Dietary Intake in Hemodialysis Patients Does Not Reflect a Heart Healthy Diet

Objective: Cardiovascular disease is highly prevalent and has a major effect on morbidity and mortality in patients undergoing maintenance hemodialysis (MHD). Dietary factors that may contribute to cardiovascular disease have not been well studied in this population. We hypothesize that dietary intake in this population does not meet the guidelines for cardiovascular risk reduction. Design: A cross-sectional study was completed using the validated ‘‘Block Dialysis 1 Food Frequency Question-naire’’ to assess dietary intake of MHD patients. Setting and Patients: A total of 70 patients undergoing MHD at our outpatient dialysis center completed the questionnaire under the supervision of a trained dietitian. The population consisted of 38 men and 32 women. Main Outcome Measure: Dietary intake was the main outcome measure, with a focus on calories, soluble ﬁber, saturated fatty acid (SFA), unsaturated fatty acid intake (UFA), and protein. Results: The mean ﬁber intake was 10.77 ( 6 5.87) g/day, and only 2 of 71 (2.9%) were in compliance with the recommended daily intake of . 25 g/day. As percentage of total calories, of the 70 patients, 5 (7.1%) had a fat intake of , 30%, 22 (31.4%) had SFA intake of , 10%, 64 (91.4%) had a UFA of # 30%, 22 (31.4%) had a protein-based diet of $ 15%, and 66 (94.3%) had a carbohydrate diet of , 60%. Conclusions: Most patients did not meet the dietary guidelines for reducing the risk of cardiovascular disease. Substituting UFA or soluble ﬁber for SFA improves low density lipoprotein (LDL

risk factors, numerous other factors including, hypoalbuminemia, elevated troponin levels, anemia, hyperhomocystinemia, increased calciumphosphate product, inflammation, increased oxidant stress, and decreased nitric oxide activity, could contribute to an increased risk of coronary artery disease (CAD) in MHD patients. 8,9Dietary factors that may contribute to cardiovascular disease and increased mortality have not been well studied in this population.In this study, our primary aim was to examine the dietary intake of a group of MHD population and to ascertain whether they meet the National Kidney Foundation (NKF), Kidney Disease Outcome Quality Initiative (K/DOQI), American Heart Association (AHA) guidelines, or other relevant dietary guidelines for cardiovascular risk reduction.We hypothesized that MHD patients ingest an atherogenic diet.Our secondary aim was to explore the relationship between background history of cardiac disease and nutritional intake.We used Block's Dialysis Food Frequency Questionnaire (FFQ) version 1 (NutritionQuest Block Dialysis, Berkeley, CA), a version of Block 2005 FFQ, developed for the dialysis population.

Patients
The study protocol was reviewed and approved by the hospital institutional review committee before beginning the study.All patients gave informed consent, in writing, before enrollment in the study.
All patients undergoing MHD at the outpatient dialysis unit at Staten Island University Hospital were considered for the study.The patients varied in race, age, socioeconomic status, and cause of renal failure, although diabetes and hypertension were predominant (see later in the text).The major inclusion criteria were: age between 18 and 90 years; at least 6 months of uninterrupted thrice-weekly MHD; ability to understand, read, and write English, and to fill out the questionnaire; and the ability to ingest 3 meals a day independently.Exclusion criteria included terminal or active cancer, active systemic inflammatory disease, AIDS, inflammatory bowel disease, severe congestive heart disease (NYHA class III, or IV), severe lung disease, the use of nonoral feedings, the inability to ingest food because of medical con-ditions, a history of dementia, and patients not oriented to person, place, or time.
As part of the standard dietary intake process, all patients underwent routine dietary education which included an assessment of height, weight, body mass index (BMI), dentition, and education on the importance of appropriate nutrition for the MHD patient.A 3-day diet diary was reviewed with the dietician to highlight specific changes necessary to each individual.Before leaving, patients were provided with a handout to guide them in choosing appropriate foods.The handout included the current recommendation of the AHA and NKF for a heart healthy diet.Further counseling was triggered by an inappropriate gain or loss of dry weight, high levels of phosphorous or potassium found on monthly laboratory results, and after review of a routine 3-day diet diary conducted in all patients every 6 months.
All patients continued their standard MHD treatments, as well as treatments of all chronic diseases and issues related to MHD including, but not limited to treatment of anemia, calcium, phosphorous and parathyroid hormone (PTH) abnormalities, and iron deficiency.

