Management of obesity after spinal cord injury: a systematic review

Abstract

Context

Individuals with chronic spinal cord injury (SCI) are susceptible to central and visceral obesity and it’s metabolic consequences; consensus based guidelines for obesity management after SCI have not yet been stablished.

Objectives

To identify and compare effective means of obesity management among SCI individuals.

Methods

This systematic review included English and non-English articles, published prior to April 2017 found in the PubMed/Medline, Embase, CINAHL Psychinfo and Cochrane databases. Studies evaluating any obesity management strategy, alone or in combination, including: diet therapy, voluntary and involuntary exercise such as neuro-muscular electric stimulation (NMES), pharmacotherapy, and surgery, among individuals with chronic SCI were included. Outcomes of interest were reductions in waist circumference, body weight (BW), body mass index (BMI) and total fat mass (TFM) and increases in total lean body mass (TLBM) from baseline. From 3,553 retrieved titles and abstracts, 34 articles underwent full text review and 23 articles were selected for data abstraction. Articles describing weight loss due to inflammation, cancer or B12 deficiency were excluded. The Downs and Black reported poor to moderate quality of the studies.

Results

Bariatric surgery produced the greatest permanent weight reduction and BMI correction followed by combinations of physical exercise and diet therapy. Generally, NMES and pharmacotherapy improved TLBM and reduced TFM but not weight.

Conclusions

The greatest weight reduction and BMI correction was produced by bariatric surgery, followed by a combination of physical exercise and diet therapy. NMES and pharmacologic treatment did not reduce weight or TFM but increased in TLBM.

Introduction

The chronic Spinal Cord Injury (SCI) population is vulnerable to obesity, due to concurrent development of sarcopenia and decreased metabolic rate, limited mobility, and diminished levels of activity.¹ The prevalence of obesity among the individuals with SCI varies from 40 to 66% across studies.² Obese individuals with SCI are susceptible to a wide range of health consequences.

Dangerous health consequences associated with obesity include, but are not limited to: an increased prevalence of angina pectoris, cerebrovascular accidents, breast cancer, carpal tunnel syndrome, cholecystitis/cholelithiasis, colon cancer, congestive heart failure, coronary artery disease, depression, glucose intolerance, diabetes mellitus, dyslipidemia, gout, nephrolithiasis, obstructive sleep apnea, osteoarthritis, peripheral vascular disease, pressure ulcers, reproductive dysfunction and social isolation.³ Adequate detection, monitoring and management of obesity in persons with SCI is of utmost importance in reducing all-cause mortality.

Detection of obesity in SCI

Obesity is defined as the excess accumulation of body fat, (more than 25% of body weight (BW) for men, and more than 30% for women,⁴ which is well characterized in the general population by Body Mass Index (BMI) due to the high correlation (r = 0.7–0.9) of BMI with fat mass. BMI is measured in (kg/m²) and is calculated by dividing an individual’s weight (kg) by his or her height squared (m²).⁵ The WHO has assigned BMI threshold values for the diagnosis of overweight (BMI 25.0–29.9) or obese (BMI ≥30) respectively.⁶

Although BMI is a good obesity screening tool in the general population, it’s application to individuals with chronic SCI is problematic as the decreases in lean mass and increases in fat mass do not necessarily result in changes in BW, a key component of the BMI calculation. In addition, measurement of height in individuals with chronic SCI is neither as feasible nor reproducible.^7,8 Therefore, the BMI thresholds for individuals with chronic SCI have been lowered to 22 Kg/m² to account for these measurement dilemmas.⁹

Total fat mass and its distribution to visceral rather than subcutaneous tissues is associated with Metabolic Syndrome, and other dangerous medical consequences of obesity.^10,11 Abdominal obesity, specifically visceral adipose tissue (VAT) increases following SCI.¹² The VAT threshold for obesity has been lowered from 130 cm² in able-bodied¹³ to 100 cm² for the SCI population.¹⁴

Waist circumference (WC) is an accurate means of predicting VAT in the able-bodied or general population. Persons with a WC > 88 cm for women and > 102 cm for men are defined as obese by the National Institute of Health (http://www.nhlbi.nih.gov). Although WC as a marker of VAT in SCI patients was arguable,⁷ newer studies have reported an increase in VAT and obesity-related cardiovascular disease risk in SCI population; with a disease- specific cut-point for obesity of WC ≥ 94cm.^15,16

