Oral protein‐based supplements versus placebo or no treatment for people with chronic kidney disease requiring dialysis

Jia Yee Mah; Suet Wan Choy; Matthew A Roberts; Anne Marie Desai; Melissa Corken; Stella M Gwini; Lawrence P McMahon

doi:10.1002/14651858.CD012616.pub2

透析を必要とする慢性腎臓病の人に対する経口タンパク質の補充とプラセボまたは無治療との比較

Declaraciones de intereses de los autores

Versión publicada: 11 mayo 2020 Historial de versiones

https://doi.org/10.1002/14651858.CD012616.pub2

Contraer todo Desplegar todo

Abstract

disponible en

Background

Malnutrition is common in patients with chronic kidney disease (CKD) on dialysis. Oral protein‐based nutritional supplements are often provided to patients whose oral intake is otherwise insufficient to meet their energy and protein needs. Evidence for the effectiveness of oral protein‐based nutritional supplements in this population is limited.

Objectives

The aims of this review were to determine the benefits and harms of using oral protein‐based nutritional supplements to improve the nutritional state of patients with CKD requiring dialysis.

Search methods

We searched the Cochrane Kidney and Transplant Register of Studies up to 12 December 2019 through contact with the Information Specialist using search terms relevant to this review. Studies in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE, conference proceedings, the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.

Selection criteria

Randomised controlled trials (RCTs) of patients with CKD requiring dialysis that compared oral protein‐based nutritional supplements to no oral protein‐based nutritional supplements or placebo.

Data collection and analysis

Two authors independently assessed studies for eligibility, risk of bias, and extracted data from individual studies. Summary estimates of effect were obtained using a random‐effects model, and results were expressed as risk ratios and their 95% confidence intervals (CI) for dichotomous outcomes, and mean difference and 95% CI for continuous outcomes.

Main results

Twenty‐two studies (1278 participants) were included in this review. All participants were adults on maintenance dialysis of whom 79% were on haemodialysis (HD) and 21% peritoneal dialysis. The follow‐up period ranged from one to 12 months. The majority of studies were at unclear risk of selection, performance, and reporting bias. The detection bias was high for self‐reported outcomes.

Oral protein‐based nutritional supplements probably lead to a higher mean change in serum albumin compared to the control group (16 studies, 790 participants: MD 0.19 g/dL, 95% CI 0.05 to 0.33; moderate certainty evidence), although there was considerable heterogeneity in the combined analysis (I² = 84%). The increase was more evident in HD participants (10 studies, 526 participants: MD 0.28 g/dL, 95% CI 0.11 to 0.46; P = 0.001 for overall effect) and malnourished participants (8 studies, 405 participants: MD 0.31 g/dL, 95% CI 0.10 to 0.52, P = 0.003 for overall effect). Oral protein‐based nutritional supplements also probably leads to a higher mean serum albumin at the end of the intervention (14 studies, 715 participants: MD 0.14 g/dL, 95% CI 0 to 0.27; moderate certainty evidence), however heterogeneity was again high (I² = 80%). Again the increase was more evident in HD participants (9 studies, 498 participants: MD 0.21 g/dL, 95% CI 0.03 to 0.38; P = 0.02 for overall effect) and malnourished participants (7 studies, 377 participants: MD 0.25 g/dL, 95% CI 0.02 to 0.47; P = 0.03 for overall effect).

Compared to placebo or no supplement, low certainty evidence showed oral protein‐based nutritional supplements may result in a higher serum prealbumin (4 studies, 225 participants: MD 2.81 mg/dL, 95% CI 2.19 to 3.43), and mid‐arm muscle circumference (4 studies, 216 participants: MD 1.33 cm, 95% CI 0.24 to 2.43) at the end of the intervention. Compared to placebo or no supplement, oral protein‐based nutritional supplements may make little or no difference to weight (8 studies, 365 participants: MD 2.83 kg, 95% CI ‐0.43 to 6.09; low certainty evidence), body mass index (9 studies, 368 participants: MD ‐0.04 kg/m², 95% CI ‐0.74 to 0.66; moderate certainty evidence) and lean mass (5 studies, 189 participants: MD 1.27 kg, 95% CI ‐1.61 to 4.51; low certainty evidence). Due to very low quality of evidence, it is uncertain whether oral protein‐based nutritional supplements affect triceps skinfold thickness, mid‐arm circumference, C‐reactive protein, Interleukin 6, serum potassium, or serum phosphate.

There may be little or no difference in the risk of developing gastrointestinal intolerance between participants who received oral protein‐based nutritional supplements compared with placebo or no supplement (6 studies, 426 participants: RR 2.81, 95% CI 0.58 to 13.65, low certainty evidence). It was not possible to draw conclusions about cost or quality of life, and deaths were not reported as a study outcome in any of the included studies.

Authors' conclusions

Overall, it is likely that oral protein‐based nutritional supplements increase both mean change in serum albumin and serum albumin at end of intervention and may improve serum prealbumin and mid‐arm muscle circumference. The improvement in serum albumin was more evident in haemodialysis and malnourished participants. However, it remains uncertain whether these results translate to improvement in nutritional status and clinically relevant outcomes such as death. Large well‐designed RCTs in this population are required.

PICO

Population

Intervention

Comparison

Outcome

El uso y la enseñanza del modelo PICO están muy extendidos en el ámbito de la atención sanitaria basada en la evidencia para formular preguntas y estrategias de búsqueda y para caracterizar estudios o metanálisis clínicos. PICO son las siglas en inglés de cuatro posibles componentes de una pregunta de investigación: paciente, población o problema; intervención; comparación; desenlace (outcome).

Para saber más sobre el uso del modelo PICO, puede consultar el Manual Cochrane.

一般語訳

disponible en

透析を必要とする慢性腎臓病の人に対するタンパク質を含む経口栄養補助食品

論点透析を必要とする慢性腎臓病の患者は、多くの理由で栄養失調を発症するリスクがあり、食欲が乏しいために食事摂取量が不足していることがよくある。経口栄養補助食品は、一般に、必要栄養量を満たすのに十分な量を食べていない人に提供される。透析患者に栄養補助食品を提供するには、カリウム、リン酸、水分制限を注意深く考慮する必要がある。

実施したこと タンパク質を含む栄養補助食品を経口投与すると、血清アルブミン濃度やその他の栄養指標が向上するかどうかを確認することを目的とした。

わかったこと このレビューには22件（合計1278人）の研究が含まれ、タンパク質を含む経口栄養補助食品の効果を調査した。すべての参加者は、維持透析を受けている成人だった（79％が血液透析、21％が腹膜透析）。研究の観察期間は1～12ヶ月間であった。調査結果によると、タンパク質を含む栄養補助食品を経口投与すると、おそらくアルブミン濃度（栄養状態の指標）がわずかに増加し、プレアルブミン濃度（アルブミン濃度よりも短期間の栄養状態を反映する指標）と中腕筋周囲（骨格筋量の指標）を改善する可能性があることが示唆された。アルブミン濃度の上昇は、血液透析を受けている参加者と栄養失調の参加者でより顕著であった。経口によるタンパク質の栄養補給がカリウムとリン酸の血中濃度に影響を与えるかどうかは不明である。タンパク質を含む経口栄養補助食品は、腹部症状を発症するリスクにほとんど差がないか、あるいは全く差がなかった。研究の質とデザインにはいくつかの違いがあった。

結論著者らは、タンパク質を含む経口栄養補助食品が、透析を必要とする患者において一部の栄養指標を改善するのに効果的であると結論付けている。ただし、これらの結果がこの集団にとって意味のある結果につながるかどうかは依然として不明である。この治療の費用対効果を見極め、気分が良くなったり、長生きしたりといったメリットを患者にもたらすことができるかどうか、さらなる研究が求められている。

Authors' conclusions

Implications for practice

The results of this systematic review found that the use of oral protein‐based nutritional supplements in people with CKD requiring dialysis contributed to an increase in serum albumin and prealbumin, as well as an increase in mid‐arm muscle circumference. The rise in serum albumin was more evident in participants who were on HD compared to PD, and those who were malnourished. However, given the limitations of serum albumin as a nutritional marker, it remains uncertain whether improvement in these outcomes translates to improved nutritional status or better clinically relevant outcomes. There may be little or no difference in the risk of developing gastrointestinal intolerance, and there were limited data available on adverse biochemical effects. This review also found insufficient data to determine the effects on death, cost, or patient centred outcomes.

Implications for research

The current data mainly involves HD patients, and further studies assessing PD patients are required. Additionally, the current studies are small, underpowered, with variable risk of biases and larger, well‐designed RCTs are required in this area. Furthermore, studies assessing hard endpoints such as death are warranted to determine if optimisation of nutritional status translates to a reduction in death date. More research into cost benefit and patient centred outcomes is also required. Given the complexity of protein‐energy wasting in dialysis patients, co‐intervention of oral protein‐based nutritional supplements with inflammatory modulators could be considered in future studies.

Summary of findings

Open in table viewer

Summary of findings 1. Summary of findings Table 1

Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Oral protein‐based nutritional supplement versus placebo or no treatment for people with CKD requiring dialysis
Patient or population: people with CKD requiring dialysis Intervention: oral protein‐based nutritional supplement Comparison: placebo or no treatment
Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with control	Risk difference with Oral protein‐based nutritional supplement
Change in serum albumin	790 (16)	⊕⊕⊕⊝ MODERATE ¹	‐‐	The mean change in serum albumin was ‐0.01 g/dL	MD 0.19 g/dL higher (0.05 higher to 0.33 higher)
Serum albumin (end of intervention)	715 (14)	⊕⊕⊕⊝ MODERATE ¹	‐‐	The mean serum albumin (end of intervention) was 3.55 g/dL	MD 0.14 g/dL higher (0 to 0.27 higher)
Serum prealbumin	225 (4)	⊕⊕⊝⊝ LOW ²	‐‐	The mean serum prealbumin was 27.18 mg/dL	MD 2.81 mg/dL higher (2.19 higher to 3.43 higher)
Death	‐‐	‐‐	‐‐	Not reported
Intolerance to therapy ‐ Gastrointestinal intolerance	426 (6)	⊕⊕⊝⊝ LOW ³	RR 2.81 (0.58 to 13.65)	20 per 1,000	57 more per 1,000 (12 fewer to 279 more)
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CKD: chronic kidney disease; CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Down‐graded because of high heterogeneity, some of which are not explained by the pre‐specified subgroups (inconsistency) ² Down‐graded because of high attrition bias in two studies, a moderate number of studies and patient assessed with results heavily influenced by one study (Benner 2010) (imprecision) ³ Down‐graded because of high risk of detection bias in most studies, high risk of attrition bias in three studies, and small number of events in low number of patients assessed with wide CI (imprecision).

