Energy Expenditure Following Spinal Cord Injury: A Delicate Balance

Abstract

Following a spinal cord injury (SCI), neurogenic obesity results from changes in body composition, physical impairment, and endometabolic physiology and when dietary intake exceeds energy expenditure. Given the postinjury reductions in lean body mass, sympathetic nervous system dysfunction, and anabolic deficiencies, energy balance is no longer in balance, and thereby an obesogenic environment is created that instigates cardiometabolic dysfunction. Accurate determination of metabolic rate can prevent excess caloric intake while promoting positive body habitus and mitigating obesity-related comorbidities. Metabolic rate as determined by indirect calorimetry (IC) has not been adopted in routine clinical care for persons with SCI despite several studies indicating its importance. This article reviews current literature on measured and predicted metabolic rate and energy expenditure after SCI and stresses the importance of IC as standard of care for persons with SCI.

Introduction

Spinal cord injury (SCI) is a medical condition caused by direct or indirect damage to the ascending and descending pathways of the spinal cord. The disruption of efferent (motor) and afferent (sensory) transmission between supraspinal regions and the periphery leads to skeletal and smooth muscle paralysis and loss of sensation below the level of injury. SCI disrupts almost every physiological system in the body through the denervation of tissues and autonomic dysregulation that is more severe with higher level of injuries. A decrease in energy expenditure instigated by a loss of lean body mass within the first year of the injury predisposes individuals with SCI to several health risks in the event that energy intake exceeds energy expenditure (Figure 1).^1,2 This positive energy balance increases the risk of obesity; coupled with SCI-related inactivity, physical deconditioning, impaired fitness, sympathetic nervous system dysfunction, diminished anabolic milieu, obligatory sarcopenia, and neurogenic osteopenia, it instigates the development of neurogenic obesity. Unless physical activity and caloric restriction are incorporated as a lifestyle modification to cause negative energy balance, neurogenic obesity leads to numerous cardiometabolic sequelae, including glucose intolerance and insulin resistance,^3–6 hyperlipidemia,^7–12 hypertension,^13,14 and coronary artery disease.^7,15

Figure 1.

The relationship between energy expenditure and energy intake, and the components that influence them after spinal cord injury. TEPA, thermic effect of physical activity.

Energy Balance

Energy balance is defined by an ever-dynamic relationship of energy intake and energy expenditure (Figure 1). Energy intake reflects the number of calories consumed thorough the ingestion and digestion of foodstuff.¹⁶ Macronutrients are dietary constituents that provide energy and include protein, carbohydrates, fats, and alcohol (although not an essential macronutrient).¹⁷

Energy expenditure represents the calories used to maintain homeostasis, perform movement, and digest food. Total daily energy expenditure (TDEE) is the aggregate energy used over a 24-hour period and is equal to the sum of the thermic effect of food (TEF), the thermic effect of physical activity (TEPA), and metabolic rate.¹⁸ The TEF is the least variable constituent of TDEE since absorption and digestion of food remains constant and is relatively uninfluenced by body composition after SCI.¹⁹ TEPA is the most variable constituent of TDEE and is dependent on lean body mass and adjustable parameters of physical activity (e.g., frequency, duration, intensity, and mode). Metabolic rate contributes the most to TDEE and is the minimum energy needed to support life. It is measured as resting metabolic rate (RMR), basal metabolic rate (BMR), or sleeping metabolic rate (SMR). SMR is the metabolic rate during normal sleep, and it decreases during sleep by around 15% in individuals without SCI.²⁰ RMR is similar to BMR, but BMR is often considered more precise due to greater testing stringency.^18,21 BMR typically accounts for 60% to 70% of TDEE in the nondisabled (ND) population, whereas in persons with SCI, it accounts for over 80% of TDEE.¹⁸ Following SCI, with the loss of metabolically active tissue, BMR is significantly decreased by as much as 14% to 27% compared to body mass index-matched ND controls,¹⁹ resulting in a proportionate decrease in TDEE by over 50% in persons with tetraplegia.¹⁶ For the purpose of this review, we defined metabolic rate as the energy expended in a supine, rested, and alert individual following a postabsorptive state in a quiet, thermoneutral environment and before any exercise or physical activity.