Total body water (TBW), total lean body mass (TLBM), total fat mass (TFM) and extra cellular water (ECW) can be reasonably well predicted by bioelectric impedance analysis (BIA) in SCI population.^17,18 Dual-Energy X-ray Absorptiometry (DEXA) is another valid and reliable method for estimation of body composition components and reference values have been defined from national health and nutrition examination survey (NHANES) in general population. Fat Mass Index (FMI) ≥ 9 kg/m² or percentage body fat ≥ 25% for males and ≥ 13 kg/m² or percentage body fat ≥35% for females indicates obesity.¹⁹ DEXA values are highly reproducible among wheel-chair athletes.²⁰

Management of obesity in SCI

Numerous interventions are used to manage obesity in the general population; these mainly include: diet therapy, medical therapy, physical therapy or various exercise programs and bariatric surgeries

Application of these therapies to individuals with SCI is inappropriate due to their mobility impairment, lifestyle issues, greater sensitivity to anti-obesity agents and other health conditions, such as, pressure ulcer or septicemia.²¹

• Diet Therapy:

The premise of all diets assumes lowering caloric intake below basic requirements for body organs/tissue and physical activity will result in weight loss. Low-calorie diets (LCDs) with 1000–1,200 kcal/day reduce weight by an average of 8% among the able-bodied population over 3–12 months. In contrast to LCD’s, Very low-calorie diets (VLCDs) of 400–500 kcal/day produce larger initial weight loss; however, the long-term (≥1 year) weight loss is equivalent to LCDs.²² Thus, healthy diets recommend decreasing caloric intake from fat to <30%, carbohydrate and protein ∼50% and 15%, respectively while increasing Fiber intake. LCDs and VLCDs may not deliver sufficient amounts of macro and micro-nutrients crucial for healthy body functioning.²³ Individuals with chronic SCI may not respond to LCD’s due to sarcopenic obesity and greater protein requirements.

• Physical Exercise:

The Resting Metabolic Rate (RMR) of skeletal muscle is relatively low although skeletal muscles comprises 40% - 50% of total BW, and dictates resting Total Energy Expenditure (TEE). Energy Expenditure in Physical Activity (EEPA) represents the most variable component of TEE. Lean Body Mass (LBM) is the main organ of the body in terms of daily EEPA.²⁴ Both RMR and EEPA are decreased in chronic SCI due to reduced LBM and low levels of physical activity.²⁵

• Behavioral Modification:

Behavior therapy is one principal component of an “effective high-intensity lifestyle intervention” within the content of overweight and obesity programs which facilitates adherence to weight management recommendations (from Reduce CAD Risk in SCI). Common components of behavioral therapy for obesity include: self-monitoring, stimulus control, slow eating, goal settings, behavioral contracting, education, increasing physical activity and social supports.²⁶ The key principle is to identify cravings and weaken or disconnect the triggering events that precipitate overeating.

• Pharmacotherapy:

Pharmacological treatments for obesity are used as adjuvant therapy among the morbidly obese after lifestyle modification in order to maintain weight loss is achieved. Anti-obesity drugs either suppress appetite and/or stimulate thermogenesis (e.g. serotonergic: fenfluramine, chatecolaminergic: phentermine) and interact with intestinal fat absorption (orlistate). Older drugs include: dinitrophenol, aminorex, amphetamines, fenfluramine/ dexfenfluramine, phenylprpoanolamoine. Phentermine and diethylpropione are approved for short-term use (< three months) in the USA.

• Surgical Therapy:

Weight-loss surgery is typically second or third line therapy after failure of lifestyle modifications in morbidly obese individuals. The National Institutes of Health consensus conference on gastrointestinal surgery for severe obesity reports weight-loss surgery may be an appropriate option for morbid obesity (BMI >40 kg/m²) or for those with moderate obesity (BMI>35 kg/m²) with two or more obesity related comorbidities.²⁷ Bariatric procedures fall into three major categories: intestinal malabsorption, gastric restriction and combined malabsorption and restriction procedures. Mal-absorptive operations depend on rearrangement of the small intestine to decrease the functional length or efficiency of the intestinal mucosa for nutrient absorption. Restrictive procedures involve restriction of the amount of food entering the foregut.²⁸

This systematic review aimed to identify and compare the effectiveness of diet therapy, physical exercises and newer modalities of passive exercises such as functional electrical stimulation (FES), pharmacological treatments and surgical therapy for management of obesity in individuals with chronic SCI. We considered five outcomes of importance including decrease in BW, BMI, total fat mass (TFM) and WC and TLBM.

Methods

Data source

The PubMed/Medline, EMBASE, CINAHL, PSYCHINFO, Cochrane CENTRAL and Cochrane database for Systematic Review databases were reviewed to identify all potential trials published prior to April 2017 in any language. The search strategy in Ovid-Medline is attached as Appendix 1. The PRISMA Flow Diagram is shown in Figure 1.

