Let Them Eat During Dialysis: An Overlooked Opportunity to Improve Outcomes in Maintenance Hemodialysis Patients

In individuals with chronic kidney disease surrogates of protein-energy wasting (PEW) including a relatively low serum albumin and fat or muscle wasting are by far the strongest death risk factor than any other condition. There are data to indicate that hypoalbuminemia responds to nutritional interventions, which may save lives in the long run. Monitored, in-center provision of high-protein meals and/or oral nutritional supplements during hemodialysis is a feasible, inexpensive and patient-friendly strategy, despite concerns such as postprandial hypotension, aspiration risk, infection control and hygiene, dialysis staff burden, diabetes and phosphorus control, and financial constraints. Adjunct pharmacologic therapies can be added including appetite stimulators (megesterol, ghrelin, and mirtazapine), anabolic hormones (testosterone and growth factors), anti-myostatin agents, and anti-oxidative and anti-inflammatory agents (pentoxiphylline and cytokine modulators) to increase efficiency of intradialytic food and oral supplementation although adequate evidence is still lacking. If more severe hypoalbuminemia (<3.0 g/dL) not amenable to oral interventions prevails or if patient is not capable of enteral interventions, e.g. due to swallowing problems, parenteral interventions such as intra-dialytic parenteral nutrition can be considered. Given the fact that meals and supplements during hemodialysis would require only a small fraction of the funds currently used for the expensive medications of dialysis patients with no proven outcome improvement, this is also an economically feasible strategy.


Introduction
Overnutrition is a major problem in the general population and a serious risk of metabolic syndrome, cardiovascular disease and chronic kidney disease (CKD), with subsequent increased death risk. In CKD patients, however, this relationship may be different, especially in those who undergo maintenance dialysis treatment. In the latter patient population the so-called "uremic malnutrition" 1 (or "malnutrition-inflammation complex" 2 or "renal cachexia" 3 ) which is recently also referred to as "protein-energy wasting" (PEW), 4 is by far the strongest risk factor for adverse outcomes and death, 5 whereas surrogates of overnutrition such as obesity or hyperlipidemia appear counterintuitively protective. 6 Similar associations have been described in individuals with other chronic disease states such as heart failure 7 or in the geriatric populations. 8 It is believed that in CKD and other chronic diseases that are associated with wasting syndrome, pathophysiologic pathways related to malnutrition act as short-term killers and render such long-term killers as obesity or hypertension practically irrelevant. In other words, dialysis patients die much faster of short-term consequences of PEW, so that they do not live long enough to die of the longterm consequences of overnutrition. This so-called timediscrepancy hypothesis 9 suggests that in CKD patient whose short-term mortality is high, interventions that can improve their nutritional status and prevent or correct wasting and sarcopenia have the potential to save lives. 10 In addition to longevity, nutritional status is a strong predictor of better healthrelated quality of life in dialysis patients. 11

PEW and Mortality
If the PEW is such as strong death risk factor, one would expect that the PEW surrogates such as low serum albumin or lower protein intake correlate with mortality. Indeed, evidence suggests they do. A low serum albumin concentration is by far the strongest predictor of mortality and poor outcomes in dialysis patients when compared to any other risk factors, 12,13 be it the traditional risk factors (hypertension, hypercholesterolemia, diabetes, obesity) or nonconventional ones (anemia measures, minerals and bone surrogates, dialysis treatment and technique). 5 The sensitivity of serum albumin to predict CKD patients outcomes is relatively high with such a granularity of as little as 0.2 g/dL or even smaller. [14][15][16][17] In other words, a dialysis patient with a baseline serum albumin of even 0.2 g/ dL higher or lower than another patient with similar demographic and comorbidity constellations has a significantly lower or higher death risk, respectively. The albumin-death association is highly incremental and linear with virtually no cutoff level below or above which the association with survival would cease or reverse. 14,15 This is in sharp contradistinction to most other outcome predictors in CKD with U-or J-shape survival associations. Even more importantly, changes in serum albumin over time are associated with proportional and reciprocal alterations in subsequent death risk, in that a rise or drop in serum albumin by as little as 0.1 g/dL over a few month period is associated with improving or worsening survival, respectively. 14 Similar mortality predictabilities have also been reported with other nutritional markers such as serum prealbumin 18 (e.g. <30 mg/dL) and the "malnutrition-inflammation score" (MIS≥5). 19 Nevertheless, serum albumin remains the simple single test that is readily available ubiquitously and has been recommended by most nutritional societies as a first line nutritional marker. Hence, as shown in Table 1, a diverse array of nutritional and dietary interventions are often considered for maintenance hemodialysis patients with serum albumin <4.0 g/dL or other signs of protein energy wasting.