Open in table viewer

Summary of findings 2. Summary of findings Table 2

Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Oral protein‐based nutritional supplement versus placebo or no treatment for people with CKD requiring dialysis
Patient or population: people with CKD requiring dialysis Intervention: oral protein‐based nutritional supplement Comparison: placebo or no treatment
Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with control	Risk difference with Oral protein‐based nutritional supplement
Weight	365 (8)	⊕⊕⊝⊝ LOW ¹	‐‐	The mean weight was 59.51 kg	MD was 2.83 kg higher (0.43 lower to 6.09 higher)
Body mass index	368 (9)	⊕⊕⊕⊝ MODERATE ²	‐‐	The mean body mass index (BMI) was 25.18 kg/m²	MD was 0.04 kg/m² lower (0.74 lower to 0.66 higher)
Triceps skinfold thickness	367 (6)	⊕⊝⊝⊝ VERY LOW ³	‐‐	The mean triceps skinfold thickness (TSF) was 14.14 mm	MD was 0.81 mm thicker (1.59 lower to 3.21 higher)
Mid‐arm circumference	97 (2)	⊕⊝⊝⊝ VERY LOW ⁴	‐‐	The mean mid‐arm circumference (MAC) was 24.99 cm	MD was 0.5 cm wider (1.56 lower to 2.56 higher)
Mid‐arm muscle circumference	216 (4)	⊕⊕⊝⊝ LOW ⁵	‐‐	The mean mid‐arm muscle circumference (MAMC) was 22.32 cm	MD 1.33 cm wider (0.24 higher to 2.43 higher)
Lean mass	189 (5)	⊕⊕⊝⊝ LOW ⁶	‐‐	The mean lean mass was 42.95 kg	MD 1.27 kg higher (1.61 lower to 4.15 higher)
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CKD: chronic kidney disease; CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Down‐graded because of high attrition bias in four studies, and moderate heterogeneity (inconsistency). CI is wide and effect shows appreciable benefit and harm, but shows benefit in subgroup treated for 6 months or more ² Down‐graded because of low number of patients assessed, and CI is wide and effect shows appreciable benefit and harm (imprecision) ³ Down‐graded because of high attrition bias in over half the studies included, high heterogeneity (inconsistency), and CI is wide and effect shows appreciable benefit and harm (imprecision) ⁴ Down‐graded because of high/unclear biases in both studies, substantial heterogeneity (inconsistency), low number of studies and patients assessed (imprecision), and CI is wide and effect shows appreciable benefit and harm ⁵ Down‐graded because of high attrition bias in three studies, and moderate number of studies and patients assessed (imprecision) ⁶ Down‐graded because of high attrition bias in two studies, and moderate number of studies and low patients assessed with wide CI (imprecision).

Open in table viewer

Summary of findings 3. Summary of findings Table 3

Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Oral protein‐based nutritional supplement versus placebo or no treatment for people with CKD requiring dialysis
Patient or population: people with CKD requiring dialysis Intervention: oral protein‐based nutritional supplement Comparison: placebo or no treatment
Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with control	Risk difference with Oral protein‐based nutritional supplement
Quality of life	662 (8)	‐‐	‐‐	Data could not be combined for meta‐analysis. See 'Effects of interventions' for results
Cost	117 (4)	‐‐	‐‐	Data could not be combined for meta‐analysis. See 'Effects of interventions' for results
CRP	436 (8)	⊕⊝⊝⊝ VERY LOW ¹	‐‐	The mean C‐reactive protein (CRP) was 4.28 mg/dL	MD 0.06 mg/dL lower (0.29 lower to 0.18 higher)
IL‐6	62 (2)	⊕⊝⊝⊝ VERY LOW ²	‐‐	The mean IL‐6 was 28.47 pg/mL	MD 5.05 pg/mL lower (23.91 lower to 13.81 higher)
Serum potassium	357 (7)	⊕⊝⊝⊝ VERY LOW ³	‐‐	The mean serum potassium was 4.90 mEq/L	MD 0.05 mEq/L higher (0.24 lower to 0.35 higher)
Serum phosphate	403 (7)	⊕⊝⊝⊝ VERY LOW ⁴	‐‐	The mean serum phosphate was 4.84 mg/dL	MD 0.03 mg/dL higher (0.42 lower to 0.47 higher)
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CKD: chronic kidney disease; CI: confidence interval; RR: risk ratio; CRP: C‐reactive protein; IL‐6: interleukin 6
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Down‐graded because of high attrition bias in three studies, surrogate marker for death (indirectness), and results heavily weighted on one study (Lee 1998) (imprecision). ² Down‐graded because of low number of studies and patients assessed with results heavily influenced by one study (IHOPE 2019). CI is wide and effect shows appreciable benefit and harm (imprecision) ³ Down‐graded because of high heterogeneity (inconsistency), surrogate marker for episodes of hyperkalaemia (indirectness), moderate number of studies and CI is wide and effect shows appreciable benefit and harm (imprecision) ⁴ Down‐graded because of moderate heterogeneity (inconsistency), surrogate marker for episodes of hyperphosphataemia (indirectness), moderate number of studies and CI is wide and effect shows appreciable benefit and harm (imprecision)

Background

Description of the condition

The number of people estimated to be receiving kidney replacement therapy worldwide increased from 1.1 million during the 1990 to 2.618 million in 2010 (Liyanage 2015; Lysaght 2002). Malnutrition is common in patients with chronic kidney disease (CKD) on dialysis. A prospective cohort study by Tan 2016 found that more than half of hospitalised patients requiring haemodialysis (HD) were malnourished. In fact, malnutrition can already be demonstrated in about 10% of kidney patients at the start of predialysis care (Westland 2015).

Several factors are thought to contribute to malnutrition in dialysis‐dependent patients. These include, but are not limited to, reduced oral intake and loss of appetite due to uraemic toxins, increased catabolism, low‐grade inflammation, oxidative stress, and the presence of other co‐morbidities (Burrowes 2003; Malgorzewicz 2016). An additional issue for hospitalised patients includes frequent fasting for tests and procedures.

Malnutrition is defined as a low‐nutrient intake or an intake that is inadequate for the nutritional needs of the individual (Fouque 2008a). In patients with kidney disease, the International Society of Renal Nutrition and Metabolism (ISRNM) recommend using 'protein‐energy wasting' to describe the more complex syndrome of muscle wasting, malnutrition, and inflammation seen in this population (Fouque 2008a). Protein‐energy wasting is defined as the state of decreased body stores of protein and energy fuels which is characterised by biochemical abnormalities (low serum albumin, transthyretin, or cholesterol), reduced body mass, reduced muscle mass, and low protein or energy intakes (Fouque 2008a).

Several nutrition assessment tools have been used in assessing malnutrition among dialysis patients, which includes the Subjective Global Assessment (SGA) (CANUSA 1996; Detsky 1987; Enia 1993; KDOQI 2000). In dialysis patients, malnutrition has been strongly associated with increased death and hospitalisations, as well as poor quality of life (QOL) (Dwyer 2002; Kalantar‐Zadeh 2001; Kalantar‐Zadeh 2011). Of the several clinical factors evaluated in the dialysis population, advanced age has been associated with a higher incidence of malnutrition (Kadiri 2011).

Description of the intervention

Dietary energy and protein intake of patients on maintenance HD are inadequate compared to the recommended standard of care, which can result in loss of lean body mass and malnutrition (Burrowes 2003). Recently, The Renal Association in the United Kingdom have recommended an energy intake of 30 to 40 kcal/kg ideal body weight (IBW)/day, and a protein intake of 0.8 to 1.0 g/kg IBW/day for patients with stage 4 to 5 CKD not on dialysis, 1.1 to 1.4 g/kg IBW/day for patients on HD, and 1.0 to 1.2 g/kg IBW/day for patients on peritoneal dialysis (PD) (Wright 2019). These recommendations are similar to previous guidelines (Chan 2012; KDOQI 2000).

Oral nutritional supplements (ONS) are often provided to patients whose oral intake is otherwise insufficient to meet their energy and protein needs (Jensen 2013). They are used in conjunction with expert advice from dietitians. ONS are typically formulated as a combination of macro and micronutrients at varying levels of concentrations, and are therefore not all nutritionally complete (BAPEN 2016; Schneyder 2014). This focus of this review is on oral protein‐based nutritional supplements.

How the intervention might work

Oral nutritional support that targets dietary protein intake have previously been reported to mitigate malnutrition (Kalantar‐Zadeh 2011). However, appropriate nutritional supplements can be difficult to administer in dialysis patients. In addition to factors such as a reduced appetite, poor nutrition is further compounded by the dictates of a stringent fluid restriction, especially in patients with minimal urine output, and electrolyte imbalance. High protein foods such as meat, dairy and nuts carry a high phosphate load, and 'healthy' foods such as fruit and vegetables are often restricted because of their potassium content. Oral protein‐based nutritional supplements prescribed to dialysis patients must therefore be suitably adjusted in order to be taken safely and repeatedly. As such, renal‐specific supplements have a higher protein and energy content, with a lower potassium and phosphate content (Williams 2009). Examples of commercially available oral protein‐based nutritional supplements prescribed to dialysis patients include Nepro, Novasource Renal, Renilon, and Suplena (Fouque 2008b; Shah 2014). Protein isolates which include, but are not limited to, whey protein and soy protein, have also been used as a source of protein supplementation in dialysis patients (Tomayko 2015a).

Reported benefits of oral protein‐based nutritional supplements in HD patients include improvement in muscle quality leading to improvements in physical function (Tomayko 2015a). The use of oral protein‐based nutritional supplements may also provide quicker nutritional repletion and a trend towards a reduced number of days of hospitalisation (Wilson 2001). Reported harms include adverse biochemistry (hyperkalaemia, hyperphosphataemia), intolerability due to taste, and cost that is often borne by the patient (Williams 2009).