Figure 1.

PRISMA Flow Diagram – study selection process.

Study selection

Inclusion and Exclusion Criteria: Articles evaluating any obesity management strategy alone or in combination including diet therapy, physical exercises, passive exercises like neuro-muscular electric stimulation (NMES), pharmacotherapy, and surgery, among individuals with chronic SCI were included. Outcomes of interest were declines in WC, BW, BMI and TFM and increases in TLBM.

Studies in which individuals had inflammation or any kind of neoplasms were excluded due to their role in unintended weight loss. Studies among participants with nutritional disturbances as a cause of SCI such as subacute combined degeneration due to vitamin B12 deficiency were also excluded.

Data extraction and quality assessment

Two reviewers (MHS, SMA) independently assessed the studies identified through the electronic searches. Discrepancy between reviewers was resolved by discussion, in the event of disagreement between the reviewers, a third reviewer was recruited prior to resolution. The quality of the selected studies was evaluated by Downs and Black (D&B) scale regardless of their designs.²⁹ Because of the nature of the interventions, which are infeasible to blinding during study implementation, we chose the D&B tool to appraise article quality, a more comprehensive approach when compared to the logistics of blinding.

Results

In total there were 3,553 titles and abstracts reviewed based on the search strategy, 3,519 were removed due to being unrelated to the study objective. Thirty four articles met the inclusion criteria for full text review. We dismissed 9 articles due to lack of data regarding the outcomes of interest, and included the remaining 25 articles; six articles related to physical exercise modalities, 11 articles related to intervention with neuromuscular electric stimulation (NMES) or functional electric stimulation (FES), and eight articles related to pharmacotherapy (3), surgery (3) or diet therapy (1) (Fig. 1). The quality of the most studies based on Downs and Black appraisal tool²⁹ was moderate (scale <20) to poor (scale <15) except for four RCT studies.

Physical exercise intervention to manage obesity in SCI

Evidence for physical exercise intervention to improve body composition in SCI individuals included one RCT,³⁰ five pre-post interventions^31–35 and one case-report,³⁶ for a total of 128 participants. These interventions consisted of arm ergometry, community-based exercise, activity-based therapy, therapeutic recreation such as bocce, canoeing, wheelchair rugby, and resistance training. The range of duration and frequency of the exercise interventions was from six weeks to six months and from 30 minutes once a week to 60 minutes five times per week. In the majority of studies, physical exercise alone did not decrease participants’ BW or BMI; however, it significantly reduced TFM and WC and increased TLBM (Table 1).

Table 1.

Initial values and amount of Changes in Anthropometric and Body Composition after Physical Exercise in Spinal Cord Injury individuals, Extracted from Selected Studies

