0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation |

Diet and Kidney Disease in High-Risk Individuals With Type 2 Diabetes Mellitus FREE

Daniela Dunkler, PhD1,2,3; Mahshid Dehghan, PhD1; Koon K. Teo, PhD1; Georg Heinze, PhD3; Peggy Gao, MSc1; Maria Kohl, MSc1,2,3; Catherine M. Clase, MB4; Johannes F. E. Mann, MD2,5; Salim Yusuf, DPhil1; Rainer Oberbauer, MD6; for the ONTARGET Investigators
[+] Author Affiliations
1Population Health Research Institute, McMaster University, Hamilton, Ontario, Canada
2Department of Nephrology, Universitaetsklinikum Erlangen, Erlangen, Germany
3Section of Clinical Biometrics, Center for Medical Statistics, Informatics, and Intelligent Systems, Medical University of Vienna, Vienna, Austria
4Department of Medicine, McMaster University, Hamilton, Ontario, Canada
5Schwabing General Hospital, and KfH Kidney Center, Munich, Germany
6Krankenhaus der Elisabethinen, Linz, Austria, and Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria
JAMA Intern Med. 2013;173(18):1682-1692. doi:10.1001/jamainternmed.2013.9051.
Text Size: A A A
Published online

Importance  Type 2 diabetes mellitus and associated chronic kidney disease (CKD) have become major public health problems. Little is known about the influence of diet on the incidence or progression of CKD among individuals with type 2 diabetes.

Objective  To examine the association between (healthy) diet, alcohol, protein, and sodium intake, and incidence or progression of CKD among individuals with type 2 diabetes.

Design, Setting, and Participants  All 6213 individuals with type 2 diabetes without macroalbuminuria from the Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial (ONTARGET) were included in this observational study. Recruitment spanned from January 2002 to July 2003, with prospective follow-up through January 2008.

Main Outcomes and Measures  Chronic kidney disease was defined as new microalbuminuria or macroalbuminuria or glomerular filtration rate decline of more than 5% per year at 5.5 years of follow-up. We assessed diet using the modified Alternate Healthy Eating Index (mAHEI). The analyses were adjusted for known risk factors, and competing risk of death was considered.

Results  After 5.5 years of follow-up, 31.7% of participants had developed CKD and 8.3% had died. Compared with participants in the least healthy tertile of mAHEI score, participants in the healthiest tertile had a lower risk of CKD (adjusted odds ratio [OR], 0.74; 95% CI, 0.64-0.84) and lower risk of mortality (OR, 0.61; 95% CI, 0.48-0.78). Participants consuming more than 3 servings of fruits per week had a lower risk of CKD compared with participants consuming these food items less frequently. Participants in the lowest tertile of total and animal protein intake had an increased risk of CKD compared with participants in the highest tertile (total protein OR, 1.16; 95% CI, 1.05-1.30). Sodium intake was not associated with CKD. Moderate alcohol intake reduced the risk of CKD (OR, 0.75; 95% CI, 0.65-0.87) and mortality (OR, 0.69; 95% CI, 0.53-0.89).

Conclusions and Relevance  A healthy diet and moderate intake of alcohol may decrease the incidence or progression of CKD among individuals with type 2 diabetes. Sodium intake, within a wide range, and normal protein intake are not associated with CKD.

Trial Registration  clinicaltrials.gov Identifier: NCT00153101

Figures in this Article

In industrialized countries, as life expectancy increases and populations increase in age, type 2 diabetes mellitus and associated chronic kidney disease (CKD) have become major public health problems. Glycemic control, antihypertensive therapy, and inhibition of the renin-angiotensin system are known to affect established diabetic renal disease. However, little is known about the long-term effect of diet on the incidence and progression of early-stage diabetic CKD. The slow progression of CKD, the competing risk of death among individuals with diabetes and CKD, and the difficulty in assessing dietary intake contribute to this lack of information.

In nephrology, restrictions of dietary protein and salt have received the most attention. In the largest randomized trial, the Modification of Diet in Renal Disease,1 low vs normal protein intake did not change the progression of CKD in individuals with nondiabetic renal disease. A meta-analysis2 suggested otherwise but concluded that the optimal amount of protein intake remains unclear.

High sodium intake may contribute to arterial hypertension and thereby renal and cardiovascular disease.3 A Cochrane review of short-term studies4 reported no effect of sodium intake on the rate of decline of glomerular filtration rate (GFR) or progression of proteinuria in individuals with advanced diabetic CKD.

Because of the lack of knowledge about the effect of dietary factors on incidence or progression of early-stage diabetic renal disease, we analyzed the association of overall diet quality and specific nutrients of interest with renal outcomes and the competing risk of death among individuals with diabetes at high vascular risk in the Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial (ONTARGET).

We hypothesized that a healthy diet, low protein and sodium intake, as well as moderate alcohol intake would be associated with a lower incidence or slower progression of early CKD and possibly death. We focused on those 4 important variables because they are potentially modifiable.

Design and Participants

The ONTARGET trial5 included 25 620 participants recruited at 733 centers in 40 countries; these participants were aged 55 years or older, had vascular disease or type 2 diabetes mellitus with end-organ damage, and thus were also at high risk for kidney disease. The composite primary outcome was cardiovascular mortality, myocardial infarction, stroke, or hospitalization for heart failure. Secondary outcomes included doubling of the serum creatinine level, the need for dialysis, and changes in urine albumin. The main finding was that combined renin-angiotensin system blockade led to more adverse events without additional benefit and that either drug alone was equally effective in preventing the combined primary outcome.6,7

To calculate the urinary albumin-creatinine ratio (UACR) (milligrams per millimole), urinary albumin and creatinine were measured centrally in a core laboratory at baseline and after 5 years of follow-up.8 Serum creatinine was measured locally at study sites at baseline and after 5.5 years. Estimated GFR was calculated using the 4-variable Modification of Diet in Renal Disease formula and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation for comparison.9,10

In this study we included only participants with a history of type 2 diabetes, normoalbuminuria or microalbuminuria at baseline, and UACR and GFR measurements at study entry and end, resulting in a study population of 6213.

Assessment of Risk Factors

We recorded food intake once at baseline using a qualitative food frequency questionnaire that has been found to be applicable to different countries (Supplement [eTable 1]).11,12 We classified the 20 food items based on their nutrient similarities, such as high-carbohydrate foods (consisting of desserts, sugar, and refined grains) or vegetables (consisting of leafy green, raw, and cooked vegetables). Frequencies of consumption were converted to servings per week. To estimate daily protein intake, we assigned a medium serving size to each food item and defined animal proteins as the sum of protein provided by meat/poultry, fish, eggs, and dairy products and plant protein as protein provided by tofu/soybean curd, legumes, and whole and refined grains. The conversion was based on the US Department of Agriculture Nutrient database (Supplement [eTable 2]).13

We assessed diet quality using the modified Alternate Healthy Eating Index (mAHEI) originally developed by McCullough and colleagues14 and applied by Dehghan and colleagues.11 Instead of portion sizes, we used frequency of consumption. The scoring system and cutoff points for food items were identical to those of the mAHEI described by Dehghan and colleagues but included multivitamins as used by McCullough and colleagues. Each participant’s points from the assessment of food items were summed to a total score. Higher scores indicated adherence to dietary guidelines and reflected high intake of healthy foods, such as vegetables, and lower intake of unhealthy foods, such as fried food.

