Effect of Malnutrition After Acute Stroke on Clinical Outcome
Background and Purpose Malnutrition has received little attention in acute stroke, although it represents a risk of decreased immunity and nosocomial infections. Our objectives were to determine the prevalence of malnutrition after 1 week of hospitalization in acute stroke and to establish its relation to the stress response and neurological outcome.
Methods The study included 104 patients with an acute stroke of less than 24 hours’ duration. Nutritional parameters (triceps skinfold thickness, midarm muscle circumference, serum albumin, and calorimetry) were evaluated at admission and after 1 week. Stress response (free urinary cortisol) was measured daily during the first week. Neurological deficit was evaluated by the Canadian Stroke Scale. Clinical outcome was estimated by the Barthel Index 1 month after the acute stroke. Patients received an oral standard diet or polymeric enteral nutrition when they had swallowing difficulties.
Results Protein-energy malnutrition was observed in 16.3% of patients at inclusion and in 26.4% after the first week, with a significant decrease in fat (P=.002) and visceral protein compartments (P=.049). Malnourished patients showed higher stress reaction and increased frequency of infections and bedsores in comparison with the appropriately nourished group. Multiple logistic regression analysis showed that malnutrition after 1 week (odds ratio, 3.5; 95% confidence interval, 1.2 to 10.2) and elevated free urinary cortisol (odds ratio, 3.3; confidence interval, 1.05 to 10.2) increased the risk of poor outcome (death or Barthel Index ≤50 on the 30th day of follow-up) independently of age and nutritional status at admission.
Conclusions Our findings suggest that protein-energy malnutrition after acute stroke is a risk factor for poor outcome. Early appropriate enteral caloric feeding did not prevent malnutrition during the first week of hospitalization.
Malnutrition has received little attention in acute stroke, although it is associated with an increased prevalence of complications, impaired immunologic function, and a high mortality rate among hospitalized patients on medical or surgical wards.1 2 3 Some studies suggest that appropriate nutritional support reduces complications and mortality.4
Malnourishment during hospitalization in acute stroke has been related to eating problems, age, poor nutritional status on admission, and immobilization in patients with impaired functional capacity.5 6 Increased energy requirements are not relevant in terms of this malnourishment, since resting energy expenditure is not high after a stroke, probably because of decreased physical activity.7 Stress response in acute stroke may lead to malnutrition by hypercatabolism and visceral consumption, and both stress and malnutrition could worsen the prognosis by decreasing cellular immunity.
Our objectives were to determine the prevalence of malnutrition in acute stroke patients after 1 week of hospitalization and to establish its relation to stress response and neurological outcome.
Subjects and Methods
Between October 1990 and January 1992, 202 patients admitted consecutively for an acute stroke were followed up to evaluate nutritional status, glucose profile, and stress response. The protocol was approved by the Ethics Committee of the Hospital Doctor Josep Trueta. Inclusion criteria were (1) age younger than 80 years; (2) admission delay of less than 24 hours from onset of symptoms; (3) persistence of neurological deficit at entry; (4) absence of subarachnoid hemorrhage, neurological sequelae of a previous stroke, liver disease, renal failure, thyroid dysfunction, past or present inflammatory diseases, hematologic malignancies, solid tumors, and medical conditions for which death was likely within 1 year; (5) no ongoing treatment with α-methyldopa, clonidine, β-blockers, benzodiazepines, or neuroleptics; (6) cranial CT scan performed during the first week; and (7) obtained informed consent. Ninety-eight patients were excluded. Only one reason for exclusion was recorded for each patient: age (23 patients), delay in admission (18), recovery from transient ischemic attack before inclusion (12), refusal to participate or transfer to another hospital (11), ongoing treatment with certain drugs (specified above) (11), sequelae of a previous stroke (8), cancer or serious diseases (2), CT scan not available (1), subarachnoid hemorrhage (4), and death within the first 24 hours before the initial nutritional evaluation (8). In excluded patients, nutritional parameters were only measured on the first day of hospitalization.