Measured Metabolic Rate

Indirect calorimetry (IC) is a noninvasive method that measures the respiratory gas volumes using a metabolic system in a dark, quiet room following a 6- to 12-hour fast.^18,21 The new generation of IC machines are more portable, accurate, sensitive, and less expensive and require a shorter calibration time than previous systems.²² IC has the benefit of being able to be used in a hospital- or rehabilitation-based setting with a transportable metabolic cart or through handheld or wearable metabolic technologies.²³ Measurements made by IC devices often provide predicted values to compare the measured quantities and ascertain fasting substrate oxidation that provide valuable information for determining whether an individual was truly in a state of rest.

Dietary inaccuracies may generate additional morbidity and adverse consequences for individuals with SCI already at elevated risk for morbidity and mortality.^24,25 Because of the risk of developing malnutrition after SCI, a carefully designed nutrition support regimen to provide optimal nutrients and calories is essential when these individuals are under hospital or rehabilitation care and living within the community. IC is the best tool to measure and monitor TDEE and hence optimize the energy prescription. One of the primary benefits to IC is its use in preventing under- or over-feeding due to its precise assessment and control of energy needs.²³ Because of the difficulties that persist in accurately predicting energy expenditure after SCI,¹⁸ and the fact that energy expenditure may significantly change during the course of recovery and treatment or based on new or preexisting comorbidities, IC is beneficial to optimize caloric needs.

Measured metabolic rate is lower in persons with SCI compared to individuals in the ND population. The mean metabolic rate for persons with longstanding SCI ranges from 1256 to 1854 kcal/day^{1,18,19,26–48} and minimally overlaps the metabolic rate of 1594 to 2249 kcal/day reported for the ND population.⁴⁹ It is well established that sex, age, height, and body weight impact metabolic rate; in persons with SCI, these factors, in addition to time since injury, level of injury, and completeness of injury, also influence metabolic rate. Monroe et al.¹ reported persons with chronic SCI compared to nondisabled controls have a reduced resting (SCI: 1756 vs ND: 2211 kcal/day) and sleeping (SCI: 1402 vs ND: 1657 kcal/day) metabolic rate. Bauman et al.²⁶ compared the BMR and RMR in pairs of identical twins and reported the metabolic rates of the twin with SCI (1387 and 1682 kcal/day, respectively) was significantly less than those of the nondisabled twin (1660 and 1854 kcal/day, respectively). Buchholz et al.¹⁹ showed RMR was significantly higher in nondisabled controls (1677 kcal/day) compared to persons with paraplegia (1472 kcal/day). The authors also reported fat free mass (FFM) was the single best predictor of RMR in both groups (r² = 0.83 for controls and 0.70 for persons with paraplegia). In another study by Buchholz et al.,⁵⁰ the authors examined RMR by injury completeness and sex and demonstrated a mean difference of 63 kcal/day between complete and incomplete injuries and 310 kcal/day between men and women with SCI. Collins et al.³² did not report significant differences by level of injury. We recently reported nonsignificant differences by sex (males: 1421 kcal/day; females: 1367 kcal/day),³⁸ and significant differences by level of injury (tetraplegia: 1224 kcal/day vs paraplegia: 1517 kcal/day²⁹; tetraplegia: 1259 kcal/day vs paraplegia: 1483 kcal/day³⁴) in motor complete SCI. The discrepancies in the results may be caused by a heterogenous sample of persons with complete and incomplete SCI, sample and population demographics (i.e., age and time since injury), sample sizes, and the methods by which metabolic rate was assessed (RMR vs BMR).

Bauman et al.³⁰ demonstrated that men with SCI on testosterone replacement therapy had a significantly increased metabolic rate (1328 vs 1440 kcal/day). Similarly, Gorgey et al.⁵¹ recently reported that men with chronic SCI had a 14% to 17% increase in BMR following neuromuscular electrical stimulation resistance training coupled with testosterone replacement therapy. Immediately following 16 weeks of exercise training, BMR significantly increased compared to baseline and then significantly decreased at a follow-up visit.³⁷ When comparing manual and power wheelchair users with SCI, persons in the former group had a higher BMR (manual: 1551 kcal/day vs power: 1340 kcal/day),⁵² although this did not reach statistical significance likely due to a limited sample size. The increase in metabolic rate can be attributed to increased lean tissue mass that results from physical exertion and/or exogenous testosterone, an anabolic hormone that significantly decreases in men after SCI.^53,54