Why it is important to do this review

The benefits of oral protein‐based nutritional supplements in patients with CKD, particularly those on dialysis, have long been debated. However, several recent, large, observational studies have demonstrated that the use of oral intra‐dialytic nutritional supplements is associated with a reduction in death of up to 35% in HD patients (Lacson 2012; Weiner 2014).

The cost of ONS is usually borne by the patients. The increasing expenditure of such supplements further complicates the picture. A recent systematic review by Elia 2016, based mostly on retrospective cost analyses, indicated that the use of nutritional supplements in the community and care homes produce an overall cost advantage. This cost advantage was primarily from reduced hospitalisations and length of stays, but there were also several clinically relevant outcomes favouring ONS reported, including an improved QOL and reduced infections (Elia 2016). Focusing on the subgroup analyses comparing the use of ONS to no ONS, there is a median cost saving of 11.5% (Elia 2016).

Baldwin 2011 conducted a systematic review examining several interventions in addition to dietary advice in adults with disease‐related malnutrition. One intervention arm of this review compared the outcome of dietary advice with dietary advice plus an ONS (Baldwin 2011). Improvements in mid‐arm muscle circumference, triceps skinfold, and grip strength were demonstrated in the supplement group (Baldwin 2011). However, for the two latter outcomes, there was significant heterogeneity in the pooled analysis (Baldwin 2011). In this intervention arm, only three of the 16 studies included patients on dialysis (Baldwin 2011). Furthermore, as this review was completed in 2011 and the terms 'dialysis or CKD' were not specifically used in their search terms, there may be studies specific to dialysis patients available now that were not included in their analysis (Baldwin 2011).

The focus of this review is the role of oral protein‐based nutritional supplements in patients with CKD on dialysis. Although the above reviews by Elia 2016 and Baldwin 2011 are not specific to dialysis patients, it certainly raises the question whether potential benefits in clinical outcomes and cost are applicable to the dialysis population more generally. Currently, ONS is recommended when nutritional intake fails to increase and remains inadequate to meet energy and protein requirements despite dietary intervention (Wright 2019). However, the availability and prescription of ONS varies widely between dialysis units. Understanding the relevant benefits, cost restrictions and potential for harm with their use will likely lead to a more consistent approach.

Objectives

The aims of this review were to determine the benefits and harms of using oral protein‐based nutritional supplements to improve the nutritional state of patients with CKD requiring dialysis.

Methods

Criteria for considering studies for this review

Types of studies

All randomised controlled trials (RCTs) and quasi‐RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) comparing oral protein‐based nutritional supplements to no oral protein‐based nutritional supplements or placebo were included without language restriction.

Types of participants

Inclusion criteria

We included adults with CKD on dialysis, which encompasses both HD and PD.

Exclusion criteria

Studies of patients with kidney disease not requiring dialysis, including conservative care and kidney transplant recipients, were excluded.

Types of interventions

Studies comparing oral protein‐based nutritional supplements to no oral protein‐based nutritional supplements or placebo were included. These supplements include renal‐specific supplements (such as Nepro, Novasource Renal, Renilon, and Suplena), essential amino acids or its derivative, as well as protein isolates including whey and soy protein. Studies where standard dietary advice was provided to the non‐intervention group were included. Studies where the control group received intensive dietary education were excluded.

In order to differentiate between products, we have separated the supplements into two groups, complete and non‐complete oral protein‐based nutritional supplements. Complete oral protein‐based nutritional supplements are those that provide all of the required macro‐ and micronutrients to support life. Products that are lacking in any one or more macro‐ or micronutrients, or there was insufficient information provided by the study author, for the purpose of this review have been termed non‐complete.

Types of outcome measures

Assessment of nutritional status in patients receiving dialysis takes into account multiple factors and no single measure of nutritional status is sufficient on its own. We have therefore included both physical and biochemical measures of nutritional status as outcomes.

Primary outcomes

Serum albumin: change in serum albumin level and serum albumin at end of intervention
Other measures of nutritional status (serum prealbumin, weight, body mass index, triceps skinfold thickness, mid‐arm circumference, mid‐arm muscle circumference, lean mass)
Death
Intolerance to therapy

Secondary outcomes

QOL
Cost
Markers of inflammation such as C‐reactive protein (CRP) or Interleukin 6 (IL‐6)
Biochemical adverse effects (e.g. hyperkalaemia, hyperphosphataemia)

Search methods for identification of studies

Electronic searches

Monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL)
Weekly searches of MEDLINE OVID SP
Handsearching of kidney‐related journals and the proceedings of major kidney conferences
Searching of the current year of EMBASE OVID SP
Weekly current awareness alerts for selected kidney and transplant journals
Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov.

Studies contained in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE based on the scope of Cochrane Kidney and Transplant. Details of search strategies, as well as a list of handsearched journals, conference proceedings and current awareness alerts, are available on the Cochrane Kidney and Transplant website.

See Appendix 1 for search terms used in strategies for this review.

Searching other resources

Reference lists of review articles, relevant studies and clinical practice guidelines
Letters seeking information about unpublished or incomplete studies to investigators known to be involved in previous studies

Data collection and analysis

Selection of studies

We used the search strategy described to obtain titles and abstracts of studies relevant to the review. Two authors independently screened the titles and abstracts, and discarded studies that were not applicable; however, studies and reviews that might include relevant data or information on studies were retained initially. Two authors independently assess retrieved abstracts and, if necessary, the full text, of these studies to determine which studies satisfied the inclusion criteria. Any differences were resolved by discussion and where necessary, by consultation with a third author.

Data extraction and management

Two authors independently extracted data using standard data extraction forms. Studies reported in non‐English language journals were translated before assessment. Where more than one publication of one study exists, reports were grouped together and the publication with the most complete data was used in the analyses. Where relevant outcomes were only published in earlier versions, these data were used. Any disagreements between authors regarding study selection or data extraction were resolved by discussion and where necessary, by consultation with a third author.

Assessment of risk of bias in included studies

The following items were independently assessed by two authors using the risk of bias assessment tool (Higgins 2011) (see Appendix 2).

Was there adequate sequence generation (selection bias)?
Was allocation adequately concealed (selection bias)?
Was knowledge of the allocated interventions adequately prevented during the study?
- Participants and personnel (performance bias)
- Outcome assessors (detection bias)
Were incomplete outcome data adequately addressed (attrition bias)?
Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?
Was the study apparently free of other problems that could put it at a risk of bias?

Measures of treatment effect

Dichotomous outcomes (e.g. episodes of treatment intolerance) were expressed as risk ratio (RR) with 95% confidence intervals (CI). Where continuous scales of measurement are used to assess the effects of treatment (e.g. change in serum albumin levels, anthropometric measures), the mean difference (MD) was used, or the standardised mean difference (SMD) if different scales were used. Cost and QOL assessments were assessed with descriptive techniques.

Unit of analysis issues

Studies with non‐standard designs were analysed in this review using the recommended methods for data extraction and analysis described by The Cochrane Collaboration (Higgins 2011).

When considering cross‐over studies, we only included data for end points reported during the first period of study in studies in which the order of receiving treatments was randomised.

When considering studies with multiple treatment groups, we only included data for the relevant treatment groups of interest to this review. We attempted where possible to combine all relevant experimental intervention groups of the study into a single group and to combine all relevant control intervention groups into a single control group to enable a single pair‐wise comparison.

Dealing with missing data

Further information required from the original author were requested by written correspondence and any relevant information obtained in this manner were included in the review. Evaluation of important numerical data such as screened, randomised patients as well as intention‐to‐treat (ITT), as‐treated and per‐protocol population were carefully performed. Attrition rates, for example drop‐outs, losses to follow‐up and withdrawals were investigated. Issues of missing data and imputation methods were critically appraised (Higgins 2011).

For the outcome of change in serum albumin, only a limited number of studies reported this. The review statistician used data from available studies to impute the missing data for this outcome.

Assessment of heterogeneity

Heterogeneity was assessed by visual inspection of the forest plot. Heterogeneity will then be analysed using a Chi² test on N‐1 degrees of freedom, with an alpha of 0.05 used for statistical significance and with the I² test (Higgins 2003). A guide to the interpretation of I² values will be as follows.

0% to 40%: might not be important
30% to 60%: may represent moderate heterogeneity
50% to 90%: may represent substantial heterogeneity
75% to 100%: considerable heterogeneity.

The importance of the observed value of I² depends on the magnitude and direction of treatment effects and the strength of evidence for heterogeneity (e.g. P‐value from the Chi² test, or a confidence interval for I²) (Higgins 2011).

Assessment of reporting biases

Funnel plots were used to assess for the potential existence of small study bias (Higgins 2011).

Data synthesis

Data were pooled using the random‐effects model but the fixed‐effect model will also be used to ensure robustness of the model chosen and susceptibility to outliers.

Subgroup analysis and investigation of heterogeneity

Subgroup analysis was used to explore possible sources of heterogeneity (e.g. participants, interventions and study quality). Pre‐specified subgroup analyses were as follows.

Age
Co‐morbidities
Severity of malnutrition at enrolment
HD versus PD
Reduced versus adequate intake
Dose of oral protein‐based nutritional supplements
Type of oral protein‐based nutritional supplements: we compared complete to non‐complete oral protein‐based nutritional supplements
Duration of oral protein‐based nutritional supplements: we compared < 6 months to ≥ 6 months of intervention

Sensitivity analysis

We performed sensitivity analyses in order to explore the influence of the following factors on effect size.