Alcohol intake was measured by the number of drinks per week, with 1 drink equaling 1.5 ounces of hard liquor or 1 glass of beer or wine. Measurements of 24-hour urinary excretion of potassium and sodium were estimated from a fasting morning urine sample as previously described.15,16

Main Outcome

The main study outcome was defined by the 3 possible states of a participant’s status after 5.5 years of follow-up: (1) alive with no incidence or progression of CKD, (2) alive with incidence or progression of CKD, or (3) death. Incidence or progression of CKD was defined as at least 1 of the following renal events: new microalbuminuria, new macroalbuminuria, GFR-decline of more than 5% per year, or end-stage renal disease. New microalbuminuria was defined as progression of UACR to above 3.4 mg/mmol and an increase of at least 30% from UACR at baseline (UACRb). To avoid random categorical changes of participants with a UACRb near the next cutoff point, the 30%-increase rule was included. New macroalbuminuria was defined as progression of UACR above 33.9 mg/mmol and an increase of at least 30% from UACRb. Without the 30%-increase criterion, the number of renal outcomes would have increased by only 15 (Supplement [eTable 3]). Glomerular filtration rate decline was measured using GFR measurements at baseline, 2 years, and 5.5 years. End-stage renal disease was defined as a GFR less than 15 mL/min/1.73 m2 or renal replacement therapy for more than 2 months. In an additional analysis, the association of overall diet quality and specific nutrients was estimated separately for the albuminuria and GFR-decline renal events. The competing nature of the outcome states was addressed by sensitivity analysis using only 2 years of follow-up; no deaths had occurred by this time point.

Statistical Analysis

Median, first, and third quartiles (interquartile range [IQR]) were used to summarize continuous variables; absolute frequencies and percentages were used to summarize categorical variables. Wilcoxon and χ2 tests were applied to compare continuous and categorical variables with normoalbuminuria or microalbuminuria at baseline. Urinary albumin-creatinine ratio and duration of type 2 diabetes were not normally distributed, so values were log transformed. Given the exploratory nature of this observational study, we did not adjust P values for multiple testing. Multinomial logit regression models were applied to identify odds ratios (ORs) and 95% CIs for the association of diet with the 3 outcome states. Fractional polynomials were used to model nonlinear relationships (P = .157). In the final models, interactions between variables were tested by including pairwise product interaction terms while controlling for a false discovery rate of 5%. Three sets of confounders were defined based on expert opinion. The set of known confounders at baseline included age, duration of type 2 diabetes, status of albuminuria, sex, ONTARGET randomization arms, GFR, and Δ-UACR to progression (dUACRtp), which was defined as the difference between the participant-specific cutoff point for diagnosis of new microalbuminuria or macroalbuminuria and UACRb on the log scale. The extended set 1 of confounders additionally included body mass index (calculated as weight in kilograms divided by height in meters squared), mean arterial blood pressure (millimeters of mercury), glucose (milligrams per deciliter) and previous use of angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers. These additional variables may be mediators for CKD rather than confounders.17 The extended set 2 additionally included tobacco use, physical activity, and educational level to study the nutrition-specific effects of a healthier lifestyle and socioeconomic status. Correlations between explanatory variables were low, with the largest absolute correlation of 0.34 between sodium and potassium excretion. For each diet variable, we created models adjusted with all sets of confounders (single-variable models). On the basis of expert opinion, a multivariable diet model with nonoverlapping diet variables adjusted with either set of confounders was devised (multivariable model). The mAHEI is an aggregate of various dimensions of diet and was therefore not included in the adjusted multivariable model. In 2 separate additional analyses, we evaluated associations of diet with the outcome states considering only albuminuria or GFR-decline renal events. With multinomial logit regression, 2 ORs were estimated for each food item: ORrenal compares participants who were alive at study’s end and experienced an incidence or progression of CKD with participants who were alive at study’s end and did not develop or have progression of CKD; ORdeath compares participants who died during follow-up with participants who were alive and did not develop CKD. For diet variables, such as fried food, that showed heavy clustering of zero values and a small range, we used binary indicators based on consumption of these foods at least once per week or less. For continuous variables, we state 2 ORs: OR2vs1 and OR3vs1, comparing participants at the median of the second and third tertiles (50.0th and 83.3rd percentile) with participants at the median of the first tertile (16.7th percentile), respectively. All available cases were used for each statistical model. A 2-sided P < .05 was considered statistically significant. We used R, version 2.12.2, for analysis and figure preparation. Multinomial logit regression was computed with the mlogit library.18 A more detailed description of statistical analysis is contained in the Supplement.

Participants’ characteristics and dietary details are reported in the Table and the Supplement (eTable 4). After 5.5 years of follow-up, 1971 participants (31.7%) experienced the combined renal end point of incidence or progression of CKD, and 516 participants (8.3%) died (Supplement [eTable 5]); 979 participants (15.8%) developed new microalbuminuria (678 [10.9%]) or macroalbuminuria (301 [4.8%]), 1270 (20.4%) experienced a more than 5% decline of GFR per year, and 33 (0.5%) developed end-stage renal disease (Figure 1). Two hundred ninety-two participants (4.7%) experienced the albuminuria and the GFR-decline event. Adjusted single-variable and multivariable models for the combined renal outcome adjusted with known confounders are presented below. Parameter estimates of the 2 separated renal outcomes, albuminuria and GFR-decline, were only marginally different from the combined renal outcome (Supplement [eTables 14 and 17]).

Table Graphic Jump LocationTable.  Clinical and Nutrition Characteristics of Participants With Diabetes, Separated by the 3 Outcome States at 5.5 Years of Follow-upa
Place holder to copy figure label and caption
Figure 1.
Flowchart of Number of Participants and Outcomes at 5.5 Years of Follow-up

GFR indicates glomerular filtration rate; ONTARGET, Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial; and UACR, urinary albumin-creatinine ratio.

Graphic Jump Location

Scores for the mAHEI, a measure of healthy diet, varied from 9.8 to 66.2 (unhealthy to healthy), with a median of 24.6. We found an almost linear association between the mAHEI and CKD in the adjusted single-variable model (Figure 2); participants with a higher mAHEI had significantly lower renal risk (ORrenal2vs1, 0.88 [95% CI, 0.82-0.94], ORrenal3vs1, 0.74 [95% CI, 0.64-0.84]) (Figure 3 and the Supplement [eTable 7]). Risk of death was 39% lower in the third tertile of mAHEI than in the first tertile (ORdeath3vs1, 0.61 [95% CI, 0.48-0.78]). The decreased risk of CKD for participants with higher mAHEI values was also detected when albuminuria and GFR-decline were analyzed separately. Because participants with a healthier diet lived longer, they had a longer period during which to develop the renal outcome, but it occurred less frequently.