Patients were admitted to the emergency unit within the first 24 hours after stroke onset. Laboratory parameters evaluated in this study were glycemia levels at admission before administration of intravenous fluids; fasting glycemia, glycated hemoglobin, fructosamine, and serum albumin within the first 24 hours and weekly during hospitalization; serum cortisol on days 1, 2, 4, and 7; and 24-hour free urinary cortisol every day during the first week, following the same method as described in a previous report.8
Nutritional status was assessed within the first 24 hours after admission and weekly during hospitalization with the use of three well-recognized, reliable nutritional parameters: TSF, MAMC, and serum albumin concentration, which represents the fat, muscle protein, and visceral compartments, respectively.9 TSF was measured with a skinfold caliper (John Bull, British Indicators Ltd), and MAMC was measured as described by the World Health Organization.10 TSF and MAMC measurements were taken from the left arm, except when the left arm was paralyzed. All anthropometric measurements were performed by the same investigator (S.S.). To minimize intraoperator variability, the mean of three consecutive measurements was recorded. Values for each variable were expressed as a percentage of the 50th percentile, adjusted by age and sex, of a large sample of a healthy population living in the area covered by our hospital.11 Weight was evaluated at admission with an electronic lift scale (Ambulift C3, Arjo). Indirect calorimetry (Calorimet CS, ICOR) was used to study energy expenditure in a sample of 46 patients. All subjects were measured after an overnight fast in a recumbent position and at least 1 hour after the last nursing or medical intervention.
Neurological deficit was evaluated by CSS score at admission, day 7, and the first month of follow-up by two neurologists (A.D., S.S.). Functional capacity was determined by the BI for daily activities on day 30. This scale was interpreted in accordance with previous studies as follows: 0 to 50, severely disabled; 55 to 90, moderately disabled; and 95 to 100, functionally independent. Mortality rate was evaluated at 3 months. Complications during the stay at the neurological unit were recorded according to predefined criteria: respiratory infection (fever, purulent sputum, or bronchial secretions with or without radiographic confirmation of pneumonia), urinary infection (bacteriologic confirmation of >100 000 organisms per milliliter of urine), and bedsores (full-thickness skin loss >2 cm in diameter). The patients received postural changes and early rehabilitation and were managed according to the recommendations of the Spanish Cerebrovascular Study Group.12
Patients were fed with an oral standard diet that supplied approximately 2000 kcal and 16 g of nitrogen per day or, when they had swallowing difficulties, with polymeric enteral nutrition (Osmolite, Abbott Laboratories) that supplied 30 kcal/kg and 14 g of nitrogen. Dysphagia was diagnosed in alert patients unable to swallow a 10-mL mouthful of water and in unconscious patients. Patients were encouraged to eat all meals served. The enteral diet was continuously infused into the stomach (or the duodenum in comatose patients) through a fine-bore nasogastric feeding tube with the aid of a peristaltic pump for a minimum of 1 week. The head of the bed was maintained elevated day and night to prevent gastroesophageal reflux and pulmonary aspiration. Oral nutrients were administered progressively after the 7th day of hospitalization in those patients who recovered swallowing function.
Protein-energy malnutrition was diagnosed when serum albumin was less than 35 g/L or when TSF or MAMC was less than the 10th percentile of our reference population. Cutoff values were TSF below 59.5% and 62.5% and MAMC below 85% and 86.4% for men and women, respectively. For the purpose of this study we focused on nutritional status at the 7th day of hospitalization because during the first week all the patients remained hospitalized and under controlled nutritional support.
Patients were classified into two groups according to functional capacity 1 month after the acute stroke: (1) good outcome, including patients moderately disabled or independent (BI >50), or (2) poor outcome (dead or BI ≤50).
The χ2 test was used to compare proportions. Depending on the normality and homogeneity of the variances, one-way ANOVA or the Mann-Whitney rank sum test was used to compare continuous variables between groups; differences in plasmatic and urinary cortisol between malnourished and nonmalnourished patients during the first week were studied with ANOVA with repeated measures. We used the t test for paired data or the Wilcoxon test for paired comparisons of the nutritional parameters between admission and day 7.