Repeating the analysis excluding unpublished studies
Repeating the analysis taking account of risk of bias, as specified

'Summary of findings' tables

We presented the main results of the review in 'Summary of findings' tables. These tables present key information concerning the quality of the evidence, the magnitude of the effects of the interventions examined, and the sum of the available data for the main outcomes (Schunemann 2011a). The 'Summary of findings' tables also include an overall grading of the evidence related to each of the main outcomes using the GRADE (Grades of Recommendation, Assessment, Development and Evaluation) approach (GRADE 2008; GRADE 2011). The GRADE approach defines the quality of a body of evidence as the extent to which one can be confident that an estimate of effect or association is close to the true quantity of specific interest. The quality of a body of evidence involves consideration of within‐trial risk of bias (methodological quality), directness of evidence, heterogeneity, precision of effect estimates and risk of publication bias (Schunemann 2011b). We presented the following outcomes in the 'Summary of findings' tables:

Summary of findings Table 1
- Change in serum albumin
- Serum albumin at end of intervention
- Serum prealbumin
- Death
- Intolerance to therapy
Summary of findings Table 2
- Weight
- Body mass index
- Triceps skinfold thickness
- Mid‐arm circumference
- Mid‐arm muscle circumference
- Lean mass
Summary of findings Table 3
- QOL
- Cost
- C‐reactive protein
- IL‐6
- Serum potassium
- Serum phosphate

Results

Description of studies

Results of the search

The process of study selection is outlined in Figure 1. The search conducted to December 2019 identified 90 records, of which 82 were identified by electronic database searches and eight from reference lists of reviews and included articles. Nineteen duplicate publications were removed. After screening titles, abstracts, and full‐text review, 22 studies (33 records) were included and 27 studies (36 records) were excluded. Two studies are awaiting assessment (NCT02371018; Salazar 2008).

Figure 1

Study flow diagram.

Additional data on outcomes of interest and on aspects of the study were sought from sixteen authors. Three authors could not be contacted due to change of place of practice or lack of contact information (Akpele 2004; Hiroshige 2001; Lee 1998). Replies were obtained from five authors. No replies were received from eight authors (Calegari 2011; IHOPE 2019; IMPROVES 2018; Ivarsen 1999; Moretti 2009; Oldrizzi 1999; Salazar 2008; Sohrabi 2016).

Included studies

Twenty‐two studies, enrolling 1278 participants, were included in this review. Detailed descriptions of the study design, participant characteristics, and oral protein‐based nutritional supplements used are provided in the Characteristics of included studies table. All participants were adults on dialysis. Sixteen of the 22 studies included HD participants (Afaghi 2016; AIONID 2013; Benner 2010; Bolasco 2011; Calegari 2011; Fitschen 2017; Fouque 2008; Herrero 2009; Hiroshige 2001; IHOPE 2019; IMPROVES 2018; Ivarsen 1999; Oldrizzi 1999; Sharma 2002c; Sohrabi 2016; Tomayko 2015), four included PD participants (Gonzalez‐Espinoza 2005; Lee 1998; Sahathevan 2018; Teixido‐Planas 2005), and two included both HD and PD participants (Eustace 2000; Moretti 2009).

The duration of treatment was different between the studies: three were for 12 months (IHOPE 2019; Oldrizzi 1999; Teixido‐Planas 2005); nine for six months (Benner 2010; Fitschen 2017; Gonzalez‐Espinoza 2005; Herrero 2009; Hiroshige 2001; IMPROVES 2018; Moretti 2009; Sahathevan 2018; Tomayko 2015); five for three months (Bolasco 2011; Calegari 2011; Fouque 2008; Ivarsen 1999; Lee 1998); one for two months (Afaghi 2016); one for one month (Sharma 2002c); one for 16 weeks (AIONID 2013); one for eight weeks (Sohrabi 2016); and one for 90 days (Eustace 2000).

There were variations with regards to the oral protein‐based nutritional supplements used. Six studies used renal‐specific supplements (AIONID 2013; Benner 2010; Fouque 2008; Herrero 2009; Sharma 2002c; Teixido‐Planas 2005), six used essential amino acids (EAA) or a derivative (Bolasco 2011; Eustace 2000; Fitschen 2017; Hiroshige 2001; Ivarsen 1999; Oldrizzi 1999), and five used soy or whey‐based supplements (IHOPE 2019; IMPROVES 2018; Sahathevan 2018; Sohrabi 2016; Tomayko 2015). Calegari 2011 and Gonzalez‐Espinoza 2005 prepared their own oral protein‐based nutritional supplement, Afaghi 2016 used both an IsoWhey and EAA oral nutritional supplement, and two used other protein supplements (Lee 1998; Moretti 2009).

Several methods were utilised to identify malnourished participants; five studies used the SGA (Calegari 2011; Gonzalez‐Espinoza 2005; Herrero 2009; Lee 1998; Sohrabi 2016), two used low normalised protein nitrogen appearance (nPNA) in addition to serum albumin and body mass index (BMI) (Bolasco 2011; Fouque 2008), and IMPROVES 2018 included participants with clinical protein‐energy wasting as per ISRNM criteria. The remaining 14 studies have been categorised as malnourishment non‐specified for the purpose of this review, which includes studies that have identified malnourished participants based on albumin, BMI alone or both.

Placebos were used in the control group in seven studies (AIONID 2013; Eustace 2000; Fitschen 2017; Hiroshige 2001; IHOPE 2019; Ivarsen 1999; Tomayko 2015).

Excluded studies

We excluded 27 studies. Three were excluded because of a wrong study design: one only randomised for one of three visits (Boudville 2003); one was cluster RCT (NOURISH 2013); and one was only a single session (Sundell 2009). Twenty‐two studies were excluded because the comparison did not meet our inclusion criteria. Two studies compared ONS versus intensive dietary intervention (Akpele 2004; Hernandez Morante 2014), seven studies compared ONS versus ONS plus exercise (ACTINUT 2013; Dong 2011; ISRCTN10251828; Majchrzak 2008; Martin‐Alemany 2016; NCT00244075; NCT02418065), one study compared the effect of high load strength training with or without ONS (Molsted 2013), two studies used non‐protein based supplement (Fukuda 2015; Hosojima 2017), one study used intradialytic parenteral nutrition (Pupim 2002), one study compared two different diets with no supplement used (Koontz 2012), five studies were comparing between different ONS (Jeloka 2013; NCT02989688; Salamon 2018; Sukkar 2012; Weinmeister 2008) and three studies prescribed ONS to all groups with a different intervention (Daud 2012; Liu 2016b; HELPS‐HD 2016).

Two studies were excluded as they were terminated (NCT00179205; NCT00367198). Further information is provided in the Characteristics of excluded studies table.

Studies awaiting classification

NCT02371018 is an open‐label, three‐arm cross‐over RCT comparing HD with nutritional supplement versus standard HD without supplement versus nutritional supplement without HD among adult maintenance HD patients. The study has been completed but the results have yet to be published. No data were available for inclusion in this review. Salazar 2008 compared ONS versus no supplement in non‐diabetic adult maintenance HD patients with nPNA < 1.0 g/kg/day. It is unclear if this was an RCT as only the abstract was available. The author was contacted for clarification. Further information is provided in the Characteristics of studies awaiting classification table.

Risk of bias in included studies

A summary of the risk of bias assessment is provided in Figure 2 and Figure 3. Details for each paper are provided in the Characteristics of included studies table.

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Random sequence generation

Seven studies adequately performed and reported on random sequence generation (AIONID 2013; Eustace 2000; Herrero 2009; IHOPE 2019; Sahathevan 2018; Sohrabi 2016; Teixido‐Planas 2005), with no information provided in the remaining fifteen studies.

Allocation concealment

Five studies reported adequate allocation concealment (AIONID 2013; Eustace 2000; IHOPE 2019; Sahathevan 2018; Teixido‐Planas 2005), with no information provided in the remaining seventeen studies.

Blinding

Performance bias

Six studies reported blinding of participants and personnel (AIONID 2013; Eustace 2000; Fitschen 2017; Hiroshige 2001; Ivarsen 1999; Tomayko 2015) which was achieved with the use of placebos that were similar to the respective supplements used. The participants in Sharma 2002c were not blinded and the authors reported a significant increase in dietary intake by the end of the study month that was not seen in the supplement group; this was judged to be at high risk of bias. The remaining fifteen studies did not provide sufficient information for assessment.

Detection bias

Biochemical outcomes

Blinding of assessment for biochemical outcomes was graded as low risk for 19 studies; six reported blinding of personnel (AIONID 2013; Eustace 2000; Fitschen 2017; Hiroshige 2001; Ivarsen 1999; Tomayko 2015), one was deemed low risk given the use of placebo (IHOPE 2019), and 12 were deemed low risk despite the absence of placebo arm (Benner 2010; Bolasco 2011; Calegari 2011; Fouque 2008; Gonzalez‐Espinoza 2005; Herrero 2009; IMPROVES 2018; Moretti 2009; Sahathevan 2018; Sharma 2002c; Sohrabi 2016; Teixido‐Planas 2005). The remaining three studies were graded as unclear risk.

Anthropometric outcomes

Blinding of assessment for anthropometric outcomes was graded as low risk for five studies as there was blinding of personnel (Eustace 2000; Fitschen 2017; Hiroshige 2001; Ivarsen 1999; Tomayko 2015). Seventeen studies were graded as unclear risk; four did not evaluate this outcome (Afaghi 2016; AIONID 2013; Benner 2010; Sohrabi 2016) and the remaining 13 did not provide sufficient information for assessment.

Quality of life and treatment tolerance

Blinding of assessment for this outcome was graded as low risk for two studies as the personnel and participants were blinded (Eustace 2000; Fitschen 2017). Nine studies were graded as high risk of bias as participants were not blinded (Afaghi 2016; Calegari 2011; Fouque 2008; Herrero 2009; IMPROVES 2018; Sahathevan 2018; Sharma 2002c; Sohrabi 2016; Teixido‐Planas 2005). The remaining 11 studies were graded as unclear risk of which eight did not evaluate this outcome (AIONID 2013; Benner 2010; Bolasco 2011; Gonzalez‐Espinoza 2005; Hiroshige 2001; Ivarsen 1999; Moretti 2009; Tomayko 2015).

Incomplete outcome data

Eleven studies were graded as low risk of bias as the number of study exclusions was small or equal between the groups, and reason for exclusions were provided (Afaghi 2016; AIONID 2013; Bolasco 2011; Calegari 2011; Eustace 2000; Fitschen 2017; Gonzalez‐Espinoza 2005; Hiroshige 2001; IHOPE 2019; Sharma 2002c; Sohrabi 2016). Eight studies were graded as high risk of bias as the number of exclusions was high or reasons for exclusions not provided (Benner 2010; Fouque 2008; Herrero 2009; IMPROVES 2018; Ivarsen 1999; Moretti 2009; Sahathevan 2018; Teixido‐Planas 2005). The remaining three studies were graded as unclear risk.

