 0.001
 Diabetes 86.7 83.7 82.6 83.6 90.8 0.013
 Cholesterol 72.7 80.2 81.2 83.0 81.5 < 0.001
Open in a new tab

Values are adjusted for age.

During a mean follow-up of 6.5 years, 5043 patients developed heart failure. For African Americans, relative to patients who had eGFRCKD-EPI ≥ 90 ml/min/1.73 m2, patients who had eGFRCKD-EPI 75–89 ml/min/1.73 m2 had a 11% [95% confidence interval (CI) 0–23%) increased risk for heart failure, those who had eGFRCKD-EPI 60–74 ml/min/1.73 m2 had a 31% (95% CI 17–46%) increased risk for heart failure, those who had eGFRCKD-EPI 30–59 ml/min/1.73 m2 had a 75% (95% CI 56–97%) increased risk for heart failure, and patients who had eGFRCKD-EPI < 30 ml/min/1.73 m2 had a 193% (95% CI 140–257%) increased risk for heart failure when adjusted for multiple covariates (Table 2). The pattern of the association between eGFRCKD-EPI and heart failure risk in White patients was similar to that found in African American patients.

Table 2.

Hazard ratios for heart failure according to estimated glomerular filtration rate categories by the CKD-EPI equation and the MDRD study equation at baseline among African American and White patients with Type 2 diabetes

eGFRCKD-EPI categories (ml/min/1.73 m2)
P for trend
≥ 90 75–89 60–74 30–59 < 30
African American patients
 No. cases 1 266 546 448 158 111
 Person-year 65 582 22 982 14 694 10 189 1 516
 Age- and sex-adjusted HR (95% CI) 1.00 1.14 (1.02–1.26) 1.38 (1.23–1.55) 1.93 (1.72–2.17) 3.42 (2.81–4.16) < 0.001
 Multivariable adjustment HR (95% CI)* 1.00 1.11 (1.00–1.23) 1.31 (1.17–1.46) 1.75 (1.56–1.97) 2.93 (2.40–3.57) < 0.001
White patients
 No. cases 719 498 429 484 84
 Person-year 33 191 17 460 13 524 8 948 878
 Age- and sex-adjusted HR (95% CI) 1.00 1.12 (1.00–1.26) 1.15 (1.02–1.31) 1.75 (1.54–1.98) 3.29 (2.61–4.14) < 0.001
 Multivariable adjustment HR (95% CI)* 1.00 1.11 (0.98–1.24) 1.08 (0.95–1.23) 1.59 (1.40–1.80) 2.92 (2.31–3.69) < 0.001
Both
 No. cases 1 985 1 044 877 942 195
 Person-year 98 773 40 442 28 218 19 138 2 395
 Age- and sex-adjusted HR (95% CI)† 1.00 1.13 (1.05–1.22) 1.27 (1.16–1.38) 1.85 (1.70–2.02) 3.39 (2.92–3.94) < 0.001
 Multivariable adjustment HR (95% CI)† 1.00 1.11 (1.03–1.20) 1.19 (1.10–1.30) 1.68 (1.54–1.83) 2.94 (2.53–3.41) < 0.001
eGFRMDRD categories (ml/min/1.73 m2)
P for trend
≥ 90 75–89 60–74 30–59 < 30
African Americans
 No. cases 1 325 541 461 408 94
 Person-year 66 697 23 739 14 418 8 863 1 248
 Age- and sex-adjusted HR (95% CI) 1.00 1.09 (0.98–1.20) 1.43 (1.28–1.60) 1.97 (1.75–2.21) 3.54 (2.87–4.37) < 0.001
 Multivariable adjustment HR (95% CI)* 1.00 1.08 (0.98–1.20) 1.34 (1.20–1.49) 1.77 (1.57–1.99) 3.17 (2.57–3.92) < 0.001
Whites
 No. cases 667 519 450 506 72
 Person-year 29 885 18 814 15 004 9 526 772
 Age- and sex-adjusted HR (95% CI) 1.00 1.08 (0.96–1.21) 1.12 (0.99–1.27) 1.73 (1.52–1.96) 3.42 (2.68–4.37) < 0.001
 Multivariable adjustment HR (95% CI)* 1.00 1.08 (0.96–1.21) 1.09 (0.96–1.23) 1.59 (1.40–1.80) 3.06 (2.39–3.92) < 0.001
Both
 No. cases 1 992 1 060 911 914 166
 Person-year 96 582 42 553 29 422 18 389 2 020
 Age- and sex-adjusted HR (95% CI)† 1.00 1.09 (1.01–1.18) 1.27 (1.17–1.38) 1.87 (1.72–2.04) 3.51 (3.00–4.12) < 0.001
 Multivariable adjustment HR (95% CI)† 1.00 1.09 (1.01–1.18) 1.21 (1.11–1.31) 1.70 (1.57–1.85) 3.11 (2.65–3.65) < 0.001
Open in a new tab
*

Adjusted for age, sex, smoking, income, type of insurance, BMI, systolic blood pressure, HbA1c, LDL cholesterol, triglycerides, myocardial infarction, use of antihypertensive drugs, use of diabetes medications, and use of cholesterol-lowering agents

†

Adjusted also for race.

