Welcome to SCIRE Professional
 

Energy Imbalance

To maintain a healthy weight, one must stay in energy balance whereby energy intake equals energy expenditure. Total daily energy expenditure is determined by three factors: resting metabolic rate, physical activity and the thermic effect of food. In addition to lifestyle practices (e.g., smoking) each of these factors is altered following a SCI, rendering it challenging for patients to achieve and maintain energy balance (De Groot et al. 2008). The resting metabolic rate of people with chronic SCI is estimated to be 14-27% lower than their non-SCI counterparts, largely due to reductions in fat-free mass and reduced sympathetic nervous system activity (Buchholz & Pencharz 2004). Physical activity levels of persons with SCI are generally lower than that of non-SCI persons (Buchholz & Pencharz 2004). In addition, a lower thermic effect of food has been reported in persons with a SCI compared to non-SCI controls (Monroe et al. 1998). Three studies have examined dietary intake and malnutrition in the SCI population (Pellicane et al. 2013; Sabour et al. 2012; Wong et al. 2012).

Table 1 Dietary Intake in Individuals with SCI

 

Author Year

Country

Research Design

Sample Size

MethodsOutcomes
Sabour et al. 2016

Iran

Observational

N=103

Population: Mean age: 39.5 yr; Gender: males=86, females=17; Injury etiology: unspecified; Level of injury: cervical=23, thoracic=63, lumbar=17; Level of severity: AIS A=76, B=13, C=4, D=10.

Intervention: Participants were assessed upon admission to a research centre.

Outcome Measures: Caloric Intake, Protein Intake, Body Mass Index (BMI), Bone Mineral Density (BMD).

1.      Measurements were taken at the femoral neck (FN), femoral trochanter (FT), femoral intertrochanteric zone (FIZ), lumbar vertebrae (LV), and hip.

2.      BMD was significantly correlated with BMI at all measured points (p<0.05).

3.      BMD was significantly greater in female participants at all measured points (p<0.05), except at the FN.

4.      BMD of the LV was significantly greater in participants with incomplete injury (p<0.05) and with paraplegia (p<0.05).

5.      BMD of the FIZ was significantly greater in participants with AIS D (p<0.05).

6.      Caloric intake was not significantly correlated with BMD at any point.

7.      Protein intake was negatively correlated with BMD of the LV (r=­0.24, p=0.03).

8.      BMD of the LV was negatively correlated with intake of tryptophan, isoleucine, lysine, cysteine, tyrosine, threonine, leucine, methionine, phenylalanine, valine, and histidine (p<0.05).

Gorgey et al. 2015

USA

Observational

N=16

Population: Mean age: 38 yr; Gender: males=16, females=0; Injury etiology: unspecified; Level of injury: C5-7=6, T3-10=10; Level of severity: AIS A=12, B=4; Time since injury: >1yr.

Intervention: Participants from the community were assessed and dietary intake was recorded for 4wk.

Outcome Measures: Dietary Record Frequency, Percentage of Macronutrients, Caloric Intake, Total Energy Expenditure (TEE), Basal Metabolic Rate (BMR), Fat-Free Mass (FFM), Fat Mass (FM).

1.      Caloric intake decreased over 4 wk, but the difference was not significant (p=0.056). There was no significant difference (p=0.93) or interaction (p=0.54) in measuring caloric intake among different dietary record frequencies (1, 3, or 5 d/wk).

2.      TEE was significantly higher than caloric intake using 1 d (p=0.001), 3 d (p=0.015), or 5 d (p=0.005) dietary frequency records.

3.      BMR was not significantly different from caloric intake for any dietary record frequency, and the two were not significantly correlated.

4.      BMR was significantly correlated with total FFM (r=0.71, p=0.005), leg FFM (r=0.55, p=0.04), and trunk FFM (r=0.62, p=0.018).

5.      Percentage of macronutrients consumed was not significantly different among dietary frequency records: fat (p=0.92), carbohydrates (p=0.50), or protein (p=0.35).

6.      Percentage of fat consumed was significantly different across 4 wk (p=0.031), particularly at 2-3 wk (p=0.034). There was no significant interaction among dietary record frequencies in measuring fat intake (p=0.80).

7.      Percentage of carbohydrates consumed was significantly different across 4 wk (p=0.032), particularly at 1-3 wk (p=0.026) and 2-3 wk (p=0.014). There was no significant interaction among dietary record frequencies in measuring carbohydrate intake (p=0.30).

8.      Percentage of protein consumed was significantly different across 4 wk (p=0.021), particularly at 1-3 wk (p=0.008). There was no significant interaction among dietary record frequencies in measuring protein intake (p=0.025).