Parameter	Population	Intervention	BW (kg)	BMI (kg/m2)	TFM	TLBM	WC (cm)
Kim, D.I. 2015 RCT D&B=21	N=15 (T=8, C=7) Mean Age= 33.1±5.4 Injury level: C5-T11 ASIA: A-B	Hand Bike Exercise, 60 minutes, 3 times/week for 6 weeks Intensity: Borg scale; 5–7	Initial: 64.8±10.5 Change: No data	Initial: 22.0±3.7 Change: T: -0.2±0.2, C:+0.3±0.4 P<0.01	Initial: -39.0±13.7% Change: (%) T: -3.5±7.2 C:-0.1±1.9 P: 0.23	Initial: 20.2±5.0 Change: (Kg) T: +1.5±3.5 C: +0.3±0.9 P: 0.39 (NS)	Initial: 88.3±13.1 Change: T:-2.6±1.7 C:+0.8±1.6 P:<0.01
Radomski, M. 2011 Pre-Post intervention D&B=17	n=13 injury level: T1-T9 all complete	Community-based program, Die/ Exercise, 12 Weeks	Initial: 95.2 Change: -5.4 P=0.037	Initial: 34.1 Change: -1.2 P= 0.031	Skinfold body fat (%); Initial: 35.8 change:-3.2 P=0.013	No data	Initial: 115.5 cm Change: -10.6 cm P: 0.005
Astorino T.A. 2015 Pre-Post intervention D&B=15	N=17 Mean Age=36.1±11.5 injury level: C4-L1 complete and incomplete	Chronic activity-based therapy (ABT)1 a minimum of 2 days/week for 6 months, intensities ranging from 5 to 8 ml/kg/minute	Initial: 76.5± 13.0 Change: +0.5 P:NS	No Data	Initial: 31.2± 12.9% Change: +1% P: NS	Initial:50.4±.09 -NS change in Incomplete injury -Decreased (−4.8 ± 5.8%) in complete injury (P<0.05)	Initial: 92.5± 14.6 Change: -1.8 P: NS
Chen Y 2006 Pre-Post D&B=15	N=16 Age= 43.8 Injury leve: Tetra and Para AISA: A-C-D	Time-calorie displacement diet + exercise + behavioral modification 12 weeks	Initial: 97.4±17.8 Change: -3.5±3.1 P=0.0004	Initial: 34.3±4.5 Change: -1.3±1.2 P=0.0005	Initial: 41.4±11.2 Change: -2.9±4.6 P=0.05	Initial: 51.8±10.2 Change: -0.8±5.1 P=0.58	Initial: 117.4±17.1 Change: -4.1±5.0 P=0.005
Neto, F.R. 2011 Pre-Post intervention D&B=13	N=53 Men Tetraplegia (C4-C8) n=20, Higher (T1-T6) n=15 and Lower (T7-L2) n=18 Paraplegia (PP) ASIA A-D	Bocce, Canoeing, Physical therapy, Resistance training, Rugby, Swimming, Table tennis,Wheelchair basketball, Wheelchair, Repulsion physical conditioning 1–5 times/week 30–60 min/session	Initial: 68.3±12.4 -Total change: NS -Increased in TP 0.8 ±1.5, P<0.05 -Decreased in PP, −1.0±1.8; P<0.05	No data	Initial: 15.2±7.1 Kg Change: −0.5±1.3 Kg P<0.05 *Significant decrease only in PP group	Initial: 53.1±6.8 Kg Change: 0.7±1.2 Kg P<0.05	No data
Gorla, J. I. 2016 Pre-Post intervention D&B=12	N=13 Mean Age= 26.6±6.0 Tetraplegia ASIA: A-B-C	Wheelchair Rugby 4 times/week (aerobic and anaerobic) average intervention time 8.1±2.5 months	Initial: 64.2±6.2 Change: -0.9 P:NS	Initial:20.6±1.9 Change:-1.2% P=0.731 (NS)	Initial: 15.191±4.603 Change: -3.0% P=0.016	Initial: 46.759±4.831 change: +2.0% P=0.308 (NS)	No data
Dolbow, David R 2010 Case report D&B=11	N=1 62 yo Male Injury level:T8-9 ASIA A Paraplegic	Arm Crank Exercise Five times/week 16 Weeks 60 min/session Workload:20–40	Initial: 79.4 Change: -3	Initial: 28.1 Change: -1.0	Initial: 28.05 Change: -2.79	Initial: 51.06 Change: -0.12	No data

Activity-based therapy (ABT): This exercise modality is of high volume (>6 hours/week) and typically includes dynamic resistance training, FES, body weight-supported treadmill training, and load bearing and/or standing. T: trial, C control, NS: not significant; TP: tetraplegia; PP: paraplegia; D&B: Downs & Black

NMES intervention to manage obesity in SCI

Available evidence for NMES as a treatment of obesity in SCI population includes two RCTs,^37,38 six pre-post studies^39–44 and 3 case-reports,^45–47 for a total of 128 individuals. Duration and frequency of NMES and FES interventions in the selected studies ranged from six weeks to 6 months, and from once a week to five times per week, respectively. Regarding the effect of NMES, one study evaluated eight weeks of five days per week, therapy for one hour per day, and reported a significant decrease on BW; the other 10 studies showed no change or non-significant changes. Two pre-post studies of six and eight weeks duration reported significant decreases in TFM (-3.5% and -1.9%) and 3 case-reports reported similar effects (from -1.2 to -1.3%), however, the other 4 pre-post studies and 2 RCTs failed to show any significant change in TFM. All studies except two RCTs (Giangregorio 2012 and Gorgey 2012) reported a noticeable increase in TLBM by NMES interventions (from 1.16 to 4.08 kg) (Table 2).

Table 2.