Selective reporting

Three studies reported all pre‐specified or expected outcomes from their study protocol (Herrero 2009; IHOPE 2019; Sahathevan 2018). Five studies did not report all outcomes identified in their study protocol and were graded as high risk of bias (Afaghi 2016; AIONID 2013; Benner 2010; Hiroshige 2001; Sohrabi 2016). The remaining 14 studies have been graded as unclear risk due to insufficient information for assessment.

Other potential sources of bias

Two studies did not provide details on baseline characteristics to determine variability, and were considered at risk of bias (Benner 2010; Ivarsen 1999). Seven studies were graded as unclear as there were statistically significant differences for certain characteristics at baseline (AIONID 2013; Calegari 2011; Eustace 2000; Herrero 2009; IHOPE 2019; IMPROVES 2018; Sahathevan 2018). Two studies were graded as unclear risk due to insufficient information for assessment (Lee 1998; Oldrizzi 1999). The remaining 11 studies have been graded as low risk as the baseline variables were similar between the groups.

Effects of interventions

See: Summary of findings 1 Summary of findings Table 1; Summary of findings 2 Summary of findings Table 2; Summary of findings 3 Summary of findings Table 3

Primary outcomes

Serum albumin

Change in serum albumin level

Sixteen studies were included in this analysis. Three studies reported change in serum albumin level (Eustace 2000; Sohrabi 2016; Teixido‐Planas 2005). Using data from the Teixido‐Planas 2005 study, the best estimated correlation between baseline and final measurements was 0.33 for the supplement group and 0.54 for the control group. This was used to derive the mean change in serum albumin with standard deviation for the remaining thirteen studies (AIONID 2013; Bolasco 2011; Calegari 2011; Fitschen 2017; Fouque 2008; Gonzalez‐Espinoza 2005; Herrero 2009; IHOPE 2019; IMPROVES 2018; Lee 1998; Moretti 2009; Sahathevan 2018; Sharma 2002c). Oral protein‐based nutritional supplements probably leads to a higher mean change in serum albumin compared to the control group (Analysis 1.1 (16 studies, 790 participants): MD 0.19 g/dL, 95% CI 0.05 to 0.33; I² = 84%; moderate certainty evidence). There was considerable heterogeneity.

Subgroup analysis

Subgroup analysis based on dialysis modality showed that the increase in mean change of albumin was demonstrated in the HD cohort (Analysis 1.2.1 (10 studies, 526 participants): MD 0.28 g/dL, 95% CI 0.11 to 0.46; P = 0.001 for overall effect) but not seen in the PD cohort (Analysis 1.2.2 (4 studies, 168 participants): MD ‐0.02 g/dL, 95% CI ‐0.22 to 0.17; P = 0.81 for overall effect). The test for subgroup differences was Chi² = 5.19, df = 1 (P = 0.02), I² = 80.7%. The increase was also more evident in participants who were malnourished at enrolment (Analysis 1.3.1 (8 studies, 405 participants): MD 0.31 g/dL, 95% CI 0.10 to 0.52; P = 0.003 for overall effect). The test for subgroup differences was Chi² = 4.18, df = 1 (P = 0.04), I² = 76.1%. There was no difference in results based on duration of intervention (Analysis 1.4) and type of oral protein‐based nutritional supplements used (Analysis 1.5).

Sensitivity analysis

Given the means supplied by Teixido‐Planas 2005 were based on ITT, the correlation could be different if the mean for only those followed up completely were available. Sensitivity analyses were hence performed; one assuming correlation is 0.2 for the supplement group and 0.3 for the control group, and one assuming correlation of 0.5 for the supplement group and 0.75 for the control group. The results were not altered on both analyses (Analysis 1.6; Analysis 1.7).

Serum albumin at end of intervention

Fourteen studies reported serum albumin at end of intervention (AIONID 2013; Bolasco 2011; Calegari 2011; Fitschen 2017; Fouque 2008; Gonzalez‐Espinoza 2005; Herrero 2009; IHOPE 2019; IMPROVES 2018; Lee 1998; Moretti 2009; Sahathevan 2018; Sharma 2002c; Teixido‐Planas 2005). Apart from the IMPROVES 2018 study, the baseline means between both groups were similar. Oral protein‐based nutritional supplements probably leads to a higher mean serum albumin at the end of the intervention compared to the control group (Analysis 1.8 (14 studies, 715 participants): MD 0.14 g/dL, 95% CI 0 to 0.27; I² = 80%; moderate certainty evidence). There was considerable heterogeneity.

Subgroup analysis

Subgroup analysis based on dialysis modality showed that HD participants who were given oral protein‐based nutritional supplements had a higher mean serum albumin at the end of the study (Analysis 1.9.1 (9 studies, 498 participants): MD 0.21 g/dL, 95% CI 0.03 to 0.38; P = 0.02 for overall effect). The test for subgroup differences was Chi² = 4.46, df = 1 (P = 0.03), I² = 77.6%. Participants who were malnourished during enrolment also had a higher mean serum albumin when given oral protein‐based nutritional supplements compared to the control group (Analysis 1.10.1 (7 studies, 377 participants): MD 0.25 g/dL, 95% CI 0.02 to 0.47; P = 0.03 for overall effect). The test for subgroup differences was Chi² = 2.67, df = 1 (P = 0.10), I² = 62.5%. Heterogeneity remains high in both analyses. There was no difference for duration of intervention and type of intervention (data not shown).

Other measures of nutritional status

Serum prealbumin

Four studies reported serum prealbumin at the end of the intervention (AIONID 2013; Benner 2010; Fouque 2008; Lee 1998). One study reported this outcome in a different format and was excluded from this meta‐analysis (Eustace 2000). Oral protein‐based nutritional supplements may result in a higher serum prealbumin at end of intervention compared to control (Analysis 1.11 (4 studies, 225 participants): MD 2.81 mg/dL, 95% CI 2.19 to 3.43; I² = 0%; low certainty evidence).

Weight

Eight studies reported weight at the end of the intervention (Bolasco 2011; Calegari 2011; Fouque 2008; Herrero 2009; Lee 1998; Sahathevan 2018; Sharma 2002c; Teixido‐Planas 2005). Three studies reported this outcome in different formats and were excluded from this meta‐analysis (Hiroshige 2001; Ivarsen 1999; Tomayko 2015). Participants given oral protein‐based nutritional supplements may have little or no difference in weight compared to the control group at end of intervention (Analysis 1.12 (8 studies, 365 participants): MD 2.83 kg, 95% CI ‐0.43 to 6.09; I² = 57%; low certainty evidence). There was moderate heterogeneity.

Subgroup analysis

Subgroup analyses were performed for dialysis modality, nutritional status at enrolment, type and duration of intervention. This did not significantly alter the results (data not shown).

Body mass index (BMI)

Nine studies reported BMI at end of intervention (Bolasco 2011; Calegari 2011; Fitschen 2017; Gonzalez‐Espinoza 2005; Herrero 2009; IHOPE 2019; Lee 1998; Sahathevan 2018; Sharma 2002c). Four studies reported this outcome in different formats and were excluded from this meta‐analysis (Eustace 2000; Fouque 2008; Moretti 2009; Sohrabi 2016). Oral protein‐based nutritional supplements probably makes little or no difference to BMI at the end of the intervention (Analysis 1.13 (9 studies, 368 participants): MD ‐0.04 kg/m², 95% CI ‐0.74 to 0.66; I² = 8%; moderate certainty evidence).

Subgroup analysis

Subgroup analyses were performed for dialysis modality, nutritional status at enrolment, type and duration of intervention. This did not significantly alter the results (data not shown).

Triceps skinfold thickness (TSF)

Six studies reported TSF at the end of the intervention (Gonzalez‐Espinoza 2005; Herrero 2009; IMPROVES 2018; Lee 1998; Sahathevan 2018; Teixido‐Planas 2005). Two studies reported this outcome in different formats and were excluded from this meta‐analysis (Calegari 2011; Ivarsen 1999). It is uncertain whether oral protein‐based nutritional supplements improves TSF compared to control group because the certainty of this evidence is very low (Analysis 1.14 (6 studies, 367 participants): MD 0.81 mm, 95% CI ‐1.59 to 3.21; I² = 66%; very low certainty evidence). Subgroup analysis was not performed due to the small number of studies.

Mid‐arm circumference (MAC)

Two studies reported MAC at the end of the intervention (Lee 1998; Sahathevan 2018). One study reported this outcome in a different format and was excluded from this meta‐analysis (IMPROVES 2018). It is uncertain whether there is any difference in MAC between the supplement and control group (Analysis 1.15 (2 studies, 97 participants): MD 0.50 cm, 95% CI ‐1.56 to 2.56, I² = 64%; very low certainty evidence).

Mid‐arm muscle circumference (MAMC)

Four studies reported MAMC at the end of the intervention (Gonzalez‐Espinoza 2005; Herrero 2009; Sahathevan 2018; Teixido‐Planas 2005). One study reported this outcome in a different format and was excluded from this meta‐analysis (IMPROVES 2018). Oral protein‐based nutritional supplements may result in a higher end of intervention MAMC compared to control (Analysis 1.16 (4 studies, 216 participants): MD 1.33 cm, 95% CI 0.24 to 2.43; I² = 29%; low certainty evidence).

Lean mass

Five studies reported lean mass at the end of the intervention (Calegari 2011; Fitschen 2017; Lee 1998; Sahathevan 2018; Teixido‐Planas 2005). Two studies reported this outcome in different formats and were excluded from this meta‐analysis (Hiroshige 2001; Tomayko 2015). Compared to the control group, oral protein‐based nutritional supplements may make little or no difference to lean mass (Analysis 1.17 (5 studies, 189 participants): MD 1.27kg, 95% CI ‐1.61 to 4.15; I² = 39%; low certainty evidence).

Death

This outcome was not reported as a study outcome by any of the included studies.

Intolerance to therapy

Gastrointestinal intolerance

The most reported therapy intolerance was gastrointestinal intolerance. Six studies contributed to this outcome (Afaghi 2016; Fouque 2008; IHOPE 2019; Sahathevan 2018; Sohrabi 2016; Teixido‐Planas 2005). There may be little or no difference in the risk of developing gastrointestinal side effects between the supplement and control group (Analysis 1.18.1 (6 studies, 426 participants): RR 2.81, 95% CI 0.58 to 13.65, I² = 48%; low certainty evidence). There was moderate heterogeneity.