Similarly, the multivariable-adjusted hazard ratios (HRs) of heart failure at five eGFRMDRD groups (≥ 90, 75–89, 60–74, 30–59 and ≤ 30 ml/min/1.73 m2) were 1.00, 1.08 (95% CI 0.98–1.20), 1.34 (95% CI 1.20–1.49), 1.77 (95% CI 1.57–1.99) and 3.17 (95% CI 2.57–3.92) among African American patients (Ptrend < 0.001), and 1.00, 1.08 (95% CI 0.96–1.21), 1.09 (95% CI 0.96–1.23), 1.59 (95% CI 1.40–1.80) and 3.06 (95% CI 2.39–3.92) among White patients (Ptrend < 0.001), respectively (Table 2). Harrell’s C for the models without eGFR but all the other covariates (model 1) were 0.683 (95% CI 0.668–0.698) for African American patients and 0.710 (95% CI 0.694–0.727) for White patients. Harrell’s C for the models with eGFRCKD-EPI and all the other covariates (model 2) were 0.697 (95% CI 0.682–0.711) for African American patients and 0.716 (95% CI 0.700–0.733) for White patients. Harrell’s C for the models with GFRMDRD and all the other covariates (model 3) were 0.694 (95% CI 0.680–0.710) for African American patients and 0.714 (95% CI 0.698–0.731) for White patients. Among African American patients, values of Harrell’s C were statistically different between model 1 and model 2 (P < 0.001), between model 1 and model 3 (P < 0.001), and between model 2 and model 3 (P = 0.007). Among White patients, values of Harrell’s C were statistically different between model 1 and model 2 (P = 0.019), between model 1 and model 3 (P = 0.108), and between model 2 and model 3 (P = 0.035).

The median value for eGFRCKD-EPI [90.0 (interquartile range, IQR, 33.6)] was higher than for eGFRMDRD [89.0 (IQR, 34.9)]. More patients (144) left the MDRD defined > 60 category than the number of new patients that enter this category when using the CKD-EPI. As a result, using eGFRCKD-EPI, the overall prevalence of impaired eGFR (i.e. < 60 ml/min/1.73 m2) was 12.5% compared with 12% using eGFRMDRD (Table 3).

Table 3.

Reclassification across estimated glomerular filtration rate categories by the CKD-EPI equation from estimated glomerular filtration rate categories based on the MDRD study equation

eGFRMDRD categories (ml/min/1.73 m2) eGFRCKD-EPI categories (ml/min/1.73 m2) Total
≥ 90 75–89 60–74 30–59 < 30
≥ 90 13 738 (47.1%) 413 (1.4%) 0 (0%) 0 (0%) 0 (0%) 14 151 (48.6%)
75–89 844 (2.9%) 5 530 (19.0%) 325 (1.1%) 0 (0%) 0 (0%) 6 699 (23.0%)
60–74 0 (0%) 417 (1.4%) 4 104 (14.1%) 279 (1.0%) 0 (0%) 4 800 (16.5%)
30–59 0 (0%) 0 (0%) 135 (0.5%) 2 885 (9.9%) 72 (0.3%) 3 092 (10.6%)
< 30 0 (0%) 0 (0%) 0 (0%) 0 (0%) 402 (1.4%) 402 (1.4%)
Total 14 582 (50.0%) 6 360 (21.8%) 4 564 (15.7%) 3 164 (10.9%) 474 (1.6%) 29 144 (100.0%)
Open in a new tab