Place holder to copy figure label and caption
Figure 2.
Single-Variable Multinomial Logit Models Adjusted With Known Confounders

Confounders (at study entry) are age, duration of type 2 diabetes mellitus, albuminuria status, glomerular filtration rate, sex, Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial randomization arms, and urinary-albumin-creatinine ratio (UACR) to progression, which was defined as the difference between the participant-specific cutoff point of developing new microalbuminuria or macroalbuminuria and UACR at baseline on the log scale. Association of modified Alternate Healthy Eating Index (mAHEI) and 24-hour urinary sodium and relative odds (solid line) with 95% CI (shaded area) with chronic kidney disease (CKD) (A) or death (B) and respective histograms. The horizontal line on the top shows tertiles, and the numbers within each tertile give the percentage of participants experiencing the respective event.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 3.
Forest Plot for Single-Variable Multinomial Logit Models Adjusted With Known Confounders

For confounders, see the legend to Figure 2. If not stated otherwise, food items are given in servings per week. Renal Outcome column gives odds ratios (ORs) comparing participants alive with chronic kidney disease (CKD) with participants alive but without CKD; Death column reports ORs comparing participants who died during follow-up with participants alive without CKD. For continuous independent variables, the ORs for the median of the second tertile (50.0th percentile [solid circle]) and the median of the third tertile (83.3rd percentile [open circle]) compared with the median of the first tertile (16.7th percentile) as reference are given. Independent variables in bold type have a significant association with CKD. The last column states the P value of inclusion of the respective variable. mAHEI indicates modified Alternate Healthy Eating Index.

Graphic Jump Location

Participants consumed a median of animal, plant, and total protein of 0.47 g/kg/d (IQR, 0.32-0.68), 0.10 g/kg/d (IQR, 0.05-0.16), and 0.58 g/kg/d (IQR, 0.42-0.82), respectively. A weak positive association was observed between the frequency of animal and plant protein consumption (Pearson correlation, 0.22). In adjusted single-variable models, consumption of total and animal protein showed a significantly decreased risk of CKD (total protein: ORrenal2vs1, 0.94 [95% CI, 0.91-0.98]; ORrenal3vs1, 0.86 [95% CI, 0.77-0.95]) (Figure 3). In the adjusted multivariable model, the negative association of higher animal protein intake and CKD was reproduced (ORrenal2vs1, 0.95 [95% CI, 0.91-0.98]; ORrenal3vs1, 0.86 [95% CI, 0.77-0.96]) (Figure 4, Figure 5, and Figure 6 and the Supplement [eTable 8]). In the separate adjusted albuminuria- and GFR-decline multivariable models, higher animal protein intake showed a nonsignificant decreased risk of CKD (Supplement [eTable 14 and eTable 17]). For plant protein consumption, no significant association with CKD was found in adjusted single-variable and multivariable models.

Place holder to copy figure label and caption
Figure 4.
Multivariable Multinomial Logit Model Adjusted With Known Confounders

Association of animal protein, fruits and fruit juices, and leafy green vegetables and relative odds with 95% CI with chronic kidney disease (CKD) (A) or death (B) and respective histograms. For confounders, see the legend to Figure 2. The horizontal line on top shows tertiles, and the numbers within each tertile give the percentage of participants experiencing the respective event. The remaining independent variables of the adjusted multivariable model are depicted in the Supplement (eFigure 3).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 5.
Multivariable Multinomial Logit Model Adjusted With Known Confounders

Association of alcohol, 24-hour urinary sodium, and 24-hour urinary potassium and relative odds with 95% CI with CKD (A) or death (B) and respective histograms. For confounders, see the legend to Figure 2. The horizontal line on top shows tertiles, and the numbers within each tertile give the percentage of participants experiencing the respective event. The remaining independent variables of the adjusted multivariable model are depicted in the Supplement (eFigure 3).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 6.
Forest Plot for the Multivariable Multinomial Logit Model Adjusted With Known Confounders

For confounders, see legend to Figure 2. If not stated otherwise, food items are given in servings per week. Renal outcome column gives odds ratios (ORs) comparing participants alive with chronic kidney disease (CKD) with participants alive but without CKD; column death reports ORs comparing participants who died during follow-up with participants alive without CKD. For continuous independent variables, the ORs for the median of the second tertile (50.0th percentile [solid circle]) and the median of the third tertile (83.3rd percentile [open circle]) compared with the median of the first tertile (16.7th percentile) as reference are given. Independent variables in bold type have a significant association with CKD. The last column states the P value of inclusion of the respective variable.

Graphic Jump Location

Consuming more high-carbohydrate foods increased the risk of CKD in adjusted single-variable and multivariable models (multivariable model: ORrenal2vs1, 1.03 [95% CI, 1.01-1.06]; ORrenal3vs1, 1.15 [95% CI, 1.01-1.32]). Although the association of high-carbohydrate foods and GFR-decline–type CKD were virtually identical, the albuminuria-type CKD showed a similar, but not significant, association with high-carbohydrate foods.

Participants consumed a median of 9 (IQR, 7-14) servings of fruits and fruit juices per week and a median of 11 (IQR, 7-17) servings of vegetables per week. In adjusted single-variable and multivariable models, frequent consumption of fruits and fruit juices reduced the risk of CKD (multivariable model: ORrenal2vs1, 0.95 [95% CI, 0.91-0.99]; ORrenal3vs1, 0.91 [95% CI, 0.84-0.99]) (Figure 4) and the risk of death (ORdeath2vs1, 0.90 [95% CI, 0.84-0.96]; ORdeath3vs1, 0.82 [95% CI, 0.72-0.94]) in a dose-dependent manner. For GFR-decline–type CKD, the effect of intake of fruits and fruit juices was similar but not significant. In the adjusted single-variable model, vegetable consumption of more than 5 servings per week reduced the risk of CKD and mortality (ORrenal2vs1, 0.94 [95% CI, 0.90-0.99]; ORrenal3vs1, 0.90 [95% CI, 0.82-0.98]; ORdeath2vs1, 0.88 [95% CI, 0.82-0.95]; and ORdeath3vs1, 0.80 [95% CI, 0.69-0.91]). The significant association with mortality remained in the multivariable models. Vegetables were further classified into leafy green, raw, and cooked varieties, which were weakly correlated with each other (maximal Pearson correlation, 0.25). Only high intake of leafy green vegetables reduced the risk of CKD in adjusted single-variable models (ORrenal2vs1, 0.93 [95% CI, 0.88-0.98]; ORrenal3vs1, 0.90 [95% CI, 0.83-0.98]).