Meals and Oral Supplements During Hemodialysis Treatment
Given the exceptionally high dietary protein requirement of dialysis patients (~1.2 g/kg/day) and given the observation that most dialysis patients eats <1.0 g/kg/day of protein, 20 an average dialysis patient needs an additional 0.2 to 0.4g/kg/day of protein supplement 21 (see Figure 1). Inadequate food intake especially during hemodialysis treatment days is a common practice among American dialysis patients, while in many other countries meals are routinely served during the hemodialysis treatment sessions. Table 2 summarizes some of the pros and cons pertaining to in-center (in the dialysis clinic) monitored eating and the provision of meals during hemodialysis treatments. In a recent online survey, when we asked nephrologists and dialysis centers in the United States as to why meal trays for patients do not exist during hemodialysis treatment, the common stated concerns include: (1) postprandial hypotension, (2) risk of choking or aspiration, (3) infection control and hygiene issues, including fear of fecal-oral transmission of such diseases as Hepatitis A, (4) staff burden and distraction, and (5) diabetes and phosphorus control (see Table 2). 21 It is not unusual to hear statements such as: "They get food everywhere and this is not fair to the next patient that has to sit in their crumbs."; "I don't want another lawsuit for choking while eating on dialysis" ; and "Having a full stomach might complicate their management." 22 Conversely, meals are routinely given to dialysis outpatients in most European and South East Asian countries. German dialysis patients eat invariably during their hemodialysis treatments and have higher serum albumin and greater survival than their American counterparts. 23 In the past, meals on dialysis were routine in the United States as well. Indeed a few Veteran Administrations hospitals still provide meal trays including breakfast, lunch or supper during all dialysis shifts, be it inpatient or outpatient.
Despite the traditional concerns of North American nephrologists and dialysis care providers, the positive development is that over the past few years increasing numbers of dialysis clinics have allowed and even encourage oral nutritional supplement during the treatment. Indeed several recent pilot and non-randomized studies have indicated that provision of oral nutritional supplements with high protein content during hemodialysis has improved serum albumin. [24][25][26][27] Indeed an elaborate metabolic study showed that oral protein intake during hemodialysis therapy is effective in opposing the catabolic effect of hemodialysis treatment that would otherwise last even hours after the therapy ended. 24 We would also argue that in addition to improving nutritional status, providing in-center meals and/or oral nutritional supplements during hemodialysis treatment would improve patient compliance and satisfaction (see Table 2). Patients may be more motivated to attend the treatments when they know that a lunchbox is awaiting them. Even though in Europe meals on dialysis rarely lead to hypotension, we would argue that it can be considered as an effective strategy against intradialytic hypertension. Many patients may already ignore the eating-prohibitory regulations of some dialysis clinics and still bring in their own foods including ones with high phosphorus content and super-sized soft drinks. Hence, we are in the position to offer them a better and more appropriate food or supplement with higher protein content and lower phosphorus to protein ratio 28 and lower potassium content. 29 The in-center food can be offered along with directly observed administration of phosphorus binder regimen and required multivitamins at the time of meal or supplement intake.
There are several studies in which oral nutrition has been provided during hemodialysis treatment including studies by Szklarek-Kubicka et al. 30 and Moreira et al. 31 In a more recent controlled trial known as the "Anti-Inflammatory and Anti-Oxidative Nutrition during Dialysis" (AIONID) Study, 32 84 adult hypoalbuminemic (albumin<4.0 g/dL) 84 hemodialysis patients were double-blindly randomized to receive 16 weeks of interventions including oral nutritional supplement (ONS), pentoxifylline, ONS with pentoxifylline, or placebos during hemodialysis treatments; and these 4 groups were associated with an average change in serum albumin of +0.21 (p=0.004), +0.14 (p=0.008), +0.18 (p=0.001) and, +0.03 g/dL (p=0.59), respectively. However, in a pre-determined internet-to-treat regression analysis only ONS during hemodialysis without pentoxifylline was associated with a significant albumin rise (+0.17±0.07 g/dL, p=0.018). 32 In two recent large observational studies, ONS during hemodialysis was associated with improved survival 33 and improved hospitalization. 34 In another recent randomized controlled trial known as "Fosrenol for Enhancing Dietary Protein Intake in Hypoalbuminemic Dialysis Patients" (FrEDI) Study 35 (ClinicalTrials.gov # NCT0111694110) in 110 hypoalbuminemic (<4.0mg/ dL) hemodialysis patients received meals during hemodialysis for 8 weeks, the intervention group received high protein meals as prepared meal boxes (50g protein, 850Cal, phosphorus to protein ratio <10mg/gm) along with 0.5 to 1.5g lanthanum carbonate (Fosrenol) titrated as needed to control phosphorus burden from the high protein meals), whereas the control group received meal boxes containing low calorie (<50 Cal) and almost no protein (<1g, such as salads) during each hemodialysis treatment. Among the 51 intervention and 55 control subjects who qualified for the intention-to-treat analyses, the combined rise in albumin ≥0.2g/dL while maintaining phosphorus in 3.5-<5.5mg/dL range was achieved in 25.5% and 9.8%, respectively (χ 2 p-value 0.036). No serious adverse events were reported, and patients reported satisfaction with high protein meals during hemodialysis. 35 Hence, in the FREDI Study provision of high protein meals combined with a potent binder during hemodialysis treatment was safe and improved serum albumin while controlling serum phosphorus. 35 In summary given the above studies, we suggest provision of maintenance meals (as in FREDI Study) 35 or balanced dietary supplement (as in AIONID Study) 32 during each and every hemodialysis treatment and dialysis clinic visit. A maintenance regimen can assure adequate protein intake and reinforce similar dietary habits at home. We also recommend the frequent intake of small amount of protein-rich liquid oral supplement with prescribed pills to replace water, which is shown to improve outcomes in geriatric and nursing home patients. 36