Initial values and amount of Changes in Anthropometric and Body Composition after Neuro-Muscular Electric Stimulation (NMES) in Spinal Cord Injury individuals

Parameter	Population	Intervention	BW (kg)	BMI (kg/m2)	TFM	TLBM (kg)	WC (cm)
Giangregorio, L 2012, RCT D&B=23	N=34 (T:17, C:17) Age: 56.6±14 Injury level: C2-T12; AISA C-D	FES walking/conventional exercise program 16 Weeks,	Initial: 81.3±13.1 Change: T: no data C: no data	Initial:26.75 Change: T: no data C: no data	Initial: 25.4±9.5 Kg Change: T: -1.3 C: +0.8 P: NS	Initial: (leg) 14.9±5.7 Kg Change: T: +2.2 C: -0.3 P: NS	No Data
Gorgey, A. 2012 RCT D&B=19	N=9 males Age = 35± 9 Injury level: C5-T11 AISA: A-B	NME resistance training, 12 Weeks, twice a week	Initial:74±14 Change: Kg T: +1 C:-1 P: NS	Initial:21±5 Change: T: 0 C: 0 P: NS	Initial: 23.3±9 Change: Kg T:-0.7 C: -1 P: NS	Initial:51.8±8 Change: Kg T: = +0.5 C: -0.4 P: NS	No data
Kim, D.I. 2014 Pre-Post D&B=15	N=12, Injury level: C6-L1 AISA: A-B-C	FES rowing 6 Weeks, 5 days/w monophasic rectangular phase, 30Hz, 10–140mA	Initial: 70.2±12.5 Change: no data P: NS	Initial: 23.4±3.7 Change: -0.4 P: 0.058	Initial: 23.9±8.5% Change: -3.5% P=0.028	Initial: 50.4±9.4% Change: +2.9% P=0.001	Initial: 83.9±9.9 Change: -2.1 P=0.059
Carty, A. 2013 Cohort (Pre-Post) D&B=14	N=14 Injury lev:T4-11 ASIA: A-B	Subtetanic NMES on lower limb muscles, 8 weeks; 1h/d, 5d/w	Initial: 74.81 ± 3.94 Change: -1.41 P:0.001	Initial: 26.32± Change: +0.1 P: 0.69 (NS)	Initial: 33.76±6.62% Change: -0.6% P: 0.15 (NS)	Initial:13.65 ± 2.91 Kg Change: +1.16 Kg P<0.001	No Data
Griffin, L. 2009, Pre-Post D&B=14	N=18, mean=40 Injury level: C4-T7 Complete and incomplete	FES cycling 50 Hz 10 weeks, 2–3 times/week	Initial: 153Ib Change: -4.54Ib P: NS	No Data	Initial:50.42Ib Change: +1.36 P: NS	Initial: 96.8Ib Change: +3.2 P<0.05	No Data
Liu, C. 2007 Pre-Post D&B=14	N=18 Mean=40±11.3 Injury level: C4-L1 AISA: A-B-C-D	FES cycling 8 Weeks	Initial: 73.8±13.9 Change: +1.2 P: NS	Initial: 25.4±3.9 Change:+0.3 P: NS	Initial: 18.6±8.6 Change: +0.1 P: NS	Initial: 51.6±7.1 Change: +1.2 P<0.05	No Data
Skold, C. 2002 Pre-Post D&B=14	N=15, males Age: 33 AISA: A-B	FES cycling 6 months, 3 times/week	No change	No data	No change	10% Increase	No data
Hjeltnes, 1997 Pre-Post D&B=14	N=5, male age: 35±3 Injury level: C5-7, ASIA A	ESLC: Electrically Stimulated Leg Cycling 8 Weeks,	Initial: 76.6±4.7 Change: No Data	Initial: 22.23±1.34 Change: No Data	Initial: 29.7±2.6% Change: -1.9% P<0.05	Initial: 66.2±2.6% Change:+2.0 P<0.05	No Data
Gorgey, A.S. 2016, case report D&B=11	N=1, 33 year old male, Injury level T6, AISA A	Surface NMES + ankle weight, 10 weeks, once a week	No change	No change	Initial: No data Change: -1.3%	Initial: No Data Change: -5.0%	No Data
Dolbow, D.R. 2014, Case Report D&B=11	N=1, 60 year old female Injury levl: T6 ASIA: A	FES cycling, 12 Months, 2.9 sessions/week	No change	No change	Initial: 36.79 Change: -1.2%	Initial: 39.22 Change: +3.01 (7.7%)	No Data
Dolbow, D.R. 2012, Case Report D&B=11	N=1, 64 year old male, motor complete C5 injury	Home-based FES cycling 9 Weeks, 3 sessions/week, distance from 3.98 to 9 km. 93% compliance	No change	No change	Initial: 29.6 Change:-1.2%	Initial: 48.94 Change: +4.08 (8.3%)	No Data

T: trial, C: control, NS: not significant, TP: tetraplegia, PP: paraplegia

Pharmacotherapy and Surgery to manage obesity in SCI

No pharmacotherapy study reported a decrease in TFM among SCI participants.^48–50 In an RCT, a 5–10 mg testosterone patch for 12 months did not result in changes in TFM and TLBM, although it showed an increase in BW in the treatment group. In contrast oral Alpha-Lipoic Acid for 12 weeks resulted in a significant decrease in BW, BMI and WC in the treatment group compared with controls (Table 3).