To determine whether malnutrition after the first week of hospitalization was an independent predictor of poor outcome, we used stepwise multiple logistic regression analysis. We assigned a value of 0 to good outcome and a value of 1 to poor outcome. Age, sex, protein-energy malnutrition at admission and after 1 week, and the mean value of daily urinary free cortisol during the first week were included as covariates. Age and urinary cortisol were categorized (0, low; 1, high); cutoff values were established at 68 years and 1430 nmol/24 h, respectively, according to the method described by Robert et al.13 Two different models were fitted, which either included or excluded CSS score and swallowing difficulties at admission as covariates.
The mean age of the 104 patients included in the study was 66±10 years, and 64% were male. Hypertension was present in 52%, diabetes mellitus in 16%, heart diseases in 28%, and previous transient ischemic attack or stroke in 18%. Cerebral hemorrhage was diagnosed in 25 patients, lacunar infarct in 17, cardioembolic infarct in 21, and thrombotic infarct in 41. Swallowing problems at admission were present in 43 patients (41.3%). The average time of hospitalization was 20±16 days. Nineteen patients (18.3%) died during the 3-month monitoring period, 13 of them in the first week and 15 in the first month. The suspected causes of death were encephalic herniation (5), thrombosis of the basilar artery (3), sepsis (2), pulmonary disease (1), congestive heart failure (1), sudden death (1), metabolic disease (1), recurrent stroke (2), and unknown (3).
Prevalence of malnutrition at admission was 16.3% in the 104 included patients and 16.7% in the 90 evaluated patients who did not meet the inclusion criteria. The proportion of malnutrition increased during hospitalization, at 26.4% after 1 week in 91 surviving patients and 35% after 2 weeks in 43 patients who remained in our hospital. The number of patients with serum albumin, TSF, and MAMC below the cutoff limits is shown in Table 1⇓. Between admission and the 7th day of hospitalization, paired tests showed a significant decrease in TSF (118±54% versus 110±42%; P=.002) and serum albumin concentration (40.7±4.6 g/L versus 39.5±5.3 g/L; P=.049) but not in MAMC (105±11% versus 105±11%). Although nutritional parameters were not significantly different at admission between patients with or without swallowing difficulties, malnutrition after the first week was more frequent (48.3%) in patients with swallowing incapacity than in those with normal swallowing function (13.6%) (P=.0004).
Clinical characteristics and laboratory parameters at admission in malnourished and nonmalnourished patients after the first week of hospitalization are shown in Tables 2⇓ and 3⇓. Malnourishment was related significantly to low CSS scores at admission, swallowing difficulties, poor nutritional status on admission, and nonlacunar infarcts. Serum albumin at admission was significantly lower, and free urinary cortisol significantly higher, in patients who showed abnormal nutritional parameters after the first week.
Energy expenditure was measured by calorimetry in 46 patients at admission and repeated in 43 patients after 1 week. Those patients with abnormal nutritional parameters after the first week of hospitalization had lower caloric requirements at admission than patients with normal nutrition (1064±483 kcal/d versus 1706±559 kcal/d; P=.007). This significant difference disappeared at the 7th day (1680±556 kcal/d versus 1580±532 kcal/d; P=NS). There was not a statistically significant correlation between the initial energy expenditure and plasmatic and urinary cortisol levels. The stress reaction during the first week after admission was closely related to malnutrition at the 7th day; free urinary cortisol values (Figure⇓) and plasmatic cortisol (P=.018) were significantly higher in malnourished patients and in both cases decreased with time (P<.001).
Urinary or respiratory infections (50% versus 24%; P=.017) and bedsores (17% versus 4%; P=.054) were more prevalent in patients with protein-energy malnutrition than in those with normal nutritional parameters. Aspirative pneumonia was recorded in 4 of the 42 patients (9.5%) who received enteral nutrition and in none of those fed orally; in 1 patient aspiration was the cause of death.
CSS score and BI at day 30 were significantly lower in the group of patients with malnutrition (Table 4⇓). Mortality after the first week of hospitalization was more frequent in patients with abnormal nutritional parameters (5 patients versus 1 patient; P=.005). The median duration of hospitalization was significantly longer in malnourished patients (28 days; range, 9 to 86 days) than in nonmalnourished patients (17 days; range, 6 to 49 days) (P=.001).