Predicted Metabolic Rate

Several prediction equations used to predict metabolic rate have been developed and validated in the nondisabled population (Table 1).^18,29,49A recent systematic review evaluated the accuracy of measured and predicted metabolic rates and raised concern over the accuracy, reliability, and practicality of the several prediction equations in research and clinical settings.¹⁸ Nevin et al.⁵⁵ reported that prediction equations derived for ND individuals overestimate metabolic rate by 4% to 92% in individuals with SCI. These equations do not account for losses in lean tissue mass and sympathetic nervous system dysfunction and do not distinguish between FFM and adipose tissue in persons with SCI. In men with paraplegia, Lee et al.⁴⁰ reported that 35%, 29%, and 35% were hypometabolic, normometabolic, and hypermetabolic, respectively, by calculating the ratio of the measured RMR to estimated RMR using the Harris-Benedict equation. However, Sedlock and Laventure⁴⁷ (Cunningham equation), Lee et al.⁴⁰ (Harris-Benedict), Buchholz et al.¹⁹ (Schofield equation), Alexander et al.²⁸ (Harris-Benedict), Liu et al.²⁷ (Harris-Benedict), Hayes et al.³⁹ (Harris-Benedict and author-developed equations), Chun et al.³¹ (author-developed equation), Pelly et al.⁴⁸ (Mifflin, Cunningham, Harris-Benedict, Schofield, and Owen), and Nightingale and Gorgey⁴⁵ (Nelson et al., Cunningham, and Chun et al. equations) reported that the various prediction equations all overestimated metabolic rate compared with the measured value. Similarly, Bauman and colleagues^26,30 have identified that the Harris-Benedict equation overestimated RMR; according to a recent systematic review, this was a statistically significant overestimation.¹⁸ Buchholz et al.,¹⁹ Chun et al.,³¹ and Nightingale and Gorgey⁴⁵ developed an SCI-specific equation to estimate metabolic rate; however, the equations do not account for TEF or TEPA. Broad et al.⁵⁶ measured metabolic rate in competitive wheelchair rugby players and demonstrated an RMR of 1735 ± 257 kcal/day that supported the statistical equivalency of the Chun et al., Cunningham, Mifflin et al., Nightingale and Gorgey, and Owen et al. prediction models as an alternative to measuring RMR. Collectively, these data mark the ND prediction equations as inadequate for persons with SCI and emphasize the importance of IC in assessing metabolic rate in this population.

Table 1.

Prediction equations for estimating metabolic rate

Equation name/author(s), year		Sex	Equation
Nondisabled equations	Cunningham, 1980	M/F	= 500 + 22 (LBM)
	Harris-Benedict, 1919	M	= 66.4730 + (13.7516 × wt) + (5.0033 × ht) – (6.7550 × age) = (1.8496 × ht) + (9.5634 × wt) + 655.0955 – (4.6756 × age)
	Hayes et al., 2002	M/F	= KAT× AT + KSM× SM + KBone× Bone + KBrain× Brain + KRM× RM
	Mifflin, 1990	M	= 10 × wt + 6.25 × ht – 5 × age + 5 = 10 × wt + 6.25 × ht – 5 × age – 161
	Nelson et al., 1992	M/F	= (108 × FFM) + (16.9 × FM)
	Owen, 1987	M	= 290 + 22.3 (LBM)
		F	= 334 + 1.97 (LBM)
	Schofield, 1985	M	= 15.057 × wt + 692.2 (age, 18 – 30 y), 11.472 × wt + 873.1 (age, 30 – 60 y), 11.711 × wt + 587.7 (age, > 60 y)
		F	= 14.818 × wt + 486.6 (age, 18 – 30 y), 8.126 × wt + 845.6 (age, 30 – 60 y), 9.082 × wt + 658.5 (age, > 60 y)
SCI-specific equations	Buchholz et al., 2003	M/F	= −3618 – 795 × age – 731 × sex + 3170 × wt – 794 × T3 + 261 × metanephrine
		M/F	= 10682 – 1238 × age – 521 × sex – 24 × ht + 87 × FFM
	Chun et al., 2017	M/F	= 24.5 × FFM + 244.4
	Nightingale-Gorgey, 2018	M	= 23.469 × FFM + 294.330 (FFM alone)
		M	= 23.995 × FFM + 6.189 × SAD + 6.384 × TAD – 6.948 × TC + 275.211 (FFM with circumferences and diameters)
		M	= 19.789 × FFM + 5.156 × wt + 8.090 × ht – 15.301 × CC – 860.546 (FFM with anthropometrics)
		M	= 13.202 × ht + 11.329 × wt – 16.729 × TAD – 1185.445 (anthropometrics alone)