9.      Percentage of fat consumed accounted for 29% of total FM (r2=0.29, p=0.037), 34% of leg FM (r2=0.34, p=0.022), and 24% of trunk FM (r2=0.24, p=0.066). It was negatively correlated with total FFM (r=­0.53, p=0.04), trunk FFM (r=­0.54, p=0.036), and BMR (r=­0.52, p=0.059).

10.   Percentage of carbohydrates was negatively correlated with % fat (r=­0.92, p<0.0001), % protein (r=­0.67, p=0.005), total FM (r=­0.56, p=0.031), leg FM (r=­0.64, p=0.01), and trunk FM (r=­0.50, p=0.059). It was positively correlated with total FFM (r=0.54, p=0.037) trunk FFM (r=0.52, p=0.046), and BMR (r=0.55, p=0.04).

11.   Percentage of protein was not correlated with FM, FFM, or BMR.

Tsunoda et al. 2015

Japan

Observational

N=841

Population: Mean age: 61 yr; Gender: males=718, females=123; Injury etiology: unspecified; Level of injury: cervical=245, thoracic=434, lumbar=162; Level of severity: unspecified; Mean time since injury: 27 yr.

Intervention: Participants from the community were assessed via questionnaires, and categorized as superior (n=413) or subordinate (n=428) based on food intake score.

Outcome Measures: Food Intake, Trans-Theoretical Model (TTM), Self-Efficacy (SE), Outcome Expectancy (OE).

1.      Food intake frequency scores between the superior and subordinate groups were significantly different in age (p<0.001), gender (p=0.002), living situation (p=0.002), and care services status (p=0.007).

2.      In univariate analysis, all food intake variables were significantly correlated (p<0.001) with TTM (OR range: 2.55-5.89) SE (OR range: 1.93-4.08), and OE (OR range: 1.61-2.76).

3.      In multivariate analysis, TTM was significantly correlated with the following food intake variables: ‘to eat vegetable dishes’ (OR=2.76, p<0.001), ‘to eat green/yellow vegetables (OR=2.29, p=0.003), ‘to eat dairy products’ (OR=2.75, p<0.001), and ‘to eat fruits’ (OR=1.87, p=0.003).

4.      In multivariate analysis, SE was significantly correlated with the following food intake variables: ‘to eat vegetable dishes’ (OR=2.12, p=0.008), ‘to eat dairy products’ (OR=1.91, p=0.001), and ‘to eat fruits’ (OR=1.97, p=0.001).

5.      In multivariate analysis, OE was not significantly correlated with any food intake variable.

Lieberman et al. 2014

USA

Observational

N=100

Population: Mean age: 45.3 yr; Gender: males=78, females=22; Injury etiology: unspecified; Level of injury: paraplegia=43, quadriplegia=57; Level of severity: AIS A=66, B=16, C=18; Mean time since injury: 15.1 yr;

Intervention: Participants from the community were assessed and compared to age- and gender-matched controls (n=100).

Outcome Measures: Nutrient Intake, Food Intake, Dietary Guideline Adherence.

1.      Nutrient intake: participants consumed significantly less calcium (means: 1049 versus 1415 mg; p=0.004) and Vitamin D (means: 223 versus 315 IU; p=0.009) when compared to controls.

2.      Food intake: participants consumed significantly fewer mean daily servings of dairy (2.10 versus 4.79, p<0.0001), fruit (2.01 versus 3.64, p=0.002), whole grains (1.20 versus 2.44, p=0.007), and sugars (1.46 versus 3.50, p=0.002) when compared to controls.

3.      Guidelines: fewer participants adhered to recommended daily servings of fruits and vegetables (>5 cups; 40.3% versus 68.7%, p<0.001), whole grains (>3 oz; 8.9% versus 21.1%, p=0.01), and dairy (>3 cups; 23.4% versus 48.6%, p<0.001).

Wong et al. 2014

UK

Observational

N=150

Population: Median age: 44 yr; Gender: males=46, females=104; Injury etiology: trauma=107, non-trauma=43; Level of injury: cervical=57, thoracic=59, lumbar=22, sacral=1; Level of severity: AIS A=70, B=10, C=28, D=31; Mean time since injury: unspecified.

Intervention: Participants were assessed upon admission to SCI centers.

Outcome Measures: Spinal Nutrition Screening Tool (SNST), Malnutrition Universal Screening Tool (MUST), Length of Stay (LOS), Mortality.

1.      44.6% of participants were at risk for undernutrition (SNST>11 / MUST>1).

2.      LOS was significantly higher in at-risk participants than those not at risk (129 versus 85 d, p=0.012).

3.      Increased LOS was associated with higher SNST score (p=0.012), higher MUST score (p=0.013), new admission (p<0.01), prior ICU stay (p<0.01), low protein (p=0.022), low albumin (p<0.01), and weight loss >10% (p<0.01).