Compared with African American patients with both eGFRCKD-EPI and eGFRMDRD ≥ 90 ml/min/1.73 m2, the multivariable-adjusted HRs were: 1.34 (95% CI 1.18–1.50) for African American patients with both eGFRCKD-EPI and eGFRMDRD 60–74 ml/min/1.73 m2; 1.40 (95% CI 1.08–1.81) for those with eGFRCKD-EPI 30–59, but eGFRMDRD 60–74 ml/min/1.73 m2; 1.79 (95% CI 1.58–2.02) for those with both eGFRCKD-EPI and eGFRMDRD 30–59 ml/min/1.73 m2; 1.90 (95% CI 1.17–3.08) for those with eGFRCKD-EPI < 30 but eGFRMDRD 30–59 ml/min/1.73 m2; and 3.18 (95% 2.57–3.94) for those with both eGFRCKD-EPI and eGFRMDRD < 30 ml/min/1.73 m2 (Table 4). Compared with White patients with both eGFRCKD-EPI and eGFRMDRD ≥ 90 ml/min/1.73 m2, the HRs were: 1.58 (95% CI 1.39–1.81) for White patients with both eGFRCKD-EPI and eGFRMDRD 30–59 ml/min/1.73 m2; 2.16 (95% CI 1.20–3.88) for those with eGFRCKD-EPI < 30, but eGFRMDRD 30–59 ml/min/1.73 m2; and 3.05 (95% 2.38–3.91) for those with both eGFRCKD-EPI and eGFRMDRD < 30 ml/min/1.73 m2 (Table 4). Stratification for age yielded similar results (Table 4).

Table 4.

Hazard ratios for heart failure according to classification to estimated glomerular filtration rate categories by the CKD-EPI equation and MDRD satudy equation at baseline among African American and White patients with diabetes

