Selenium for malnutrition in hemodialysis patients: have we considered all of the elements?

Although death directly attributable to malnutrition per se is uncommon in hemodialysis patients, markers of poor nutritional status (such as hypoalbuminemia and abnormal anthropometric measures) are associated with excess mortality in this population [1–3]. Many studies have investigated the link between biochemical markers of inflammation and evidence of malnutrition with cardiovascular disease [4, 5] and other causes of morbidity and mortality. It is important to distinguish between malnutrition from inadequate caloric intake and malnutrition from factors unique to end-stage renal disease—such as uremic toxins, acidosis, inflammation and iatrogenic causes such as nutrient loss during dialysis or exposure to dialysis membranes [6]. In recognition of these latter processes, the term protein energy wasting (PEW) is recommended to indicate malnutrition in chronic kidney disease [7]. The term, malnutrition inflammation complex syndrome (MICS) is used to emphasize the key role of inflammation in PEW and its association with poor outcomes [8]. Although the exact mechanisms are unclear, a common pathway—oxidative stress—may mediate many of the factors contributing to MICS. Oxidative stress occurs when reactive oxygen species overwhelm antioxidant defenses and damage various molecular structures. Although treatment with antioxidants is generally viewed as beneficial, it is plausible that unnecessary supplementation with these agents could disturb the balance between harmful and beneficial effects of free radicals and therefore should be reserved for those with deficiencies in antioxidant defense [9]. For this reason, hemodialysis patients may prove optimal candidates for antioxidant therapy, a hypothesis that is supported by higher markers of oxidative stress compared with the general population [10, 11] and is appealing since the limited data on interventions aimed at traditional cardiovascular risk factors in this population have been disappointing [12]. However, in spite of this theoretical potential [13, 14], the role of antioxidant therapy in modifying clinical outcomes in this population has not been extensively evaluated [15]. Antioxidants are either present endogenously or are obtained from the diet. Selenium is an essential trace element obtained from meat, seafood, grains and nuts with a range of biochemical functions [16]. Its antioxidant function is a constituent of the antioxidant enzyme glutathione peroxidase which removes damaging hydrogen peroxide and lipid hydroperoxides from free radicals [17]. Selenium supplementation in humans increases antioxidant activity [18] and decreases lipid peroxidation [19]. Although the clinical importance of selenium supplementation is controversial, selenium deficiency has been associated with an increased risk of cancer [20] and coronary artery disease [21] in the general population. Serum selenium levels are markedly lower in hemodialysis patients [22] when compared with non-dialysis patients; however, the consequences of such deficiency are unknown. In this issue of NDT, Salehi et al. [23] report the results of a double-blind randomized controlled trial comparing daily oral selenium supplementation to placebo for improving the nutritional status of chronic hemodialysis patients. The investigators randomized 80 patients receiving chronic hemodialysis at a single center in Iran to either selenium or placebo daily for 12 weeks. The primary outcome was the subjective global assessment (SGA). Another nutritional scoring system, the Malnutrition Inflammation Score (MIS), was used as a secondary outcome. The MIS is similar to the SGA, but also includes the number of major comorbid conditions, body mass index (BMI), albumin and total iron-binding capacity. Additional outcomes included malondialdehyde (MDA; a biomarker of lipid oxidation), interleukin-6 (IL-6) and high-sensitivity C-reactive protein (hsCRP). All measurements were taken at the beginning and at the end of the 12-week trial. Selenium levels were not measured as a part of this study. The significant improvement in SGA score from the baseline in the selenium group [3.89 (95% CI: 5.32, 2.65)] was substantially better than the improvement of only 1.35 points (0.30, 2.40) in the placebo group (P<0.001). MIS also significantly improved compared with placebo. Although MDA and IL-6 levels decreased significantly in the selenium group compared with controls (suggesting favorable effects on lipid peroxidation and inflammation), there was no significant change in hsCRP or serum albumin among selenium recipients. Further, although MDA is often used as a marker of oxidative stress [15], the reliability of the assay in this study was based on an animal model. Unfortunately, on-therapy changes in BMI were not reported. Salehi et al. [23] should be congratulated on their selection of a clinical outcome (nutritional status). Unfortunately, interpretation of their results is not straightforward. The conventional SGA classifies patients into one of three nutritional categories (well