Other Nutritional Interventions
In addition to meals and nutritional supplements during hemodialysis, there are other potential interventions that can be used in conjunction or alone to improve the nutritional status of dialysis patients. These include, but are not limited to appetite stimulators with or without antidepressant properties such as megesterol 37 ghrelin, 38 and mirtazapine; 39 anabolic hormones such as testosterone and growth factors; 40 and anti-oxidative and antiinflammatory agents such as pentoxiphylline and cytokine modulatory agents, 41,42 or omega-3 fatty acid 43 (see Table 1). Intradialytic exercise with or without concomitant nutritional supplementation has been proposed as a potential therapy although long-term efficacy of this strategy needs to be confirmed. 44,45 If more severe hypoalbuminemia (e.g. <3.0 g/dL) prevails that is not amenable to oral interventions even with adjunct pharmacologic therapy, or if patient is not capable of enteral interventions, parenteral interventions should be considered such as intra-dialytic parenteral nutrition (IDPN). 46,47 The IDPN is especially effective with such low serum albumin values. 48 Finally, nonnutritional interventions should also be considered such as dialysis treatment modalities and techniques that leads to less inflammation or protein loss. 49,50

Impact of Nutritional Interventions on Outcomes
An important question that is still unanswered is whether the PEW-albumin-death association a causal association (and amenable to interventions listed in Table 1) or an epiphenomenon? Whereas the debate continues as to how to find the correct answer to this question, [51][52][53] in our opinion a more clinically relevant and time-sensitive question is the following: "Can a nutritional intervention increase serum albumin in CKD patients and by doing so improve survival and quality of life?" We believe that the answer is positive based on a number of experimental data, 48, 54-59 even though to date no single well-designed and well-performed randomized controlled trial with adequate sample size has been performed to answer this simple question. Indeed the entire field of Nutritional support (such as in terminal cancer patients, post-surgical patients or geriatric or disabled populations) is based on the premise that independent of the cause of wasting and cachexia, provision of nutritional support improves patient's immediate and short-term outcomes, or we would not be practicing it over the past few decades. 60 Whereas we do not deny the paucity of the controlled trials and the difficulties sounding the feasibility of nutritional interventions and testing their effects on hard outcomes, 61 it is our opinion that keeping hemodialysis patients hungry during dialysis treatment is not an appropriate action, both clinically and ethically.

Conclusion Remarks
There appears to be a consensus pertaining to the important role of favorable nutritional status in dialysis patient outcomes within the nephrology community. As we have moved towards longer hemodialysis sessions 62 and in anticipation of drastic changes in practice pattern and dialysis patient care in many countries, we need to rethink the pros and cons of provision of meals and oral supplements during dialysis treatment. While this is a routine practice in Europe and most other countries, Northern American dialysis patients are deprived of nutritional intervention during dialysis. There is consistent, strong and robust association of nutritional status, and in particular serum albumin level, with survival in CKD patients along with data from several studies indicating improvement in response to intradialytic nutritional supplementation. Hence, providing intradialytic meals or oral nutritional supplements to dialysis patients and other nutritional interventions are the most promising intervention to increase serum albumin and to improve longevity and quality of life in this patient population. Since provision of meals and oral supplements would require only a small fraction of the funds currently used for the expensive medications given to dialysis patients with no proven outcome modification, this is also an economically feasible strategy 63   Table 1 Suggested intervention for maintenance hemodialysis patients with serum albumin <4.0 g/dL or other signs of protein energy wasting

Oral Nutritional Interventions
In-center, intra-dialytic administration Advantages Disadvantages -Meals during dialysis treatment Preferred as routine for all HD patients See Table 2 See Table 2 -Oral nutritional supplements Preferred esp. if meals not effective See Table 2 See Table 2 -