Alcohol was consumed by 32.6% of participants (n = 2024), who had a median intake of 5 (IQR, 2-8) alcoholic drinks per week. The association between alcohol intake and CKD was J-shaped (Figure 5). In adjusted single-variable and multivariable models, participants with alcohol intake of 5 drinks per week had lower risk of CKD (multivariable model: ORrenal3vs1, 0.75 [95% CI, 0.65-0.87]) and mortality (ORdeath3vs1, 0.69 [95% CI, 0.53-0.89]) compared with nondrinkers.

Median estimated measurements of 24-hour potassium and sodium urinary excretion were 2.13 g (IQR, 1.82-2.53) and 4.89 g (IQR, 3.89-5.95), respectively). Higher potassium was associated with reduced risk of CKD in adjusted single-variable and multivariable models (multivariable model: ORrenal2vs1, 0.90 [95% CI, 0.85-0.95]; ORrenal3vs1, 0.78 [95% CI, 0.69-0.88]); sodium was not associated with risk of CKD. Sodium excretion had a U-shaped association with mortality. Estimates of sodium and potassium remained substantially unchanged for the 2 separate renal events.

Albuminuria at baseline was independently associated with CKD (ORrenal, 1.16 [95% CI, 1.01-1.34]) (Supplement [eFigure 2 and eTable 20]). No effect modification with mAHEI and nutrients was found. Likewise, no effect modification was observed between renin-angiotensin–blocking medication and nutrients. When adjusted with the extended sets of confounders or when the use of diuretics was included, estimates were not materially different from the above when single-variable and multivariable models were adjusted with known confounders (Supplement [eTables 9-12]). The adjusted multivariable model using only 2 years’ follow-up (n = 5847) showed estimates for CKD similar to the 5-year model, indicating that our estimates were not sensitive to the competing risk of mortality (Supplement [eTable 13 and eFigure 4]). Results for individual food items are provided in the Supplement (eTable 7).

In this large international study, we found that a healthier diet was associated with lower incidence or slower progression of early CKD and lower mortality among individuals with type 2 diabetes. In fact, the healthier the diet, assessed by the mAHEI score, the lower was the risk of incidence or progression of CKD. A diet rich in fruits and vegetables and moderate consumption of alcohol reduced the risk of incidence or progression of CKD. However, high protein and sodium intake, often cited as modifiable risk factors for causing renal disease, were not associated with early-stage CKD in this high-risk population. In fact, low intake of total and animal proteins exhibited an increased risk of CKD and mortality. We confirm the U-shaped association of urinary sodium excretion and mortality in individuals with type 2 diabetes mellitus.15

Consistent with our findings, the Nurses’ Health Study19 reported that a healthier (Dietary Approach to Stop Hypertension [DASH]) diet was associated with less albuminuria and loss of GFR compared with a Western-style diet. However, of the 3121 individuals analyzed in the Nurses’ Health Study, only 23% had type 2 diabetes, and death was not considered as a competing risk. Both the mAHEI and DASH diet measure high-quality diet and agree in the classification of individuals with similar underlying dietary patterns. De Koning and colleagues20 reported a high correlation between AHEI and DASH indices, and both strongly predicted a lower risk of new-onset type 2 diabetes among men from the Health Professionals Follow-up study. We speculate that a combination of effects of a healthy diet may slow the complex process of atherosclerosis. In other studies,2124 a healthy diet improved endothelial function and reduced inflammation, although it has been difficult to pinpoint specific food items. These endothelial effects may especially be important in individuals with increased cardiovascular and renal risk, such as those who have type 2 diabetes.25,26

The negative association of a low protein diet and the incidence or progression of CKD in our study was unexpected because a meta-analysis2 of interventional studies on protein restriction in advanced CKD suggested that a low-protein diet reduced the hazard of renal death by 32% compared with higher or unrestricted protein intake. However, in this meta-analysis, studies on late-stage CKD were analyzed. Similar to our findings, results from the Prevention of Renal and Vascular End-Stage Disease study27 did not show an inverse association of low protein intake and GFR-decline in the general population, and low protein intake was associated with an increased mortality. In addition, a recent study28 in obese individuals confirmed that high-protein diet was not associated with low GFR or albuminuria.

Urinary sodium excretion was not associated with CKD, even after controlling for the effects of sodium on blood pressure. A meta-analysis4 of mainly short-term interventional studies in patients with diabetic nephropathy showed that dietary sodium reduction to less than 2 g/d lowered blood pressure but did not affect kidney disease. Vegter and colleagues29 analyzed GFR-decline in 500 participants in the Ramipril Efficacy in Nephropathy (REIN) trials who had advanced nondiabetic CKD and reported, for univariate analysis, an almost linear relationship between sodium excretion and end-stage renal disease. However, after controlling for baseline proteinuria, this relationship lost significance.

In contrast to our findings for CKD, mortality showed a U-shaped association with sodium intake among people with type 2 diabetes, confirming data for the whole population of the ONTARGET trial15 and a report by Ekinci and colleagues30 for individuals with type 2 diabetes. The relative hazard of sodium for mortality was independent of blood pressure and cardiovascular comorbidities. In the present analysis, estimated sodium excretions less than 3 g/d and more than 7 g/d were both associated with increased mortality.

We detected a protective effect of fruits and fruit juices for individuals who consumed more than 3 servings per week on both CKD and mortality. In individuals with established early CKD due to hypertension, a short-term dietary intervention rich in fruits and vegetables reduced albuminuria in stage 2 but not stage 1 CKD.31 The mechanism of benefit may be improved endothelial function of the microvasculature.32 In addition, we showed that leafy green vegetables, but not raw or cooked vegetables, were associated with decreased incidence or progression of CKD. A likely explanation may be the high potassium content of these food items; potassium content has been shown previously to decrease the risk of cardiovascular disease.33 Accordingly, the consumption of more vegetable servings per week was associated with a decreased risk of mortality.

It is known that moderate and excessive alcohol consumption are associated with reduced mortality and a tendency toward increased mortality, respectively.34 Our data support such a U-shaped relationship for incidence or progression of CKD in individuals with type 2 diabetes. Moderate alcohol consumption may serve as a proxy for social integration and overall subjective well-being but may also act through direct vascular effects.35

This study encompasses all limitations of an observational investigation. We addressed this issue by careful selection of confounders. The study has high internal validity because all risk factors were collected at baseline and preceded outcomes. External validity applies to a population with similar demographics. Because albuminuria was measured only 3 times (baseline, 2 years, and 5 years), biological variability may have led to misclassification. However, technical variability was minimized by central analysis. Accordingly, spot estimates of urinary electrolytes may not precisely reflect 24-hour consumption, and a single dietary assessment may not be representative. We cannot exclude some underreporting in protein intake, but the reported amount has been shown to be representative for this population.36

In summary, our findings indicate that a healthy diet is associated with a reduced risk of developing CKD and slower progression of early kidney disease among individuals with type 2 diabetes mellitus. If the associations identified would be causal, then for each 1000 individuals with type 2 diabetes and vascular disease adhering to a healthy diet, 131 would be expected to experience incidence or progression of CKD within the next 5 years compared with 151 individuals on an unhealthy diet. We also found a clear association of healthy diet with decreased mortality. Neither a low protein nor a low sodium diet, the 2 main nutritional recommendations in individuals with CKD, reduced the incidence and progression of the disease. Although this finding was not causally shown, it may be legitimate to advise individuals with type 2 diabetes and vascular disease to adhere to a healthy diet avoiding extremes of protein and salt intake to reduce their high risk for renal disease and death. This parallels the observations with cardiovascular disease; therefore, the same approach to diet is appropriate to prevent cardiovascular disease and renal disease, further confirming the parallels in the etiology of both conditions.11 This recommendation is based on the above associations, the extensive adjustments made for confounding, and the expectation that randomized trials of dietary interventions are unlikely to be forthcoming.