nourished, moderately malnourished and severely malnourished) based on the history of weight loss, anorexia, gastrointestinal symptoms, functional status and the physical examination findings of edema and loss of fat and muscle. It is unclear whether SGA is a sensitive measure of the nutritional status in hemodialysis patients; the current recommendations from the European Guidelines on Nutrition suggest that the SGA be used to diagnose only severe nutrition [24]. Although the quantitative version of the SGA [25] used by Salehi et al. may be more sensitive for the detection of clinically meaningful changes in the nutritional status, it has not apparently been evaluated in longitudinal studies of hemodialysis patients—making it difficult to assess the clinical significance of the observed change in SGA score. Results obtained with the MIS were consistent with those for SGA—but skeptics will note that the MIS is a relatively new nutritional scoring system and thus will also pose potential challenges in interpretation when used in hemodialysis patients. Presenting results separately for the SGA subscales might also have been informative—since it would allow readers to speculate as to the link between the clinical and biochemical outcomes. For example, an improvement in the anorexia subscale could theoretically relate to a decrease in pro-inflammatory cytokines as a result of reduced oxidative stress. Selenium was noted to have gastrointestinal side effects and it would have been interesting to see how this affected the anorexia subscale. As the investigators mention, perhaps the most significant limitation of their study is that selenium levels were not measured. Presentation of the relationship between levels of MDA and serum concentrations of selenium would have helped to elucidate the putative mechanism of antioxidant action. Furthermore, potential interactions between the outcomes and baseline selenium levels (if present) could help to indicate which patients should be targeted in future studies. Given that the nutritional status of at least some malnourished patients might improve with augmented calorie intake [26, 27], a tool that could accurately identify patients requiring novel interventions such as selenium supplementation would be clinically useful. Antioxidant therapy is potentially harmful, and until more is known about its long-term effects, it should be reserved for those who might benefit the most. Selenium is no exception to this general rule and has the potential to be highly toxic in hemodialysis patients; it has a high level of bioavailability, is excreted in urine and has a narrow therapeutic window [28]. The study by Salehi et al. [23] also raises the question of which antioxidant to supplement. In the context of mitigating oxidative stress, perhaps it would be best to supplement multiple potential deficiencies. For example, comparable to selenium, zinc has antioxidant properties and is associated with malabsorption [29] and a relative deficiency of zinc has been noted in hemodialysis patients [22]. Without relying on baseline measurements to guide therapy, perhaps the approach taken by the investigators would have been better rationalized by broader supplementation. Although Salehi et al. [23] did not confirm that selenium supplementation leads to clinically meaningful improvement in the nutritional status in hemodialysis patients, their study is important because it draws attention to the potential role of trace element supplementation in hemodialysis patients. The results should not change clinical practice just yet, but there are several implications for researchers. First, further studies are required to identify a tool that is sufficiently sensitive to identify meaningful changes in the nutritional status in response to interventions in hemodialysis patients. Second, the definition of MICS is useful for defining a distinct syndrome that is associated with poor outcomes but is limited in terms of addressing etiology. Salehi et al. refer to oxidative stress and inflammation as one process—yet, inflammation likely represents one of the pathways through which oxidative stress is mediated. These mechanisms require further evaluation in hemodialysis patients. The degree to which a lack of calories contributes to oxidative stress (independent of inflammation) is unclear and could be evaluated through an evaluation of standard nutritional therapy with and without antioxidant supplementation. Third, future studies of trace element supplementation should assess the effect of treatment on serum levels of the element in question. Finally, although the current knowledge regarding the role of antioxidants in hemodialysis patients is incomplete, randomized trials of nutritional interventions should evaluate clinical outcomes such as quality of life and mortality. None declared. (See related article by Salehi et al. Selenium supplementation improves the nutritional status of hemodialysis patients: a randomized, double-blind, placebo-controlled trial. Nephrol Dial Transplant 2013; 28: 716–723.)

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.