Dietary Assessment Instrument
To assess dietary intake, we used the Block Dialysis FFQ.The FFQ is an extensive questionnaire designed to measure nutrient and calorie intake on the basis of the description of food and vitamins consumed by the patient.1][12][13] The questionnaire is 8 pages long and contains approximately 150 multiple choice questions regarding which foods are eaten, the frequency of eating each item, and the size of the portion generally eaten.Each variable has multiple choices to be answered, and the answers are appropriate to the food.Food is grouped together in ways that are sensible and familiar to patients to increase the ease and accuracy of the questionnaire.The questionnaire also has a section of questions designed to assess the patients' perception of overall health and well being, as well as certain demographic information.
The questionnaire has gone through several revisions to improve the accuracy of the assessment.The Block dialysis 1 FFQ was designed specifically to assess dietary intake of patients on dialysis, and there are several important differences to note.The list of food in the multiple choice questions was developed by examining 24-hour diet diaries, specifically those of patients who were on dialysis.Foods listed together in a single question were placed together to more accurately reflect intake of minerals most important to patients who are on dialysis.Portion sizes were adapted to reflect dialysis diet exchange serving sizes for protein foods, and questions were added to capture dietary intake from high-and low-protein liquid diets used in the care of patients suffering from kidney disease.A question was added to specifically address vitamins designed for dialysis patients, such as Nephro-Vite, Dialyvite, and so forth.Although the final product is in fact very similar to the original Block FFQ, the specifications allowed for a better assessment of intake most important to the dialysis population.
The questionnaire was administered by a group of medical residents trained to administer the FFQ.The subjects were given the questionnaire at the hemodialysis (HD) sessions, and were permitted to fill it out there, or complete it at home and return it at their next scheduled HD session.Any questions not addressed were reviewed with the patient by the researchers and the correct answer was filled in.In this manner, all questionnaires were filled out completely.

Study Population
A total of 70 individuals agreed to participate in the study.The studied population consisted of 38 (54%) men and 32 (56%) women.The mean age was 61.47 years (6standard deviation: 614.42 years), ranging from 19 to 87 years.This included 44 (62.9%) nonHispanic whites, 20 (28.6%) African Americans, and 6 (8.6%) Hispanics.The mean dry weight was 75 kg (6standard deviation: 622.P 5.004), and were more frequently on clopidogrel (48.8% vs. 13.8%;P 5.0025) and statins (65.9% vs. 31%; P 5.007).Although there was a trend toward more diabetes, higher BMI, and lower LDL and high density lipoprotein (HDL) in the group with cardiac disease, this did not meet statistical significance.There was no difference between the 2 groups with regard to hemoglobin, serum albumin, serum calcium, serum phosphorus, and serum ferritin (Table 1).

Statistical Methods
The primary objective of the statistical analysis was to provide descriptive statistics on dietary intake in the studied subjects; thus, the sample size used for this study was not based on any formal power calculations.
Overall mean values of demographic variables and dietary information from the FFQ were calculated, and values were also tabulated according to the presence or absence of background cardiac disease.Comparison between the 2 groups was performed using the nonparametric Mann-Whitney test.Comparisons were also performed according to various demographic characteristics of the patient (e.g., age group (. or ,60), ethnicity, presence of diabetes, etc.).
This study is purely descriptive and was designed to provide guidance to the existing trends in dietary intake and compliance with current guidelines.This will assist the investigators in designing a future study examining whether dietary counseling helps patients meet dietary recommendations.

Results
Participating MHD patients were divided into 2 groups on the basis of the existence of known cardiac disease at baseline (assessed by reviewing the record in the patients chart) or lack thereof.The rationale was that there might be an association between poor diet and cardiac disease, or possibly that those with cardiac disease might be more compliant with dietary recommendations.
Table 2 displays the dietary results of the FFQ, and includes data with and without supplementation with vitamins.The total average kilocalories were 1,626.1 6 672.84, with approximately half in the form of 201.28 6 90.79 g of carbohydrates, 15% in the form of 54.79 6 23.36 g of protein, and 40% in the form of 69.58 6 30.13 g of fat.‡Kopple JD: The National Kidney Foundation K/DOQI clinical practice guidelines for dietary protein intake for chronic dialysis patients.Am J Kidney Dis 38:S68-S73, 2001 (4 suppl 1).
The average intake of sodium was 2,502.0 6 1,094.7 mg, potassium was 1,695.1 6 701.55 mg, and phosphorus was 841.23 6 339.44 mg.The average total fiber intake was 10.77 6 5.87 g per day, and saturated fat was 19.38 6 8.34 g per day.The daily intake of calcium was less than 2 g/day.Vitamin B1, niacin, vitamin C, and selenium were within recommended range without supplementation, and exceeded the range with supplementation.Riboflavin intake exceeded recommendations even without supplementation.Folate and copper did not meet the recommendation range even with supplementation.Vitamin B6 and zinc intake did not meet the recommendation range without supplementation, and were within range with supplementation.Vitamins E and B12 were within range without supplementation, and exceeded range with supplementation.
Table 3 compares the group with known history of cardiac disease to those without, and none of the intake values are statistically different in the 2 groups.
Table 4 shows the proportion of total patients who met the AHA dietary guidelines for cardiovascular risk reduction for various dietary variables, as well as the percentage of those in each group who meet the guidelines.Once again, there was no difference between those with baseline cardiac disease as compared with those without the disease.Overall, only 7.1% met the guideline for the daily total fat intake (TFI), 31.4% for saturated fatty acid (SFA) intake, 91.4% for unsaturated fatty acid intake (UFA) intake, 31.4% for protein intake, and 48.6% for sodium intake.A maximum of 78.6% met the guideline for cholesterol intake and 94.3% for carbohydrate intake.Only 2.7% met the guideline for fiber intake, and it is notable that nobody in the baseline cardiac disease group met this guideline.