Table 3.

Initial values and amount of Changes in Anthropometric and Body Composition after Pharmacotherapy^a and Surgery^b in Spinal Cord Injury individuals

Parameter	Population	Intervention	BW (kg)	BMI (kg/m2)	TFM (Kg)	TLBM (kg)	WC (cm)
aBauman, W.A 2011 RCT D&B=27	N=22 Men Age: 43±6 AISA: A-B-C	Testosterone patch, 5–10mg 12 Months	Initial: 82.9±12.6 Change: T: +3.2 C:+0.1 P: NS	Initial T: 25 ± 3	Initial: 29.8±10.6 Change: T:+0.3 C:+1.5 P:NS	Initial: 49.6±7.6 Change: T:+3.5 C:-1.4 P:NS	No Data
aMohammadi, V 2015 RCT D&B=23	N=58 Men No more data	alpha-Lipoic Acid, Orally 12 Weeks	Initial: 77.11±14.58 Change: T:−3.7±4.4 C: +0.38±1.4 P=0.0001	Initial: 27.77±4.33 Change: T:−1.0±0.6 C:0.09±0.6 P=0.0001	No Data	No Data	Initial T: 101.79 Change T:−3.7± 2.7, C:0.4±1.4) P=0.0001
aHallstead, L.S. 2010, Prospective repeated-measures D&B=15	N= 10 Men Age: 32.5 Injury level:C4–8 AISA: A-B	20 mg, Orally Oxandrolone 8 Weeks	Initial: 78.8	Initial : 23.8	Initial : 26.7 Change: -1.5% P: NS	Initial T:48.7±8.5 Change: +0.9% P=0.02	No Data
bLutrzykowski, M. 2008 Case report D&B=13	N=2 (Only 1 SCI) Age: 49	Open duodenal switch procedure 4 years follow up	Initial: 134 Change: -65	Initial: 47.7 Change: −22.7 (-91.6%)	No Data	No Data	Initial: 101.8±10.9 Change: −3.7±2.7
bGros Herguido,N 2014 Case report D&B=11	N=2 Males Age: 37 and 47	Laparoscopic gastric bypass, 24 months follow up	Initial: 150 Change:-66	Initial: 44.08 kg Change: -21 kg	No Data	No Data	Initial: 140 Change: -53
bWong, S. 2013 Case report D&B=11	N=1, Age: 28 Injury level: T12 AISA: C	Roux-en-Y gastric bypass followed 7 months	Initial: 180.3 Change:-66	Initial:59.8 kg After 7m: 49.8 Change: -10 kg -16.7%	No Data	No Data	Initial:165 Change :-19 (-11.5%)
bAlaedeen, D.I. 2006 Case Report D&B=11	N=1 Male Age=51 Injury level: T7	Roux-en-Y gastric bypass	Initial: 169 Change:-52	Initial:48 kg Change: -15	No Data	No Data	No Data

Among all intervention modalities to manage obesity after SCI, surgery showed a significant effect in TFM reduction (from -10 to -22.7 Kg); however, all four studies were case-reports,^51–54 and were appraised as low-quality with a high potential for bias (Table 3).

Discussion

The present review indicates bariatric surgeries including restrictive and mal-absorptive procedures produced the greatest sustained weight loss and BMI correction in individuals with chronic SCI. However, these surgical interventions are only indicated for morbidly obese patients (BMI > 40). Similar to the general non-SCI population, bariatric surgery provides durable treatment of obesity and obesity-associated morbidities. Moreover, surgery is not a simple “cure” for obesity. An individual’s commitment to dietary and lifestyle changes are necessary for successful long-term outcomes. Surgical interventions for obesity are invasive with many anticipated complications and/or long term side effects, including: protein-calorie malnutrition, vitamin B12 deficiency, folate, calcium and vitamin D insufficiency, post-surgical infection, emboli and anastomosis leak.^55–61 These side effects are of utmost importance in individuals with SCI as they are at risk of protein calorie malnutrition. Energy and calcium intake are shown to be lower as active handicapped individuals,⁶² also very low intakes of vitamins C, D, E, folic acid, pantothenic acid, biotin, potassium and iron are reported for the SCI population.⁶³ On the other hand, it seems that larger amounts of Vitamins A, C and E and Copper and Zinc are required due to the high incidence of infections and pressure ulcers in this population.^64,65 Thus, most clinicians typically consider surgical interventions as a last resort to reduce morbid obesity.²⁷