We observed malnutrition in 17 of 41 patients with poor outcome (41%) and in 7 of 50 with good outcome (14%) (P=.003). Multiple logistic regression analysis showed that malnutrition after the first week of hospitalization (OR, 3.5; 95% CI, 1.2 to 10.2) and increased free urinary cortisol (OR, 3.3; 95% CI, 1.05 to 10.2) predicted poor outcome independently of age, sex, and nutritional status at admission. When the prognostic variables CSS (OR, 10.5; 95% CI, 3 to 37) and swallowing disability at admission (OR, 5.9; 95% CI, 1.6 to 22) were included in the logistic regression, malnutrition and average free urinary cortisol during the first week were not selected by the model as independent predictors of poor prognosis.
To our knowledge this is the first study in acute stroke to analyze the influence of stress response on nutritional parameters and the effect of malnutrition during the first week of hospitalization on clinical outcome.
Nutritional status in cerebrovascular diseases has been reported rarely. Two previous studies, using similar nutritional parameters after stroke onset, found abnormal nutritional status in 16% and 8% of patients.6 14 Malnutrition was observed in 16.3% of our patients at inclusion, in 26.4% after the first week, and in 35% after the second week from admission. This malnourishment during hospitalization has been observed in other studies5 6 14 and was evident in 49% of the stroke patients admitted to a rehabilitation service.15 Body fat and visceral protein were the nutritional compartments more significantly decreased in our study, as in previous reports.6 14 The most important factors related to malnourishment were stroke severity (CSS score ≤5) and swallowing difficulties. Individuals immobilized as a result of low functional capacity lose body cell mass irrespective of nutritional intake because of reduced synthesis of proteins,16 but why feeding dependence is associated with malnutrition has not been clarified.6
An important finding in our study was that nutritional parameters deteriorated despite aggressive early enteral nutrition in patients with swallowing problems. In a sample of 46 patients, the total number of calories administered was greater than that calculated by calorimetry, and patients with malnourishment had a lower resting energy expenditure than those who were well nourished. Therefore, nutritional deterioration during hospitalization may be attributed to factors other than inadequate number of calories administered. Our results demonstrated that malnourished patients had a higher stress reaction; therefore, patients with acute stroke must be considered moderately hypercatabolic but with low caloric requirements. Catabolic disease alters body composition rapidly, with a gradual shrinkage of body fat and body cell mass compartments.17 The neuroendocrine response to injury modifies the metabolism of carbohydrates, inducing mobilization of fat stores and consequently a decrease in TSF. Nearly every aspect of the immune system is damaged by inadequate nutrition3 and stress reaction.18 Immunosuppression may worsen the prognosis in poststroke recovery, with an increased susceptibility to infections and bedsores, which occurred in our patients.
In this study malnutrition was a significant predictor of poor outcome when swallowing disability was not included as a covariate in the logistic regression model. Dysphagia increased sixfold the risk of poor outcome after we controlled for age, sex, nutritional status, and CSS score at inclusion. Because dysphagia was related to a decrease in level of consciousness in several patients, we believe it is a powerful sign of stroke severity. However, swallowing problems may contribute to poor outcome independently of other markers of overall stroke severity and initial coma,19 and dysphagia after hemispheric or brain stem strokes may lead to aspiration pneumonia.20
The role of therapeutic nutritional intervention in stroke outcome remains unclear. One important issue in acute stroke is when to begin nutritional intervention in comatose patients or in those unable to swallow. Until now, no nutritional guidelines have been recommended by experts in stroke management. Lack of caloric intake until the patient is conscious and has recovered from swallowing problems may lead to malnutrition and consequently poor outcome, infections, and bedsores. However, in our study the relationship between poor prognosis and malnutrition was dependent on stroke severity and dysphagia. Since more than half of patients with dysphagia improve by the end of the first week,19 it might be reasonable to delay