Accepted for Publication: May 28, 2013.

Corresponding Author: Rainer Oberbauer, MD, Department of Internal Medicine III, Room 6J, Medical University of Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria (Rainer.Oberbauer@meduniwien.ac.at).

Published Online: August 12, 2013. doi:10.1001/jamainternmed.2013.9051.

Author Contributions: Drs Dunkler, Teo, Yusuf, and Oberbauer and Ms Gao had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Dunkler, Dehghan, Heinze, Clase, Mann, Yusuf, Oberbauer.

Acquisition of data: Dehghan, Teo, Mann.

Analysis and interpretation of data: Dunkler, Dehghan, Heinze, Gao, Kohl, Clase, Mann, Yusuf, Oberbauer.

Drafting of the manuscript: Dunkler, Dehghan, Gao, Oberbauer.

Critical revision of the manuscript for important intellectual content: Dunkler, Dehghan, Teo, Heinze, Kohl, Clase, Mann, Yusuf, Oberbauer.

Statistical analysis: Dunkler, Dehghan, Heinze, Gao, Kohl, Oberbauer.

Obtained funding: Mann, Yusuf, Oberbauer.

Administrative, technical, and material support: Teo, Mann.

Study supervision: Heinze, Mann, Oberbauer.

Conflict of Interest Disclosures: None reported.

Funding/Support: The ONTARGET trial was sponsored by Boehringer-Ingelheim. The observational study presented herein was funded by SysKid, a collaborative FP7 research project to fight CKD (grant HEALTH-F2-2009-241544).

Role of the Sponsors: The study sponsor had no role in study design, analysis, or interpretation of data.