eGFRMDRD categories (ml/min/1.73 m2) eGFRCKD-EPI categories (ml/min/1.73 m2)
≥ 90 75–89 60–74 30–59 < 30
Both African American and White Patients
 ≥ 90 No. cases 1 902 90 – – –
Person-year 93 647 2 936 – – –
Age- and sex-adjusted HR (95% CI)† 1.00 1.03 (0.83–1.28) – – –
Multivariable adjustment HR (95% CI)† 1.00 0.95 (0.77–1.18) – – –
 75–89 No. cases 83 901 76 – –
Person-year 5 126 35 142 2 284 – –
Age- and sex-adjusted HR (95% CI)† 0.89 (0.72–1.11) 1.12 (1.04–1.22) 1.02 (0.80–1.29) – –
Multivariable adjustment HR (95% CI)† 0.94 (0.75–1.17) 1.11 (1.02–1.20) 1.01 (0.80–1.28) – –
 60–74 No. cases – 53 780 78 –
Person-year – 2 365 25 210 1 847 –
Age- and sex-adjusted HR (95% CI)† – 1.20 (0.91–1.58) 1.28 (1.17–1.39) 1.34 (1.06–1.69) –
Multivariable adjustment HR (95% CI)† – 1.26 (0.96–1.66) 1.20 (1.10–1.30) 1.25 (0.99–1.58) –
 30–59 No. cases – – 21 864 29
Person-year – – 723 17 291 375
Age- and sex-adjusted HR (95% CI)† – – 1.37 (0.89–2.11) 1.88 (1.73–2.06) 2.58 (1.38–3.73)
Multivariable adjustment HR (95% CI)† – – 1.38 (0.89–2.12) 1.70 (1.56–1.86) 2.08 (1.44–3.02)
 < 30 No. cases – – – – 166
Person-year – – – – 2 020
Age- and sex-adjusted HR (95% CI)† – – – – 3.53 (3.01–4.14)
Multivariable adjustment HR (95% CI)† – – – – 3.10 (2.64–3.65)
African American patients
≥ 90 No. cases 1 248 77 – – –
Person-year 64 032 2 665 – – –
Age- and sex-adjusted HR (95% CI)† 1.00 1.10 (0.87–1.40) – – –
Multivariable adjustment HR (95% CI)† 1.00 1.02 (0.81–1.30) – – –
 75–89 No. cases 18 468 55 – –
Person-year 1 550 20 257 1 932 – –
Age- and sex-adjusted HR (95% CI)† 0.74 (0.47–1.18) 1.13 (1.01–1.25) 1.01 (0.76–1.33) – –
Multivariable adjustment HR (95% CI)† 0.71 (0.45–1.14) 1.12 (1.00–1.24) 1.04 (0.79–1.38) – –
 60–74 No. cases – 1 393 67 –
Person-year – 60 12 714 1 643 –
Age- and sex-adjusted HR (95% CI)† – 1.24 (0.18–8.84) 1.44 (1.28–1.61) 1.52 (1.18–1.96) –
Multivariable adjustment HR (95% CI)† – 1.20 (0.16–8.56) 1.34 (1.18–1.50) 1.40 (1.08–1.81) –
 30–59 No. cases – – – 391 17
Person-year – – – 8 547 268
Age- and sex-adjusted HR (95% CI)† – – – 1.98 (1.75–2.23) 2.61 (1.61–4.22)
Multivariable adjustment HR (95% CI)† – – – 1.79 (1.58–2.02) 1.90 (1.17–3.08)
 < 30 No. cases – – – – 94
Person-year – – – – 1 248
Age- and sex-adjusted HR (95% CI)† – – – – 3.57 (2.89–4.41)
Multivariable adjustment HR (95% CI)† – – – – 3.18 (2.57–3.94)
White patients
≥ 90 No. cases 654 13 – – –
Person-year 29 615 271 – – –
Age- and sex-adjusted HR (95% CI)† 1.00 1.05 (0.60–1.83) – – –
Multivariable adjustment HR (95% CI)† 1.00 0.95 (0.55–1.66) – – –
 75–89 No. cases 65 433 21 – –
Person-year 3 576 14 885 353 – –
Age- and sex-adjusted HR (95% CI)† 0.94 (0.73–1.22) 1.09 (0.97–1.24) 1.36 (0.87–2.11) – –
Multivariable adjustment HR (95% CI)† 1.02 (0.79–1.32) 1.08 (0.96–1.23) 1.14 (0.73–1.79) – –
 60–74 No. cases – 52 387 11 –
Person-year – 2 304 12 496 204 –
Age- and sex-adjusted HR (95% CI)† – 1.25 (0.95–1.67) 1.11 (0.98–1.27) 0.95 (0.52–1.75) –
Multivariable adjustment HR (95% CI)† – 1.33 (1.00–1.77) 1.06 (0.93–1.21) 1.07 (0.59–1.98) –
 30–59 No. cases – – 21 473 12
Person-year – – 675 8 744 106
Age- and sex-adjusted HR (95% CI)† – – 1.40 (0.91–2.17) 1.74 (1.53–1.99) 2.35 (1.32–4.21)
Multivariable adjustment HR (95% CI)† – – 1.38 (0.89–2.14) 1.58 (1.39–1.81) 2.16 (1.20–3.88)
 < 30 No. cases – – – – 72
Person-year – – – – 772
Age- and sex-adjusted HR (95% CI)† – – – – 3.44 (2.69–4.39)
Multivariable adjustment HR (95% CI)† – – – – 3.05 (2.38–3.91)
Age ≥ 60 years
 ≥ 90 No. cases 184 66 – – –
Person-year 6 412 2 028 – – –
Age- and sex-adjusted HR (95% CI)† 1.00 1.08 (0.81–1.43) – – –
Multivariable adjustment HR (95% CI)† 1.00 0.95 (0.72–1.27) – – –
 75–89 No. cases 19 215 62 – –
Person-year 642 6 139 1 669 – –
Age- and sex-adjusted HR (95% CI)† 0.97 (0.60–1.58) 1.11 (0.91–1.36) 1.10 (0.82–1.48) – –
Multivariable adjustment HR (95% CI)† 0.98 (0.60–1.58) 1.08 (0.88–1.32) 1.06 (0.79–1.42) – –
 60–74 No. cases – – 245 54 –
Person-year – – 5 906 1 203 –
Age- and sex-adjusted HR (95% CI)† – – 1.28 (1.05–1.56) 1.29 (0.94–1.76) –
Multivariable adjustment HR (95% CI)† – – 1.15 (0.95–1.41) 1.17 (0.85–1.60) –
 30–59 No. cases – – 6 366 17
Person-year – – 179 6 929 187
Age- and sex-adjusted HR (95% CI)† – – 1.18 (0.52–2.68) 1.57 (1.31–1.89) 2.31 (1.39–3.83)
Multivariable adjustment HR (95% CI)† – – 1.19 (0.52–2.69) 1.43 (1.19–1.72) 1.82 (1.10–3.04)
 < 30 No. cases – – – – 63
Person-year – – – – 504
Age- and sex-adjusted HR (95% CI)† – – – – 3.90 (2.92–5.20)
Multivariable adjustment HR (95% CI)† – – – – 2.87 (2.14–3.86)
Age < 60 years
 ≥ 90 No. cases 1 718 24 – – –
Person-year 87 235 908 – – –
Age- and sex-adjusted HR (95% CI)† 1.00 1.05 (0.70–1.57) – – –
Multivariable adjustment HR (95% CI)† 1.00 1.02 (0.68–1.53) – – –
 75–89 No. cases 64 686 14 – –
Person-year 4 484 29 003 615 – –
Age- and sex-adjusted HR (95% CI)† 0.90 (0.70–1.16) 1.12 (1.02–1.22) 0.86 (0.51–1.46) – –
Multivariable adjustment HR (95% CI)† 0.93 (0.72–1.20) 1.10 (1.00–1.20) 0.89 (0.52–1.50) – –
 60–74 No. cases – 53 535 24 –
Person-year – 2 362 19 304 643 –
Age- and sex-adjusted HR (95% CI)† – 1.26 (0.95–1.66) 1.24 (1.12–1.37) 1.55 (1.04–2.33) –
Multivariable adjustment HR (95% CI)† – 1.32 (1.00–1.75) 1.17 (1.06–1.30) 1.66 (1.10–2.49) –
 30–59 No. cases – – 15 498 12
Person-year – – 544 10 362 187
Age- and sex-adjusted HR (95% CI)† – – 1.41 (0.85–2.35) 2.13 (1.92–2.36) 2.64 (1.49–4.66)
Multivariable adjustment HR (95% CI)† – – 1.41 (0.84–2.34) 1.86 (1.68–2.06) 2.01 (1.14–3.55)
 < 30 No. cases – – – – 103
Person-year – – – – 1 516
Age- and sex-adjusted HR (95% CI)† – – – – 3.28 (2.69–4.00)
Multivariable adjustment HR (95% CI)† – – – – 3.11 (2.54–3.80)
Open in a new tab
*