nutritional support until the second week in patients with swallowing problems. This decision should be made after the advantages and disadvantages of early enteral feeding are considered. Several advantages of early enteral nutrition have been reported: (1) It is now considered important to provide fuel to the intestine to keep the local defense barrier of the intestine intact and prevent bacterial translocation and sepsis of enteral origin21 ; (2) prospective randomized studies have shown the beneficial effect of an enteral diet in critically ill patients22 ; and (3) a recent meta-analysis comparing early enteral feeding with parenteral feeding in high-risk surgical patients showed reduced septic morbidity rates when enteral feeding was initiated early.23 Potential disadvantages also must be considered: (1) A large, positive caloric balance during the acute catabolic phase of injury or sepsis appears to increase the risk of diet-induced thermogenesis24 ; (2) several reports indicate that aggressive nutritional support does not prevent substantial body protein loss during severe catabolic illnesses25 ; and (3) enteral nutrition may contribute to aspirative pneumonia, a potentially severe complication of intragastric infusion of nutrients that can be avoided by jejunal tube feeding.26
Our study found that malnutrition was associated with increased stress reaction during the first week, higher frequency of respiratory and urinary infections and bedsores, greater mortality, worse outcome, and a longer duration of hospitalization. Thus, malnutrition was an important predictor of poor prognosis. Early appropriate enteral caloric feeding did not prevent malnutrition during the first week of hospitalization. Based on the available data, we believe that early enteral nutrition can only be recommended in malnourished patients who are unable to swallow, with the tube positioned beyond the Treitz angle to prevent aspiration. We believe that the data are not sufficient to warrant a general recommendation of immediate or delayed nutritional support in stroke patients with dysphagia or unconscious patients. A controlled trial comparing early versus delayed enteral feeding in stroke patients with swallowing difficulties is needed.
Selected Abbreviations and Acronyms
|CSS||=||Canadian Stroke Scale|
|MAMC||=||midarm muscle circumference|
|TSF||=||triceps skinfold thickness|
This study was supported by grant 90/0421 from the Fondo de Investigaciones Sanitarias de la Seguridad Social.
- Received November 24, 1995.
- Revision received February 26, 1996.
- Accepted February 27, 1996.
- Copyright © 1996 by American Heart Association
Reilly JJ, Hull SF, Albert N, Waller A, Bringardener S. Economic impact of malnutrition: a model system for hospitalized patients. JPEN J Parenter Enteral Nutr.. 1988;12:371-376.
Dominioni L, Dionigi R. Immunological function and nutritional assessment. JPEN J Parenter Enteral Nutr.. 1987;11:70S-72S.
Unosson M, Ek A-C, Bjurulf P, von Schenck H, Larsson J. Feeding dependence and nutritional status after acute stroke. Stroke. 1994;25:366-371.
Weekes E, Elia M. Resting energy expenditure and body composition following cerebro-vascular accident. Clin Nutr.. 1992;11:18-22.
Dávalos A, Fernandez-Real JM, Ricart W, Soler S, Molins A, Planas E, Genís D. Iron-related damage in acute ischemic stroke. Stroke. 1994;25:1543-1546.
Jaliffe DB. The assessment of nutritional status of the community. Geneva, Switzerland: World Health Organization; 1966. WHO Monograph Series No. 53.
Dávalos A, Castillo J, Martinez-Vila E, for the Cerebrovascular Diseases Study Group of the Spanish Society of Neurology. Delay in neurological attention and stroke outcome. Stroke. 1995;26:2233-2237.
Clutter WE, Bier DM, Shah SD, Cryer PE. Epinephrine plasma metabolic clearance rates and physiologic thresholds for metabolic and hemodynamic actions in man. J Clin Invest. 1980;66:94-101.
Barer DH. The natural history and functional consequences of dysphagia after hemispheric stroke. J Neurol Neurosurg Psychiatry. 1989;52:236-241.
Lipman TO. Bacterial translocation and enteral nutrition in humans: an outsider looks in. JPEN J Parenter Enteral Nutr.. 1995;19:156-165.
Kinney JM. Clinical biochemistry: implications for nutritional support. JPEN J Parenter Enteral Nutr.. 1990;14:148S-156S.
Montecalvo MA, Steger KA, Farber HW, Smith BF, Dennis RC, Fitzpatrick GF, Pollack SD, Korsberg TZ, Birkett DH, Hirsch EF, Craven DE, and the Critical Care Research Team. Nutritional outcome and pneumonia in critical care patients randomized to gastric versus jejunal tube feedings. Crit Care Med. 1992;20:1377-1387.