Klahr  S, Levey  AS, Beck  GJ,  et al; Modification of Diet in Renal Disease Study Group.  The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. N Engl J Med. 1994;330(13):877-884.
PubMed   |  Link to Article
Fouque  D, Laville  M.  Low protein diets for chronic kidney disease in non diabetic adults. Cochrane Database Syst Rev. 2009;(3):CD001892.
PubMed
He  FJ, MacGregor  GA.  A comprehensive review on salt and health and current experience of worldwide salt reduction programmes. J Hum Hypertens. 2009;23(6):363-384.
PubMed   |  Link to Article
Suckling  RJ, He  FJ, Macgregor  GA.  Altered dietary salt intake for preventing and treating diabetic kidney disease. Cochrane Database Syst Rev. 2010;(12):CD006763.
PubMed
Teo  KK, Yusuf  S, Sleight  P,  et al; ONTARGET/TRANSCEND Investigators.  Rationale, design, and baseline characteristics of 2 large, simple, randomized trials evaluating telmisartan, ramipril, and their combination in high-risk patients: the Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial/Telmisartan Randomized Assessment Study in ACE Intolerant Subjects with Cardiovascular Disease (ONTARGET/TRANSCEND) trials. Am Heart J. 2004;148(1):52-61.
PubMed   |  Link to Article
Yusuf  S, Teo  KK, Pogue  J,  et al; ONTARGET Investigators.  Telmisartan, ramipril, or both in patients at high risk for vascular events. N Engl J Med. 2008;358(15):1547-1559.
PubMed   |  Link to Article
Mann  JFE, Schmieder  RE, McQueen  M,  et al; ONTARGET Investigators.  Renal outcomes with telmisartan, ramipril, or both, in people at high vascular risk (the ONTARGET study): a multicentre, randomised, double-blind, controlled trial. Lancet. 2008;372(9638):547-553.
PubMed   |  Link to Article
Schmieder  RE, Mann  JFE, Schumacher  H,  et al; ONTARGET Investigators.  Changes in albuminuria predict mortality and morbidity in patients with vascular disease. J Am Soc Nephrol. 2011;22(7):1353-1364.
PubMed   |  Link to Article
Levey  AS, Coresh  J, Balk  E,  et al; National Kidney Foundation.  National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med. 2003;139(2):137-147.
PubMed   |  Link to Article
Levey  AS, Stevens  LA, Schmid  CH,  et al; CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration).  A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-612.
PubMed   |  Link to Article
Dehghan  M, Mente  A, Teo  KK,  et al; Ongoing Telmisartan Alone and in Combination With Ramipril Global End Point Trial (ONTARGET)/Telmisartan Randomized Assessment Study in ACEI Intolerant Subjects With Cardiovascular Disease (TRANSCEND) Trial Investigators.  Relationship between healthy diet and risk of cardiovascular disease among patients on drug therapies for secondary prevention: a prospective cohort study of 31 546 high-risk individuals from 40 countries. Circulation. 2012;126(23):2705-2712.
PubMed   |  Link to Article
Iqbal  R, Anand  S, Ounpuu  S,  et al; INTERHEART Study Investigators.  Dietary patterns and the risk of acute myocardial infarction in 52 countries: results of the INTERHEART study. Circulation. 2008;118(19):1929-1937.
PubMed   |  Link to Article
US Department of Agriculture, Agricultural Research Service, 2005. USDA national nutrient database for standard reference (release 18). http://www.ars.usda.gov/Services/docs.htm?docid=13747. Accessed June 24, 2013.
McCullough  ML, Feskanich  D, Stampfer  MJ,  et al.  Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance. Am J Clin Nutr. 2002;76(6):1261-1271.
PubMed
O’Donnell  MJ, Yusuf  S, Mente  A,  et al.  Urinary sodium and potassium excretion and risk of cardiovascular events. JAMA. 2011;306(20):2229-2238.
PubMed   |  Link to Article
Kawasaki  T, Itoh  K, Uezono  K, Sasaki  H.  A simple method for estimating 24 h urinary sodium and potassium excretion from second morning voiding urine specimen in adults. Clin Exp Pharmacol Physiol. 1993;20(1):7-14.
PubMed   |  Link to Article
Babyak  MA.  Understanding confounding and mediation. Evid Based Ment Health. 2009;12(3):68-71.
PubMed   |  Link to Article
Croissant Y. Estimation of multinomial logit models in R: the mlogit packages. R package version 0.2-1. 2011. http://cran.r-project.org/. Accessed March 18, 2013.
Lin  J, Fung  TT, Hu  FB, Curhan  GC.  Association of dietary patterns with albuminuria and kidney function decline in older white women: a subgroup analysis from the Nurses’ Health Study. Am J Kidney Dis. 2011;57(2):245-254.
PubMed   |  Link to Article
de Koning  L, Chiuve  SE, Fung  TT, Willett  WC, Rimm  EB, Hu  FB.  Diet-quality scores and the risk of type 2 diabetes in men. Diabetes Care. 2011;34(5):1150-1156.
PubMed   |  Link to Article
Esposito  K, Marfella  R, Ciotola  M,  et al.  Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA. 2004;292(12):1440-1446.
PubMed   |  Link to Article
Yusuf  S, Dagenais  G, Pogue  J, Bosch  J, Sleight  P; The Heart Outcomes Prevention Evaluation Study Investigators.  Vitamin E supplementation and cardiovascular events in high-risk patients. N Engl J Med. 2000;342(3):154-160.
PubMed   |  Link to Article
Lonn  E, Yusuf  S, Arnold  MJ,  et al; Heart Outcomes Prevention Evaluation (HOPE) 2 Investigators.  Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med. 2006;354(15):1567-1577.
PubMed   |  Link to Article
Bosch  J, Gerstein  HC, Dagenais  GR,  et al; ORIGIN Trial Investigators.  n-3 Fatty acids and cardiovascular outcomes in patients with dysglycemia. N Engl J Med. 2012;367(4):309-318.
PubMed   |  Link to Article
Dandona  P, Aljada  A, Chaudhuri  A, Mohanty  P, Garg  R.  Metabolic syndrome: a comprehensive perspective based on interactions between obesity, diabetes, and inflammation. Circulation. 2005;111(11):1448-1454.
PubMed   |  Link to Article
Meigs  JB, Hu  FB, Rifai  N, Manson  JE.  Biomarkers of endothelial dysfunction and risk of type 2 diabetes mellitus. JAMA. 2004;291(16):1978-1986.
PubMed   |  Link to Article
Halbesma  N, Bakker  SJL, Jansen  DF,  et al; PREVEND Study Group.  High protein intake associates with cardiovascular events but not with loss of renal function. J Am Soc Nephrol. 2009;20(8):1797-1804.
PubMed   |  Link to Article
Friedman  AN, Ogden  LG, Foster  GD,  et al.  Comparative effects of low-carbohydrate high-protein versus low-fat diets on the kidney. Clin J Am Soc Nephrol. 2012;7(7):1103-1111.
PubMed   |  Link to Article
Vegter  S, Perna  A, Postma  MJ, Navis  G, Remuzzi  G, Ruggenenti  P.  Sodium intake, ACE inhibition, and progression to ESRD. J Am Soc Nephrol. 2012;23(1):165-173.
PubMed   |  Link to Article
Ekinci  EI, Clarke  S, Thomas  MC,  et al.  Dietary salt intake and mortality in patients with type 2 diabetes. Diabetes Care. 2011;34(3):703-709.
PubMed   |  Link to Article
Goraya  N, Simoni  J, Jo  C, Wesson  DE.  Dietary acid reduction with fruits and vegetables or bicarbonate attenuates kidney injury in patients with a moderately reduced glomerular filtration rate due to hypertensive nephropathy. Kidney Int. 2012;81(1):86-93.
PubMed   |  Link to Article
Buscemi  S, Rosafio  G, Arcoleo  G,  et al.  Effects of red orange juice intake on endothelial function and inflammatory markers in adult subjects with increased cardiovascular risk. Am J Clin Nutr. 2012;95(5):1089-1095.
PubMed   |  Link to Article
Cook  NR, Obarzanek  E, Cutler  JA,  et al; Trials of Hypertension Prevention Collaborative Research Group.  Joint effects of sodium and potassium intake on subsequent cardiovascular disease: the Trials of Hypertension Prevention follow-up study. Arch Intern Med. 2009;169(1):32-40.
PubMed   |  Link to Article
Klatsky  AL, Armstrong  MA, Friedman  GD.  Alcohol and mortality. Ann Intern Med. 1992;117(8):646-654.
PubMed   |  Link to Article
Collins  MA, Neafsey  EJ, Mukamal  KJ,  et al.  Alcohol in moderation, cardioprotection, and neuroprotection: epidemiological considerations and mechanistic studies. Alcohol Clin Exp Res. 2009;33(2):206-219.
PubMed   |  Link to Article
Campbell  WW, Johnson  CA, McCabe  GP, Carnell  NS.  Dietary protein requirements of younger and older adults. Am J Clin Nutr. 2008;88(5):1322-1329.
PubMed

Figures

Place holder to copy figure label and caption
Figure 1.
Flowchart of Number of Participants and Outcomes at 5.5 Years of Follow-up

GFR indicates glomerular filtration rate; ONTARGET, Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial; and UACR, urinary albumin-creatinine ratio.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.
Single-Variable Multinomial Logit Models Adjusted With Known Confounders

Confounders (at study entry) are age, duration of type 2 diabetes mellitus, albuminuria status, glomerular filtration rate, sex, Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial randomization arms, and urinary-albumin-creatinine ratio (UACR) to progression, which was defined as the difference between the participant-specific cutoff point of developing new microalbuminuria or macroalbuminuria and UACR at baseline on the log scale. Association of modified Alternate Healthy Eating Index (mAHEI) and 24-hour urinary sodium and relative odds (solid line) with 95% CI (shaded area) with chronic kidney disease (CKD) (A) or death (B) and respective histograms. The horizontal line on the top shows tertiles, and the numbers within each tertile give the percentage of participants experiencing the respective event.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 3.
Forest Plot for Single-Variable Multinomial Logit Models Adjusted With Known Confounders

For confounders, see the legend to Figure 2. If not stated otherwise, food items are given in servings per week. Renal Outcome column gives odds ratios (ORs) comparing participants alive with chronic kidney disease (CKD) with participants alive but without CKD; Death column reports ORs comparing participants who died during follow-up with participants alive without CKD. For continuous independent variables, the ORs for the median of the second tertile (50.0th percentile [solid circle]) and the median of the third tertile (83.3rd percentile [open circle]) compared with the median of the first tertile (16.7th percentile) as reference are given. Independent variables in bold type have a significant association with CKD. The last column states the P value of inclusion of the respective variable. mAHEI indicates modified Alternate Healthy Eating Index.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 4.
Multivariable Multinomial Logit Model Adjusted With Known Confounders

Association of animal protein, fruits and fruit juices, and leafy green vegetables and relative odds with 95% CI with chronic kidney disease (CKD) (A) or death (B) and respective histograms. For confounders, see the legend to Figure 2. The horizontal line on top shows tertiles, and the numbers within each tertile give the percentage of participants experiencing the respective event. The remaining independent variables of the adjusted multivariable model are depicted in the Supplement (eFigure 3).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 5.
Multivariable Multinomial Logit Model Adjusted With Known Confounders

Association of alcohol, 24-hour urinary sodium, and 24-hour urinary potassium and relative odds with 95% CI with CKD (A) or death (B) and respective histograms. For confounders, see the legend to Figure 2. The horizontal line on top shows tertiles, and the numbers within each tertile give the percentage of participants experiencing the respective event. The remaining independent variables of the adjusted multivariable model are depicted in the Supplement (eFigure 3).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 6.
Forest Plot for the Multivariable Multinomial Logit Model Adjusted With Known Confounders

For confounders, see legend to Figure 2. If not stated otherwise, food items are given in servings per week. Renal outcome column gives odds ratios (ORs) comparing participants alive with chronic kidney disease (CKD) with participants alive but without CKD; column death reports ORs comparing participants who died during follow-up with participants alive without CKD. For continuous independent variables, the ORs for the median of the second tertile (50.0th percentile [solid circle]) and the median of the third tertile (83.3rd percentile [open circle]) compared with the median of the first tertile (16.7th percentile) as reference are given. Independent variables in bold type have a significant association with CKD. The last column states the P value of inclusion of the respective variable.

Graphic Jump Location

Tables

Table Graphic Jump LocationTable.  Clinical and Nutrition Characteristics of Participants With Diabetes, Separated by the 3 Outcome States at 5.5 Years of Follow-upa

References

Klahr  S, Levey  AS, Beck  GJ,  et al; Modification of Diet in Renal Disease Study Group.  The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. N Engl J Med. 1994;330(13):877-884.
PubMed   |  Link to Article
Fouque  D, Laville  M.  Low protein diets for chronic kidney disease in non diabetic adults. Cochrane Database Syst Rev. 2009;(3):CD001892.
PubMed
He  FJ, MacGregor  GA.  A comprehensive review on salt and health and current experience of worldwide salt reduction programmes. J Hum Hypertens. 2009;23(6):363-384.
PubMed   |  Link to Article
Suckling  RJ, He  FJ, Macgregor  GA.  Altered dietary salt intake for preventing and treating diabetic kidney disease. Cochrane Database Syst Rev. 2010;(12):CD006763.
PubMed
Teo  KK, Yusuf  S, Sleight  P,  et al; ONTARGET/TRANSCEND Investigators.  Rationale, design, and baseline characteristics of 2 large, simple, randomized trials evaluating telmisartan, ramipril, and their combination in high-risk patients: the Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial/Telmisartan Randomized Assessment Study in ACE Intolerant Subjects with Cardiovascular Disease (ONTARGET/TRANSCEND) trials. Am Heart J. 2004;148(1):52-61.
PubMed   |  Link to Article
Yusuf  S, Teo  KK, Pogue  J,  et al; ONTARGET Investigators.  Telmisartan, ramipril, or both in patients at high risk for vascular events. N Engl J Med. 2008;358(15):1547-1559.
PubMed   |  Link to Article
Mann  JFE, Schmieder  RE, McQueen  M,  et al; ONTARGET Investigators.  Renal outcomes with telmisartan, ramipril, or both, in people at high vascular risk (the ONTARGET study): a multicentre, randomised, double-blind, controlled trial. Lancet. 2008;372(9638):547-553.
PubMed   |  Link to Article
Schmieder  RE, Mann  JFE, Schumacher  H,  et al; ONTARGET Investigators.  Changes in albuminuria predict mortality and morbidity in patients with vascular disease. J Am Soc Nephrol. 2011;22(7):1353-1364.
PubMed   |  Link to Article
Levey  AS, Coresh  J, Balk  E,  et al; National Kidney Foundation.  National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med. 2003;139(2):137-147.
PubMed   |  Link to Article
Levey  AS, Stevens  LA, Schmid  CH,  et al; CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration).  A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-612.
PubMed   |  Link to Article
Dehghan  M, Mente  A, Teo  KK,  et al; Ongoing Telmisartan Alone and in Combination With Ramipril Global End Point Trial (ONTARGET)/Telmisartan Randomized Assessment Study in ACEI Intolerant Subjects With Cardiovascular Disease (TRANSCEND) Trial Investigators.  Relationship between healthy diet and risk of cardiovascular disease among patients on drug therapies for secondary prevention: a prospective cohort study of 31 546 high-risk individuals from 40 countries. Circulation. 2012;126(23):2705-2712.
PubMed   |  Link to Article
Iqbal  R, Anand  S, Ounpuu  S,  et al; INTERHEART Study Investigators.  Dietary patterns and the risk of acute myocardial infarction in 52 countries: results of the INTERHEART study. Circulation. 2008;118(19):1929-1937.
PubMed   |  Link to Article
US Department of Agriculture, Agricultural Research Service, 2005. USDA national nutrient database for standard reference (release 18). http://www.ars.usda.gov/Services/docs.htm?docid=13747. Accessed June 24, 2013.
McCullough  ML, Feskanich  D, Stampfer  MJ,  et al.  Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance. Am J Clin Nutr. 2002;76(6):1261-1271.
PubMed
O’Donnell  MJ, Yusuf  S, Mente  A,  et al.  Urinary sodium and potassium excretion and risk of cardiovascular events. JAMA. 2011;306(20):2229-2238.
PubMed   |  Link to Article
Kawasaki  T, Itoh  K, Uezono  K, Sasaki  H.  A simple method for estimating 24 h urinary sodium and potassium excretion from second morning voiding urine specimen in adults. Clin Exp Pharmacol Physiol. 1993;20(1):7-14.
PubMed   |  Link to Article
Babyak  MA.  Understanding confounding and mediation. Evid Based Ment Health. 2009;12(3):68-71.
PubMed   |  Link to Article
Croissant Y. Estimation of multinomial logit models in R: the mlogit packages. R package version 0.2-1. 2011. http://cran.r-project.org/. Accessed March 18, 2013.
Lin  J, Fung  TT, Hu  FB, Curhan  GC.  Association of dietary patterns with albuminuria and kidney function decline in older white women: a subgroup analysis from the Nurses’ Health Study. Am J Kidney Dis. 2011;57(2):245-254.
PubMed   |  Link to Article
de Koning  L, Chiuve  SE, Fung  TT, Willett  WC, Rimm  EB, Hu  FB.  Diet-quality scores and the risk of type 2 diabetes in men. Diabetes Care. 2011;34(5):1150-1156.
PubMed   |  Link to Article
Esposito  K, Marfella  R, Ciotola  M,  et al.  Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA. 2004;292(12):1440-1446.
PubMed   |  Link to Article
Yusuf  S, Dagenais  G, Pogue  J, Bosch  J, Sleight  P; The Heart Outcomes Prevention Evaluation Study Investigators.  Vitamin E supplementation and cardiovascular events in high-risk patients. N Engl J Med. 2000;342(3):154-160.
PubMed   |  Link to Article
Lonn  E, Yusuf  S, Arnold  MJ,  et al; Heart Outcomes Prevention Evaluation (HOPE) 2 Investigators.  Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med. 2006;354(15):1567-1577.
PubMed   |  Link to Article
Bosch  J, Gerstein  HC, Dagenais  GR,  et al; ORIGIN Trial Investigators.  n-3 Fatty acids and cardiovascular outcomes in patients with dysglycemia. N Engl J Med. 2012;367(4):309-318.
PubMed   |  Link to Article
Dandona  P, Aljada  A, Chaudhuri  A, Mohanty  P, Garg  R.  Metabolic syndrome: a comprehensive perspective based on interactions between obesity, diabetes, and inflammation. Circulation. 2005;111(11):1448-1454.
PubMed   |  Link to Article
Meigs  JB, Hu  FB, Rifai  N, Manson  JE.  Biomarkers of endothelial dysfunction and risk of type 2 diabetes mellitus. JAMA. 2004;291(16):1978-1986.
PubMed   |  Link to Article
Halbesma  N, Bakker  SJL, Jansen  DF,  et al; PREVEND Study Group.  High protein intake associates with cardiovascular events but not with loss of renal function. J Am Soc Nephrol. 2009;20(8):1797-1804.
PubMed   |  Link to Article
Friedman  AN, Ogden  LG, Foster  GD,  et al.  Comparative effects of low-carbohydrate high-protein versus low-fat diets on the kidney. Clin J Am Soc Nephrol. 2012;7(7):1103-1111.
PubMed   |  Link to Article
Vegter  S, Perna  A, Postma  MJ, Navis  G, Remuzzi  G, Ruggenenti  P.  Sodium intake, ACE inhibition, and progression to ESRD. J Am Soc Nephrol. 2012;23(1):165-173.
PubMed   |  Link to Article
Ekinci  EI, Clarke  S, Thomas  MC,  et al.  Dietary salt intake and mortality in patients with type 2 diabetes. Diabetes Care. 2011;34(3):703-709.
PubMed   |  Link to Article
Goraya  N, Simoni  J, Jo  C, Wesson  DE.  Dietary acid reduction with fruits and vegetables or bicarbonate attenuates kidney injury in patients with a moderately reduced glomerular filtration rate due to hypertensive nephropathy. Kidney Int. 2012;81(1):86-93.
PubMed   |  Link to Article
Buscemi  S, Rosafio  G, Arcoleo  G,  et al.  Effects of red orange juice intake on endothelial function and inflammatory markers in adult subjects with increased cardiovascular risk. Am J Clin Nutr. 2012;95(5):1089-1095.
PubMed   |  Link to Article
Cook  NR, Obarzanek  E, Cutler  JA,  et al; Trials of Hypertension Prevention Collaborative Research Group.  Joint effects of sodium and potassium intake on subsequent cardiovascular disease: the Trials of Hypertension Prevention follow-up study. Arch Intern Med. 2009;169(1):32-40.
PubMed   |  Link to Article
Klatsky  AL, Armstrong  MA, Friedman  GD.  Alcohol and mortality. Ann Intern Med. 1992;117(8):646-654.
PubMed   |  Link to Article
Collins  MA, Neafsey  EJ, Mukamal  KJ,  et al.  Alcohol in moderation, cardioprotection, and neuroprotection: epidemiological considerations and mechanistic studies. Alcohol Clin Exp Res. 2009;33(2):206-219.
PubMed   |  Link to Article
Campbell  WW, Johnson  CA, McCabe  GP, Carnell  NS.  Dietary protein requirements of younger and older adults. Am J Clin Nutr. 2008;88(5):1322-1329.
PubMed