Adjusted for age, sex, smoking, income, type of insurance, BMI, systolic blood pressure, HbA1c, LDL cholesterol, triglycerides, myocardial infarction, use of antihypertensive drugs, use of diabetes medications, and use of cholesterol-lowering agents

†

Also adjusted for race.

Discussion

This study demonstrated that both reduced eGFRCKD-EPI and reduced eGFRMDRD (< 75 ml/min/1.73 m2) were significantly associated with an increased risk of incident heart failure among patients with Type 2 diabetes. However, compared with eGFRMDRD, eGFRCKD-EPI adds more predictive power to a model with only conventional covariates. Also, eGFRCKD-EPI provides better risk stratification when eGFR < 75 ml/min/1.73 m2, because patients reclassified downward by the CKD-EPI equation showed higher heart failure risk than those who were not reclassified.

Although the association of eGFR with all-cause mortality, cardiovascular disease mortality or end-stage renal disease has been extensively studied in patients with and without diabetes [30] few studies have investigated the association between eGFR and incident cardiovascular disease risk in patients with diabetes [31,32]. This risk association may be of particular interest because, in patients with Type 2 diabetes, the additional development of diabetic kidney disease would markedly amplify their risk for cardiovascular disease [32,33]. Two studies assessed the association of eGFRMDRD with composite cardiovascular disease end points including cardiovascular disease death and incident cardiovascular disease in patients with Type 2 diabetes [28,29]. Both studies found that risk of the composite cardiovascular disease end points increased at eGFRMDRD < 60 ml/min/1.73 m2, when compared with eGFR ≥ 90 ml/min/1.73 m2. In the current study, besides eGFRMDRD < 60 ml/min/1.73 m2, even mildly decreased eGFRMDRD (60–74 ml/min/1.73 m2) predicts heart failure risk, which indicated that eGFR might be a more sensitive marker for incident cardiovascular disease than for cardiovascular disease mortality. In this study, for the first time, eGFRCKD-EPI < 75 ml/min/1.73 m2 was also found to be associated with an increased risk of heart failure, which suggested that like eGFRMDRD, eGFRCKD-EPI can be also used for cardiovascular disease risk stratification.

Moreover, the increment in prognostic utility of eGFR in heart failure was investigated. The result indicated that, among African American patients, both eGFRCKD-EPI and eGFRMDRD added more predictive value in heart failure risk beyond other heart failure risk factors, i.e. age, sex, smoking, income, type of insurance, BMI, systolic blood pressure, HbA1c, LDL cholesterol, triglycerides, myocardial infarction, use of anti-hypertensive drugs, use of diabetes medications and use of cholesterol-lowering agents. However, among White patients, only eGFRCKD-EPI added more predictive power to other covariates in predicting heart failure. Besides, we also compared the model with eGFRCKD-EPI (model 2) with the model with eGFRMDRD (model 3), the model with eGFRCKD-EPI had significantly higher predictive power than the model with eGFRMDRD (model 3), which indicated that eGFRCKD-EPI was a better predictor for future heart failure than eGFRMDRD.