4.      Mortality rate at 1 yr was significantly higher in at-risk participants than those not at risk (10.2% versus 1.4%, p=0.036).

5.      Higher mortality was associated with age >60 yr (p<0.01), readmission (p=0.018), pressure ulcers (p=0.028), and mechanical ventilation (p=0.025).

6.      In univariate analyses, predictors of LOS were SNST score (p=0.003), MUST score (p=0.003), injury level (p=0.027), admission type (p<0.001), mechanical ventilation usage (p=0.003), prior ITU stay (p<0.001), serum protein (p=0.002), and serum albumin (p<0.001).

7.      In multivariate analysis, predictors of LOS were admission type (B=81.23, p<0.001) and serum albumin (B=­3.62, p=0.013).

Pellicane et al. 2013

USA

Observational

N=78

Population: SCI (n=16): Mean age=41.1±21.2 yr; Gender: males=13, females=3; Level of injury: tetraplegia=8, paraplegia=8; Other injury etiologies: TBI=9, stroke=43, Parkinson’s disease (PD)=10.

Treatment: Rehabilitation inpatients were assessed by a Registered Dietitian for dietary intake once weekly.

Outcome Measures: Calorie and protein intake.

1.     Total calorie intake was significantly higher in individuals with SCI compared to stroke (p<0.003) and PD (p<0.45).

2.     Calorie intake per body weight (cal/kg) was significantly higher in individuals with SCI compared to stroke (p<0.025).

3.     There were no significant differences in total protein intake between varying etiologies.

4.     Age (p<0.001), gender (p=0.023), were significant predictors of calorie and protein intake; admission weight also predicted calorie intake (p=0.025).

Krempien & Barr 2012

Canada

Observational

N=32

 

 

 

Population: Mean age: 30.6 yr; Gender: males=24, females=8; Injury etiology: unspecified; Level of injury: paraplegia=12, quadriplegia=20; Level of severity: unspecified; Time since injury: unspecified.

Intervention: Participants with professional athletic history were assessed.

Outcome Measures: Three-Factor Eating Questionnaire (TFEQ), Body Mass Index (BMI), Sum of Skinfolds (SoS), Dietary Intake.

1.      Participants with low dietary restraint (<11; n=16) had significantly lower TFEQ disinhibition score (2.1 versus 3.5, p<0.05) and percentage of energy from protein (16.9% versus 18.4%, p<0.05) than those with high dietary restraint.

2.      There were no significant differences in BMI, SoS, or other dietary intakes (i.e. calories, carbohydrates, fat, fibre) between high and low dietary restraint groups.

3.      TFEQ dietary restraint score was not significantly associated with BMI, SoS, or dietary intakes (p>0.05).

4.      TFEQ disinhibition score was significantly associated with SoS (r=0.513, p=0.003).

5.      TFEQ hunger score was significantly associated with intake of calories (r=0.354, p=0.047), carbohydrates (r=0.361, p=0.042), and protein (r=0.456, p=0.009).

Sabour et al. 2012

Iran

Observational

N=162

 

Population: Mean age=34.2±0.7 yr; Gender: males=131, females=31; Level of injury: tetraplegia=94, paraplegia=68; Time since injury=8.0±0.5 yr.

Treatment: Face-to-face interviews examining habitual daily food intake patterns.

Outcome Measures: Macronutrient intake, simple carbohydrate intake, total calorie intake.

1.     Percentages of total energy intake derived from macronutrients were 53% vs. 52% carbohydrate, 10% vs. 11% protein, and 37% vs. 39% fat for men and women, respectively.

2.     There was excessive consumption of simple carbohydrates (102.2±40.4 g/d).

3.     Males consumed a greater number of calories than women (p<0.05).

4.     No difference in total intake between those with tetraplegia versus paraplegia.

5.     Individuals with incomplete injuries consumed significantly more monounsaturated fatty acids than those with complete injuries (p=0.03).

6.     Age, education and gender significantly predicted calorie intake; time since injury, education, and gender were significant predictors for carbohydrate intake.

7.     Smoking and level of injury were not related to any dietary variable, and there were no significant predictors for dietary protein and simple carbohydrate intake.

Wong et al. 2012

UK
Observational

N=150

Population: Age: <60 yr=109, >60 yr=38; Level of injury: C=41.1%, T=42.4%, L=15.8%, S=0.7%; Severity of injury: AISA A=50.4%, B=7.2%, C=20.1%, D=22.3%.

Treatment: Assessment of nutritional risk on admission to SCI centers.