Note:AT = adipose tissue; CC = calf circumference (cm); FM = fat mass; FFM = fat free mass (kg); F = female; Ht = height (cm); K = constant for metabolic rate of organ/tissue at resting state; LBM = lean body mass; M = male; metanephrine (μmol/L) = a biproduct of catecholamines breakdown; RM = residual = mass; RMR = resting metabolic rate; SAD = sagittal abdominal diameter (cm); SM = skeletal muscle; sex = 0 for men and 1 for women; T3 = triiodothyronine (nmol/L), a thyroid hormone; TAD = transverse abdominal diameter (cm); TC = thigh circumference (cm); Wt = weight (kg).

Energy Expenditure

Similar to metabolic rate, TDEE can be measured or estimated. Buchholz et al.⁵⁰ reported significant differences in TDEE between men (2490 kcal/day) and women (1870 kcal/day) with SCI and between persons with complete (2072 kcal/day) and incomplete (2582 kcal/day) injuries. Nightingale et al.⁴⁴ reported an average energy expenditure of 2103 kcal/day in men with chronic paraplegia, while Monroe et al.¹ showed that the 24-hour TDEE was significantly lower in persons with SCI than in nondisabled controls (1870 vs. 2376 kcal/day). Monroe and colleagues¹ also reported that adjusting for FFM, fat mass, and age showed that 24-hour energy expenditure was still significantly lower (−180 kcal/day) in SCI than in control subjects. These studies demonstrate that the accurate determination of metabolic rate is vital, especially as it is often used to determine caloric intake in individuals with SCI. In the nondisabled population, TDEE is estimated using the product of BMR and 1.2, where 1.2 is the activity correction factor for TEF and TEPA. However, this estimation often overestimates energy expenditure after SCI.^29,49 We recently established an SCI-specific correction factor of 1.15 to estimate TDEE.²⁹ The new correction factor was developed using the SCI-specific and ND metabolic equivalents of 2.7 and 3.5 mL/kg/min, respectively, to minimize TDEE overestimation using ND prediction equations. Its use offers promise for more precisely estimating energy expenditure and caloric intake after SCI, thereby potentially reducing the burden of obesity and cardiometabolic-related comorbidities.

Future Direction

Evaluating energy balance as a multidimensional system, in place of independent components that complement one another, marks the next chapter of research on energy metabolism. Emerging evidence suggests energy balance is more complex than “calories in and calories out” as aspects such as energy usage and storage must also be considered.^57,58 A deeper understanding of energy influx and efflux as it relates to energy balance after SCI can provide new insights, preventative strategies, and treatments to address the obesity epidemic and attenuate cardiometabolic risk profiles. Improving the accuracy of measurements for assessing caloric intake (e.g., dietary recall questionnaires) and energy balance is warranted, because these tools are needed to guide future, more detailed recommendations.

Conclusion

There is a considerable amount of variation in metabolic rates reported in the literature. Several ND predictive equations presumably overestimate metabolic rate and the caloric requirements for individuals with SCI. SCI-specific equations, while promising, do not account for differences in neurological parameters and individual variation. As a result, IC remains the gold standard assessment of metabolic rate. Through its use we can prevent the over- and under-feeding of persons with SCI by providing an “individualized” measurement and accounting for specific injury characteristics. By measuring metabolic rate and recommending healthy eating habits and physical activity, we may ultimately mitigate the high rates of morbidity and mortality caused by obesity and obesity-induced cardiometabolic dysfunction in persons with SCI.