By showing that, when eGFR below 75 ml/min/1.73 m2, patients reclassified downward by the CKD-EPI equation showed higher risk than those who were not reclassified, our study demonstrated that eGFRCKD-EPI may provide more accurate heart failure risk stratification than eGFRMDRD. However, it is unclear whether our finding could be attributable to a higher accuracy of the CKD-EPI equation than the MDRD equation. Because the ‘gold’ standard – the direct measured GFR was not available, our study cannot verify whether the CKD-EPI equation provides a more accurate GFR estimate than the MDRD study equation in patients with Type 2 diabetes [12]. Actually, results regarding the performance of the CKD-EPI equation in estimating GFR in patients with diabetes were mixed [34–36]. Two studies [31,32] showed that the CKD-EPI equation did not exhibit better performance than the MDRD study equation in estimating GFR, whereas another study [36] demonstrated that the CKD-EPI equation is more accurate overall and across subgroup with diabetes. Because of the small sample size in these studies, it is crucial to test the performance of the CKD-EPI equation in a bigger diabetic cohort. Of note, our study did not find a lower prevalence of eGFR < 60 ml/min/1.73 m2 using eGFRCKD-EPI compared with when using eGFRMDRD, which is inconsistent with previous studies [7,13,14]. Differences in the characteristics of the study populations may contribute to this inconsistency: compared with previous cohorts [7,13,14], the current Type 2 diabetes cohort had a higher proportion of African American patients, and patients were mainly from low income class. There are several strengths in our study, including the large sample size, high proportion of African American patients, long follow-up time, and the use of administrative databases to avoid the problem of differential recall bias. In addition, patients in this study used the same public healthcare system and have the same socio-economic status, which minimizes the influence from the accessibility of health care, particularly when comparing African American and White patients. One limitation of our study is that our analysis was not performed on a representative sample of the state of Louisiana’s population, which limits the generalizability of this study; however, LSUHCSD hospitals are public hospitals and cover over 1.6 million patients, most of whom are low-income persons in Louisiana. A second limitation is that even though our analyses were adjusted for an extensive set of confounding factors, residual confounding due to the measurement error in the assessment of confounding factors, unmeasured factors such as physical activity, education, dietary factors, and family history of diabetes and other chronic diseases cannot be excluded.

In conclusion, we found that impaired kidney function (i.e. GFR< 60 ml/min/1.73 m2), even mildly decreased GFR (60–74 ml/min/1.73 m2) estimated by both equations is associated with an increased risk of heart failure in low-income patients with Type 2 diabetes. Compared with GFR estimated using the MDRD study equation, GFR estimated using the CKD-EPI equation added more predictive power to the model with the other risk factors. Also, eGFRCKD-EPI provided more accurate heart failure risk stratification than eGFRMDRD in this low-income cohort.

What’s new?

  • The is the first large prospective study to assess the risk prediction and risk stratification performances of the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and the Modification of Diet in Renal Disease (MDRD) equation for estimated glomerular filtration rate ((eGFRCKD-EPI vs. eGFRMDRD) on heart failure in low-income patients with Type 2 diabetes.

  • The study showed that impaired kidney function (i.e. GFR < 60 ml/min/1.73 m2), and even mildly decreased GFR (60–74 ml/min/1.73 m2) estimated by both equations is associated with an increased risk of heart failure.

  • Compared with eGFRMDRD, eGFRCKD-EPI adds more predictive power to a model with only conventional covariates.

  • Also, eGFRCKD-EPI may provide more accurate heart failure risk stratification than eGFRMDRD.

Acknowledgments

Funding sources

This work was supported by Louisiana State University’s Improving Clinical Outcomes Network (LSU ICON), the Louisiana Clinical Data Research Network (LACDRN), and 1 U54 GM104940 from the National Institute of General Medical Sciences of the National Institutes of Health which funds the Louisiana Clinical and Translational Science (LA CaTS) Center.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References