Secondary outcomes

Quality of life

Eight studies reported this outcome (Calegari 2011; Eustace 2000; Fitschen 2017; Fouque 2008; Herrero 2009; IHOPE 2019; IMPROVES 2018; Sahathevan 2018). Only three studies used placebos in their control groups (Eustace 2000; Fitschen 2017; IHOPE 2019). Data were not combined for analysis as three different QOL instruments were used and the data were not reported in a way that allowed meta‐analysis. In one study, the Kidney Disease Quality of Life (KDQOL) questionnaire was used (Fitschen 2017). No significant differences in QOL measures were observed between the two groups, however patient satisfaction with care reduced over time during the six month follow‐up period. Two studies used the Short Form 12 (SF12) questionnaire (Eustace 2000; IHOPE 2019). In the study by Eustace 2000, there was a mean difference in change in mental health score of 5.0 (95% CI 0.8 to 9.2, P = 0.02) between the supplement and placebo groups. There was no significant difference in physical health score between the two groups. IHOPE 2019 reported a decline in mental health status over time, but there was no change in physical status over time.

Five studies used the SF‐36 QOL questionnaire. Data however could not be combined as two studies reported changes in scores from baseline (Fouque 2008; Sahathevan 2018), one study reported the baseline and end readings (Calegari 2011), one study reported improvement in scores but only provided the P values (IMPROVES 2018), and the final study provided the baseline readings only and reported no improvement at end of study (Herrero 2009). In Fouque 2008, although there was no significant difference between the changes in scores, they reported a trend towards improvement in the individual component scores in the supplement group, and deterioration in the control group. Sahathevan 2018 reported no change in scores within the supplement group. In contrast, the control group had a significant decrease in SF‐36 physical component score of 6.62 (mean change = ‐6.62 ± 16.53, P = 0.020) and SF‐36 total score of 4.89 (mean change = ‐4.89 ± 14.46, P = 0.047). Calegari 2011 reported significant difference between groups in the physical role functioning and bodily pain domains with improvements seen in the supplement group. IMPROVES 2018 reported a significant improvement in the final score in the supplement group however the readings were not provided. In contrast, there was no change in scores reported by Herrero 2009, again no readings were provided.

Cost

Four studies reported on cost of their individual oral protein‐based nutritional supplements (Calegari 2011; Eustace 2000; Gonzalez‐Espinoza 2005; Sharma 2002c). Data were not combined for analysis as the data were not reported in a way that allowed meta‐analysis. The Aminess N supplement prescribed in Eustace 2000 costs approximately 5 US dollars a day. The commercially available supplement prescribed in Sharma 2002c cost 2 US dollars a day, whereas its home‐prepared blend cost 0.25 US dollars a day. Gonzalez‐Espinoza 2005 prescribed an egg albumin‐based protein supplement that cost 0.60 US dollars per day. The monthly cost of the supplement prescribed in Calegari 2011 was 2 US dollars.

Markers of inflammation

C‐reactive protein (CRP)

Eight studies reported CRP at the end of the intervention (AIONID 2013; Bolasco 2011; Fitschen 2017; Fouque 2008; IHOPE 2019; IMPROVES 2018; Lee 1998; Sahathevan 2018). Three studies reported this outcome in different formats and were excluded from this meta‐analysis (Herrero 2009; Sohrabi 2016; Tomayko 2015). It is uncertain whether there is any difference in CRP between the supplement and control group (Analysis 1.19 (8 studies, 436 participants): MD ‐0.06 mg/dL, 95% CI ‐0.29 to 0.18; I² = 9%; very low certainty evidence).

Subgroup analysis

Subgroup analyses were performed for dialysis modality, nutritional status at enrolment, type and duration of intervention. This did not significantly alter the results (data not shown).

Interleukin 6 (IL‐6)

Two studies reported IL‐6 at the end of the intervention (AIONID 2013; IHOPE 2019). Two studies reported this outcome in different formats and were excluded from this meta‐analysis (Sohrabi 2016; Tomayko 2015). It is uncertain whether there is any difference in IL‐6 between the supplement and control group because the certainty of this evidence is very low (Analysis 1.20 (2 studies, 62 participants): MD ‐5.05 pg/mL, 95% CI ‐23.91 to 13.81, I² = 28%; very low certainty evidence).

Biochemical adverse effects

Serum potassium

Seven studies reported serum potassium at the end of the intervention (AIONID 2013; Calegari 2011; Fitschen 2017; Fouque 2008; Gonzalez‐Espinoza 2005; IMPROVES 2018; Sharma 2002c). Two studies reported this outcome in different formats and were excluded from this meta‐analysis (Ivarsen 1999; Tomayko 2015). It is uncertain whether oral protein‐based nutritional supplements results in any difference in serum potassium at end of intervention compared to the control group (Analysis 1.21 (7 studies, 357 participants): MD 0.05 mEq/L, 95% CI ‐0.24 to 0.35; I² = 70%; very low certainty evidence).

Subgroup analyses

Subgroup analyses were performed for dialysis modality, nutritional status at enrolment, type and duration of intervention. This did not significantly alter the results (data not shown).

Serum phosphate

Seven studies reported serum phosphate at the end of the intervention (AIONID 2013; Calegari 2011; Fitschen 2017; Fouque 2008; IMPROVES 2018; Sahathevan 2018; Sharma 2002c). Three studies reported this outcome in different formats and were excluded from this meta‐analysis (Gonzalez‐Espinoza 2005; Sohrabi 2016; Tomayko 2015). It is uncertain whether oral protein‐based nutritional supplements results in any difference in serum phosphate at end of intervention compared to the control group (Analysis 1.22 (7 studies, 403 participants): MD 0.03 mg/dL, 95% CI ‐0.42 to 0.47; I² = 49%; very low certainty evidence).

Sensitivity analysis

Only seven studies used placebos in their control groups (AIONID 2013; Eustace 2000; Fitschen 2017; Hiroshige 2001; IHOPE 2019; Ivarsen 1999; Tomayko 2015). Exclusion of studies without placebos resulted in a higher end study serum phosphate in the participants who received oral protein‐based nutritional supplements compared with placebo (Analysis 1.23, MD 0.74mg/dL, 95% CI 0.08 to 1.41).

Other subgroup analysis

Apart from the subgroup analyses already detailed, further subgroup analyses were not performed for the remaining pre‐specified groups due to the lack of data to allow analysis.

Other sensitivity analyses

Apart from serum phosphate, the exclusion of studies without placebos did not result in any significant effect on the other outcomes.

Exclusion of unpublished studies (Benner 2010; Herrero 2009; Ivarsen 1999; Oldrizzi 1999) did not result in any significant change in results.

Discussion

Summary of main results

This review included 1278 participants from twenty‐two studies which compared oral protein‐based nutritional supplements to placebo or no supplement in patients with CKD requiring dialysis. All participants were adults on maintenance dialysis of whom 79% were on HD and 21% PD. The follow‐up period ranged from one to 12 months. In our review, although there was some variability in the ONS prescribed in the individual studies, all ONS were protein‐based.

This review demonstrated that giving oral protein‐based nutritional supplements to patients with CKD requiring dialysis probably leads to a higher change in serum albumin compared with the control group. There was considerable heterogeneity and subgroup analysis demonstrated that the mean change of serum albumin was more evident in participants receiving HD than those receiving PD (Analysis 1.2), and in participants who were malnourished at study entry compared to not malnourished (Analysis 1.3).

Consistent with this, use of oral protein‐based nutritional supplements probably leads to higher end of intervention serum albumin, which in subgroup analysis was more pronounced in participants on HD than PD (Analysis 1.9), and to a lesser extent malnourished versus not malnourished participants (Analysis 1.10). Oral protein‐based nutritional supplements may also result in a high serum prealbumin at end of intervention compared to the control group, but this was strongly influenced by one study (Benner 2010).

With regards to the anthropometric outcomes, oral protein‐based nutritional supplements may result in a higher MAMC at end of intervention. Oral protein‐based nutritional supplements made little or no difference to weight, BMI and lean mass. It is uncertain whether the administration of oral protein‐based nutritional supplements changes TSF or MAC.

Due to very low quality of evidence, it is uncertain whether oral protein‐based nutritional supplements changes CRP, IL‐6, serum potassium, or serum phosphate. The serum phosphate at the end of the intervention was lower in the control group when limited to placebo‐controlled studies only, however this only encompassed two studies and should be interpreted with caution.

Oral protein‐based nutritional supplements may result in little or no difference in the risk of gastrointestinal intolerance compared to the control group, however the studies were limited by high attrition bias and lack of blinding. It was not possible to draw conclusions about cost or QOL, and deaths were not reported as a study outcome in any of the included studies.

Overall completeness and applicability of evidence

This review found that oral protein‐based nutritional supplements probably improves some nutritional markers, in particular serum albumin and prealbumin. In this review, MAMC was the only anthropometric measure which may increase in participants receiving oral protein‐based nutritional supplements.

A higher increase in end intervention serum albumin and mean change in serum albumin was observed among participants receiving HD compared to PD. This may reflect the low level of albumin loss from participants receiving PD coupled with the extra calories provided by glucose‐based PD fluids. A similar difference was evident for participants who were given oral protein‐based nutritional supplements who were malnourished at time of enrolment. This may reflect improvement in those at risk and more likely to benefit, or regression to the mean if low serum albumin was a criteria for classifying them as malnourished. Current practice recommendations support the need for nutritional support in dialysis patients who are unable to meet their protein and energy requirements and are hence at higher risk of malnutrition (KDOQI 2000).

However, there was substantial heterogeneity in our findings that may not be entirely explained by results of subgroup analyses. There were variations in the type and duration of intervention used between the studies that may contribute to the heterogeneity observed. This was addressed by using a random‐effects model and subgroup analyses, although the latter was not always possible due to the small number of studies. Furthermore, given most studies did not collect or report on dietary intake data, unmeasured increases in nutritional intake may be a confounding factor.