Following bariatric surgery, combinations of physical exercise and diet therapy had the greatest effect in decreasing BW and BMI and remarkably improved body composition (increases in TLBM and reductions in TFM and WC), although the methodological quality of the five studies included in this review were evaluated as low. The advantage of diet therapy is that it is feasible to consider the special requirements and concerns of individuals in the SCI population. An anti-atherosclerotic diet, which basically lowers the amounts of saturated and trans fatty acids with a higher proportion of omega 3 fatty acids, limited salt, increased fiber and anti-oxidants intakes are recommended for these individuals, as cardiovascular diseases are the primary cause of morbidity and mortality among SCI out-patients.⁶⁶ The findings of this review are in accordance with Espirito Santo et al. findings that body weight- support treadmill training increased muscular trophism,⁶⁷ and reports from Nash et al. that a combination of circuit training and Mediterranean diet reduces dyslipidaemia (Total chol/HDL-C ratio or LDL-C levels).⁶⁸

NMES and FES

Selected NMES and FES studies were difficult to interpret collectively because of controversial results, small sample size, short duration of follow-up and the variability in the intervention and outcome measure selection. In general a higher frequency of training each week had more favourable results. For example, the trials with five times per week intervention showed more significant results compared to trials with the interventions less than five times per week, regardless of the study duration. However, the feasibility of doing an exercise program five times per week is poor.⁶⁹ Although NMES did not reduce BW or TFM, it increased regional lean mass and reduced regional fat mass. Similarly Gorgey et al. in a 2015 review reported that NMES or FES training can increase skeletal muscle size and soft tissue lean mass.⁷⁰ One may argue that active voluntary aerobic exercises with calorie restrictions, improves body composition more efficiently than passive non-voluntary interventions such as FES or NMES, which have greater local rather than systemic effects.

Exercise therapy

Aerobic exercise results in weight loss and decreases in TFM in several ways: First, it causes loss of energy as heat during ATP synthesis in the mitochondria and ATP hydrolysis during muscular contraction and therefore, increases energy expenditure above the individuals’ basal energy expenditure. Secondly, basal metabolic rate (BMR) and thermic effect of food (TEF) increases with PhA. Thirdly, fat oxidation is increased during PhA, especially in endurance trained subjects, due to an increase in the oxidative capacity of muscles. Consequently, the ability to oxidize lipids, due to upregulation of the enzyme AMP-activated protein kinase in skeletal muscles increases; so maximal oxygen consumption and the proportion of fat to carbohydrate oxidation increases.⁷¹ Many hormonal and metabolic pathways are activated during exercise, including reductions in blood insulin levels or increases in Vo2 peak.⁷² If individuals sustain a low caloric diet, to maintain a low blood glucose level, insulin production by the pancreas will decrease and result in a rise in hormone sensitive lipoprotein lipase activity in fat tissues and a subsequent reduction of TFM. These processes may not occur to the same extent with passive interventions such as FES or NMES, which may explain why the lack of a substantial change in TFM although regional increase in the lean mass are reported where the stimulus was applied.

Passive modalities have also been shown to have some side effects. Bhambhani et al. in a study conducted in Alberta, reported FES exercise in SCI individuals elicits some degree of muscular deoxygenation;⁷³ Villareal et al., in a clinical trial reported non-exercise-induced weight loss is associated with a decrease in bone mineral density (BMD) at clinically important sites of fracture, and suggested that exercise should be an important component of a weight loss program.⁷⁴ This consideration is of importance in the SCI population where many patients have comorbid low bone mineral density and excess adiposity.

Active aerobic or resistance exercise such as wheelchair propulsion in individuals with paraplegic can cause shoulder pain or rotator cuff injury⁷⁵ and carpal tunnel syndrome.^76,77 So health care providers who prescribe exercises should educate patients regarding the correct way to perform exercise and monitor patients for side effects or advers consequences.^78,79

Pharmacologic therapy

Pharmacologic treatment with anabolic agents - testosterone or oxandrolone- failed to show any significant change in BW, BMI, TLBM and TFM in SCI individuals. One RCT with alpha-lipoic acid showed a significant decrease in BW, BMI and WC but did not report data on TLBM and TFM.