  • 1.Zhang Q-L, Rothenbacher D. Prevalence of chronic kidney disease in population-based studies: systematic review. BMC Public Health. 2008;8 doi: 10.1186/1471-2458-8-117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Khan S, Amedia CA. Economic burden of chronic kidney disease. J Eval Clin Pract. 2008;14:422–434. doi: 10.1111/j.1365-2753.2007.00883.x. [DOI] [PubMed] [Google Scholar]
  • 3.Go AS, Chertow GM, Fan DJ, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med. 2004;351:1296–1305. doi: 10.1056/NEJMoa041031. [DOI] [PubMed] [Google Scholar]
  • 4.Schiffrin EL, Lipman ML, Mann JFE. Chronic kidney disease – effects on the cardiovascular system. Circulation. 2007;116:85–97. doi: 10.1161/CIRCULATIONAHA.106.678342. [DOI] [PubMed] [Google Scholar]
  • 5.Vassalotti JA, Stevens LA, Levey AS. Testing for chronic kidney disease: a position statement from the National Kidney Foundation. Am J Kidney Dis. 2007;50:169–180. doi: 10.1053/j.ajkd.2007.06.013. [DOI] [PubMed] [Google Scholar]
  • 6.Crowe E, Halpin D, Stevens P. Guideline Dev G. Guidelines – early identification and management of chronic kidney disease: summary of NICE guidance. BMJ. 2008;337 doi: 10.1136/bmj.a1530. [DOI] [PubMed] [Google Scholar]
  • 7.Matsushita K, Mahmoodi BK, Woodward M, Emberson JR, Jafar TH, Jee SH, et al. Comparison of risk prediction using the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate. JAMA. 2012;307:1941–1951. doi: 10.1001/jama.2012.3954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.National Kidney Foundation. Clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Diseases. 2002;39:S1–266. [PubMed] [Google Scholar]
  • 9.Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med. 1999;130:461–470. doi: 10.7326/0003-4819-130-6-199903160-00002. [DOI] [PubMed] [Google Scholar]
  • 10.Stevens LA, Coresh J, Feldman HI, Greene T, Lash JP, Nelson RG, et al. Evaluation of the modification of diet in renal disease study equation in a large diverse population. J Am Soc Nephrol. 2007;18:2749–2757. doi: 10.1681/ASN.2007020199. [DOI] [PubMed] [Google Scholar]
  • 11.Levey AS, Stevens LA, Schmid CH, Zhang Y, Castro AF, III, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:U604–U607. doi: 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Levey AS, Stevens LA. Estimating GFR using the CKD Epidemiology Collaboration (CKD-EPI) creatinine equation: more accurate GFR estimates, lower CKD prevalence estimates, and better risk predictions. Am J Kidney Dis. 2010;55:622–627. doi: 10.1053/j.ajkd.2010.02.337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Matsushita K, Selvin E, Bash LD, Astor BC, Coresh J. Risk implications of the new CKD Epidemiology Collaboration (CKD-EPI) equation compared with the MDRD study equation for estimated GFR: the Atherosclerosis Risk in Communities (ARIC) study. Am J Kidney Dis. 2010;55:648–659. doi: 10.1053/j.ajkd.2009.12.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Skali H, Uno H, Levey AS, Inker LA, Pfeffer MA, Solomon SD. Prognostic assessment of estimated glomerular filtration rate by the new Chronic Kidney Disease Epidemiology Collaboration equation in comparison with the Modification of Diet in Renal Disease study equation. Am Heart J. 2011;162:548–554. doi: 10.1016/j.ahj.2011.06.006. [DOI] [PubMed] [Google Scholar]
  • 15.Pugliese G, Solini A, Bonora E, Orsi E, Zerbini G, Giorgino F, et al. The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation provides a better definition of cardiovascular burden associated with CKD than the Modification of Diet in Renal Disease (MDRD) Study formula in subjects with type 2 diabetes. Atherosclerosis. 2011;218:194–199. doi: 10.1016/j.atherosclerosis.2011.04.035. [DOI] [PubMed] [Google Scholar]
  • 16.Stevens LA, Li S, Tamura MK, Chen S-C, Vassalotti JA, Norris KC, et al. Comparison of the CKD Epidemiology Collaboration (CKD-EPI) and Modification of Diet in Renal Disease (MDRD) study equations: risk factors for and complications of CKD and mortality in the Kidney Early Evaluation Program (KEEP) Am J Kidney Dis. 2011;57:S9–S16. doi: 10.1053/j.ajkd.2010.11.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hu G, Jousilahti P, Tuomilehto J. Joint effects of history of hypertension at baseline and type 2 diabetes at baseline and during follow-up on the risk of coronary heart disease. Eur Heart J. 2007;28:3059–3066. doi: 10.1093/eurheartj/ehm501. [DOI] [PubMed] [Google Scholar]
  • 18.Li W, Wang Y, Chen L, Horswell R, Xiao K, Besse J, et al. Increasing prevalence of diabetes in middle or low income residents in Louisiana from 2000 to 2009. Diabetes Res Clin Pract. 2011;94:262–268. doi: 10.1016/j.diabres.2011.08.010. [DOI] [PubMed] [Google Scholar]