Correspondence

CME
Also Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
Please click the checkbox indicating that you have read the full article in order to submit your answers.
Your answers have been saved for later.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
Submit a Comment

Multimedia

Supplement.

eAppendix. The ONTARGET Investigators

eTable 1. Description of food items in the food frequency questionnaire

eTable 2. Assumed protein content per serving size and conversion between servings and grams based on USDA United State Department of Agriculture National Nutrient database for standard reference

eTable 3. Changes in the number of participants with new microalbuminuria or macroalbuminuria at study end when the minimum increase in UACR between baseline and 5-year follow-up measurement is changed

eTable 4. Clinical and nutrition characteristics of participants with type 2 diabetes mellitus, separated by the 3 outcome states at 5.5 years of follow-up

eTable 5. Distribution of the 3 outcome states at the 5.5 years of follow-up separated by normoalbuminuria and microalbuminuria at baseline

eTable 6. Comparison of albuminuria (UACR) and GFR-decline renal events

eTable 7. Combined renal outcome: single-variable models adjusted with known confounders

eTable 8. Combined renal outcome: multivariable model adjusted with known confounders

eTable 9. Combined renal outcome: single-variable models adjusted with the extended set of confounders 1

eTable 10. Combined renal outcome: multivariable model adjusted with the extended set of confounders 1

eTable 11. Combined renal outcome: single-variable models adjusted with the extended set of confounders 2

eTable 12. Combined renal outcome: multivariable model adjusted with the extended set of confounders 2

eTable 13. Combined renal outcome: multivariable logistic model adjusted with known confounders

eTable 14. Albuminuria outcome: single-variable model with mAHEI and multivariable model adjusted with known confounders

eTable 15. Albuminuria outcome: single-variable model with mAHEI and multivariable model adjusted with the extended set of confounders 1

eTable 16. Albuminuria outcome: single-variable model with mAHEI and multivariable model adjusted with the extended set of confounders 2

eTable 17. GFR-decline outcome: single-variable model with mAHEI and multivariable model adjusted with known confounders

eTable 18. GFR-decline outcome: single-variable model with mAHEI and multivariable model adjusted with the extended set of confounders 1

eTable 19. GFR-decline outcome: single-variable model with mAHEI and multivariable model adjusted with the extended set of confounders 2

eTable 20. Combined renal outcome: multinomial logit model including only variables from the set of known confounders

eTable 21. Combined renal outcome: multinomial logit model including only variables from the set of extended confounders 1

eTable 22. Combined renal outcome: multinomial logit model including only variables from the set of extended confounders 2

eFigure 1. Combined renal outcome: single-variable models adjusted with known confounders

eFigure 2. Combined renal outcome: single-variable model with mAHEI adjusted with known confounders, separated for participant’s albuminuria status at baseline

eFigure 3. Combined renal outcome: multivariable model adjusted with known confounders

eFigure 4. Comparison of estimates of multivariable models adjusted with known confounders after 2 and 5.5 years of follow-up

eReferences.

Supplemental Content

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 9

Related Content

Customize your page view by dragging & repositioning the boxes below.

See Also...
Articles Related By Topic
Related Collections
PubMed Articles