Discussion
Examining the dietary intake of MHD patients through FFQ in a community dialysis center, we found that their dietary energy intake is similar to the dietary intake described in previous studies using the FFQ. 10,14The dietary energy intake remains significantly lower than that recommended by the NKF (K/DOQI) for MHD (35 kcal/kg/day for patients who are ,60 years of age and 30 kcal/kg for patients $60 years of age; Table 2). 15Low body weights are associated with increased morbidity and mortality in HD. 15,16 Clinical factors like old age, presence of clinical CAD, and DM have been reported to predict malnutrition. 17Other important contributing factors are anorexia from uremic toxicity in underdialyzed patients, acidemia, and comorbid illnesses. 15,18The recommended dietary protein intake in MHD patients is at least 1.2 g/kg.Our data shows that MHD patients do not meet these recommendations.A similar observation was observed previously by Kalantar-Zadeh et al. using an older version of the FFQ 10 Protein-calorie malnutrition, also known as PEW, is an important factor predicting mortality in patients with end-stage renal disease on MHD. 4,19Patients undergoing MHD are advised to eat a high protein diet to avoid loss of muscle mass and hypoalbuminemia, which is associated with an increased mortality in this population. 3,4For this reason, a diet rich in protein but not in saturated fat or cholesterol is recommended. 19,20ur data indicate that the majority of patients undergoing MHD consume an atherogenic diet.There is currently no consensus on the optimal dietary lipid composition in patients undergoing MHD.Several studies support limiting the TFI to 35% and the SFA to 7% of the total calorie intake in dialysis patients (Table 2). 21,22The high prevalence of atherosclerotic disease in dialysis patients might be partly because of a diet rich in SFA (the principle dietary determinant of LDL cholesterol levels). 23However, there is no prospec-tive study proving that dietary modifications have any benefit for MHD patients.The AHA strongly recommends limiting the TFI to 30% and the SFA to 10% or even 7% of the total calorie intake in patients with high LDL levels or CAD (Table 4). 19,24nly 7% and 31% of our patients met the AHA recommendations for TFI and SFA respectively.The AHA and nutrition experts advise substituting UFA for SFA, with a 30% upper limit of UFA of the total energy intake. 19In the vast majority of our patients, UFA intake was less than that range, which makes this option feasible (Table 4).Almost all our patients met the AHA and the national dietary recommendation of carbohydrate and cholesterol intake in patients undergoing MHD (Tables 2 and 4). 19atients undergoing MHD frequently have hypertriglyceridemia and low serum HDL, which are more common than hypercholesterolemia. [25][26][27] Although reduction of LDL cholesterol levels can be achieved by substituting carbohydrates for SFA, this can further worsen triglyceride and HDL levels. 19Substituting UFA or soluble fiber for SFA improves LDL cholesterol levels without negative effects on other lipid parameters. 19This is feasible in our patients because most ingest less than the upper recommended limit for UFA.Dietary counseling can be directed to increase fiber intake by adding low potassium whole grains, fruits, and vegetables.
The mean daily dietary vitamin C intake in our studied population exceeds the minimum recommended for dialysis patients.A study performed by Kalantar et al. using the older version of the FFQ concluded that the daily intake of vitamin C is lower in MHD patients compared with control group.This was suggested as a potential atherogenic factor in this population.The older FFQ version used in that study did not have the ability to quantify the additional intake from the vitamin supplements.Adding the supplements to the dietary intake clearly shows that our patients consumed high levels of vitamin C, which might be cardioprotective and beneficial in promoting intestinal iron absorption and reducing iron deficiency anemia 28 (Table 2).Riboflavin, thiamin, and niacin serum levels are usually normal in MHD patients, although case reports have linked niacin deficiency to Wernicke's encephalopathy. 29The average total intake exceeds that recommended for this group (Table 2).Vitamin B6, which helps in the metabolism of homocysteine, is usually removed with hemodialyis; thus, supplementing this vitamin is advised in this population. 30,31With supplementation, our patients met the daily requirement of vitamin B6 (Table 2).
The levels of vitamin B12, folate, and vitamin E of our patients exceed the levels recommended by the K/DOQI clinical practice guidelines.Vitamin B12 deficiency, through rapid metabolism of homocysteine and increased cysteine levels, could constitute a risk factor for CAD. 32Although the role of vitamin E in atherogenesis has not yet been entirely studied in human beings, a small study showed its benefits as an antioxidant in animals. 33Currently, the AHA believes that there are insufficient data to justify the use of vitamin E supplements in CAD prevention. 19Vitamin D supplements are routinely used in MHD patients to prevent or treat hyperparathyroidism or improve calcium homeostasis.Animal studies showed that phosphate loading and the severity of uremia play an important role in controlling arterial medial calcification in mice. 34Despite this interesting observation, studies in human beings are currently lacking.The phosphorus intake of our patient was within the current recommendation of less than 1,000 mg/day.Serum magnesium level has also been related to noncardiovascular mortality in HD patients. 35The recommended intake is 300 to 400 mg/day and the average in our patient is only 225 mg/day.
