Diet Patterns—A Neglected Aspect of Hemodialysis Care

Good nutrition is an important part in providing care for patients on chronic dialysis. These patients have complex nutritional challenges ranging from controlling electrolyte and metabolic abnormalities associated with CKD to changing the lipid and carbohydrate composition to reduce cardiovascular risk and limiting protein intake while still meeting total energy requirements. Unfortunately, there is a lack of definitive evidence to firmly support guidelines for best CKD dietary practices.¹ Thus, the study by Saglimbene et al.² is a valuable addition to the CKD nutrition literature; however, the results overall are a letdown for those wanting more evidence to support current dietary advice.

Unlike nutrition recommendations for CKD, the scientific literature supporting public nutrition advice is vast. In recent years, dietary guidelines have moved away from focusing on single foods or nutrients to a more encompassing diet pattern approach.³ Although there is some trial evidence supporting this shift,^4–6 the bulk of information on healthy eating patterns comes from diet assessment studies. Different methods (principal components factor analysis⁷ and dietary indices^8,9) have been used to identify primary dietary patterns that affect health. However, the most enduring method has been the creation of diet scores that indicate the degree of adherence to the Mediterranean diet or Dietary Approaches to Stop Hypertension (DASH) diet and then use that information to examine associations with health outcomes.

In this issue of the Journal of the American Society of Nephrology, Saglimbene et al.² report the results of a prospective, multinational cohort study that was undertaken by investigators from ten European countries and one South American country to examine the association of dietary patterns with mortality in adults on long-term hemodialysis. Nutrient intake was estimated by a food frequency questionnaire (FFQ) in a prevalent hemodialysis population (9757 people; mean age of 63.1 years old; 58% men). Population-based diet quality scores reflecting adherence to the Mediterranean diet⁸ and the DASH-style eating plan⁹ were calculated from this information, with higher scores indicating food intake more consistent with these diets. Outcomes were cardiovascular (interestingly not cerebrovascular) mortality and all-cause mortality; they were derived from death certificates, and the cause was adjudicated by the participants’ treating clinicians. The main finding of the study was that higher Mediterranean and DASH diet scores were not associated with total or cardiovascular mortality. The results are disappointing given the increased attention that dietary patterns have received in reducing cardiovascular events in both generally healthy populations and individuals with preexisting cardiovascular disease.³

The lack of association observed in this observational study may suggest that there is none. However, there are also several other possible explanations, including limitations of their dietary assessment tool and their approach to evaluating usual food consumption. In this regard, the authors did mention ascertainment errors and a single dietary measurement. The European Prospective Investigation into Cancer and Nutrition (EPIC) Study FFQ used to derive the Mediterranean scores is well established, and it has been validated and used extensively in many studies.⁸ Unfortunately, the instrument does not consider differences in portion size, assuming that people eat the same portion. This contrasts with the FFQ by Willett et al.,¹⁰ which asks people whether they eat a small, medium, or large portion of each food item. The lack of information on portion size may contribute to more random measurement error in the assessment of dietary intake.

Like the EPIC FFQ, the Global Allergy and Asthma European Network (GA²LEN) FFQ used in this study does not take portion size into account. Importantly, it has not been validated against more rigorous diet assessment methods, such as food records or 24-hour dietary recall.¹¹ Dietary fat consumption, as determined from the GA²LEN FFQ, was validated against plasma phosphatidylcholine fatty acids. The validation paper reported reasonably good correlations with dietary intake of polyunsaturated fat but not saturated or monounsaturated fatty acids. Additionally, FFQs are not designed to provide a precise measure of sodium intake,⁹ a hallmark of the DASH Sodium Trial.¹² Despite this limitation, a sodium component was included in the construction of the DASH score used in this study, and it was recognized that its inclusion may weaken results, likely owing to greater misclassification. Fortunately, newer digital tools are emerging to quantify daily sodium consumption,¹³ which will be particularly useful in studies where urine collections, an alternative method to assess daily salt consumption,¹⁴ cannot be used.

The authors conducted a series of subgroup analyses to examine associations of each diet score with cardiovascular and total mortality across strata of age, sex, smoking history, and myocardial infarction. Most of these secondary analyses showed no significant association between dietary patterns and outcome events, except for a significant inverse association between the DASH diet score and total mortality in the younger population (≤60 years old). The authors provide no biologically plausible explanation for this finding and posited a play of chance, which is certainly a possibility in the absence of adjustments for multiple testing. It is also possible that the diet scores are a marker of health conscientiousness that was not fully taken into account in the analyses (i.e., “healthy-user bias”).¹⁵ Whatever the explanation, their observation needs further investigation in other populations.

The authors are to be commended for reporting associations for each dietary component. The field of nutrition has evolved rapidly in the past few decades, and new data have questioned old dogmas. An objective assessment of the components of each diet score (the Mediterranean diet and the DASH) is important so that readers can see which dietary components might explain the association (or lack of association) of diet scores with outcome events. What is striking in their analysis is the observation that both diet scores showed a significant increase in cardiovascular disease mortality with dairy products. The mechanisms through which dairy products may influence health and longevity are undoubtedly complex, and they may not be solely related to the effect of dairy products on total saturated fatty intake.¹⁶

In a population-based cohort study from Sweden, premature death was unacceptably high in hemodialysis, with crude rates being from more than eightfold higher in patients 70 years old and older to almost 50-fold higher in individuals age 18–49 years old compared with the general population.¹⁷ The results of this study echo those in other jurisdictions.¹ Patients on hemodialysis are particularly at high risk to experiencing a cardiovascular event, and as the authors showed, cardiovascular disease is the leading cause of death in this vulnerable population. Good nutrition will undoubtedly be a part of solutions to lower mortality in these patients. In this regard, insights from this study are extremely valuable in constructing effective nutrition interventions. However, the results also point out the pitfalls of relying too heavily on associations to determine what constitutes a healthy diet, and they underscore the need for well designed, randomized, controlled trials with clinical outcomes to properly guide nutrition recommendations.

Disclosures

A.M. is the recipient of a 2-year operating grant from the Dairy Farmers of Canada.