This review did not identify improvement in the other physical and biochemical measures of nutritional status outlined. Although serum albumin is the most commonly reported outcome in studies of ONS, its use as a marker of nutritional status is limited. Serum albumin can be affected by various factors including inflammation, fluid status, and presence of systemic disease (Jones 1997; Sahathevan 2018). Therefore, it is uncertain if the improvement in serum albumin identified in this review truly reflects nutritional repletion. Death was not reported as a study outcome in any of the included studies. It has been well‐established that hypoalbuminaemia has a strong and consistent association with death in previous studies of dialysis patients (Iseki 1993; Kalantar‐Zadeh 2005a; Kaysen 1998; Lowrie 1990). However, improvement in serum albumin has not been demonstrated to correlate with hard endpoints such as death (Weiner 2014). Further RCTs are required to determine if improvement in serum albumin with oral protein‐based nutritional supplements or the use of oral protein‐based nutritional supplements translates to improvement in nutritional status and clinically relevant outcomes such as death and hospitalisations.

About 79% of all participants included in this review were on HD. Given the small number of PD participants, and the lack of an increase in albumin in the PD subgroup, the findings of this review must be applied cautiously to this group.

Although we initially planned to determine the incidence of hyperkalaemia and hyperphosphataemia, this data was not reported in the included studies. Given that serum potassium and phosphate at end of intervention was not dissimilar between the two groups, it is unlikely that there was an increase in episodes of hyperkalaemia or hyperphosphataemia with the use of oral protein‐based nutritional supplements. However, there was very low certainty evidence for both these outcomes. A sensitivity analysis using only placebo‐controlled RCTs demonstrated that supplement group had a slightly higher serum phosphate than the control group. This finding should be interpreted with caution given the analysis was limited to two studies. There was little or no difference in the risk of gastrointestinal intolerance between the two groups. However, there was high attrition bias and lack of blinding in the studies reporting this outcome.

Quality of the evidence

Of the twenty‐two studies included in this review, one was only available as abstract. Adequate sequence generation and allocation concealment was only demonstrated in a quarter of the studies, with the remaining being unclear. Very few studies were adequately blinded which would result in high detection bias particularly for the self‐reported outcomes such as treatment intolerance and QOL measures. Furthermore, participants in the control arm could change their dietary habits given the lack of blinding as demonstrated in the study by Sharma 2002c, which could act as a confounder. Most studies were also small and inadequately powered, with some outcomes consisting only of a small number of studies.

The certainty of the evidence for biochemical outcomes ranged from very low to moderate; there was considerable heterogeneity as well as small number of studies and patient numbers (summary of findings Table 1; summary of findings Table 3). Similarly, the certainty of evidence for anthropometric outcomes ranged from very low to moderate (summary of findings Table 2). For patient reported outcomes such as intolerance to therapy, the certainty of evidence was considered low because of small patient numbers and events included in the analysis, and high risk of detection and attrition bias in many studies (summary of findings Table 1).

Potential biases in the review process

The main limitation of this review is the low number of studies and small number of participants in individual studies resulting in a small sample size. To minimise the likelihood of bias, an exhaustive search for published studies was performed with broad search terms. Information was also obtained from authors of two unpublished studies for inclusion in this review. The results of funnel plots suggest no publication bias for the effects of oral protein‐based nutritional supplements on both mean change in serum albumin and end of intervention serum albumin (Figure 4; Figure 5).

Figure 4

Funnel plot of comparison: 1 Oral protein‐based nutritional supplement versus control, outcome: 1.1 Change in serum albumin [g/dL].

Figure 5

Funnel plot of comparison: 1 Oral protein‐based nutritional supplement versus control, outcome: 1.8 Serum albumin (end of intervention) [g/dL].

Agreements and disagreements with other studies or reviews

The findings of this review suggest a role of oral protein‐based nutritional supplements in improving some measures of nutritional status of CKD patients requiring dialysis, in particular serum albumin and prealbumin. This is consistent with two previous systematic reviews. The review by Stratton 2005 conducted over 10 years ago was comprised of five RCTs and thirteen non‐RCTs examining the potential benefits of enteral support, which includes oral supplements, tube feeding or both. They found a significant increase in serum albumin of 0.23 g/dL. A more recent review demonstrated a greater mean change in serum albumin (0.158g/dL) in the supplement group compared to the control group (Liu 2018). This result is similar to our finding of 0.19 g/dL. Compared with the review by Liu 2018, our review had a higher number of participants (1278 versus 589), although it is important to note that 11 of the 15 studies in Liu 2018 were also included in our review. Both previous reviews found little change in serum potassium or phosphate and had insufficient comparable data on death or QOL ‐ again, comparable with the outcomes of our review. However, our findings also suggest oral protein‐based nutritional supplements may improve MAMC, a finding which was not demonstrated in other reviews. To our knowledge, there are no systematic reviews on the cost of oral protein‐based nutritional supplements in CKD patients on dialysis.

Figure 1

Study flow diagram.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 4

Funnel plot of comparison: 1 Oral protein‐based nutritional supplement versus control, outcome: 1.1 Change in serum albumin [g/dL].

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 5

Funnel plot of comparison: 1 Oral protein‐based nutritional supplement versus control, outcome: 1.8 Serum albumin (end of intervention) [g/dL].

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.1

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 1: Change in serum albumin

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.2

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 2: Change in serum albumin: Subgroup analysis for dialysis modality

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.3

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 3: Change in serum albumin: Subgroup analysis for nutritional status

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.4

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 4: Change in serum albumin: subgroup analysis for duration of intervention

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.5

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 5: Change in serum albumin: subgroup analysis for type of intervention

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.6

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 6: Change in serum albumin: sensitivity analysis ‐ correlation 0.2 for supplement group, 0.3 for control group

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.7

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 7: Change in serum albumin: sensitivity analysis ‐ correlation 0.5 for supplement group, 0.75 for control group

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.8

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 8: Serum albumin (end of intervention)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.9

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 9: Serum albumin: Subgroup analysis for dialysis modality

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.10

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 10: Serum albumin: subgroup analysis for nutritional status

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.11

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 11: Serum prealbumin

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.12

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 12: Weight

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.13

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 13: Body mass index (BMI)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.14

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 14: Triceps skinfold thickness (TSF)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.15

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 15: Mid‐arm circumference (MAC)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.16

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 16: Mid‐arm muscle circumference (MAMC)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.17

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 17: Lean mass

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.18

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 18: Intolerance to therapy

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.19

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 19: C‐reactive protein (CRP)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.20

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 20: IL‐6

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.21

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 21: Serum potassium

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.22

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 22: Serum phosphate

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.23

Comparison 1: Oral protein‐based nutritional supplement versus control, Outcome 23: Serum phosphate: sensitivity analysis ‐ placebo‐controlled studies only

Ir a la figura de la revisiónAbrir en una pestaña nueva

Summary of findings 1. Summary of findings Table 1

Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Oral protein‐based nutritional supplement versus placebo or no treatment for people with CKD requiring dialysis
Patient or population: people with CKD requiring dialysis Intervention: oral protein‐based nutritional supplement Comparison: placebo or no treatment
Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with control	Risk difference with Oral protein‐based nutritional supplement
Change in serum albumin	790 (16)	⊕⊕⊕⊝ MODERATE ¹	‐‐	The mean change in serum albumin was ‐0.01 g/dL	MD 0.19 g/dL higher (0.05 higher to 0.33 higher)
Serum albumin (end of intervention)	715 (14)	⊕⊕⊕⊝ MODERATE ¹	‐‐	The mean serum albumin (end of intervention) was 3.55 g/dL	MD 0.14 g/dL higher (0 to 0.27 higher)
Serum prealbumin	225 (4)	⊕⊕⊝⊝ LOW ²	‐‐	The mean serum prealbumin was 27.18 mg/dL	MD 2.81 mg/dL higher (2.19 higher to 3.43 higher)
Death	‐‐	‐‐	‐‐	Not reported
Intolerance to therapy ‐ Gastrointestinal intolerance	426 (6)	⊕⊕⊝⊝ LOW ³	RR 2.81 (0.58 to 13.65)	20 per 1,000	57 more per 1,000 (12 fewer to 279 more)
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CKD: chronic kidney disease; CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Down‐graded because of high heterogeneity, some of which are not explained by the pre‐specified subgroups (inconsistency) ² Down‐graded because of high attrition bias in two studies, a moderate number of studies and patient assessed with results heavily influenced by one study (Benner 2010) (imprecision) ³ Down‐graded because of high risk of detection bias in most studies, high risk of attrition bias in three studies, and small number of events in low number of patients assessed with wide CI (imprecision).

Summary of findings 1. Summary of findings Table 1

Ir a la tabla de la revisión

Summary of findings 2. Summary of findings Table 2

Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Oral protein‐based nutritional supplement versus placebo or no treatment for people with CKD requiring dialysis
Patient or population: people with CKD requiring dialysis Intervention: oral protein‐based nutritional supplement Comparison: placebo or no treatment
Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with control	Risk difference with Oral protein‐based nutritional supplement
Weight	365 (8)	⊕⊕⊝⊝ LOW ¹	‐‐	The mean weight was 59.51 kg	MD was 2.83 kg higher (0.43 lower to 6.09 higher)
Body mass index	368 (9)	⊕⊕⊕⊝ MODERATE ²	‐‐	The mean body mass index (BMI) was 25.18 kg/m²	MD was 0.04 kg/m² lower (0.74 lower to 0.66 higher)
Triceps skinfold thickness	367 (6)	⊕⊝⊝⊝ VERY LOW ³	‐‐	The mean triceps skinfold thickness (TSF) was 14.14 mm	MD was 0.81 mm thicker (1.59 lower to 3.21 higher)
Mid‐arm circumference	97 (2)	⊕⊝⊝⊝ VERY LOW ⁴	‐‐	The mean mid‐arm circumference (MAC) was 24.99 cm	MD was 0.5 cm wider (1.56 lower to 2.56 higher)
Mid‐arm muscle circumference	216 (4)	⊕⊕⊝⊝ LOW ⁵	‐‐	The mean mid‐arm muscle circumference (MAMC) was 22.32 cm	MD 1.33 cm wider (0.24 higher to 2.43 higher)
Lean mass	189 (5)	⊕⊕⊝⊝ LOW ⁶	‐‐	The mean lean mass was 42.95 kg	MD 1.27 kg higher (1.61 lower to 4.15 higher)
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CKD: chronic kidney disease; CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Down‐graded because of high attrition bias in four studies, and moderate heterogeneity (inconsistency). CI is wide and effect shows appreciable benefit and harm, but shows benefit in subgroup treated for 6 months or more ² Down‐graded because of low number of patients assessed, and CI is wide and effect shows appreciable benefit and harm (imprecision) ³ Down‐graded because of high attrition bias in over half the studies included, high heterogeneity (inconsistency), and CI is wide and effect shows appreciable benefit and harm (imprecision) ⁴ Down‐graded because of high/unclear biases in both studies, substantial heterogeneity (inconsistency), low number of studies and patients assessed (imprecision), and CI is wide and effect shows appreciable benefit and harm ⁵ Down‐graded because of high attrition bias in three studies, and moderate number of studies and patients assessed (imprecision) ⁶ Down‐graded because of high attrition bias in two studies, and moderate number of studies and low patients assessed with wide CI (imprecision).

Summary of findings 2. Summary of findings Table 2

Ir a la tabla de la revisión

Summary of findings 3. Summary of findings Table 3

Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects^ (95% CI)**
Oral protein‐based nutritional supplement versus placebo or no treatment for people with CKD requiring dialysis
Patient or population: people with CKD requiring dialysis Intervention: oral protein‐based nutritional supplement Comparison: placebo or no treatment
Outcomes	No. of participants (studies)	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Risk with control	Risk difference with Oral protein‐based nutritional supplement
Quality of life	662 (8)	‐‐	‐‐	Data could not be combined for meta‐analysis. See 'Effects of interventions' for results
Cost	117 (4)	‐‐	‐‐	Data could not be combined for meta‐analysis. See 'Effects of interventions' for results
CRP	436 (8)	⊕⊝⊝⊝ VERY LOW ¹	‐‐	The mean C‐reactive protein (CRP) was 4.28 mg/dL	MD 0.06 mg/dL lower (0.29 lower to 0.18 higher)
IL‐6	62 (2)	⊕⊝⊝⊝ VERY LOW ²	‐‐	The mean IL‐6 was 28.47 pg/mL	MD 5.05 pg/mL lower (23.91 lower to 13.81 higher)
Serum potassium	357 (7)	⊕⊝⊝⊝ VERY LOW ³	‐‐	The mean serum potassium was 4.90 mEq/L	MD 0.05 mEq/L higher (0.24 lower to 0.35 higher)
Serum phosphate	403 (7)	⊕⊝⊝⊝ VERY LOW ⁴	‐‐	The mean serum phosphate was 4.84 mg/dL	MD 0.03 mg/dL higher (0.42 lower to 0.47 higher)
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CKD: chronic kidney disease; CI: confidence interval; RR: risk ratio; CRP: C‐reactive protein; IL‐6: interleukin 6
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect
¹ Down‐graded because of high attrition bias in three studies, surrogate marker for death (indirectness), and results heavily weighted on one study (Lee 1998) (imprecision). ² Down‐graded because of low number of studies and patients assessed with results heavily influenced by one study (IHOPE 2019). CI is wide and effect shows appreciable benefit and harm (imprecision) ³ Down‐graded because of high heterogeneity (inconsistency), surrogate marker for episodes of hyperkalaemia (indirectness), moderate number of studies and CI is wide and effect shows appreciable benefit and harm (imprecision) ⁴ Down‐graded because of moderate heterogeneity (inconsistency), surrogate marker for episodes of hyperphosphataemia (indirectness), moderate number of studies and CI is wide and effect shows appreciable benefit and harm (imprecision)

Summary of findings 3. Summary of findings Table 3

Ir a la tabla de la revisión

Comparison 1. Oral protein‐based nutritional supplement versus control

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Change in serum albumin Show forest plot	16	790	Mean Difference (IV, Random, 95% CI)	0.19 [0.05, 0.33]

1.2 Change in serum albumin: Subgroup analysis for dialysis modality Show forest plot	14	694	Mean Difference (IV, Random, 95% CI)	0.21 [0.06, 0.37]

1.2.1 HD participants only	10	526	Mean Difference (IV, Random, 95% CI)	0.28 [0.11, 0.46]
1.2.2 PD participants only	4	168	Mean Difference (IV, Random, 95% CI)	‐0.02 [‐0.22, 0.17]
1.3 Change in serum albumin: Subgroup analysis for nutritional status Show forest plot	16	790	Mean Difference (IV, Random, 95% CI)	0.19 [0.05, 0.33]

1.3.1 Malnourished	8	405	Mean Difference (IV, Random, 95% CI)	0.31 [0.10, 0.52]
1.3.2 Malnourishment not specified	8	385	Mean Difference (IV, Random, 95% CI)	0.07 [‐0.04, 0.18]
1.4 Change in serum albumin: subgroup analysis for duration of intervention Show forest plot	16	790	Mean Difference (IV, Random, 95% CI)	0.19 [0.05, 0.33]

1.4.1 < 6 months	8	293	Mean Difference (IV, Random, 95% CI)	0.26 [0.13, 0.39]
1.4.2 ≥ 6 months	8	497	Mean Difference (IV, Random, 95% CI)	0.13 [‐0.11, 0.38]
1.5 Change in serum albumin: subgroup analysis for type of intervention Show forest plot	16	790	Mean Difference (IV, Random, 95% CI)	0.19 [0.05, 0.33]

1.5.1 Complete nutritional supplement	5	246	Mean Difference (IV, Random, 95% CI)	0.14 [0.03, 0.24]
1.5.2 Non‐complete nutritional supplement	11	544	Mean Difference (IV, Random, 95% CI)	0.21 [0.01, 0.40]
1.6 Change in serum albumin: sensitivity analysis ‐ correlation 0.2 for supplement group, 0.3 for control group Show forest plot	16	790	Mean Difference (IV, Random, 95% CI)	0.19 [0.06, 0.33]

1.7 Change in serum albumin: sensitivity analysis ‐ correlation 0.5 for supplement group, 0.75 for control group Show forest plot	16	790	Mean Difference (IV, Random, 95% CI)	0.19 [0.05, 0.33]

1.8 Serum albumin (end of intervention) Show forest plot	14	715	Mean Difference (IV, Random, 95% CI)	0.14 [0.00, 0.27]

1.9 Serum albumin: Subgroup analysis for dialysis modality Show forest plot	13	666	Mean Difference (IV, Random, 95% CI)	0.15 [0.01, 0.29]

1.9.1 HD participants only	9	498	Mean Difference (IV, Random, 95% CI)	0.21 [0.03, 0.38]
1.9.2 PD participants only	4	168	Mean Difference (IV, Random, 95% CI)	‐0.04 [‐0.18, 0.11]
1.10 Serum albumin: subgroup analysis for nutritional status Show forest plot	14	715	Mean Difference (IV, Random, 95% CI)	0.14 [0.00, 0.27]

1.10.1 Malnourished	7	377	Mean Difference (IV, Random, 95% CI)	0.25 [0.02, 0.47]
1.10.2 Malnourishment not specified	7	338	Mean Difference (IV, Random, 95% CI)	0.03 [‐0.09, 0.15]
1.11 Serum prealbumin Show forest plot	4	225	Mean Difference (IV, Random, 95% CI)	2.81 [2.19, 3.43]

1.12 Weight Show forest plot	8	365	Mean Difference (IV, Random, 95% CI)	2.83 [‐0.43, 6.09]

1.13 Body mass index (BMI) Show forest plot	9	368	Mean Difference (IV, Random, 95% CI)	‐0.04 [‐0.74, 0.66]

1.14 Triceps skinfold thickness (TSF) Show forest plot	6	367	Mean Difference (IV, Random, 95% CI)	0.81 [‐1.59, 3.21]

1.15 Mid‐arm circumference (MAC) Show forest plot	2	97	Mean Difference (IV, Random, 95% CI)	0.50 [‐1.56, 2.56]

1.16 Mid‐arm muscle circumference (MAMC) Show forest plot	4	216	Mean Difference (IV, Random, 95% CI)	1.33 [0.24, 2.43]

1.17 Lean mass Show forest plot	5	189	Mean Difference (IV, Random, 95% CI)	1.27 [‐1.61, 4.15]

1.18 Intolerance to therapy Show forest plot	6		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only

1.18.1 Gastrointestinal intolerance	6	426	Risk Ratio (M‐H, Random, 95% CI)	2.81 [0.58, 13.65]
1.19 C‐reactive protein (CRP) Show forest plot	8	436	Mean Difference (IV, Random, 95% CI)	‐0.06 [‐0.29, 0.18]

1.20 IL‐6 Show forest plot	2	62	Mean Difference (IV, Random, 95% CI)	‐5.05 [‐23.91, 13.81]

1.21 Serum potassium Show forest plot	7	357	Mean Difference (IV, Random, 95% CI)	0.05 [‐0.24, 0.35]

1.22 Serum phosphate Show forest plot	7	403	Mean Difference (IV, Random, 95% CI)	0.03 [‐0.42, 0.47]

1.23 Serum phosphate: sensitivity analysis ‐ placebo‐controlled studies only Show forest plot	2	76	Mean Difference (IV, Random, 95% CI)	0.74 [0.08, 1.41]

Comparison 1. Oral protein‐based nutritional supplement versus control

Ir a la tabla de la revisión

	Idioma de la Revisión Cochrane Escoja su idioma de preferencia para las revisiones Cochrane y otros contenidos. Las secciones sin una traducción aparecerán en inglés.

	Idioma de la web Escoja su idioma de preferencia para la web de la Biblioteca Cochrane.

Idioma de la Revisión Cochrane

Idioma de la web

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

PICO

PICO

Population

Intervention

Comparison

Outcome

一般語訳

透析を必要とする慢性腎臓病の人に対するタンパク質を含む経口栄養補助食品

Resumen visual

Authors' conclusions

Implications for practice

Implications for research

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Inclusion criteria

Exclusion criteria

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

'Summary of findings' tables

Results

Description of studies

Results of the search

Included studies

Excluded studies

Studies awaiting classification

Risk of bias in included studies

Allocation

Random sequence generation

Allocation concealment

Blinding

Performance bias

Detection bias

Biochemical outcomes

Anthropometric outcomes

Quality of life and treatment tolerance

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Primary outcomes

Serum albumin

Change in serum albumin level

Serum albumin at end of intervention

Other measures of nutritional status