The daily sodium intake of our patient was 2.5 g.Despite the recommendations of AHA and the K/DOQI of less than 2 g of sodium for patients with cardiovascular disease, it is still a difficult goal to achieve.Hyperkalemia in HD patients is a risk factor for arrhythmia, whereas hypokalemia can be a major contributing factor to sudden cardiac death. 36e intake of potassium by our patients is well within the range recommended by Kopple and Coburn in their studies. 37Our study showed that daily fiber intake is lower than that recommended by the K/DOQI and the AHA.Our results are similar to those reported by Kalantar-Zadech et al. 10 Besides its effect on lowering cholesterol level and reducing the risk of cancer, several studies have shown a causative relation between low dietary fiber and atherosclerosis. 38ajor efforts and education are required to stimulate MHD patients to increase their fiber intake.
Vitamin A intake (including alpha, beta carotene, and cryptoxanthin) is still a controversial topic.Although our patients' intake of vitamin A exceeded the recommendations, this may not be beneficial.A recent study showed that vitamin A acts as an oxidant and pro-oxidant. 39n addition to the traditional risk factors for CAD, nontraditional factors related to malnutrition and inflammation are increasingly being studied in this population.The syndrome, known as the MIA syndrome, is an association between malnutrition, inflammation, and accelerated atherosclerosis.Malnutrition and inflammation are commonly found in uremic patients, with a potential interplay between the 2 conditions. 40,41Moreover, markers of inflammation and malnutrition were linked to long-term mortality in few studies. 42nalysis of the dietary intake shows similarity in those with history of cardiac disease at baseline compared with those without.A study performed on peritoneal dialysis patients found few differences in dietary daily intake between the 2 groups without drawing a major conclusion from their findings. 14The relatively small sample size in our study might explain the inability to detect any difference between the 2 groups.
Our study has several limitations.It is a singlecenter study and might not reflect the dietary intake in different dialysis populations.We did not have a control group for comparison, as did several other studies.We aimed to determine whether the dietary intake in MHD patients meets the AHA, K/DOQI, and other MHD nutrition guidelines.In addition, we aim to repeat the FFQ after counseling our patients to improve their diet and determine whether this simple measure can help improve compliance with the current guidelines.Although it is a well-validated questionnaire, the FFQ can potentially underestimate intake, as MHD patients tend to underreport their daily intake.We did not validate the current version of the FFQ, for example, by performing 3-day food records or 24-hour diet recalls and comparing that with the results we obtained.The main strength of our study is that we used the most updated version of the FFQ; this was developed specifically to assess dietary intake in the dialysis population.This fact makes our data more consistent and comparable with the current dietary recommendations than previous reports.

Conclusion
CAD is present in more than 40% of patients at the onset of dialysis. 6Although dietary factors can play an important role in atherogenesis in MHD, several other conventional and nonconventional risk factors can contribute to CAD in patients undergoing MHD. 8,9The analysis of dietary intake in our patients indicates that the majority did not meet the AHA, N/DOQI, and other MHD dietary guidelines for reducing the risk of cardiovascular disease.This fact is evident when comparing the total fat, SFA, and fiber intake.The low caloric intake raises awareness about the prevalence of malnutrition in dialysis patients.Increasing the protein intake is necessary to avoid loss of muscle mass and hypoalbuminemia, which is associated with increased mortality in this population.Although the dietary intake of a few vitamins is lower than recommended, appropriate levels are achieved, and sometimes exceeded with supplementation.Considering our findings, studies should be performed to see whether targeted counseling can improve the nutritional intake and overall cardiovascular outcomes of MHD patients.

Table 2 .
Dietary Intake With and Without Vitamin Supplements in Comparison With the Current Recommendations

Table 3 .
Dietary Intake of Patients With and Without Background History of Cardiac Disease

Table 4 .
The AHA Recommendation for Dietary Intake