Outcome Measures: Malnutrition Universal Screening Tool, Body Mass Index (BMI)

1.     At the time of hospital admission, 40.0% of the sample were found to be nutritionally ‘at risk’ and 21.4% were assessed as being ‘at high risk’ of malnutrition.

2.     The highest prevalence of nutritional risk was found in groups with prior intensive care unit stays (p=0.035), mechanical ventilation (p=0.183) and ‘artificial’ nutritional support at the time of arrival (<0.001).

3.     Nutritional risk showed no significant difference with increased age (p=0.913).

4.     Compared with ‘no-risk’ patients, at-risk patients were found to have significantly lower concentrations of total protein, albumin, Hb, creatinine and Mg, with lower BMI and less appetite.

5.     ‘At-risk’ patients were found to be receiving more prescribed medications.

Discussion

To ensure adequate dietary intake in a SCI population, regulating the constituents of one’s diet is important. Sabour et al. (2016) found that a higher protein intake of essential amino acids is associated with a lower bone mineral density in the lumbar vertebrae of a SCI population. Lieberman et al. (2014) evaluated dietary guideline adherence in individuals with SCI, and found that they consume fewer daily servings of fruit, dairy and whole grains when compared to age-matched controls. This is of concern as Tsunoda et al. (2015) identified in their study of a Japanese SCI population, that total consumption of vegetables, dairy products and fruits is a differentiating factor between those with superior and subordinate healthy diet habits.

Pellicane et al. (2013) found that among four populations (i.e., SCI, stroke, traumatic brain injury, and Parkinson’s disease), mean caloric intake, but not protein intake, was significantly higher in the SCI population compared to the others (p=0.004). Both Pellicane et al. (2013) and Sabour et al. (2012) reported that age and gender were significant predictors of calorie and protein intake. Further, Sabour et al. (2012) found that simple carbohydrate consumption was excessive among their sample. There were no differences in calorie intake between those with tetraplegia versus paraplegia.

Excessive or limited dietary intake can leave individuals at risk for malnutrition. Wong et al. (2012) examined rates of malnutrition among individuals with SCI on admission to hospital. The authors reported that 40.0% of the sample were found to be nutritionally ‘at risk’ and 21.4% were assessed as being ‘at high risk’ of malnutrition. Wong et al. (2014) also demonstrated that undernutrition is associated with worse clinical outcomes in the year after a SCI. Patients that were undernourished had significantly longer length of stays in rehabilitation (p=0.012), and a greater 12-month mortality rate (p=0.036). Thus, there are a significant number of individuals at risk of developing further nutrition-related complications post SCI.

A study by Krempien and Barr (2012) examined eating behaviours and attitudes of professional Canadian Paralympians with a SCI. The authors found that in reference to average individuals with SCI, these athletes had good control of: eating to maintain body weight and composition, knowledge of the types of food they were eating, and were less responsive to physiological hunger cues.

Given alterations in resting energy expenditure, it can be challenging to accurately estimate daily energy requirements for individuals with post-acute SCI. Equations validated and used in non-SCI populations to predict resting metabolic rate overestimate actual measured energy needs in the SCI population (Buchholz & Pencharz 2004). However, Gorgey et al. (2015) found that in a chronic SCI population, caloric intake was on average much lower than the total energy expenditure and basal metabolic rate of this population. Therefore, it has been suggested that energy needs following SCI are best assessed by indirect calorimetry using a metabolic cart (Hadley 2002). Because not all health care centers have access to metabolic carts to measure resting metabolic rate, validated equations specific to the SCI population are needed.

Conclusions

There is level 5 evidence (from one observational study; Sabour et al. 2016) that elevated protein intake can lower bone mineral densities in individuals with SCI.

There is level 5 evidence (from two observational studies; Tsunoda et al. 2015; Lieberman et al. 2014) that consumption of whole grains, vegetables, fruits and dairy products are important in maintaining adequate dietary intake.

There is level 5 evidence (from two observational studies; Pellicane et al. 2013; Sabour et al. 2012) that age and gender, but not level of injury, predict total caloric intake in individuals with SCI; further, level 5 evidence (from one observational study; Gorgey et al. 2015) suggests that individuals with chronic SCI often have a negative energy balance, consuming fewer calories than they burn.

There is level 5 evidence (from two observational studies; Wong et al. 2014; Wong et al. 2012) that individuals with SCI are at a significant risk for malnutrition and are at risk of worse clinical outcomes in the first year after injury.

  • Adequate dietary consumption is important in maintaining bone health.
  • Age and gender, but not level of injury, predict total caloric intake in individuals with SCI.

  • Individuals with SCI are at a significant risk for malnutrition.