Medications for weight loss have not been very effective in the general population. Sibutramin was withdrawn from the market in 2010; Orlistat -a gastro intestinal lipase inhibitor- causes fat-soluble vitamin deficiencies (vitamin A, D, E and K) and gastrointestinal side effects such as diarrhea, steatorrhea, bloating, flatulence, fecal urgency and incontinence. These adverse GI symptoms will not be well tolerated in individuals with SCI. Phentermine, a non-adrenergic compound licensed for short term use in the US, can result in a racing heart and raise diastolic blood pressure, which is relatively contraindicated in the SCI population who are at risk of cardiovascular events such as autonomic dysreflexia, arterial and ventricular arrhythmia and coronary artery diseases. However, the food and drug administration (FDA) has recently approved the serotonin agonist lorcaserin and a combination of low dose phentermine /topiramate, raising the possibility of improving current paradigms for the treatment of obesity.⁸⁰

Limitations

Some significant limitations should be taken into account when considering the clinical implications of this review. First, most studies suffered from a small sample size and a high risk of bias due to the study design (i.e., case reports versus gold standard clinical trials). For example the favorable results of the surgical interventions were obtained from case reports and not large prospective studies. Second, there was no consistency in outcome selection across the different obesity interventions. Third, there was variability in the choice of obesity biomarkers and thus a limited ability to pool data nor generalize the findings based on obesity severity. Future studies should prioritize decreases in visceral adipose tissue and increases in TFM (treatment of sarcopenic obesity) as the key outcome of importance. Understanding the role of the adipokines (adiponection, leptin, etc.) in predicting therapeutic responsiveness or lack of responsiveness is an important research priority for the field.

Conclusion

Although most of the studies selected for inclusion in this systematic review were low to moderate in terms of methodological quality, the greatest permanent weight reduction and BMI correction was produced by bariatric surgery, followed by a combination of long term physical exercise and diet therapy. Generally, NMES and pharmacologic treatment did not reduce weight or TFM, but resulted in increases in TLBM. Due to the link between adiposity and all-cause mortality, obesity is a legitimate therapeutic target. Based on feasibility and associated risk, trial of diet and exercise therapy is recommended prior to definitive bariatric surgery.

Appendix 1: Search Strategy

1. exp Spinal Cord Injuries/

2. exp Spinal Cord Vascular Diseases/

3. exp Spinal Injuries/

4. exp Spinal Fractures/

5. exp Trauma, Nervous System/

6. Spinal Cord Compression/

7. exp Myelitis/

8. Epidural Abscess/

9. Syringomyelia/

10. exp Paraplegia/

11. (postoperative adj4 (spinal cord injur* or SCI)).af.

12. (spinal cord adj3 (contusion* or laceration or transaction* or trauma* or injur* or damage*)).tw.

13. (SCI or spinal cord injur*).tw.

14. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13

15. Overweight/

16. Obesity/ or Obesity, Morbid/ or Obesity, Abdominal/

17. ((subcutaneous or intra-abdominal) adj3 fat).tw.

18. visceral fat.tw.

19. body mass index.tw.

20. waist circumference.tw.

21. android obesity.tw.

22. Adiposity/

23. Weight Gain/

24. exp Body Composition/

25. Waist-Hip Ratio/

26. Skinfold Thickness/

27. adiposity.tw.

28. weight gain.tw.

29. 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28

30. 14 and 29

31. prevention.tw.

32. treatment.tw.

33. exp Nutrition Therapy/

34. Physical Fitness/

35. “Physical Education and Training"/

36. exp Exercise Therapy/

37. exp Exercise/

38. Wheelchairs/

39. exp Ergometry/

40. Swimming/

41. swim*.tw.

42. strengthening program.mp.

43. (exercise adj4 (aerobic or resistance or therapy or physical or program or cardiopulmonary)).tw.

44. (train* adj3 (ergometry or flexibility or wheelchair*)).tw.

45. ((wheelchair* or wheel-chair*) adj3 (exercise or ergometry or propulsion)).tw.

46. (pharmacotherapy adj3 (overweight or obesity or fitness)).tw.

47. (surgery adj3 (obesity or morbid obesityor overweight or fitness)).tw.

48. exp Bariatric Surgery/

49. exp Anti-Obesity Agents/

50. 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49

51. 30 and 50

Disclaimer statements

Contributors None.

Funding This study performed as a non-granted systematic review by Neural Engineering and Therapeutics team, Toronto Rehabilitation Institute.

Declaration of interest The authors declare no conflict of interest.

Conflicts of interest None.

Ethics approval None.

Data archiving There were no data to deposit.

ORCID

SM. Alaviniaa http://orcid.org/0000-0002-5503-9362