  • 19.Wang Y, Chen L, Xiao K, Horswell R, Besse J, Johnson J, et al. Increasing incidence of gestational diabetes mellitus in Louisiana, 1997–2009. J Womens Health. 2012;21:319–325. doi: 10.1089/jwh.2011.2838. [DOI] [PubMed] [Google Scholar]
  • 20.Wang Y, Chen L, Horswell R, Xiao K, Besse J, Johnson J, et al. Racial differences in the association between gestational diabetes mellitus and risk of Type 2 diabetes. J Womens Health. 2012;21:628–633. doi: 10.1089/jwh.2011.3318. [DOI] [PubMed] [Google Scholar]
  • 21.Hu G, Horswell R, Wang Y, Li W, Besse J, Xiao K, et al. Body mass index and the risk of dementia among Louisiana low income diabetic patients. PLoS One. 2012;7:e44537. doi: 10.1371/journal.pone.0044537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Wang Y, Katzmarzyk PT, Horswell R, Li W, Xiao K, Besse J, et al. Racial disparities in diabetic complications in an underinsured population. J Clin Endocrinol Metab. 2012;97:4446–4453. doi: 10.1210/jc.2012-2378. [DOI] [PubMed] [Google Scholar]
  • 23.Zhao W, Katzmarzyk PT, Horswell R, Wang Y, Li W, Johnson J, et al. Aggressive blood pressure control increases coronary heart disease risk among diabetic patients. Diabetes Care. 2013;36:3287–3296. doi: 10.2337/dc13-0189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zhao W, Katzmarzyk PT, Horswell R, Wang Y, Johnson J, Cefalu WT, et al. Blood pressure and stroke risk among diabetic patients. J Clim Endocrinol Metab. 2013;98:3653–3662. doi: 10.1210/jc.2013-1757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Gavin JR, Alberti K, Davidson MB, DeFronzo RA, Drash A, Gabbe SG, et al. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 1997;20:1183–1197. doi: 10.2337/diacare.20.7.1183. [DOI] [PubMed] [Google Scholar]
  • 26.Levey AS, Coresh J, Greene T, Marsh J, Stevens LA, Kusek JW, et al. Expressing the modification of diet in renal disease study equation for estimating glomerular filtration rate with standardized serum creatinine values. Clin Chem. 2007;53:766–772. doi: 10.1373/clinchem.2006.077180. [DOI] [PubMed] [Google Scholar]
  • 27.Ho KK, Anderson KM, Kannel WB, Grossman W, Levy D. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation. 1993;88:107–115. doi: 10.1161/01.cir.88.1.107. [DOI] [PubMed] [Google Scholar]
  • 28.McKee PA, Castelli WP, McNamara PM, Kannel WB. Natural history of congestive heart failure: Framingham study. N Engl J Med. 1971;285:1441–1446. doi: 10.1056/NEJM197112232852601. [DOI] [PubMed] [Google Scholar]
  • 29.Newson RB. Comparing the predictive powers of survival models using Harrell’s C or Somers’ D. Stata J. 2010;10:339–358. [Google Scholar]
  • 30.Fox CS, Matsushita K, Woodward M, Bilo HJG, Chalmers J, Lambers Heerspink HJ, et al. Associations of kidney disease measures with mortality and end-stage renal disease in individuals with and without diabetes: a meta-analysis. Lancet. 2012;380:1662–1673. doi: 10.1016/S0140-6736(12)61350-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.So WY, Kong APS, Ma RCW, Ozaki R, Szeto CC, Chan NN, et al. Glomerular filtration rate, cardiorenal end points, and all-cause mortality in type 2 diabetic patients. Diabetes Care. 2006;29:2046–2052. doi: 10.2337/dc06-0248. [DOI] [PubMed] [Google Scholar]
  • 32.Ninomiya T, Perkovic V, de Galan BE, Zoungas S, Pillai A, Jardine M, et al. Albuminuria and kidney function independently predict cardiovascular and renal outcomes in diabetes. J Am Soc Nephrol. 2009;20:1813–1821. doi: 10.1681/ASN.2008121270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.de Boer IH, Rue TC, Hall YN, Heagerty PJ, Weiss NS, Himmelfarb J. Temporal trends in the prevalence of diabetic kidney disease in the United States. JAMA. 2011;305:2532–2539. doi: 10.1001/jama.2011.861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Silveiro SP, Araujo GN, Ferreira MN, Souza FDS, Yamaguchi HM, Camargo EG. Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation pronouncedly underestimates glomerular filtration rate in Type 2 diabetes. Diabetes Care. 2011;34:2353–2355. doi: 10.2337/dc11-1282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Rognant N, Lemoine S, Laville M, Hadj-Aissa A, Dubourg L. Performance of the Chronic Kidney Disease Epidemiology Collaboration equation to estimate glomerular filtration rate in diabetic patients. Diabetes Care. 2011;34:1320–1322. doi: 10.2337/dc11-0203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Stevens LA, Schmid CH, Greene T, Zhang Y, Beck GJ, Froissart M, et al. Comparative performance of the CKD Epidemiology Collaboration (CKD-EPI) and the Modification of Diet in Renal Disease (MDRD) study equations for estimating GFR levels above 60 ml/min/1. 73 m(2) Am J Kidney Dis. 2010;56:486–495. doi: 10.1053/j.ajkd.2010.03.026. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES