EBRSR [Evidence-Based Review of Stroke Rehabilitation]

16 Nutritional Interventions Following Stroke

Norine Foley MSc, Robert Teasell MD, Marina Richardson MSc, Joshua Wiener BHSc, Hillel Finestone MD

Last Updated: March 2018

Abstract Nutritional status following stroke can have a negative impact on functional recovery and mortality. Complications associated with malnutrition include a greater incidence of infections and pressure sores, and longer lengths of hospital stays. Clinical nutritional management requires effective methods of assessment, an understanding of the underlying causes of nutritional deficiencies, and effective methods of administering nutrients via feeding techniques and supplementation. In this review, the prevalence of malnutrition post stroke is evaluated and markers used to identify deficiencies are discussed. A summarization of potential causes of nutritional deficiencies is provided, including metabolic rate, nutrient intake, and gastrointestinal impairments. Interventions of enteral feeding and oral supplementation, as well as treatments for dysphagia, are discussed.

16. Nutritional Interventions Following Stroke pg. 1 of 42 www.ebrsr.com Key Points Prevalence of Malnutrition

• Malnutrition is a relatively common problem post stroke.

Development of Malnutrition

• It is unclear whether stroke directly contributes to the development of malnutrition.

• It is unclear as to whether significant gastrointestinal impairments develop post stroke.

• Patients may consume fewer calories and protein than required post stroke.

• Impaired glucose regulation post stroke is associated with greater risk of mortality.

• Treatments such as insulin injections and aerobic exercise may be effective in improving glucose regulation but not functional outcomes post stroke.

• A single dose of Vitamin D2 (100,000 units) post stroke may significantly raise Vitamin D levels for up to 16 weeks.

• A monthly dose of oral Vitamin D (60,00 units) post stroke, along with a single injection of Vitamin D (600,000 units) and a daily dose of oral calcium (1g), may significantly raise Vitamin D levels and reduce risk of mortality.

• Statin treatment post stroke may significantly reduce cholesterol and low-density lipoprotein (LDL) levels, increase high-density lipoprotein (HDL) levels, and prevent recurrent stroke. More research is needed to verify this effect.

Nutritional Interventions

• Gastric feeding tubes are associated with fewer complications and greater consumed intake post stroke compared to nasogastric tubes, but no significant difference has been found regarding poor outcome or mortality.

• Oral nutritional supplementation post stroke may improve calorie-protein intake, decrease lipid levels, and increase HDL levels, although it may not prevent poor outcome or mortality.

• High-intensity dysphagia therapy may improve swallowing function and reduce time to resuming a normal diet post stroke, although it may not prevent poor outcome or mortality.

• Rates of one-year survival and functional recovery in patients with PEG feeding tubes within the community post stroke are highly varied.

• Long-term NG tube feeding may result in greater malnourishment, although its use is usually restricted to the first week post stroke.

• Use of total parenteral nutrition has not been studied in the stroke population.

Dr. Robert Teasell Parkwood Institute, 550 Wellington Road, London, Ontario, Canada, N6C 0A7 Phone: 519.685.4000 ● Web: www.ebrsr.com ● Email: [email protected]

16. Nutritional Interventions Following Stroke pg. 2 of 42 www.ebrsr.com Table of Contents Abstract ...... 1 Key Points ...... 2 Table of Contents ...... 3 16.1 Assessment of Nutritional Status Following Stroke ...... 4 16.2 Prevalence of Malnutrition Following Stroke ...... 5 16.3 Factors Associated With the Development of Malnutrition ...... 11 16.3.1 Metabolic Rate Following Stroke ...... 11 16.3.2 Gastrointestinal Impairments Following Stroke ...... 12 16.3.3 Nutrient Intake Following Stroke ...... 14 16.3.4 Glucose Regulation Following Stroke ...... 14 16.3.5 Vitamin D Deficiency Following Stroke ...... 16 16.3.6 Lipid Profiles Following Stroke ...... 18 16.4 Nutritional Interventions Following Stroke ...... 19 16.4.1 Enteral Feeding ...... 19 16.4.2 Oral Supplementation ...... 23 16.4.3 Dysphagia Treatment ...... 27 16.4.4 Long-Term Enteral Feeding ...... 29 16.4.5 Total Parenteral Nutrition ...... 30 16.4.6 Effect of Nutritional Interventions on Changes in Nutritional Parameters ...... 30 16.5 Cochrane Reviews of Nutritional Interventions Following Stroke ...... 31 Summary ...... 33 References ...... 35

16. Nutritional Interventions Following Stroke pg. 3 of 42 www.ebrsr.com 16.1 Assessment of Nutritional Status Following Stroke Decline in nutritional status following stroke is important given its potential negative impact on functional recovery and mortality in multiple medical and surgical populations. Preliminary results from the international FOOD Trial reported that poor nutritional status was associated with an increase in the odds of death and dependency at 6 months after adjusting for a number of confounders (OR 1.82, 95%CI 1.34-2.47) (FOOD Trial Collaboration, 2003).

Poor nutrition has been found to predict lower functional status following stroke. In a study focusing on the functional consequences of malnutrition in stroke rehabilitation, patients’ serum albumin used as an marker of nutritional status was associated with poorer functional mobility, increased complications, and lower self-care scores (Aptaker et al., 1994). Gariballa et al. (1998a) investigated the associations between a variety of anthropometric and biochemical parameters assessed on admission to hospital and outcome following stroke among 201 patients. After adjusting for age, sex, medications, stroke severity, and comorbid conditions, serum albumin was related to an increase in death at 3 months. Each decline of 1 g/L in serum albumin was associated with a 1.13-fold increase in death at follow-up.

Davalos et al. (1996) reported that malnutrition after the first week of stroke among 104 patients was associated with an increased risk of poor outcome (death or dependency) at one month, a greater incidence of infections and pressure sores, and longer lengths of hospital stays. Patients who were considered malnourished had an elevated risk of death or poor outcome at 30 day follow-up (OR 3.5, 95%CI 1.2-10.2). Types of food consumed may also influence mortality risk, as Sharma et al. (2013) revealed that a higher level of meat consumption was associated with an elevated risk of stroke mortality among female participants. Diet has also been used to assess for risk of developing a stroke, such as a study that found a higher Mediterranean Diet Score (MeDi) score was significantly associated with a lower risk of stroke among males (Chan et al., 2013).

Currently, there is no universally accepted gold standard for the assessment of nutritional status. The identification of malnutrition is typically based on the evaluation of a combination of biochemical and anthropometric markers, and is inferred from a single or multiple values falling outside of specific population reference ranges or below a certain percentile within these ranges. Since the combination of markers used and the cut-off values are chosen arbitrarily, reports of malnutrition are widely varied. As a result, the true incidence of malnutrition following stroke is likely unknown. Table 16.1.1 presents some the more commonly used biochemical indicators used as well as their limitations.

Table 16.1.1 Biochemical Markers of Nutritional Status (American Dietetic Association, 2000) Measure Limitations Serum Albumin Large body pool Poor specificity to nutritional changes Not specific to nutritional status ↓ with acute illness Serum Transferrin Not specific to nutritional status ↓ with acute illness Thyroxin Binding Prealbumin Not specific to nutritional status ↓ with acute illness Retinol Binding Protein Not specific to nutritional status Total Lymphocyte Count Poor sensitivity and specificity

16. Nutritional Interventions Following Stroke pg. 4 of 42 www.ebrsr.com Unfortunately, many of these nutrition-sensitive markers are affected independently by factors associated with stroke (or any other acute illness), complicating the process of evaluating the response to nutritional interventions. Although both albumin and prealbumin are used extensively in nutritional assessment, the hepatic production of these two proteins is known to be down-regulated during periods of acute illness, independent of nutritional status (Fleck, 1989; Gabay & Kushner, 1999). Hypoalbuminemia has been repeatedly shown to be associated with morbidity and mortality, but the causal mechanism is not clear. For example, Akner and Cederholm (2001) did not demonstrate a relationship between protein and caloric intake and serum albumin in their institutionalized elderly population. Further difficulties arise due to the fact that biochemical markers of nutrition can change rapidly, whereas a change in nutritional status is considered more latent and takes longer to manifest. In addition, the presence of concurrent infection or elevations in temperature can affect serum markers, mimicking signs of malnutrition.

Measures of nutrition assessment also include indicators of skeletal muscle mass and subcutaneous fat stores including weight, mid-arm muscle circumference, and skinfold thickness. While a decline of these indicators may be associated with the development of malnutrition, factors secondary to stroke may also affect the sensitivity of these measures. Skeletal muscle losses may occur over prolonged periods of time as a result of atrophy, secondary to immobility (Deitrick et al., 1948; Schonheyder et al., 1954). It may be difficult to differentiate between these losses and those that are associated with inadequate food intake, adding to the difficulties of evaluating nutritional status. Malnutrition is considered to be a state that develops over time in response to inadequate intake relative to need, and results in gradual weight loss with associated losses of skeletal muscle mass and subcutaneous fat stores. Since the identification of malnutrition among the majority of studies was made on the basis of both anthropometric and biochemical markers, it is possible that studies reporting a higher percentage of malnourished patients conducted measurements at a later point in the hospitalization period and reflected non-nutritional changes in body composition.

16.2 Prevalence of Malnutrition Following Stroke The prevalence of malnutrition following stroke has been reported to be between 6% and 62%. If widened to include secondary criteria from two studies, the range of estimates broadened to 1.3% and 73%. Some of this variability can be attributed to differences in patient characteristics and the timing of assessments among studies. However, a substantial proportion of the variation in estimates may also be explained by the heterogeneity of nutritional assessment. Among the 22 trials reviewed below, 18 different assessment methods were used. Only 5 trials used previously validated assessment methods: an “informal assessment”, the Subjective Global Assessment (SGA), and the Mini Nutritional Assessment (MNA). The assessment methods used in the remaining studies used had not been previously validated.

The aforementioned valid assessment tools were created for different purposes. The informal assessment was developed to classify patients into groups based on nutritional state within the context of the large, multi-centred FOOD Trials. The SGA was designed for use in the prediction of risk for complications following general surgery based on pre-operative nutritional state. The MNA was developed as a screening and assessment tool to identify geriatric patients at risk for malnutrition. Both the SGA and MNA have been validated subsequently for use in other disease or injury states, but further validation of these tools for the assessment of individuals with stroke is lacking. In a study by Kim et al. (2013), a group of 35 patients with stroke aged 60 to 89 years were assessed for nutritional status using laboratory measures, as well as the SGA and MNA. There was strong correlation between SGA and objective measures when patient nutrition status was classified (normal, mildly malnourished, moderately malnourished, or severely malnourished, r=0. 449), and between MNA and objective

16. Nutritional Interventions Following Stroke pg. 5 of 42 www.ebrsr.com measures when patient nutrition status was dichotomized (well-nourished vs. malnourished or at risk of malnutrition, r=0.520). The criteria used to detect malnutrition in studies of patients post stroke, and the prevalence of malnutrition detected within those studies, are presented in Table 16.2.1.

Table 16.2.1 Summary of Studies Examining Malnutrition Prevalence Post Stroke Author (Year) Study Type (PEDro Score) Criteria Prevalence Sample Size Axelsson et al. (1988) ≥2 variables below reference limit: • 16% within 4d of symptom onset Observational • Serum albumin <38g/L (male), <37g/L (female) • 22% at hospital discharge (n=78) NInitial=100 • Serum prealbumin <0.18g/L • Serum transferrin <1.7g/L (male), <1.5g/L (female) • Body weight <80% of reference population • Tricep skinfold thickness: 4 levels based on age • Arm muscle circumference: 4 levels based on age DePippo et al. (1994) ≥1 variable below reference limit: • 6.1% at any point between hospital RCT (5) • Serum albumin < 25g/L admission and discharge NInitial=115 • Sustained ketonuria without glycosuria >2wk (TPSmedian=4.6wk) Unossen et al. (1994a) ≥3 variables below reference limit (1 from each of • 8% within 2d of symptom onset Observational biochemical, anthropometric, and skin measures): NInitial=50 • Serum albumin <36g/L • Serum prealbumin <0.20g/L (male), <0.18g/L (female) • Body weight <80% of reference population • Arm muscle circumference: 4 levels based on age and gender • Tricep skinfold <6mm (male), <12mm (female) • Delayed hypersensitivity skin testing <10mm Finestone et al. (1995) ≥2 variables below reference limit: • 49% at hospital admission Observational • Serum albumin <35g/L (TPSmean=22d) NInitial=49 • Serum transferrin <2.0g/L • 34% at 1mo (n=32) • Total lymphocyte count <1800n/mm3 • 22% at 2mo (n=9) • Body weight <90% of reference population or <95% • 19% at 2-4mo (n=42) of usual weight • Body mass index <20 • Sum of 4 skinfolds <5th percentile of reference population • Midarm muscle circumference <5th percentile of reference population Davalos et al. (1996) ≥1 variable below reference limit: • 16.3% within 24hr of hospital Observational • Serum albumin <35g/L admission NInitial=104 • Midarm muscle circumference <5th percentile of • 26.4% at 1wk (n=91) reference population • 35% at 2wk (n=43) • Tricep skinfold <10th percentile of reference population Choi-Kwon et al. (1998) ≥3 variables below reference limit (1 from each of • 87% at acute period (<3mo) Observational biochemical, anthropometric, and skin measures): NInitial=88 • Serum albumin <35g/L • Total lymphocyte count <1500/mm3 • Lean body mass, abdominal skinfold, subscapular skinfold, triceps skinfold <80% of reference population

16. Nutritional Interventions Following Stroke pg. 6 of 42 www.ebrsr.com • Body mass index <20 Aquilani et al. (1999) ≥2 variables below reference limit (including 1 weight • 30% at hospital admission Cohort measure): (TPSmean=30d) N=150 • Serum albumin <35g/L • Total lymphocyte count <1800n/mm3 • Body weight <90% of reference population or <95% of usual weight • Arm muscle area <5th percentile of reference population Westergren et al. (2001a) ≥2 variables below reference limit (including 1 weight • 8% within 24hr of symptom onset Observational measure): • 29% at 1mo N=24 • Serum albumin <36g/L • 33% at 3mo • Body mass index <20 • Body weight <80% of reference population or <95% of usual weight • Subnormal triceps skinfold or mid-upper arm muscle circumference Westergren et al. (2001a) Subjective Global Assessment (Modified Version): • 32% within 6d of hospital admission Observational A: Well nourished N=162 B: Well nourished but at risk of malnourishment C: Suspected malnourishment D: Severely malnourished (C, D = Malnourished) Davis et al. (2004) Subjective Global Assessment: • 16% within 24hr of symptom onset Observational A: Well nourished N=185 B: Moderately malnourished or suspected malnourishment C: Severely malnourished (B, C = Malnourished) Dennis et al. (2005a) Clinical judgement: • 8.6% at hospital admission RCT (7) Malnourished, normal, or overweight N=859 Dennis et al. (2005b) Clinical judgement: • 7.8% within 7d of symptom onset RCT (7) Malnourished, normal, or overweight N=4023 Martineau et al. (2005) Subjective Global Assessment (Patient-Generated): • 19.2% within 2d of symptom onset Case Series A: Well nourished N=73 B: Moderately malnourished or suspected malnourishment C: Severely malnourished (B, C = Malnourished)

Hama et al. (2005) • Serum albumin <40g/L • 1d of hospital admission (TPSmean=44d) Cohort • Body mass index (BMI) <19 • 22% based on albumin N=51 • 57% based on BMI Crary et al. (2006) • Mini Nutritional Assessment score <23.5 • 26.3% at hospital admission Observational N=76 Brynningsen et al. (2007) ≥2 variables below reference limit: • 35% within 1wk of symptom onset Cohort • Serum albumin <550micromol/L • 33% at 5wk NInitial=100 • Serum transferrin <49micromol/L • 20% at 3mo • Arm muscle circumference <10th percentile of • 22% at 6mo (n=89) reference population

16. Nutritional Interventions Following Stroke pg. 7 of 42 www.ebrsr.com • Tricep skinfold <10th percentile of reference population Poels et al. (2006) Primary criteria Primary Cohort • ≥1 variable below reference limit: • 35% at hospital admission NInitial=69 • Body weight <95% at 1mo, <90% at 6mo (TPSmean=34d) • BMI <18 for <65yr, <22 for ≥65yr • 3% at 4wk (n=60) Secondary criteria Secondary • ≥1 variable below reference limit: • 73% at hospital admission • Serum albumin <35g/L • 54% at 4wk • Fat free mass ≤6kg/m2 (male), 15 kg/m2 (female) • Tricep skinfold <90% of 12.5mm (male), 16.5 mm (female) • Mid arm muscle circumference <90% of 25.3cm (male), 23.3 cm (female) Yoo et al. (2008) ≥1 variable below reference limit: • 12.2% within 24 hr of symptom onset Cohort • Serum albumin <30g/L • 19.8% at 1wk N=131 • Serum prealbumin <0.10g/L • Serum transferrin <1.5g/L • Body weight <95% at 1wk, <90% at 3mo • Body weight <80% of reference population Chai et al. (2008) • Serum albumin <35g/L • 8.2% at chronic period (>6mo) Observational • Body mass index <18.5 NInitial=61 Lim & Choue (2010) Subjective Global Assessment (Patient-Generated): • 74% at hospital admission Cohort A: Well nourished (TPSmean=60d) NInitial=73 B: Moderately malnourished or suspected malnourishment C: Severely malnourished (B, C = Malnourished) Crary et al. (2013) • Serum prealbumin <0.05g/L • 32% at hospital admission Cohort • 33% at 7d NInitial=67 Mosselman et al. (2013) • Mini Nutritional Assessment score <17 • 5% at 2-5d after hospital admission Cohort • 26% between 9-12d (n=23) NInitial=73 Paquerau et al. (2014) • Mini Nutritional Assessment (Short Form) • 47.9% at acute period (<3mo) Observational • Body mass index NInitial=71 • Weight

Dysphagia Eleven studies recruited patients both with and without dysphagia. Greater proportions of patients with dysphagia were classified as malnourished compared to patients with normal swallowing function in four trials: 19/59 vs 10/14, p=0.007 (Martineau et al. 2005); 16/24 vs 15/67, p<0.001 (Davalos et al., 1996); 15/23 vs 9/26, p=0.032 (Finestone et al. 1995); 4/5 vs 17/56, p=0.044 (Chai et al., 2008). Poels et al. (2006) also reported that a greater proportion of subjects with dysphagia was malnourished, although the result was not statistically significant (4/20 vs 4/40, p= 0.233). Crary et al. (2006) did not demonstrate a significant association between the dysphagia and malnutrition assessed within several days of stroke (OR 1.0, 95%CI 0.4-2.8). In a subsequent study by Crary et al. (2013), this finding was repeated, and a significant relationship between dysphagia and dehydration was reported.

In a systematic review of 8 studies, Foley et al. (2009a) reported that the odds of being malnourished were increased given the presence of dysphagia following stroke. However, they also suggested that

16. Nutritional Interventions Following Stroke pg. 8 of 42 www.ebrsr.com the relationship may not be causal. While stroke size and location are the greatest determinants of swallowing function, the presence of dysphagia is itself an indicator of greater stroke severity. As well, the type of treatment for dysphagia (e.g. feeding tube) may affect study results concerning its role in the development of malnutrition.

Type and Severity of Stroke The relationship between stroke severity and malnutrition was examined in three studies (Davis et al., 2004; Dennis et al., 2005a; Yoo et al., 2008). In all of these studies, severity was assessed using the National Institutes of Health Stroke Scale (NIHSS) and was examined during the first several days following acute stroke. Increasing stroke severity was associated with baseline malnutrition in one of these studies (Yoo et al., 2008). Only one study examined the relationship between stroke type and malnutrition (Choi-Kwon et al., 1998). The prevalence of malnutrition reported in this study was much higher among patients suffering from intracerebral hemorrhagic than ischemic stroke, though the authors suggested that the result was likely attributable to differences in pre-existing malnutrition between groups. Diet may also play a role in stroke, as Tuttolomundo et al. (2015) reported a negative correlation between NIHSS score and Mediterranean Diet Score (MeDi) score, with greater severity found to be associated with a lower-standard dietary score. Compared with lacunar and cardioembolic infarcts, atherosclerotic stroke was associated with a lower MeDi score. However, Mangat et al. (2013) did not find any significant relationship between dietary patterns and stroke characteristics (type, severity, and prognosis).

Variability of Assessment Criteria With the single exception of the two related FOOD trials, no two studies used the same criteria. Serum albumin and measures of weight were common to almost all of the studies. However, the cut-off points used to distinguish well-nourished from malnourished patients were chosen arbitrarily, as were the choice of reference populations, which can affect the degree and proportion of patients who are considered to be malnourished. For instance, most studies used a cut-off point for serum albumin level of 3.5 g/L, while Hama et al. (2005) used a higher point of 4.0 g/L and DePippo et al. (1994)used a much lower point of 2.5 g/L.

There was variability in the percentiles chosen for anthropometric cut-off points, if they were identified. Choi-Kwon et al. (1998) used an undefined reference population, while Unosson et al. (1994b) chose specific, age-dependent cut-off points that did not appear to be based on population norms. Three studies used population-based national survey results as their reference standards (Axelsson et al., 1988; Davalos et al., 1996; Finestone et al., 1995). However, the cut-off values chosen were inconsistent: one used the 5th percentile (Finestone et al., 1995) and another used the 10th percentile (Davalos et al., 1996). Axelsson et al. (1988) and Choi-Kwon et al. (1998) simply made reference to “low values” when they defined their cut-off threshold. Implicit in the assessment of malnutrition is that thin people, relative to the rest of the reference population, are malnourished. The anthropometric and biochemical markers used in the previously summarized studies are presented in Table 16.2.2.

Individual Nutrition Markers Used in Reviewed Studies Biochemical Markers Study Albumin Transferrin TLC Prealbumin Ketonuria Axelsson et al. (1988) X X X DePippo et al. (1994) X X Unosson et al. (1994a) X X X Finestone et al. (1995) X X X Davalos et al. (1996) X

16. Nutritional Interventions Following Stroke pg. 9 of 42 www.ebrsr.com Choi-Kwon et al. (1998) X X Aquilani et al. (1999) X X Westergren et al. (2001a) X Hama et al. (2005) X Poels et al. (2006) X Brynningsen et al. (2007) X X Yoo et al. (2008) X X X Chai et al. (2008) X Crary et al. (2013) X Anthropometric Markers Study Weight Fat (TSF, SS) Muscle (AMC, MAMC) Axelsson et al. (1988) X X X DePippo et al. (1994) Unosson et al. (1994a) X X X Finestone et al. (1995) X X X Davalos et al. (1996) X X Choi-Kwon et al. (1998) X X X Aquilani et al. (1999) X X Westergren et al. (2001a) X X X Hama et al. (2005) X Poels et al. (2006) X X X Brynningsen et al. (2007) X X Yoo et al. (2008) X Chai et al. (2008) X Paquerau et al. (2014) X Composite Clinical Assessments Westergren et al. (2001b), Davis et al. (2004), Dennis et al. (2005a, 2005b), Martineau et al. (2005), Crary et al. (2006), Lim & Choue (2010), Mosselman et al. (2013), Paquerau et al. (2014) TLC=Total Lymphocyte Count SS=Skinfold Sum MAMC=Midarm Muscle Circumference TSF=Triceps Skinfold AMC=Arm Muscle Circumference

In a systematic review of eighteen studies, Foley et al. (2009b) reported that the prevalence of malnutrition following stroke ranged from 6.1% to 62%. Seventeen different methods of nutritional assessment were used, but only four trials used previously validated assessment methods: the informal assessment, SGA, and MNA. The nutritional assessment methods used in the remaining studies used had not been previously validated. The authors suggested that the wide array of nutritional assessment tools contributed to the wide range of malnutrition estimates.

Timing of Assessment The timing of the assessments may also explain the variability in the reported prevalence of malnutrition. Among the studies that assessed nutritional status within 1 week of stroke, the frequency of malnutrition was <20% in 10 of the studies (Axelsson et al., 1988; Davalos et al., 1996; Davis et al., 2004; Dennis et al., 2005a, 2005b; Martineau et al., 2005; Unosson et al., 1994b; Westergren et al., 2001a; Westergren et al., 2001b; Yoo et al., 2008). Frequency of malnutrition was 26% at 10 days after admission to hospital in 1 study (Mosselman et al., 2013). Among the trials that assessed nutritional state between 22 and 44 days following stroke, the frequency of malnutrition ranged from 30% to 49% in 4 trials (Aquilani et al., 1999; Finestone et al., 1995; Hama et al., 2005; Poels et al., 2006).

16. Nutritional Interventions Following Stroke pg. 10 of 42 www.ebrsr.com Foley et al. (2009a) noted that the pre-stroke nutritional status of patients may be a potential confounding issue. In their systematic review, none of the studies were able to determine nutritional status prior to stroke, although five studies used anthropometric measures shortly after onset that may be comparable to pre-stroke measurements. Moreover, it remains unclear as to whether malnutrition was pre-existing at the time of assessment or if malnutrition was the consequence of stroke.

Conclusions Regarding the Incidence of Malnutrition

The prevalence of malnutrition varies from 6 - 62% post stroke, depending on timing of assessment and criteria used to define malnutrition.

There is currently no “gold standard” for the assessment of nutritional status, and various methods of detection may be used.

Malnutrition is a relatively common problem post stroke.

16.3 Factors Associated With the Development of Malnutrition Malnutrition in any disease state develops over a period of time as a result of either increased metabolism or inadequate nutritional intake. There may be occasions when these two factors are superimposed. The development of malnutrition may also be hastened if the gastrointestinal tract is compromised and nutrient absorption is impaired. The evidence with respect to the potential contributions of these mechanisms is reviewed.

16.3.1 Metabolic Rate Following Stroke Increased metabolic has been attributed to the effects of the acute phase response, mediated largely through the effects of cytokines and counter-regulatory hormones, following injury or disease (Staal-van den Brekel et al., 1995; Young et al., 1985). Elevations of peripheral plasma catecholamines, cortisol, glucagons, IL-6, IL-1RA and acute phase proteins have been well described following stroke (Beamer et al., 1995; Fassbender et al., 1994a; Fassbender et al., 1994b; Ferrarese et al., 1999; Muir et al., 1999; Murros et al., 1993; Syrjanen et al., 1989). Prolonged elevations of these compounds may lead to the depletion of muscle and fat, which may contribute to the development of malnutrition. Szczudlik et al. (2004) found that morning IL-6 levels independently predicted evening/night cortisol levels, and both were found to be higher in stroke patients than healthy controls. The authors postulated that ischemic stroke may stimulate the release of IL-6 and may be responsible for regulation of the hypothalamic- pituitary-adrenal axis. Ormstad et al. (2013) suggested that significant reductions in tryptophan and tyrosine may be indicative of a reduced capacity for 5-hydroxytryptamine (5HT) and catecholamines within the brain. Furthermore, IL-10 was found to be positively correlated with tryptophan levels therefore suggesting that higher levels of IL-10 facilitate higher tryptophan availability for 5HT synthesis, suggesting that anti-inflammatory cytokines may prevent a reduction in 5HT synthesis (Ormstad et al., 2013). The levels of various acute phase reactants following stroke are summarized in Table 16.3.1.1.

16. Nutritional Interventions Following Stroke pg. 11 of 42 www.ebrsr.com

Table 16.3.1.1 Summary of Studies Examining Acute Phase Reactants Post Stroke Author (Year) Sample Size Time Post Stroke Indicator (Response) Syrjanen et al. (1989) 50 Within 72hr • CRP () • SAA () • ACT () Murros et al. (1993) 105 Up to 1wk • Cortisol () Fassbender et al. (1994a) 23 Up to 7d • ACTH (-) • Cortisol ( until 5d) Fassbender et al. (1994b) 19 Up to 3d • IL-1ß () • IL-6 () • TNF () Beamer et al. (1995) 50 Up to 6d • IL-1RA () • IL-6 () • CRP (-) • Fibrinogen () Ferrarese et al. (1999) 40 Up to 90d • IL-6 ( 1-30d) • TNF ( 1-90d) Muir et al. (1999) 228 Within 72hr • CRP () Selakovic et al. (2002) 53 Within 48hr • Cortisol () Szczudlik et al. (2004) 22 Within 24hr • IL-6 () • Cortisol () Ben-Assayag et al. (2007) 219 Up to 6mo • CRP () Camerlingo et al. (2011) 387 Within 3hr • CRP () Hasani et al. (2011) 120 Within 24hr • CRP () Ormstad et al. (2013) 45 Within 72hr • IL-6 () • IL-10 () • Tryptophan (↓) • Tyrosine (↓) Shaafi et al. (2014) 45 Up to 90d • IL-6 () (-) indicates no change ACT (α-antichymotrypsin) IL-1ß (Interleukin-1ß) CRP (C-reactive protein) ACTH (adrenocorticotropic hormone) IL-6 (Interleukin-6) SAA (serum amyloid A protein) TNF (tumour necrosis factor) IL-1RA (Interleukin-1 receptor antagonist)

Conclusions Regarding Metabolism Following Stroke

There is insufficient evidence regarding malnutrition during the acute phase of stroke.

It is unclear whether stroke directly contributes to the development of malnutrition.

16.3.2 Gastrointestinal Impairments Following Stroke The major effect on the gastrointestinal tract following stroke is impairment of oral, pharyngeal, and esophageal functions, which is manifested as dysphagia (see Chapter 15 Dysphagia and Aspiration). Dysphagia may resolve spontaneously in the days following stroke or persist for much longer periods. For a minority of patients, severe dysphagia precludes safe oral feeding and so alternative strategies are required. In 3 studies, the odds of developing malnutrition increased significantly following acute hospital admission at both 1 week post stroke and upon admission to an inpatient rehabilitation unit at approximately 3 weeks post stroke (Davalos et al., 1996; Finestone et al., 1995; FOOD Trial Collaboration, 2003). Although the mechanism was not explored, decreased intake or delayed enteral

16. Nutritional Interventions Following Stroke pg. 12 of 42 www.ebrsr.com feeding may have contributed to a decline in nutritional status. At admission to hospital shortly after stroke, nutritional status was unrelated to dysphagia in 2 other studies (Crary et al., 2006; Crary et al., 2013). The latter study reported a relationship between dysphagia and dehydration at both hospital admission and at 7 days post stroke (Crary et al., 2013). The association between dysphagia and malnutrition is shown in Figure 16.3.2.1.

Figure 16.3.2.1 Association between Dysphagia and Malnutrition

Stroke patients do not appear to be at increased risk for stress ulcer formation or gastric bleeding (Ullman & Reding, 1996). Ogata et al. (2014) reported that gastrointestinal bleeding is a rare occurrence, with a rate of 1.4% in their multicentre study. The authors highlighted significant risk factors for gastrointestinal bleeding including absence of dyslipidemia, history of peptic ulcer, and severity of neurological deficit; no association was found with non-steroidal anti-inflammatory, anticoagulant, and antiplatelet pharmacology. In their systematic review, Schaller et al. (2006) suggested that gastric hemorrhage, small/large intestine dysfunction, gastric motility, and dysphagia may occur post stroke, as they are all modulated through the central nervous system. However, the authors argued that improvements in cerebral and molecular scanning, particularly regarding changes in pharmacodynamics, may elucidate the mechanisms behind these complications. Camara-Lemarroy et al. (2014) speculated that intestinal infection may be the result of post-stroke immunodepression, which could affect barrier function as well as increase septicemia and bacterial translocation. Further research is required into intestinal infections following stroke.

Although constipation is a frequently cited complaint following stroke, this may be due to a variety of factors arising secondary to stroke, including reduced mobility, decreased fluid intake, and medication usage. Stroke is not known to cause constipation (Johanson et al., 1992; Sonnenberg et al., 1994), but Lim et al. (2015) reported that 33% of stroke patients developed new-onset constipation, with 39% of these patients presenting within two days of admission. The authors found that patients who experienced dysphagia, used antacids, had a greater length of stay, and required the use of a bedpan were at greater risk of developing constipation. Scivolotto et al. (1997) reported that 48% of stroke patients developed heartburn, 30% developed constipation, 27% experienced abdominal pain, 15.6%

16. Nutritional Interventions Following Stroke pg. 13 of 42 www.ebrsr.com developed dysphagia, and 10% reported faecal incontinence with constipation found to be secondary to stroke. However, constipation can be difficult to characterise due to the subjectivity of symptoms (Schaller et al., 2006).

Conclusions Regarding Gastrointestinal Impairments Following Stroke

There is insufficient evidence regarding the development of significant gastrointestinal impairments post stroke.

There is limited evidence suggesting that constipation can develop post stroke.

It is unclear as to whether significant gastrointestinal impairments develop post stroke.

16.3.3 Nutrient Intake Following Stroke Individuals with stroke may be particularly vulnerable to calorie-protein malnutrition due to a variety of factors that affect their ability or willingness to self-feed. In a review, Finestone et al. (2003) noted that cognitive changes to attention, concentration, and memory may affect eating behaviours post stroke. Self-feeding ability may be attenuated by upper extremity paresis/paralysis, visuospatial-perceptual deficits, left-right disorientation, hemispatial neglect, apraxia, and agnosia. Sensory disturbances and mood disorders, such as depression, may affect desire to self-feed.

However, there are few studies that describe the protein and calorie intakes of hospitalized individuals with stroke. Gariballa (2001) reported that the average two-week calorie intake of post-stroke patients consuming a regular hospital diet and without dysphagia was 1338 kilocalories (Kcals), which represented 74% of their predicted requirement. However, this level of intake was not significantly different from matched controls who consumed 1317 Kcals, or 73% of requirement, suggesting that the intakes of patients with stroke were similar to those of other hospitalized patients. In another study, Foley et al. (2006) reported that hospitalized patients consumed an average of 85% of calorie requirements and 86% of protein requirements during the first 21 days post stroke, regardless of diet type (oral or non-oral) and texture (regular or texture-modified). A recent study by Murray et al. (2015) reported that patients without dysphagia post stroke consumed 67% of their daily recommended intake and 44% exhibited dehydration based on their blood urea nitrogen/creatine ratio (>20:1) at 7 days post stroke. Moreover, 16% of patients demonstrated one or more of the following: constipation, dehydration, hypernatremia, and urinary tract infection.

Conclusions Regarding Nutrient Intake Following Stroke

Patients consume 67% of their daily recommended intake during the first week post stroke, and up to 85% of their calorie requirements and 86% of their protein requirements during the first few weeks post stroke.

Patients may consume fewer calories and protein than required post stroke.

16.3.4 Glucose Regulation Following Stroke Impaired glucose regulation is the result of impairments to glucose tolerance and fasting glucose, characterized by insulin resistance and elevated glucose levels (hyperglycaemia) (Engberg et al., 2009). Hyperglycaemia following stroke may be the result of previous unrecognised diabetes or glucose metabolism disorders, or as a stress reaction mechanism (Dziedzic et al., 2010; Lindsberg et al., 2011).

16. Nutritional Interventions Following Stroke pg. 14 of 42 www.ebrsr.com The hazards of hyperglycaemia post stroke include poor functional outcome, increased risk of cardiovascular disease, future occurrence of parenchymal hematoma, and mortality (Lindsberg et al., 2011). Dziedzic et al. (2010) found that hyperglycaemia 24 hours after admission was associated with significantly increased risk of reported post stroke, regardless of glucose levels upon admission. However, the authors found that normalization of glucose levels after admission was associated with greater survival. Laird et al. (2014) recommended that monitoring be conducted every 4 hours for the first 72 hours after admission, irrespective of a prior history of diabetes.

Blood glucose levels can increase within the first 12 hours of a stroke, with a reported increase of 5.8- 6.1mmol/L in mild-moderate stroke and 6.2-6.7mmol/L in severe stroke (Christensen & Boysen, 2002). Correlation between fasting blood glucose (FBG) levels and stroke severity revealed greater severity with elevated FBG level; each mg/dL increase of FBG resulted in a 0.25 decrease on FIM score (Martin et al., 2012). Grant and Ali (2010) suggested that the intercept time-point for a positive 30-day prognosis is 10.4mmol/L, and that patients above this threshold may require insulin intervention. In addition to insulin, other treatments for blood glucose regulation post stroke may include exercise, dietary changes, and diabetic pharmacology such as metformin. Studies examining glucose regulation following stroke are summarized in Table 16.3.4.1.

Table 16.3.4.1 Summary of Studies Evaluating Glucose Regulation Post Stroke Author (Year) Study Design (PEDro Score) Intervention Outcomes Sample Size Gray et al. (2007) E: Glucose-Potassium insulin infusions • Mean plasma glucose levels (+) RCT (7) C: Saline infusions • Systolic blood pressure (+) NStart=933 • Diastolic blood pressure (-) NEnd=754 • Mortality rate (-) • Modified Rankin Scale (-) • Barthel Index (-) • European Stroke Scale (-) Den Hertog et al. (2015) E: Metformin (500mg/d-1000mg 2/d) • 2hr post-load glucose levels (-) RCT (7) C: No medication • Fasting glucose levels (-) NStart=40 • Waist circumference (-) NEnd=39 • Body Mass Index (-)

Ivey et al. (2007) E: Aerobic treadmill training • Fasting insulin levels (+) RCT (5) C: Conventional physiotherapy • Total 3hr insulin area (+) NStart=69 • Incremental 3hr insulin area (+) NEnd=46 • Fasting glucose levels (-) • Body weight (-) • Percentage of body fat (-) Jia et al. (2014) E: Impaired glucose regulation • Mortality (+) Case Control C: Normal glucose metabolism • Stroke recurrence (-) NStart=2639 • Modified Rankin Scale (-) NEnd=2167 Tapia-Perez et al. (2014) E: Deceased within 30d post admission • Elevated plasma glucose levels (+) Case Control C: Alive after 30d post admission • Enteral nutrition (+) NStart=116 NEnd=116 E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

16. Nutritional Interventions Following Stroke pg. 15 of 42 www.ebrsr.com

Discussion Elevated glucose levels may be detrimental to patient health post stroke. Tapia-Perez et al. (2014) reported that patients who died within 30 days of hospital admission for stroke exhibited significantly higher glucose levels, although no difference in mortality was reported between diabetic and non- diabetic patients. Similarly, Jia et al. (2014) found that mortality was significantly higher for patients with higher glucose levels due to impaired glucose regulation than those with normal glucose regulation post stroke. Stroke recurrence and dependency, however, were not significantly different between groups.

Three RCTs investigated different methods of reducing blood glucose levels. Den Hertog et al. (2015) found that patients who received Metformin demonstrated a significant reduction in glucose levels, but did not significantly differ from a non-medicated control group. As well, minor gastrointestinal side effects such as nausea and diarrhea were reported along with headache, tiredness, and dizziness. Gray et al. (2007) reported a reduction in plasma glucose levels and systolic blood pressure after treatment of glucose-potassium insulin infusions, but found no significant improvements in functional outcomes. Ivey et al. (2007) found a significant reduction in insulin levels after a treadmill exercise program, but no significant reduction in glucose levels were reported. The authors recommended further study into the relationship between levels of fitness, insulin, and glucose. A combination of exercise and pharmacological intervention may be of interest in determining the ideal approach to regulating glucose and preventing negative outcomes.

Conclusions Regarding Glucose Regulation Following Stroke

There is Level 1b evidence that glucose-potassium insulin injections significantly reduce glucose levels and systolic blood pressure post stroke; no clinical benefits were observed.

There is Level 1b evidence that administration of Metformin is ineffective in reducing glucose levels post stroke.

There is Level 2 evidence that treadmill exercise significantly reduces insulin levels but not glucose levels post stroke.

There is Level 3 evidence that patients with impaired glucose regulation post stroke are at significantly greater risk for mortality than those with normal glucose regulation; no differences in dependency or stroke recurrence were observed.

Impaired glucose regulation post stroke is associated with greater risk of mortality.

Treatments such as insulin injections and aerobic exercise may be effective in improving glucose regulation but not functional outcomes post stroke.

16.3.5 Vitamin D Deficiency Following Stroke Following stroke, patients may be susceptible to Vitamin D deficiency, which can affect future prognosis and outcomes. Vitamin D has been found to be beneficial for factors linked with stroke such as hypertension, diabetes, and chronic inflammation (Tu et al., 2014). The vascular effects of Vitamin D include modulating smooth muscle cell proliferation, inflammation, and thrombosis, and thus a deficiency can trigger inflammation and vascular remodeling by hyperparathyroidism (Wang et al., 2014).

16. Nutritional Interventions Following Stroke pg. 16 of 42 www.ebrsr.com Vitamin D levels may be a prognostic indicator of favourable outcomes, recovery, and survival after discharge (Wang et al., 2014). Tu et al. (2014) reported that patients with stroke demonstrated significantly lower Vitamin D levels than healthy controls at 90-day follow-up, and that decreased Vitamin D levels were associated with increased stroke severity. In addition, patients with favourable outcomes had significantly higher levels of Vitamin D compared to those with unfavourable outcomes, including mortality. In a similar study, Turetsky et al. (2015) revealed that the risk for unfavourable outcome doubled with each 10-ng/mL decrease in Vitamin D at 90 days post stroke. Vitamin D deficiency may also prove to be a risk factor in the development of stroke. Recent studies have found it to be an independent predictor of fatal stroke (Kojima et al., 2012; Pilz et al., 2008), as well as an independent predictor of stroke severity upon admission (Wang et al., 2014). However, Gupta et al. (2014) reported no significant association between Vitamin D deficiency and occurrence of stroke when comparing patients with stroke to healthy controls. An RCT examining Vitamin D supplementation following stroke is summarized in Table 16.3.5.1.

Table 16.3.5.1 Summary of Study Evaluating Vitamin D Supplementation Post Stroke Author (Year) Study Design (PEDro Score) Intervention Outcomes Sample Size Witham et al. (2012) E: Single Vitamin D2 dose • 25-hydroxyvitamin D levels (8 & 16wks) (+) RCT (7) C: Placebo • Flow mediated dilation change (8wks) (+) NStart=58 • Flow mediated dilation change (16wks) (-) NEnd=55 • Diastolic blood pressure (8 & 16 wks) (-) • Systolic blood pressure (8 & 16 wks) (-) • Cholesterol levels (8 & 16 wks) (-) • Albumin levels (8 & 16 wks) (-) Gupta et al. (2016) E: Vitamin D + Calcium • 25-hydroxyvitamin D levels (+) RCT (5) C: No supplement • Mortality (+) NStart=53 • Modified Rankin Scale (-) NEnd=44 E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

Discussion An RCT conducted by Witham et al. (2012) found that a single dose of Vitman D (100,000 units) had a significant increase on 25-hydroxyvitamin D levels for up to 16 weeks after treatment. However, there was no effect on blood pressure or potential biomarkers of recurrent stroke (i.e. cholesterol and albumin). The authors suggested that the lack of effectiveness in mitigating recurrent stroke may be due to the single dosage plan or concomitant statin use. Given that patients were normotensive, the authors believed that Vitamin D supplementation may still be of use for hypertensive patients in improving stroke-related outcomes.

More recently, an RCT by Gupta et al. (2016) examined the effect of combined Vitamin D and calcium supplementation, compared to usual care, in individuals with Vitamin D deficiency post stroke. Participants received a single intramuscular injection of Vitamin D (600,000 units), along with monthly doses of oral Vitamin D (60,000 units) and daily doses of oral calcium (1g) for a total of six months. Upon trial completion, the treatment group had significantly higher levels of 25-hydroxyvitamin D and a significantly higher probability of survival than controls. The treatment group also had a significantly higher proportion of participants achieving good outcome than controls, but the difference was not

16. Nutritional Interventions Following Stroke pg. 17 of 42 www.ebrsr.com significant. The authors noted that the study was underpowered and thus recommended a large RCT be conducted to substantiate its findings.

Conclusions Regarding Vitamin D Deficiency

There is Level 1b evidence that a single dose of Vitamin D2 (100,000 units) significantly increases 25- hydroxyvitamin D levels for up to 16 weeks when compared to placebo; no effects on blood pressure, cholesterol levels, and albumin levels were observed.

There is Level 1b evidence that a monthly dose of oral Vitamin D (60,000 units), along with a single injection of Vitamin D (600,000 units) and a daily dose of oral calcium (1g), increases 25- hydroxyvitamin D levels and reduces risk of mortality when compared to no supplementation; no effects on odds of good outcome were observed.

A single dose of Vitamin D2 (100,000 units) post stroke may significantly raise Vitamin D levels for up to 16 weeks.

A monthly dose of oral Vitamin D (60,00 units) post stroke, along with a single injection of Vitamin D (600,000 units) and a daily dose of oral calcium (1g), may significantly raise Vitamin D levels and reduce risk of mortality.

16.3.6 Lipid Profiles Following Stroke Lipids can have a detrimental impact on the integrity of small cerebral vessels (Athyros et al., 2010), and thus may play a key role in nutrition following stroke. Larsson et al. (2013) found that women with high dietary cholesterol were at an increased risk of stroke, whereas fat intake was not associated with stroke or any stroke subtype. Compared to healthy participants, patients with stroke showed higher levels of triglycerides, total cholesterol, low-density lipoprotein (LDL), and decreased levels of high- density lipoprotein (HDL) (Bharosay et al., 2014). Compared to patients with transient ischemic attack, patients with stroke only showed higher levels of total cholesterol (Li et al., 2014). In comparing type of stroke, Mahmood et al. (2010) reported significantly abnormal levels of total cholesterol and HDL among patients with ischemic stroke compared to patients with haemorrhagic stroke. The authors noted that previous research found a decrease in HDL at the time of an ischemic stroke, and that low HDL levels may be an acute phase reactant.

A longitudinal study by Willey et al. (2009) recorded serum lipid levels among non-stroke volunteers and, after controlling for those who had received statin treatment, reported an increased risk of stroke for higher LDL levels. The authors suggested that impact of statin treatment may be of significance in preventing the recurrence of stroke in addition to preventing first-time strokes. Lalouschek et al. (2003) revealed that patients with elevated total cholesterol levels, prior statin treatment, experienced hypertension, and coronary artery disease were most likely to be prescribed statin treatment post stroke. However, statins were prescribed significantly less frequently when total cholesterol and LDL levels were not measured or obtained, and so the authors recommended that screening and monitoring be implemented prior to discharge. An RCT examining statin treatment for lipid levels following stroke is summarized in Table 16.3.6.1.

Table 16.3.6.1 Summary of Study Evaluating Statin Treatment Post Stroke Author (Year) Study Design (PEDro Score) Intervention Outcomes Sample Size

16. Nutritional Interventions Following Stroke pg. 18 of 42 www.ebrsr.com Amarenco et al. (2006) E: Atorvastatin (80mg/d) • High-density lipoprotein (+) RCT (7) C: No statin treatment • Low-density lipoprotein (+) NStart=4731 • Total cholesterol levels (+) NEnd=4525 • Triglyceride levels (+) • Recurrent stroke (+) E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

Discussion Statins have been used a form of lipid treatment following stroke, as they decrease total cholesterol and LDL and increase HD (Yaghi & Elkind, 2016), and potentially decrease platelet aggregation thereby reducing the risk of haemorrhage (Athyros et al., 2010). An RCT conducted by Amarenco et al. (2006) revealed that patients treated with Atorvastatin exhibited significantly lower levels of triglycerides, total cholesterol, and LDL, while also demonstrating significantly higher HDL levels and significantly reduced risk of stroke. These results support the hypothesis that the reduction of cholesterols through statin treatment can prevent the recurrence of stroke.

A meta-analysis of 78 trials conducted by De Caterina et al. (2010) revealed that statins were a significantly effective method of treatment in reducing the risk of stroke, but the use of fibrates and dietary modifications were not efficacious. Although no cholesterol-lowering interventions were found to be significantly effective in preventing fatal stroke, the risk of non-fatal stroke was significantly reduced by statins. The authors reported that reducing LDL levels are greatly beneficial in preventing stroke. However, given the lack of non-statin trials in the research, the authors recommended that their findings be interpreted with caution. Future research should aim to investigate non-statin cholesterol- lowering treatments and alternative treatments that may be able to reduce the risk of fatal strokes.

Conclusions Regarding Lipid Profiles Following Stroke

There is Level 1b evidence that Atorvastatin 80mg/d is effective in reducing total cholesterol and LDL levels and increasing HDL levels post stroke.

Statin treatment post stroke may significantly reduce total cholesterol and LDL levels, increase HDL levels, and prevent recurrent stroke. More research is needed to verify this effect.

16.4 Nutritional Interventions Following Stroke Nutritional interventions associated with stroke are usually directed at the improvement of nutrient intakes through oral supplementation, enteral feeding or dysphagia therapy. The goal of all nutritional interventions is the prevention or correction of malnutrition.

16.4.1 Enteral Feeding Enteral, or “tube”, feeding may represent a sole or supplemental source of feeding for patients post stroke. Generally, tube feeding as the sole source of nutrient intake is reserved for dysphagic patients for whom oral feeding is considered unsafe. However, “failure to thrive” non-dysphagic patients may also be candidates for enteral feeding in the presence of prolonged and inadequate oral intake. In these patients, the use of feeding tubes has been shown to reverse malnutrition (Finestone et al., 1995). Therefore the use of tube feeding can prevent or reverse the effects of malnutrition in patients who are unable to safely eat and those who are unwilling to eat.

16. Nutritional Interventions Following Stroke pg. 19 of 42 www.ebrsr.com

The Post-Stroke Rehabilitation Outcomes Project, which retrospectively studied the outcomes of 919 patients from six inpatient rehabilitation sites, provided evidence of tube feeding as an effective intervention (James et al., 2005). Patients with both moderate and severe stroke who had received tube feeding during hospital stay, but were not discharged with a feeding tube, achieved greater increases in total Functional Independence Measure (FIM) and experienced greater improvement in stroke severity by discharge. Among 143 patients admitted to an inpatient rehabilitation facility with an enteral feeding tube in place due to dysphagia, 20% were able to have their tubes removed prior to discharge having resumed full oral intake (Krieger et al., 2010). As well, 65% of patients were able to return to some type of oral feeding prior to discharge, and those who were able to have their feeding tube removed were more likely to be discharged. From a cohort of 4,972 patients consecutively admitted for ischemic stroke, 14.5% could not eat orally on day 10 post stroke (Nakajima et al., 2012). Of the 512 dysphagic patients who responded at a 3-month follow-up, 27% had resumed oral intake. A National Institutes of Health Stroke Score (NIHSS) ≤17 on day 10 was the strongest predictor or ability to eat orally at 3 months. As well, NIHSS scores were found to be the most predictive factor for feeding tube placement among a group of 77 patients following severe stroke (Kumar et al., 2012).

In cases where tube feeding is indicated, it can be achieved through either a temporary or permanent access. A temporary nasogastric feeding tube is often indicated in patients who are expected to return to a full oral diet within one month, while permanent tube feeding is indicated when a prolonged period of non-oral intake (>1 month) is anticipated (Kirby et al., 1995). Available techniques for feeding tube placement include percutaneous gastrostomy/jejunostomy/gastrojejunostomy, which are endoscopically or radiologically inserted, and surgically placed gastrostomy/jejunostomy. In patients with stroke, surgically placed feeding tubes are uncommon, and placement of feeding tubes into the stomach supersedes those in the small bowel. As well, a greater percentage of feeding tubes are placed by gastroenterologists compared to radiologists, although regional differences may exist due to issues of financial remuneration, availability of facilities, and technical expertise. Studies examining enteral feeding following stroke are summarized in Table 16.4.1.1.

Table 16.4.1.1 Summary of Studies Evaluating Enteral Feeding Post Stroke Author (Year) Study Design (PEDro Score) Intervention Outcomes Sample Size Dennis et al. (2005b) Trial 1: Trial 1: RCT (8) E: Early Enteral Feeding (<72hr) • Modified Ranking Scale & Mortality (-) NStart=859 C: Delayed Enteral Feeding (>7d) • Mortality (-) NEnd=859 Trial 2: Trial 2: • Modified Rankin Scale & Mortality (-) E1: Nasogastric Tube • Mortality (-) E2: Percutaneous Endoscopic Gastrostomy Zheng et al. (2015) E: Early Enteral Feeding (<72hr) • Malnutrition rate (+) RCT (6) C: Family-Managed Nutrition • Mortality rate (+) NStart=146 • Infection rate (+) NEnd=146 • Triceps skinfold (+) • Arm muscle circumference (+) • Hemoglobin (+) • Albumin (+) • Triglyceride (+) Park et al. (1992) E1: Nasogastric Tube • Recurrent blockage/dislodgement (+) RCT (6) E2: Percutaneous Endoscopic Gastrostomy • Total prescribed volume of intake (+)

16. Nutritional Interventions Following Stroke pg. 20 of 42 www.ebrsr.com NStart=40 • Weight gain at 1wk (+) NEnd=38 • Weight gain at 4wk (-) NStroke=18 • Triceps skinfold thickness (-) • Midarm circumference (-) Norton et al. (1996) E1: Nasogastric Tube • Mortality (+) RCT (6) E2: Percutaneous Endoscopic Gastrostomy • Albumin (+) NStart=23 • Weight gain (+) NEnd=22 • Length of stay (+) • Total prescribed volume of intake (+) • Recurrent blockage/dislodgement (-) Hamidon et al. (2006) E1: Nasogastric Tube • Recurrent tube blockage/dislodgement (+) RCT (6) E2: Percutaneous Endoscopic Gastrostomy • Albumin (+) NStart=23 • Triceps skinfold thickness (-) NEnd=22 • Biceps skinfold thickness (-) • Midarm circumference (-) Gulsen et al. (2014) E: Enteral Protein Supplement (2g/kg/d of • Body Mass Index (-) RCT (6) protein) • Triceps skinfold thickness (-) NStart=20 C: Standard Enteral Nutrition (1g/kg/d of • Subscapular skinfold thickness (-) NEnd=20 protein) • Biochemical complications (-) • Inflammatory parameters (-) • Nitrogen balance (-) • Medication usage (-) Bakiner et al. (2013) E: Early Enteral Feeding (<24hr) • National Institute of Health Stroke score (-) PCT C: Late Enteral Feeding (>48hr) • Length of stay (-) NStart=24 • Plasma GLP-1 levels pre-feed at 1d (-) NEnd=24 • Plasma GLP-1 levels post-feed at 1d (-) • Plasma GLP-1 levels pre-feed at 3d (-) • Plasma GLP-1 levels post-feed at 3d (-) (GLP-1=Glucagon-Like Peptide-1) Maeshima et al. (2013) E1: Oral Intake • Functional Independence Measure (FIM) (+) Case Control E2: Enteral Tube Feeding • FIM gains (+) NStart=334 • FIM efficiency (+) NEnd=334 Kim & Byeon (2014) E1: General Diet • Caloric Intake (+) Case Control E2: Dysphagic Diet • Body Mass Index (+) NStart=261 E3: Enteral Tube Feeding • Pre-albumin (+) NEnd=261 • Albumin (+) • Total lymphocyte count (+) • Total protein level (+) • Complications (-) San Luis et al. (2013) E: Percutaneous Endoscopic Gastrostomy • NIH Stroke Score on admission (+) Case Control C: No Enteral Feeding • Number of patients requiring Intra-arterial tPA (+) NStart=157 • Number of patients requiring Intravenous tPA (+) NEnd=157 • Inability to complete first swallow test (+) • In-hospital aspiration pneumonia (+) • Pneumonia readmission at 30d (+) • Pneumonia readmission at 60d (+) • Pneumonia readmission at 90d (-) (tPA=Tissue Plasminogen Activator) Nyswonger & Helmchen E1: Early Enteral Feeding (<72hr) • Length of stay (+) (1992) E2: Late Enteral Feeding (>72hr) Case Control NStart=52

16. Nutritional Interventions Following Stroke pg. 21 of 42 www.ebrsr.com NEnd=52 E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

Discussion Compared to those with either a general diet or dysphagic diet, patients with tube feeding showed no significant difference in indicators of nutritional status at time of admission for stroke (Kim et al. 2014). However, at 7 days following admission, tube feeding was associated with suboptimal levels of pre- albumin, albumin, total protein, and total lymphocyte count that indicated malnutrition. As well, patients with tube feeding showed significantly lower Body Mass Index and daily caloric intake compared to the other diets. Gulsen et al. (2014) examined nutrient supplementation during tube feeding, but found no significant difference in indicators of nutritional status between patients receiving protein supplement and those receiving standard nutrition. The authors suggested that future research into nutrient supplementation aim for a longer study period and a larger sample size in order to fully understand its effect during tube feeding.

Several factors can predict the likelihood of a patient requiring tube feeding following stroke. Maeshima et al. (2013) found that older age and lower FIM gain during acute care were significant independent predictors for tube feeding at discharge from stroke rehabilitation. However, these findings challenge those of San Luis et al. (2013), who reported that age was not a significant factor in determining mode of feeding. The authors also reported that a higher NIHSS on admission, inability to complete a first swallow test, and presence of in-hospital aspiration pneumonia were predictors of percutaneous endoscopic gastrostomy (PEG) placement. As well, patients who were not given PEG showed an increased aspiration pneumonia readmission rate at 90-day follow-up. The authors recommended that early and accurate identification of patients requiring PEG is necessary in order to reduce length of stay and complications. While some studies have reported benefits associated with early enteral feeding, including reduced rate of poor outcomes (Zheng et al., 2015) and shorter length of stay (Nyswonger & Helmchen, 1992), other studies failed to find such an association (Bakiner et al., 2013; Dennis et al., 2005a).

Four RCTs examined differences between NG and PEG feeding for various outcomes following stroke. Both Park et al. (1992) and Hamidon et al. (2006) reported a significantly greater number of treatment failures (e.g. blockage, dislodgment) associated with NG feeding. Patients with PEG feeding consumed a significantly greater proportion of their prescribed intake in the former study and showed significantly higher serum albumin levels in the latter study. Norton et al. (1996) did not find greater treatment failure in NG feeding, but patients with PEG feeding were found to have consumed greater intake, higher serum albumin levels, and shorter length of stay. The authors also found that a significantly greater number of patients fed by NG had died at 6 weeks compared to those fed by PEG. However, these findings may have been influenced by differences between the two groups regarding a number of clinical variables including severity of stroke, recovery of swallowing, and rate of discharge.

The purported benefits of PEG over NG tube feeding reported in these studies were not supported by the results from one of the FOOD Trials (Dennis et al., 2005b). In fact, a greater proportion of patients with PEG tube feeding were either dead or dependent at 6 months, although the results were not statistically significant. In addition, delayed tube feeding (>7 days) was not significantly associated with worse outcomes compared to early feeding (<72 hours). While prescribed/consumed volume of intake was not reported in these trials, they are set apart from the aforementioned studies due to their strong methodology, clinically relevant outcomes, large sample sizes, and nearly complete follow-up. As shown

16. Nutritional Interventions Following Stroke pg. 22 of 42 www.ebrsr.com in Figure 16.4.1.1, a pooled analysis based on the FOOD Trial (Dennis et al., 2005b) and Norton et al. (1996) found no increased risk of either mortality or poor outcome associated with type of feeding tube.

Figure 16.4.1.1 Effectiveness of Enteral Feeding (PEG vs NG)

Conclusions Regarding Enteral Feeding Following Stroke

There is Level 1b evidence that gastric tube feeding such as PEG is associated with fewer mechanical complications and greater consumed intake post stroke compared to NG feeding.

There is Level 1b evidence that enteral protein supplementation post stroke does not differ significantly from standard enteral nutrition in its effect on malnutrition, based on biochemistry and/or body composition.

There is conflicting Level 1b and Level 2 evidence that early enteral feeding does not differ significantly from late or delayed enteral feeding in its effects on poor outcome post stroke.

Gastric feeding tubes are associated with fewer complications and greater consumed intake post stroke compared to nasogastric tubes, but no significant difference has been found regarding poor outcome or mortality.

16.4.2 Oral Supplementation Oral supplementation may be indicated for patients who are safe with oral intake but fail to consume sufficient quantities to meet their nutritional requirements and/or have pre-existing nutritional deficits. Theoretically, oral supplementation can effectively improve nutritional intake, leading to improvements in nutritional parameters and ultimately functional outcomes. In a review of treatments for calorie- protein malnutrition in chronic non-malignant disorders, Akner and Cederholm (2001) identified 3 RCTs treating both dysphagic and non-dysphagic patients post stroke (DePippo et al., 1994; Gariballa et al., 1998b; Norton et al., 1996), as well as 4 “uncontrolled trials” (Davalos et al., 1996; Elmstahl et al., 1999; Nyswonger & Helmchen, 1992; Wanklyn et al., 1995). The authors concluded that the efficacy of calorie- protein malnutrition treatments post stroke could not be measured due to a “striking lack of published articles”. However, more recent studies have examined various outcomes associated with calorie- protein supplementation following stroke, which are summarized in Table 16.4.2.1

16. Nutritional Interventions Following Stroke pg. 23 of 42 www.ebrsr.com

Table 16.4.2.1 Summary of Studies Evaluating Oral Supplementation Post Stroke Author (Year) Study Design (PEDro Score) Intervention Outcomes Sample Size Rabadi et al. (2008) E: Intensive nutritional supplement • Functional Independence Measure (FIM) RCT (9) (720Kcal, 33g protein, 1/d) Total Score (+) NStart=116 C: Standard nutritional supplement • FIM Motor Score (+) NEnd=102 (381Kcal, 15g protein, 1/d) • Walk Test, 2-Minute (+) • Walk Test, 6-Minute (+) • Discharge destination (+) • FIM Cognitive Score (-) • Length of stay (-) • Weight (-) • Albumin (-) • Pre-albumin (-) • Transferrin (-) Dennis et al. (2005a) E: Hospital diet + Nutritional supplement • Modified Rankin Scale & Mortality (-) RCT (8) (540Kcal, 62.5g protein, 1/d) • Mortality (-) NStart=4023 C: Hospital diet only NEnd=4004 Manolescu et al. (2013) E: Standard care + ALAnerv nutritional • Total lipids (+) RCT (8) supplement (2/d) • High-density lipoprotein (+) NStart=28 C: Standard care only • Low-density lipoprotein (-) NEnd=28 • Total cholesterol (-) • Triacylglycerol (-) • Phospholipids (-) Aquilani et al. (2015) E: Essential amino acids supplement (8g/d) • Neutrophil/Lymphocyte Ratio (+) RCT (6) C: Placebo NStart=42 NEnd=42 Gariballa et al. (1998b) E: Hospital diet + Nutritional supplement • Albumin (+) RCT (6) (600Kcal, 20g protein, 1/d) • Iron (+) NStart=42 C: Hospital diet only • Protein intake (+) NEnd=40 • Calorie intake (+) • Barthel Index (-) • Length of stay (-) • Discharge destination (-) • Infective complications (-) • Mortality (-) Aquilani et al. (2008a) E: Regular diet + Nutritional supplement • Mini Mental State Examination Log (+) RCT (6) (+250Kcal, +20g protein, 1/d) • Daily protein intake (+) NStart=48 C: Regular diet only • Daily calorie intake (+) NEnd=48 • Daily lipid intake (-) • Daily carbohydrate intake (-) • Weight (-) • Body Mass Index (-) • Mini Mental State Examination (-) Aquilani et al. (2008a) E: Regular diet + Nutritional supplement • National Institute of Health Stroke Scale (+) RCT (6) (additional protein) • Carbohydrate-protein ratio (+) NStart=42 C: Regular diet only • Daily protein intake (%kcal) (+) NEnd=41 • Daily protein intake (g or g/kg) (-) • Body Mass Index (-)

16. Nutritional Interventions Following Stroke pg. 24 of 42 www.ebrsr.com • Weight (-) Oprea et al. (2013) E: Conventional rehabilitation + ALAnerv • Glucose levels (+) RCT (6) supplement (2/d) • Total Lipids (+) NStart=28 C: Conventional rehabilitation only • High-density lipoprotein (+) NEnd=28 • Low-density lipoprotein (+) • Barthel Index (+) • Lipid hydroperoxides (-) • Protein carbonyl (-) Ha et al. (2010a) E: Individualized care (calorie and protein • EuroQol – Visual Analog Scale (+) RCT (5) supplements) • Handgrip strength (+) NStart=170 C: Routine care • Calorie intake (+) NEnd=124 • Protein intake (-) • Body weight (-) • Length of stay (-) • EuroQol – 5 Dimensions (-) Ha et al. (2010b) E: Individualized care (calorie and protein • Body Mass Index (+) RCT (5) supplements) • Weight (+) NStart=170 C: Routine care • Mean weight loss (female) (+) NEnd=124 • Mean fat loss (female) (+) • Mean weight loss (male) (-) • Mean fat loss (male) (-) • Triceps skinfold thickness (-) • Mid-upper arm circumference (-) E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

Discussion Compared to a regular diet, Gariballa et al. (1998b) found that a diet with calorie-protein supplementation significantly improved nutritional parameters (serum levels of albumin and iron) in malnourished patients following stroke. No significant differences were found between groups in terms of non-nutritional parameters including length of stay, infective complications, severity, and mortality. Given that malnutrition was assessed using only anthropometric measures, the study may have inflated the number of malnourished patients, for whom supplementation would show the greatest effects. As well, the length of intervention (4 weeks) and measure of severity (Barthel Index) may have lacked the sensitivity to detect clinically significant changes. In a similar study, Dennis et al. (2005b) compared a regular diet and supplemented diet in patients post stroke over the course of 6 months. The authors reported no significant difference between groups in the proportion of patients who experienced death or disability at follow-up, or in the odds of experiencing mortality or poor outcome. In contrast to the previous study, the amount of supplementation and length of intervention were greater, and patients were not necessarily malnourished. The volume of prescribed/consumed intake was not reported, as in the other FOOD Trials, and so it is unclear whether patients actually received adequate nutrition from their diet. As shown in Figure 16.4.2.1, a pooled analysis of the two studies found no protective effects associated with oral supplementation on either mortality or poor outcome.

16. Nutritional Interventions Following Stroke pg. 25 of 42 www.ebrsr.com

Figure 16.4.2.1 Effectiveness of Oral Supplementation

Nutritional supplementation may be beneficial in enhancing cognitive recovery following stroke. Aquilani et al. (2008a) hypothesized that an increased supply of amino acids may assist in reactivating the synthesis of neural proteins, thereby increasing neuron energy and neurotransmitter function. The authors reported a significantly greater improvement in log transformed Mini Mental State Examination (MMSE) scores after a supplemented diet than after a regular diet, although there was no significant difference between groups without data transformation. In a related study, Aquilani et al. (2008b) demonstrated that 3 weeks of a protein-supplemented diet resulted in greater neurological recovery as measured by the National Institutes of Health Stroke Score (NIHSS). The authors suggested that increased protein synthesis may “induce axonal sprouting and formation of new cortical connections” in both perilesional and remote areas of the brain, and thus improve training-induced plasticity. More recently, Aquilani et al. (2015) found that essential amino acid supplementation significantly decreased the ratio of neutrophil to lymphocyte in patients with subacute ischemic stroke. The neutrophil/lymphocyte ratio was significantly higher than normal values upon admission, and was found to be inversely correlated with functional independence and dysphagia outcome upon discharge.

Within a group who experienced significant weight loss during the acute period following stroke, Rabadi et al. (2008) randomized patients to receive either standard oral supplement or intensive oral supplement for the duration of their hospital stay. The authors found no significant difference between groups in nutritional parameters (weight, pre-albumin, albumin, transferrin), but reported significantly greater improvements in motor function (Functional Independence Measure and Walk Test) within the intensive supplement group at discharge. However, the authors failed to report the actual intake of patients during the study and utilized an improper tool for measuring stroke severity at baseline (Mini Mental State Examination). In order to evaluate whether weight loss could be minimized or prevented post stroke, Ha et al. (2010b) compared an individualized nutritional program to routine care during the acute period. Although there was no significant difference between groups in the proportion of patients who lost 5% or more of their weight, patients receiving routine care were required to complete dietary records, which may have increased attention to their own intake and thus biased such findings. In a separate analysis that excluded these patients, results showed a significantly lower proportion of patients with individualized care that lost at least 5% of their body weight. Further, female patients appeared to benefit preferentially from the treatment, as they experienced less weight loss and reductions in fat stores compared with male patients (2010a).

16. Nutritional Interventions Following Stroke pg. 26 of 42 www.ebrsr.com

ALAnerv, a supplement comprised of vitamins, minerals, and fatty acids, has been investigated as a potential treatment in regulating nutritional intake. Manolescu et al. (2013) reported significantly higher levels of total lipids and high-density lipoproteins (HDL) compared to a control group that did not receive ALAnerv, but no difference in total cholesterol and low-density lipoproteins (LDL) were reported. Noting the short duration of treatment in their study, the authors suggested that a longer period may better elucidate the efficacy of ALAnerv as a supplement for patients with stroke. Oprea et al. (2013) also investigated the use of ALAnerv, and found that patients treated with the supplement exhibited significantly higher levels of total lipids and HDL, and significantly lower levels of glucose and LDL, than those without the supplement. The authors noted LDL as a key biomarker, for increased oxidized LDL concentration has been found to result in poorer prognosis. Future research should investigate the use of alpha-lipoic acid (ALA) in stroke populations.

Conclusions Regarding Oral Supplementation Following Stroke

There is Level 1a evidence that oral nutritional supplementation improves the calorie-protein intake of patients post stroke.

There is Level 1a evidence that oral nutritional supplementation does not reduce the risk of death or dependency post stroke.

There is Level 1a evidence that the ALAnerv nutritional supplement may significantly reduce total lipid levels and increase HDL levels compared to conventional treatment post stroke.

Oral nutritional supplementation may improve calorie-protein intake, decrease lipid levels, and increase HDL levels post stroke, although it may not prevent poor outcome or mortality.

16.4.3 Dysphagia Treatment Dysphagia treatment programs may improve the nutritional status of patients following stroke. These programs can include a combination of oral motor exercise, swallowing techniques, body positioning, and diet modification. In a low level evidence study, Elmstahl et al. (1999) reported that 2 months of dysphagia therapy showed significant increases in two nutritional indicators (albumin and total iron- binding capacity). Further studies evaluating different dysphagia treatment programs following stroke are summarized in Table 16.4.3.1.

Table 16.4.3.1 Summary of Studies Evaluating Dysphagia Treatment Post Stroke Author (Year) Study Design (PEDro Score) Intervention Outcomes Sample Size Carnaby et al. (2006) E1: Low-intensity dysphagia therapy • Time returning to normal diet (+) RCT (8) E2: High-intensity dysphagia therapy • Functional swallowing ability (+) NStart=306 C: Standard care • Occurrence of chest infection (+) NEnd=243 • Modified Rankin Scale (-) • Mortality (-) DePippo et al. (1994) E1: Dysphagia therapy + Patient-chosen diet • Time to occurrence of pneumonia (+) RCT (5) E2: Dysphagia therapy + Prescribed diet • Occurrence of pneumonia (-) NStart=115 E3: Dysphagia therapy + Prescribed diet + Meal • Occurrence of dehydration (-) NEnd=114 time management/therapy • Recurrent upper-airway obstruction (-) • Calorie-nitrogen deficit (-) • Mortality (-)

16. Nutritional Interventions Following Stroke pg. 27 of 42 www.ebrsr.com Lin et al. (2003) E: General swallowing therapy • Volume per swallow (+) PCT C: No therapy • Volume per second (+) NStart=61 • Coughing/choking during meals (+) NEnd=49 • Body weight (+) • Mid-arm circumference (+) • Albumin (-) • Body Mass Index (-) • Coughing/choking during swallow test (-) E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

Discussion Three trials have examined the impact of formal dysphagia treatment programs on nutritional status, although nutritional indicators were secondary outcomes in all of the trials. Using a swallowing training program, Lin et al. (2003) reported significant improvements in coughing/choking frequency, volume consumed, and other nutritional parameters. Carnaby et al. (2005) examined the impact of two levels of dysphagia therapy on reducing the need for a modified diet. Both therapies provided instruction on compensatory swallowing strategies, swallowing exercises, and regular evaluation of dietary modifications but with different levels of intensity. Compared to those receiving standard care, patients receiving dysphagia therapy were more likely to have returned to an unmodified diet at 6 months, but no significant differences were found between the low and high intensity groups. In order to evaluate the effect of diet during dysphagia therapy, DePippo et al. (1994) compared three types of diet: patient- chosen, clinician-prescribed, and clinician-prescribed with additional management and support. The authors reported no significant difference between groups in terms of nutrition (calorie-nitrogen deficit), function (airway obstruction), complications (pneumonia and dehydration), and mortality. A pooled comparison of the two prescribed diets to the patient-chosen diet showed no significant difference as well. However, the 2 week treatment period may have been too short to demonstrate an observable effect. The results of the pooled analysis are shown in Figure 16.4.3.1.

Figure 16.4.3.1 Effectiveness of Dysphagia Treatment

In a review, Rowat (2014) noted that patients with post-stroke dysphagia are more likely to be dehydrated than malnourished during their first week of onset. Although there is no gold standard for maintaining hydration, it has been reported that intravenous hydration, enteral hydration, or a combination of both methods are acceptable treatments. However, determining the type, volume, and duration of hydration is difficult given the absence of a measurement or assessment to be completed. Further research is required to ascertain hydration protocols that maintain safety and efficacy.

16. Nutritional Interventions Following Stroke pg. 28 of 42 www.ebrsr.com

Conclusions Regarding Dysphagia Treatment Following Stroke

There is Level 1b evidence that high-intensity dysphagia therapy results in improved swallowing function and less time to resuming a normal diet post stroke.

There is Level 1b and Level 2 evidence that dysphagia therapy does not reduce the risk of death or dependency post stroke, regardless of treatment intensity or diet type.

High-intensity dysphagia therapy may improve swallowing function and reduce time to resuming a normal diet post stroke, although it may not prevent poor outcome or mortality.

16.4.4 Long-Term Enteral Feeding There has been a trend toward enteral feeding in patients who require non-oral feeding for extended periods of time, following discharge from hospital or rehabilitation unit into the community. However, given the lack of literature regarding its long-term usage, it is unclear whether enteral feeding slows the rate of decline or prolongs the time before death among these patients. There have been retrospective case studies examining long-term percutaneous endoscopic gastrostomy (PEG) tube feeding post stroke, as summarized in Table 16.4.4.1. More recently, a study evaluating long-term nasogastric (NG) tube feeding post stroke was conducted, as summarized in Table 16.4.4.2.

Table 16.4.4.1 Summary of Studies Evaluating Long-Term Gastrostomy Use Author (Year) Sample Size Time of Insertion Outcome Wanklyn et al. (1995) 37 26d (median) • 16% alive at 1yr • 14% aspiration pneumonia • 2% functional improvement James et al. (1998) 126 22d (median) • 47% alive at 1yr • 18% aspiration pneumonia • 4% functional improvement Wijdicks & McMahon (1999) 74 11d (median) • 67% alive at 1yr • 6% aspiration pneumonia • 28% functional improvement

Table 16.4.4.2 Summary of Study Evaluating Long-Term Nasogastric Tube Use Author (Year) Study Design Intervention Outcomes Sample Size Shah et al. (2012) E: Nasogastric tube • Mid-arm circumference (+) Cohort C: Oral feeding • Mid-arm muscle circumference (+) N=140 • Triceps skinfold thickness (+) • Body Mass Index (+) • Subjective Global Assessment (+) E indicates experimental group; C indicates control group + indicates statistical significance between treatment groups - indicates no statistical significance between treatment groups

Discussion Complications from long-term NG use included intentional dislodgement (42.9%), accidental dislodgement (8.6%), aspiration (8.6%), and traumatic tube insertion (4.3%), which can lead to malnourishment and dehydration. However, it should be noted that NG feeding is generally restricted to

16. Nutritional Interventions Following Stroke pg. 29 of 42 www.ebrsr.com the first weeks following stroke. On average, PEG feeding tubes were placed within the first month following stroke (James et al., 1998; Wanklyn et al., 1995; Wijdicks & McMahon, 1999). The one-year survival rate ranged from 16% to 67%, while the functional recovery rate ranged from 2 to 28%. Aspiration pneumonia was a common complication, with rates ranging from 6% to 18%. Compared to oral feeding, Shah et al. (2012) reported that long-term NG tube feeding was associated with increased malnourishment and a great number of complications. Malnutrition was assessed using a number of anthropometric measures including skinfold thickness, arm circumference, and Body Mass Index. In order to identify functional improvements and potential complications, greater support and increased monitoring of patients requiring enteral feeding follow discharge is recommended. Moreover, further research is required to develop better understanding of enteral tube use in the community.

Conclusions Regarding Long-Term Enteral Feeding

The one-year survival rate of patients with PEG feeding tubes post stroke varied from 16 - 67%, and functional recovery was reported in 2 - 28% of these patients.

There is Level 2 evidence that long-term NG tube feeding post stroke results in greater levels of malnutrition than oral feeding.

Rates of one-year survival and functional recovery in patients with PEG feeding tubes within the community post stroke are highly varied.

Long-term NG tube feeding may result in greater malnourishment, although its use is usually restricted to the first week post stroke.

16.4.5 Total Parenteral Nutrition Total Parenteral Nutrition (TPN) is a form of aggressive nutritional intervention often reserved for patients with a non-functioning gastrointestinal tract. However, it is rarely indicated for patients following stroke, unless they present with a pre-existing medical condition that precludes safe oral or enteral feeding. Enteral feeding may be contraindicated in patients with inflammatory bowel disease, or in dysphagic patients with a central venous line refusing an enteral feeding. Given the limited use of TPN, there have been no studies investigating its efficacy post stroke.

Conclusions Regarding Total Parenteral Nutrition

There is currently no evidence regarding the efficacy of total parenteral nutrition in the treatment of patients post stroke.

Use of total parenteral nutrition has not been studied in the stroke population.

16.4.6 Effect of Nutritional Interventions on Changes in Nutritional Parameters The results from four trials that assessed changes in nutritional parameters following any form of nutritional intervention were pooled (DePippo et al., 1994; Gariballa et al., 1996; Norton et al., 1996; Park et al., 1992). Sufficient data was presented in these studies to enable the generation of either an unadjusted odds ratio or a weighted mean difference for at least one of the outcomes. The results of the pooled analysis are presented in Figure 16.4.4.1.

The following outcomes were assessed:

16. Nutritional Interventions Following Stroke pg. 30 of 42 www.ebrsr.com 1. Changes in serum albumin concentration 2. Changes in weight 3. Changes in mid-arm circumference 4. Changes in triceps skinfold measurement

Figure 16.4.4.1 Changes in Nutritional Indices Associated with Nutritional Intervention

Each study reported a positive treatment effect associated with some form of nutritional intervention. Significant improvements in serum albumin, mid-arm circumference, and weight among patients in the treatment groups were reported in all of the studies. Overall, there was a statistically significant improvement in nutrition markers following a period of nutritional intervention (WMD 2.44, 95%CI 0.06- 4.83) with a significant overall effect (Z=2.01, p=0.04) and significant heterogeneity (Chi2=33.2, df=6, p<0.00001). The heterogeneity could be explained by the differences in interventions, sample sizes, treatment durations, and follow-up periods.

16.5 Cochrane Reviews of Nutritional Interventions Following Stroke There is currently one Cochrane Review evaluating nutritional interventions following stroke, which is summarized in Table 16.5.1.

Author (Year) Country Methods Results Title Geeganage et al. Studies: 33 RCTs Nutritional Supplementation: (2012) (8 studies, 4391 patients) UK Patients: 6,779, mean age 71yr Primary outcomes: No effect on death or dependency/disability.

16. Nutritional Interventions Following Stroke pg. 31 of 42 www.ebrsr.com Interventions for Secondary outcomes: Fewer pressure sores (OR 0.56, 95%CI 0.32- dysphagia and Type of stroke: ischemic and/or 0.96, p=0.03), greater calorie intake (MD 430.18, 95%CI 141.61- nutritional hemorrhagic 718.75, p=0.003), and greater protein intake (MD 17.28, 95%CI support in acute 1.99-32.56, p=0.03). No statistically significant differences for and subacute Time post stroke: <6 months (acute length of stay or albumin concentration. stroke and/or subacute) Route of Feeding: Interventions: route of feeding, timing (5 studies, 455 patients) of feeding, swallowing therapy, fluid Percutaneous Endoscopic Gastrostomy (PEG) vs Nasogastric (NG) supplementation, and/or nutritional Primary outcomes: No effect on death or dependency/disability supplementation Secondary outcomes: Fewer treatment failures (OR 0.09, 95%CI 0.01-0.51, p=0.007) and higher albumin concentrations (MD 4.92, 95%CI 0.19-9.65) for PEG over NG. No statistically significant differences for dysphagia, weight, mid-arm circumference, institutionalization, length of stay, infection, pneumonia, pressure sores, or case fatality.

Note: Results from only Nutritional Therapies section are reported. Refer to Chapter 15 for other results.

Overall, Geeganage et al. (2012) found no statistically significant evidence for decreases in death or dependency using nutritional interventions. However, nutritional supplementation resulted in greater protein-calorie intake and fewer complications; only single studies evaluated the timing of feeding and the effects of fluid supplementation. As well, the use of a percutaneous endoscopic gastrostomy (PEG) tube compared to a nasogastric (NG) tube resulted in fewer treatment failures. These results comparing PEG and NG tubes were supported in a non-stroke specific Cochrane Review by Gomes et al. (2012), which included several similar stroke-specific studies to the aforementioned Cochrane Review (Bath et al., 2000; Dennis et al., 2005a; Hamidon et al., 2006; Norton et al., 1996; Park et al., 1992).

16. Nutritional Interventions Following Stroke pg. 32 of 42 www.ebrsr.com Summary 1. The prevalence of malnutrition varies from 6% to 62% post stroke, depending on timing of assessment and criteria used to define malnutrition.

2. There is currently no “gold standard” for the assessment of nutritional status, and various methods of detection may be used.

3. There is insufficient evidence regarding malnutrition during the acute phase of stroke.

4. There is insufficient evidence regarding the development of significant gastrointestinal impairments post stroke.

5. There is limited evidence suggesting that constipation can develop post stroke.

6. Patients consume 67% of their daily recommended intake during the first week post stroke, and up to 85% of their calorie requirements and 86% of their protein requirements during the first few weeks post stroke.

7. There is Level 1b evidence that glucose-potassium insulin injections significantly reduce glucose levels and systolic blood pressure post stroke; no clinical benefits were observed.

8. There is Level 1b evidence that administration of Metformin is ineffective in reducing glucose levels post stroke.

9. There is Level 2 evidence that treadmill exercise significantly reduces insulin levels but not glucose levels post stroke.

10. There is Level 3 evidence that patients with impaired glucose regulation post stroke are at significantly greater risk for mortality than those with normal glucose regulation; no differences in dependency or stroke recurrence were observed.

11. There is Level 1b evidence that a single dose of Vitamin D2 (100,000 units) significantly increases 25-hydroxyvitamin D levels for up to 16 weeks when compared to placebo; no effects on blood pressure, cholesterol levels, and albumin levels were observed.

12. There is Level 1b evidence that a monthly dose of oral Vitamin D (60,000 units), along with a single injection of Vitamin D (600,000 units) and a daily dose of oral calcium (1g), increases 25- hydroxyvitamin D levels and reduces risk of mortality when compared to no supplementation; no effects on odds of good outcome were observed.

13. There is Level 1b evidence that Atorvastatin 80mg/d is effective in reducing total cholesterol and low-density lipoprotein (LDL) levels and increasing high-density lipoprotein (HDL) levels post stroke.

14. There is Level 1b evidence that gastric tube feeding such as percutaneous endoscopic gastrostomy (PEG) is associated with fewer mechanical complications and greater consumed intake post stroke compared to nasogastric (NG) tube feeding.

16. Nutritional Interventions Following Stroke pg. 33 of 42 www.ebrsr.com 15. There is Level 1b evidence that enteral protein supplementation post stroke does not differ significantly from standard enteral nutrition in its effect on malnutrition, based on biochemistry and/or body composition.

16. There is conflicting Level 1b and Level 2 evidence that early enteral feeding does not differ significantly from late or delayed enteral feeding in its effects on poor outcome post stroke.

17. There is Level 1a evidence that oral nutritional supplementation improves the calorie-protein intake of patients post stroke.

18. There is Level 1a evidence that oral nutritional supplementation does not reduce the risk of death or dependency post stroke.

19. There is Level 1a evidence that the ALAnerv nutritional supplement may significantly reduce total lipid levels and increase HDL levels compared to conventional treatment post stroke.

20. There is Level 1b evidence that high-intensity dysphagia therapy results in improved swallowing function and less time to resuming a normal diet post stroke.

21. There is Level 1b and Level 2 evidence that dysphagia therapy does not reduce the risk of death or dependency post stroke, regardless of treatment intensity or diet type.

22. The one-year survival rate of patients with PEG feeding tubes post stroke varied from 16 - 67%, and functional recovery was reported in 2 - 28% of these patients.

23. There is Level 2 evidence that long-term NG tube feeding post stroke results in greater levels of malnutrition than oral feeding.

24. There is currently no evidence regarding the efficacy of total parenteral nutrition in the treatment of patients post stroke.

16. Nutritional Interventions Following Stroke pg. 34 of 42 www.ebrsr.com References Akner, G., & Cederholm, T. (2001). Treatment of protein-energy malnutrition in chronic nonmalignant disorders. Am.J.Clin Nutr., 74(1), 6-24. Amarenco, P., Bogousslavsky, J., Callahan, A., 3rd, Goldstein, L. B., Hennerici, M., Rudolph, A. E., Sillesen, H., Simunovic, L., Szarek, M., Welch, K. M., & Zivin, J. A. (2006). High-dose atorvastatin after stroke or transient ischemic attack. N Engl J Med, 355(6), 549-559. American Dietetic Association. (2000). Manual of clinical dietetics: American Dietetic Association. Aptaker, R. L., Roth, E. J., Reichhardt, G., Duerden, M. E., & Levy, C. E. (1994). Serum albumin level as a predictor of geriatric stroke rehabilitation outcome. Arch.Phys.Med.Rehabil., 75(1), 80-84. Aquilani, R., Emilio, B., Dossena, M., Baiardi, P., Testa, A., Boschi, F., Viglio, S., Iadarola, P., Pasini, E., & Verri, M. (2015). Correlation of deglutition in subacute ischemic stroke patients with peripheral blood adaptive immunity: Essential amino acid improvement. International Journal of Immunopathology and Pharmacology, 28(4), 576-583. Aquilani, R., Galli, M., Guarnaschelli, C., Fugazza, G., Lorenzoni, M., Varalda, E., Arrigoni, N., Achilli, M. P., Zclaschi, G. P., & Crespi, M. G. (1999). Prevalence of malnutrition and inadequate food intake in self-feeding rehabilitation patients with stroke. Europa Medicophysica, 35, 75-82. Aquilani, R., Scocchi, M., Boschi, F., Viglio, S., Iadarola, P., Pastoris, O., & Verri, M. (2008a). Effect of calorie-protein supplementation on the cognitive recovery of patients with subacute stroke. Nutr.Neurosci., 11(5), 235-240. Aquilani, R., Scocchi, M., Iadarola, P., Franciscone, P., Verri, M., Boschi, F., Pasini, E., & Viglio, S. (2008b). Protein supplementation may enhance the spontaneous recovery of neurological alterations in patients with ischaemic stroke. Clin Rehabil., 22(12), 1042-1050. Athyros, V. G., Tziomalos, K., Karagiannis, A., Wierzbicki, A. S., & Mikhailidis, D. P. (2010). Aggressive statin treatment, very low serum cholesterol levels and haemorrhagic stroke: is there an association? Curr Opin Cardiol, 25(4), 406-410. Axelsson, K., Asplund, K., Norberg, A., & Alafuzoff, I. (1988). Nutritional status in patients with acute stroke. Acta Med.Scand., 224(3), 217-224. Bakiner, O., Bozkirli, E., Giray, S., Arlier, Z., Kozanoglu, I., Sezgin, N., Sariturk, C., & Ertorer, E. (2013). Impact of early versus late enteral nutrition on cell mediated immunity and its relationship with glucagon like peptide-1 in intensive care unit patients: a prospective study. Crit Care, 17(3), R123. Bath, P. M., Bath, F. J., & Smithard, D. G. (2000). Interventions for dysphagia in acute stroke. Cochrane.Database.Syst.Rev.(2), CD000323. Beamer, N. B., Coull, B. M., Clark, W. M., Hazel, J. S., & Silberger, J. R. (1995). Interleukin-6 and interleukin-1 receptor antagonist in acute stroke. Ann.Neurol., 37(6), 800-805. Ben-Assayag, E., Shenhar-Tsarfaty, S., Bova, I., Berliner, S., Shopin, L., Peretz, H., Usher, S., Shapira, I., & Bornstein, N. M. (2007). Triggered C-reactive protein (CRP) concentrations and the CRP gene - 717A>G polymorphism in acute stroke or transient ischemic attack. Eur J Neurol, 14(3), 315-320. Bharosay, A., Bharosay, V. V., Bandyopadhyay, D., Sodani, A., Varma, M., & Baruah, H. (2014). Effect of lipid profile upon prognosis in ischemic and haemorrhagic cerebrovascular stroke. Indian Journal of Clinical Biochemistry, 29(3), 372-376. Brynningsen, P. K., Damsgaard, E. M., & Husted, S. E. (2007). Improved nutritional status in elderly patients 6 months after stroke. J.Nutr.Health Aging, 11(1), 75-79. Camara-Lemarroy, C. R., Ibarra-Yruegas, B. E., & Gongora-Rivera, F. (2014). Gastrointestinal complications after ischemic stroke. J Neurol Sci, 346(1-2), 20-25. Camerlingo, M., Valente, L., Tognozzi, M., Beretta, G. L., Moschini, L., & Cesana, B. M. (2011). C-reactive protein levels in the first three hours after acute cerebral infarction. Int J Neurosci, 121(2), 65- 68.

16. Nutritional Interventions Following Stroke pg. 35 of 42 www.ebrsr.com Carnaby-Mann, G., & Crary, M. (2005). Pill swallowing by adults with dysphagia. Archives of Otolaryngology Head & Neck Surgery, 131(11), 970. Carnaby, G., Hankey, G. J., & Pizzi, J. (2006). Behavioural intervention for dysphagia in acute stroke: a randomised controlled trial. Lancet Neurol., 5(1), 31-37. Chai, J., Chu, F. C., Chow, T. W., & Shum, N. C. (2008). Prevalence of malnutrition and its risk factors in stroke patients residing in an infirmary. Singapore Med.J., 49(4), 290-296. Chan, R., Chan, D., & Woo, J. (2013). The association of a priori and a posterior dietary patterns with the risk of incident stroke in chinese older people in Hong Kong. Journal of Nutrition, Health and Aging, 17(10), 866-874. Choi-Kwon, S., Yang, Y. H., Kim, E. K., Jeon, M. Y., & Kim, J. S. (1998). Nutritional status in acute stroke: undernutrition versus overnutrition in different stroke subtypes. Acta Neurol Scand, 98(3), 187- 192. Christensen, H., & Boysen, G. (2002). Blood glucose increases early after stroke onset: a study on serial measurements of blood glucose in acute stroke. Eur J Neurol, 9(3), 297-301. Crary, M. A., Carnaby-Mann, G. D., Miller, L., Antonios, N., & Silliman, S. (2006). Dysphagia and nutritional status at the time of hospital admission for ischemic stroke. J.Stroke Cerebrovasc.Dis., 15(4), 164-171. Crary, M. A., Humphrey, J. L., Carnaby-Mann, G., Sambandam, R., Miller, L., & Silliman, S. (2013). Dysphagia, nutrition, and hydration in ischemic stroke patients at admission and discharge from acute care. Dysphagia, 28(1), 69-76. Davalos, A., Ricart, W., Gonzalez-Huix, F., Soler, S., Marrugat, J., Molins, A., Suner, R., & Genis, D. (1996). Effect of malnutrition after acute stroke on clinical outcome. Stroke, 27(6), 1028-1032. Davis, J. P., Wong, A. A., Schluter, P. J., Henderson, R. D., O'Sullivan, J. D., & Read, S. J. (2004). Impact of premorbid undernutrition on outcome in stroke patients. Stroke, 35(8), 1930-1934. De Caterina, R., Scarano, M., Marfisi, R., Lucisano, G., Palma, F., Tatasciore, A., & Marchioli, R. (2010). Cholesterol-lowering interventions and stroke: insights from a meta-analysis of randomized controlled trials. J Am Coll Cardiol, 55(3), 198-211. Deitrick, J. E., Whedon, G. D., & Shorr, E. (1948). Effects of mobilization upon various metabolic and physiological functions. Am.J.Med., 4, 3-36. den Hertog, H. M., Vermeer, S. E., Zandbergen, A. A. M., Achterberg, S., Dippel, D. W. J., Algra, A., Kappelle, L. J., & Koudstaal, P. J. (2015). Safety and feasibiLIty of Metformin in patients with Impaired glucose Tolerance and a recent TIA or minor ischemic stroke (LIMIT) trial - a multicenter, randomized, open-label phase II trial. International Journal of Stroke, 10(1), 105- 109. Dennis, M. S., Lewis, S. C., & Warlow, C. (2005a). Effect of timing and method of enteral tube feeding for dysphagic stroke patients (FOOD): a multicentre randomised controlled trial. Lancet, 365(9461), 764-772. Dennis, M. S., Lewis, S. C., & Warlow, C. (2005b). Routine oral nutritional supplementation for stroke patients in hospital (FOOD): a multicentre randomised controlled trial. Lancet, 365(9461), 755- 763. DePippo, K. L., Holas, M. A., Reding, M. J., Mandel, F. S., & Lesser, M. L. (1994). Dysphagia therapy following stroke: a controlled trial. Neurology, 44(9), 1655-1660. Dziedzic, T., Pera, J., Trabka-Janik, E., Szczudlik, A., & Slowik, A. (2010). The impact of postadmission glycemia on stroke outcome: glucose normalisation is associated with better survival. Atherosclerosis, 211(2), 584-588. Elmstahl, S., Bulow, M., Ekberg, O., Petersson, M., & Tegner, H. (1999). Treatment of dysphagia improves nutritional conditions in stroke patients. Dysphagia, 14(2), 61-66.

16. Nutritional Interventions Following Stroke pg. 36 of 42 www.ebrsr.com Engberg, S., Vistisen, D., Lau, C., Glümer, C., Jørgensen, T., Pedersen, O., & Borch-Johnsen, K. (2009). Progression to impaired glucose regulation and diabetes in the population-based Inter99 study. Diabetes care, 32(4), 606-611. Fassbender, K., Rossol, S., Kammer, T., Daffertshofer, M., Wirth, S., Dollman, M., & Hennerici, M. (1994a). Proinflammatory cytokines in serum of patients with acute cerebral ischemia: kinetics of secretion and relation to the extent of brain damage and outcome of disease. J.Neurol.Sci., 122(2), 135-139. Fassbender, K., Schmidt, R., Mossner, R., Daffertshofer, M., & Hennerici, M. (1994b). Pattern of activation of the hypothalamic-pituitary-adrenal axis in acute stroke. Relation to acute confusional state, extent of brain damage, and clinical outcome. Stroke, 25(6), 1105-1108. Ferrarese, C., Mascarucci, P., Zoia, C., Cavarretta, R., Frigo, M., Begni, B., Sarinella, F., Frattola, L., & De Simoni, M. G. (1999). Increased cytokine release from peripheral blood cells after acute stroke. J.Cereb.Blood Flow Metab, 19(9), 1004-1009. Finestone, H. M., Greene-Finestone, L. S., Foley, N. C., & Woodbury, M. G. (2003). Measuring longitudinally the metabolic demands of stroke patients: resting energy expenditure is not elevated. Stroke, 34(2), 502-507. Finestone, H. M., Greene-Finestone, L. S., Wilson, E. S., & Teasell, R. W. (1995). Malnutrition in stroke patients on the rehabilitation service and at follow-up: prevalence and predictors. Arch.Phys.Med.Rehabil., 76(4), 310-316. Fleck, A. (1989). Clinical and nutritional aspects of changes in acute-phase proteins during inflammation. Proc.Nutr.Soc., 48(3), 347-354. Foley, N., Finestone, H., Woodbury, M. G., Teasell, R., & Greene, F. L. (2006). Energy and protein intakes of acute stroke patients. J.Nutr.Health Aging, 10(3), 171-175. Foley, N. C., Martin, R. E., Salter, K. L., & Teasell, R. W. (2009a). A review of the relationship between dysphagia and malnutrition following stroke. J.Rehabil.Med., 41(9), 707-713. Foley, N. C., Salter, K. L., Robertson, J., Teasell, R. W., & Woodbury, M. G. (2009b). Which reported estimate of the prevalence of malnutrition after stroke is valid? Stroke, 40(3), e66-e74. FOOD Trial Collaboration. (2003). Poor nutritional status on admission predicts poor outcomes after stroke: observational data from the FOOD trial. Stroke., 34(6), 1450-1456. Gabay, C., & Kushner, I. (1999). Acute-phase proteins and other systemic responses to inflammation. N.Engl.J.Med., 340(6), 448-454. Gariballa, S. E. (2001). Malnutrition in hospitalized elderly patients: when does it matter? Clinical nutrition, 20(6), 487-491. Gariballa, S. E., Fotherby, M. D., Parker, S. G., & Castleden, C. M. (1996). Changes in Serum Lipid following Acute Stroke. Clinical Science, 90(Pt s34), 11P-11P. Gariballa, S. E., Parker, S. G., Taub, N., & Castleden, C. M. (1998a). Influence of nutritional status on clinical outcome after acute stroke. Am.J.Clin Nutr., 68(2), 275-281. Gariballa, S. E., Parker, S. G., Taub, N., & Castleden, C. M. (1998b). A randomized, controlled, a single- blind trial of nutritional supplementation after acute stroke. JPEN J.Parenter.Enteral Nutr., 22(5), 315-319. Geeganage, C., Beavan, J., Ellender, S., & Bath, P. M. (2012). Interventions for dysphagia and nutritional support in acute and subacute stroke. Cochrane Database of Systematic Reviews(10). Gomes, C. A., Jr., Lustosa, S. A., Matos, D., Andriolo, R. B., Waisberg, D. R., & Waisberg, J. (2012). Percutaneous endoscopic gastrostomy versus nasogastric tube feeding for adults with swallowing disturbances. Cochrane Database Syst Rev, 3, CD008096. Grant, P. S., & Khalid, A. (2010). A Prospective, Observational Study of the Management of Hyperglycaemia in Acute Stroke. What is the Optimum Level of Blood Glucose at which to Intervene? The British Journal of Diabetes & Vascular Disease, 10(6), 287-291.

16. Nutritional Interventions Following Stroke pg. 37 of 42 www.ebrsr.com Gray, C. S., Hildreth, A. J., Sandercock, P. A., O'Connell, J. E., Johnston, D. E., Cartlidge, N. E., Bamford, J. M., James, O. F., & Alberti, K. G. (2007). Glucose-potassium-insulin infusions in the management of post-stroke hyperglycaemia: the UK Glucose Insulin in Stroke Trial (GIST-UK). Lancet Neurol, 6(5), 397-406. Gulsen Atalay, B., Yagmur, C., Nursal, T., Atalay, H., Sezgin, N., Giray, S., & Selcuk, H. (2014). Effects of enteral protein supplements on nutritional status and clinical outcome in unconscious elderly malnourished patients. Nutrition & Dietetics, 71(2), 66-72. Gupta, A., Prabhakar, S., Modi, M., Bhadada, S. K., Kalaivani, M., Lal, V., & Khurana, D. (2016). Effect of Vitamin D and calcium supplementation on ischaemic stroke outcome: a randomised controlled open-label trial. Int J Clin Pract, 70(9), 764-770. Gupta, P., Yadav, S., & Singal, K. K. (2014). Cerebroprotein hydrolysate: Innovation in the treatment of neurodegenerative disorders. Journal, Indian Academy of Clinical Medicine, 15(2), 132-133. Ha, L., Hauge, T., & Iversen, P. O. (2010b). Body composition in older acute stroke patients after treatment with individualized, nutritional supplementation while in hospital. BMC.Geriatr., 10, 75. Ha, L., Hauge, T., Spenning, A. B., & Iversen, P. O. (2010a). Individual, nutritional support prevents undernutrition, increases muscle strength and improves QoL among elderly at nutritional risk hospitalized for acute stroke: a randomized, controlled trial. Clin Nutr., 29(5), 567-573. Hama, S., Kitaoka, T., Shigenobu, M., Watanabe, A., Imura, I., Seno, H., Tominaga, A., Arita, K., & Kurisu, K. (2005). Malnutrition and nonthyroidal illness syndrome after stroke. Metabolism, 54(6), 699- 704. Hamidon, B. B., Abdullah, S. A., Zawawi, M. F., Sukumar, N., Aminuddin, A., & Raymond, A. A. (2006). A prospective comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding in patients with acute dysphagic stroke. Med.J.Malaysia, 61(1), 59-66. Hasani, S. A., Ziai, S. A., Mehrpour, M., Amiri, M., & Motamed, M. R. (2011). Acute phase reactants as a prognostic factor in acute stroke. Basic and Clinical Neuroscience, 3(1), 30-34. Ivey, F. M., Ryan, A. S., Hafer-Macko, C. E., Goldberg, A. P., & Macko, R. F. (2007). Treadmill aerobic training improves glucose tolerance and indices of insulin sensitivity in disabled stroke survivors: a preliminary report. Stroke, 38(10), 2752-2758. James, A., Kapur, K., & Hawthorne, A. B. (1998). Long-term outcome of percutaneous endoscopic gastrostomy feeding in patients with dysphagic stroke. Age Ageing, 27(6), 671-676. James, R., Gines, D., Menlove, A., Horn, S. D., Gassaway, J., & Smout, R. J. (2005). Nutrition support (tube feeding) as a rehabilitation intervention. Arch.Phys.Med.Rehabil., 86(12 Suppl 2), S82-S92. Jia, Q., Liu, G., Zheng, H., Zhao, X., Wang, C., Wang, Y., Liu, L., & Wang, Y. (2014). Impaired glucose regulation predicted 1-year mortality of chinese patients with ischemic stroke: data from abnormal glucose regulation in patients with acute stroke across china. Stroke (00392499), 45(5), 1498-1500. Johanson, J. F., Sonnenberg, A., Koch, T. R., & McCarty, D. J. (1992). Association of constipation with neurologic diseases. Dig.Dis.Sci., 37(2), 179-186. Kim, E. J., Yoon, Y. H., Kim, W. H., Lee, K. L., & Park, J. M. (2013). The clinical significance of the mini- nutritional assessment and the scored patient-generated subjective global assessment in elderly patients with stroke. Ann Rehabil Med, 37(1), 66-71. Kim, S., & Byeon, Y. (2014). Comparison of nutritional status indicators according to feeding methods in patients with acute stroke. Nutritional Neuroscience, 17(3), 138-144. Kirby, D. F., Delegge, M. H., & Fleming, C. R. (1995). American Gastroenterological Association technical review on tube feeding for enteral nutrition. Gastroenterology, 108(4), 1282-1301. Kojima, G., Bell, C., Abbott, R. D., Launer, L., Chen, R., Motonaga, H., Ross, G. W., Curb, J. D., & Masaki, K. (2012). Low dietary vitamin D predicts 34-year incident stroke: the Honolulu Heart Program. Stroke, 43(8), 2163-2167.

16. Nutritional Interventions Following Stroke pg. 38 of 42 www.ebrsr.com Krieger, R. P., Brady, S., Stewart, R. J., Terry, A., & Brady, J. J. (2010). Predictors of returning to oral feedings after feeding tube placement for patients poststroke during inpatient rehabilitation. Top.Stroke Rehabil., 17(3), 197-203. Kumar, S., Langmore, S., Goddeau, R. P., Jr., Alhazzani, A., Selim, M., Caplan, L. R., Zhu, L., Safdar, A., Wagner, C., Frayne, C., Searls, D. E., & Schlaug, G. (2012). Predictors of percutaneous endoscopic gastrostomy tube placement in patients with severe dysphagia from an acute-subacute hemispheric infarction. J.Stroke Cerebrovasc.Dis., 21(2), 114-120. Laird, E. A., Coates, V. E., Ryan, A. A., McCarron, M. O., Lyttle, D., & McCrum-Gardner, E. (2014). Hypoglycaemia risk among a hospitalised stroke patient cohort: A case for increased vigilance in glucose monitoring. Journal of Clinical Neuroscience, 21(2), 232-235. Lalouschek, W., Lang, W., Greisenegger, S., & Mullner, M. (2003). Determination of lipid profiles and use of statins in patients with ischemic stroke or transient ischemic attack. Stroke, 34(1), 105-110. Larsson, S. C. (2013). Dietary fats and other nutrients on stroke. Current Opinion in Lipidology, 24(1), 41- 48. Li, X., Song, G., Jin, Y., Liu, H., Li, C., Han, C., & Ren, S. (2014). Higher level of heme oxygenase-1 in patients with stroke than TIA. Journal of Thoracic Disease, 6(6), 772-777. Lim, H. J., & Choue, R. (2010). Nutritional status assessed by the Patient-Generated Subjective Global Assessment (PG-SGA) is associated with qualities of diet and life in Korean cerebral infarction patients. Nutrition, 26(7-8), 766-771. Lin, L. C., Wang, S. C., Chen, S. H., Wang, T. G., Chen, M. Y., & Wu, S. C. (2003). Efficacy of swallowing training for residents following stroke. J.Adv.Nurs., 44(5), 469-478. Lindsberg, P. J., Tuomi, T., & Kaste, M. (2011). Oral glucose tolerance test should be performed after stroke and transient ischemic attack. International Journal of Stroke, 6(4), 317-320. Maeshima, S., Osawa, A., Hayashi, T., & Tanahashi, N. (2013). Factors Associated with Prognosis of Eating and Swallowing Disability after Stroke: A Study from a Community-based Stroke Care System. Journal of Stroke & Cerebrovascular Diseases, 22(7), 926-U594. Mahmood, A., Sharif, M. A., Khan, M. N., & Ali, U. Z. (2010). Comparison of serum lipid profile in ischaemic and haemorrhagic stroke. J Coll Physicians Surg Pak, 20(5), 317-320. Mangat, A., Grewal, D., Kaur, P., Jyotsna, R., Singh, R., & Pandian, J. D. (2013). Dietary patterns in stroke patients in Northwest India. Nutritional Neuroscience, 16(6), 288-292. Manolescu, B. N., Berteanu, M., & Cintezã, D. (2013). Effect of the nutritional supplement ALAnerv® on the serum PON1 activity in post-acute stroke patients. Pharmacological Reports, 65(3), 743-750. Martin, R. J., Ratan, R. R., Reding, M. J., & Olsen, T. S. (2012). Higher Blood Glucose within the Normal Range Is Associated with More Severe Strokes. Stroke research and treatment, 2012. Martineau, J., Bauer, J. D., Isenring, E., & Cohen, S. (2005). Malnutrition determined by the patient- generated subjective global assessment is associated with poor outcomes in acute stroke patients. Clin Nutr., 24(6), 1073-1077. Mosselman, M. J., Kruitwagen, C., Schuurmans, M. J., & Hafsteinsdottir, T. B. (2013). Malnutrition and Risk of Malnutrition in Patients With Stroke: Prevalence During Hospital Stay. Journal of Neuroscience Nursing, 45(4), 194-204. Muir, K. W., Weir, C. J., Alwan, W., Squire, I. B., & Lees, K. R. (1999). C-reactive protein and outcome after ischemic stroke. Stroke, 30(5), 981-985. Muò, R., Cancialosi, P., Galimberti, L., Cacciola, B. C., Gilardone, M., & Schindler, A. (2015). Validation of the Italian version of the American Speech-Language and Hearing Association—Functional Assessment of Communication Skills for adults (I-ASHA-FACS). Aphasiology, 29(9), 1110-1130. Murray, J., Doeltgen, S., Miller, M., & Scholten, I. (2015). A Descriptive Study of the Fluid Intake, Hydration, and Health Status of Rehabilitation Inpatients without Dysphagia Following Stroke. J Nutr Gerontol Geriatr, 34(3), 292-304.

16. Nutritional Interventions Following Stroke pg. 39 of 42 www.ebrsr.com Murros, K., Fogelholm, R., Kettunen, S., & Vuorela, A. L. (1993). Serum cortisol and outcome of ischemic brain infarction. J.Neurol.Sci., 116(1), 12-17. Nakajima, M., Inatomi, Y., Yonehara, T., Hashimoto, Y., & Hirano, T. (2012). Acquisition of oral intake in severely dysphagic patients with acute stroke: a single-center, observational study involving a database of 4972 consecutive stroke patients. J.Neurol.Sci., 323(1-2), 56-60. Norton, B., Homer-Ward, M., Donnelly, M. T., Long, R. G., & Holmes, G. K. (1996). A randomised prospective comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding after acute dysphagic stroke. BMJ, 312(7022), 13-16. Nyswonger, G. D., & Helmchen, R. H. (1992). Early enteral nutrition and length of stay in stroke patients. J.Neurosci.Nurs., 24(4), 220-223. Ogata, T., Kamouchi, M., Matsuo, R., Hata, J., Kuroda, J., Ago, T., Sugimori, H., Inoue, T., & Kitazono, T. (2014). Gastrointestinal bleeding in acute ischemic stroke: recent trends from the fukuoka stroke registry. Cerebrovasc Dis Extra, 4(2), 156-164. Oprea, E., Berteanu, M., Cinteza, D., & Manolescu, B. N. (2013). The effect of the ALAnerv nutritional supplement on some oxidative stress markers in postacute stroke patients undergoing rehabilitation. Applied Physiology Nutrition and Metabolism-Physiologie Appliquee Nutrition Et Metabolisme, 38(6), 613-620. Ormstad, H., Verkerk, R., Aass, H. C., Amthor, K. F., & Sandvik, L. (2013). Inflammation-induced catabolism of tryptophan and tyrosine in acute ischemic stroke. J Mol Neurosci, 51(3), 893-902. Paquereau, J., Allart, E., Romon, M., & Rousseaux, M. (2014). The long-term nutritional status in stroke patients and its predictive factors. Journal of Stroke and Cerebrovascular Diseases, 23(6), 1628- 1633. Park, R. H., Allison, M. C., Lang, J., Spence, E., Morris, A. J., Danesh, B. J., Russell, R. I., & Mills, P. R. (1992). Randomised comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding in patients with persisting neurological dysphagia. BMJ, 304(6839), 1406-1409. Pilz, S., Dobnig, H., Fischer, J. E., Wellnitz, B., Seelhorst, U., Boehm, B. O., & Marz, W. (2008). Low vitamin d levels predict stroke in patients referred to coronary angiography. Stroke, 39(9), 2611-2613. Poels, B. J., Brinkman-Zijlker, H. G., Dijkstra, P. U., & Postema, K. (2006). Malnutrition, eating difficulties and feeding dependence in a stroke rehabilitation centre. Disabil.Rehabil., 28(10), 637-643. Rabadi, M. H., Coar, P. L., Lukin, M., Lesser, M., & Blass, J. P. (2008). Intensive nutritional supplements can improve outcomes in stroke rehabilitation. Neurology, 71(23), 1856-1861. Rowat, A. (2014). Dysphagia, nutrition and hydration post stroke. British Journal of Nursing, 23(12), 634- 634. San Luis, C. O., Staff, I., Ollenschleger, M. D., Fortunato, G. J., & McCullough, L. D. (2013). Percutaneous endoscopic gastrostomy tube placement in left versus right middle cerebral artery stroke: effects of laterality. Neurorehabilitation, 33(2), 201-208. Schaller, B. J., Graf, R., & Jacobs, A. H. (2006). Pathophysiological changes of the gastrointestinal tract in ischemic stroke. Am J Gastroenterol, 101(7), 1655-1665. Schonheyder, F., Heilskov, N. C., & Olesen, K. (1954). Isotopic studies on the mechanism of negative nitrogen balance produced by immobilization. Scand.J.Clin Lab Invest, 178-188. Scivoletto, G., Petrelli, A., Lucente, L. D., & Castellano, V. (1997). Psychological investigation of spinal cord injury patients. Spinal cord, 35(8), 516-520. Selakovic, V. M., Jovanovic, M. D., & Jovicic, A. (2002). Changes of cortisol levels and index of lipid peroxidation in cerebrospinal fluid of patients in the acute phase of completed stroke. Vojnosanit Pregl, 59(5), 485-491. Shaafi, S., Sharifipour, E., Rahmanifar, R., Hejazi, S., Andalib, S., Nikanfar, M., Baradarn, B., & Mehdizadeh, R. (2014). Interleukin-6, a reliable prognostic factor for ischemic stroke. Iran J Neurol, 13(2), 70-76.

16. Nutritional Interventions Following Stroke pg. 40 of 42 www.ebrsr.com Sharma, S., Cruickshank, J. K., Green, D. M., Vik, S., Tome, A., & Kolonel, L. N. (2013). Impact of Diet on Mortality From Stroke: Results From the US Multiethnic Cohort Study. Journal of the American College of Nutrition, 32(3), 151-159. Sonnenberg, A., Tsou, V. T., & Muller, A. D. (1994). The "institutional colon": a frequent colonic dysmotility in psychiatric and neurologic disease. Am.J.Gastroenterol., 89(1), 62-66. Staal-van den Brekel, A., Dentener, M. A., Schols, A. M., Buurman, W. A., & Wouters, E. F. (1995). Increased resting energy expenditure and weight loss are related to a systemic inflammatory response in lung cancer patients. Journal of clinical oncology, 13(10), 2600-2605. Syrjanen, J., Teppo, A. M., Valtonen, V. V., Iivanainen, M., & Maury, C. P. (1989). Acute phase response in cerebral infarction. J.Clin Pathol., 42(1), 63-68. Szczudlik, A., Dziedzic, T., Bartus, S., Slowik, A., & Kieltyka, A. (2004). Serum interleukin-6 predicts cortisol release in acute stroke patients. J Endocrinol Invest, 27(1), 37-41. Tapia-Pérez, J. H., Gehring, S., Zilke, R., & Schneider, T. (2014). Effect of increased glucose levels on short-term outcome in hypertensive spontaneous intracerebral hemorrhage. Clinical Neurology and Neurosurgery, 118, 37-43. Tu, W. J., Zhao, S. J., Xu, D. J., & Chen, H. (2014). Serum 25-hydroxyvitamin D predicts the short-term outcomes of Chinese patients with acute ischaemic stroke. Clinical Science, 126(5-6), 339-346. Turetsky, A., Goddeau, R. P., Jr., & Henninger, N. (2015). Low Serum Vitamin D Is Independently Associated with Larger Lesion Volumes after Ischemic Stroke. J Stroke Cerebrovasc Dis, 24(7), 1555-1563. Tuttolomondo, A., Casuccio, A., Butta, C., Pecoraro, R., Di Raimondo, D., Della Corte, V., Arnao, V., Clemente, G., Maida, C., Simonetta, I., Miceli, G., Lucifora, B., Cirrincione, A., Di Bona, D., Corpora, F., Maugeri, R., Iacopino, D. G., & Pinto, A. (2015). Mediterranean Diet in patients with acute ischemic stroke: Relationships between Mediterranean Diet score, diagnostic subtype, and stroke severity index. Atherosclerosis, 243(1), 260-267. Ullman, T., & Reding, M. (1996). Gastrointestinal dysfunction in stroke. Semin.Neurol., 16(3), 269-275. Unosson, M., Ek, A. C., Bjurulf, P., von Schenck, H., & Larsson, J. (1994a). Feeding dependence and nutritional status after acute stroke. Stroke, 25(2), 366-371. Unosson, M., Ek, A. C., Bjurulf, P., von, S. H., & Larsson, J. (1994b). Feeding dependence and nutritional status after acute stroke. Stroke, 25(2), 366-371. Wang, J., Luo, B., Xie, Y., Hu, H. Y., Feng, L., & Li, Z. N. (2014). Evaluation methods on the nutritional status of stroke patients. European Review for Medical and Pharmacological Sciences, 18(24), 3902-3907. Wanklyn, P., Cox, N., & Belfield, P. (1995). Outcome in patients who require a gastrostomy after stroke. Age Ageing, 24(6), 510-514. Westergren, A., Karlsson, S., Andersson, P., Ohlsson, O., & Hallberg, I. R. (2001a). Eating difficulties, need for assisted eating, nutritional status and pressure ulcers in patients admitted for stroke rehabilitation. J.Clin Nurs., 10(2), 257-269. Westergren, A., Ohlsson, O., & Rahm, H., I. (2001b). Eating difficulties, complications and nursing interventions during a period of three months after a stroke. J.Adv.Nurs., 35(3), 416-426. Wijdicks, E. F., & McMahon, M. M. (1999). Percutaneous endoscopic gastrostomy after acute stroke: complications and outcome. Cerebrovasc.Dis., 9(2), 109-111. Willey, J. Z., Xu, Q., Boden-Albala, B., Paik, M. C., Moon, Y. P., Sacco, R. L., & Elkind, M. S. (2009). Lipid profile components and risk of ischemic stroke: the Northern Manhattan Study (NOMAS). Arch Neurol, 66(11), 1400-1406. Witham, M. D., Dove, F. J., Sugden, J. A., Doney, A. S., & Struthers, A. D. (2012). The effect of vitamin D replacement on markers of vascular health in stroke patients - a randomised controlled trial. Nutr Metab Cardiovasc Dis, 22(10), 864-870.

16. Nutritional Interventions Following Stroke pg. 41 of 42 www.ebrsr.com Yaghi, S., & Elkind, M. S. (2016). Lipid Control and Beyond: Current and Future Indications for Statin Therapy in Stroke. Curr Treat Options Cardiovasc Med, 18(4), 27. Yoo, S. H., Kim, J. S., Kwon, S. U., Yun, S. C., Koh, J. Y., & Kang, D. W. (2008). Undernutrition as a predictor of poor clinical outcomes in acute ischemic stroke patients. Arch.Neurol., 65(1), 39-43. Young, B., Ott, L., Norton, J., Tibbs, P., Rapp, R., McClain, C., & Dempsey, R. (1985). Metabolic and nutritional sequelae in the non-steroid treated head injury patient. Neurosurgery, 17(5), 784- 791. Zaherah Mohamed Shah, F., Suraiya, H. S., Poi, P. J., Tan, K. S., Lai, P. S., Ramakrishnan, K., & Mahadeva, S. (2012). Long-term nasogastric tube feeding in elderly stroke patients--an assessment of nutritional adequacy and attitudes to gastrostomy feeding in Asians. J Nutr Health Aging, 16(8), 701-706. Zheng, T., Zhu, X., Liang, H., Huang, H., Yang, J., & Wang, S. (2015). Impact of early enteral nutrition on short term prognosis after acute stroke. Journal of Clinical Neuroscience, 22(9), 1473-1476.

16. Nutritional Interventions Following Stroke pg. 42 of 42 www.ebrsr.com