Pathophysiology of Neurogenic Obesity After Spinal Cord Injury

Abstract

Individuals with a spinal cord injury (SCI) have a unique physiology characterized by sarcopenia, neurogenic osteoporosis, neurogenic anabolic deficiency, sympathetic dysfunction, and blunted satiety associated with their SCI, all of which alter energy balance and subsequently body composition. The distinct properties of “neurogenic obesity” place this population at great risk for metabolic dysfunction, including systemic inflammation, hyperglycemia, dyslipidemia, and hypertension. The purpose of this article is to demonstrate the relationship between neurogenic obesity and the metabolic syndrome after SCI, highlighting the mechanisms associated with adipose tissue pathology and those respective comorbidities. Additionally, representative studies of persons with SCI will be provided to elucidate the severity of the problem and to prompt greater vigilance among SCI specialists as well as primary care providers in order to better manage the epidemic from a public health perspective.

Introduction

“The World Health Organization (WHO) takes the position that obesity is a chronic, relapsing, progressive disease process and emphasizes the need for immediate action for prevention and control of this global epidemic.”¹ According to WHO, the principal environmental agent mediating the disease is food, followed closely by a decline in the level of physical activity. The WHO characterizes obesity according to body mass index (BMI), with BMI ≥25 kg/m² being overweight and BMI ≥30 kg/m² considered obese.² The genetically susceptible host^3,4 accumulates fat in white adipocyte tissue (WAT), which mainly mediates energy homeostasis but becomes dysfunctional and increases inappropriately in the obese individual.5 This “adipocyte hypertrophy” triggers adipose hyperplasia, low-grade inflammation, insufficient angiogenesis, and excess collagen deposition within the adipose stroma leading to the release of adipokines, systemic inflammation, insulin resistance, dyslipidemia, and hypertension (HTN) (Figure 1). Two prominent depots occur: visceral adipose tissue (VAT) located predominantly in the mesentery and greater omentum, and subcutaneous adipose tissue (SAT) principally found in the gluteofemoral, back, and anterolateral abdomen. VAT has a greater volume of large adipocytes that are associated with dyslipidemia, that is, increased nonesterified free fatty acids (NEFAs), increased total cholesterol (TC), increased low-density lipoprotein cholesterol (LDL-C), increased apolipoprotein B (apo B), and increased triglycerides (TG) but decreased high-density lipoprotein cholesterol (HDL-C), whereas SAT has a greater influence on insulin resistance.⁵ WAT includes various cell types within its stroma, including adipocyte progenitors, fibroblasts, endothelial cells, and proinflammatory immune cells called M1 macrophages that produce proinflammatory cytokines such as tumor necrosis factor-α (TNFα), interleukin-6 (IL-6), interleukin-1 b (IL-1b), monocyte chemoattractant protein 1 (MCP-1), and nuclear factor kappa-light-chainenhancer of activated B cells (NFκB). WAT also produces angiotensinogen, angiotensin II, and the adipokines leptin, adiponectin, and resistin that modulate appetite, energy metabolism, and insulin resistance, respectively.⁵

Figure 1.

Pathophysiology of obesity and metabolic syndrome. Environmental factors influence gene expression inducing gain of adipose tissue (AT). When subcutaneous adipose tissue capacity is reached, free-fatty acids (FFA) mobilize and are deposited in visceral and ectopic fat (e.g., liver, skeletal muscle, and heart). FFA deposition in muscle inhibits insulin-mediated glucose uptake (i.e., insulin resistance, IR), which reduces lipolysis and increases nonesterified FFA flux to the liver, resulting in hepatic IR (enhancing gluconeogenesis), hepatic lipogenesis, and atherogenic dyslipidemia. Hepatic glucagon resistance to amino acid (AA) production reduces ureagenesis, resulting in hyperaminoacidemia and glucagon resistance (GR); increased glucagon production from pancreatic α-cells accelerates hepatic gluconeogenesis. FFA deposit in the pancreas where β-cell dysfunction is caused by lipotoxicity; hyperglycemia and IR results. Hyperinsulinemia stimulates sodium reabsorption and increased sympathetic nervous system activity, contributing to hypertension (HTN). AT becomes more IR and releases proinflammatory adipokines, decreasing the anti-inflammatory adiponectin. Triglycerides and toxic metabolites in the liver induce lipotoxicity, mitochondrial dysfunction, and endoplasmic reticulum stress, resulting in hepatic damage, apoptosis, and fibrosis (nonalcoholic liver disease, NAFLD). The damaged hepatocytes release dipeptidyl peptidase 4 (DPP4), which stimulates AT macrophage inflammation, promoting further IR. Adapted from Godoy-Matos AF, Silva Junior WS, Valerio CM. NAFLD as a continuum: From obesity to metabolic syndrome and diabetes. Diabetol Metab Syndr. 2020;12:60. https://doi.org/10.1186/s13098-020-00570-y, licensed under a Creative Commons Attribution 4.0 International License: https://creativecommons.org/licenses/by/4.0/.

Although the term sarcopenia (reduced skeletal muscle mass) is often attributed to Dr. Irwin Rosenberg in 2011,⁶ sarcopenic obesity was described much earlier by a group of University of California, Los Angeles, scientists assessing premenopausal women at risk for breast cancer.⁷ The term sarcopenic obesity, characterized by the coexistence of excess fat mass and sarcopenia, has recently undergone critical appraisal for both definition and diagnostic criteria.⁸ In a systematic review beginning with 2335 references with 75 meeting eligibility criteria, the favored definition of sarcopenic obesity was based on the coexistence of obesity and sarcopenia, whereas diagnostic criteria favored appendicular skeletal mass (ASM) divided by weight (ASM/wt) or adjusted by height in meters squared (ASM/h²) in combination with BMI ≥30 kg/m² as representative of sarcopenia and obesity, respectively.⁸ Of note, functional diagnoses were also considered,⁸ including measures of functional strength (hand dynamometry or knee extensors) and physical performance (gait speed, timed up-and-go) that would preclude inclusion of persons with spinal cord injury (SCI).

We first introduced the term neurogenic obesity in a review of systemic inflammation for persons with SCI to characterize not just sarcopenic obesity due to muscle atrophy, but also the neurogenic osteoporosis, neurogenic anabolic deficiency, sympathetic dysfunction, and blunted satiety associated with SCI, all of which alter energy balance.⁹ Total daily energy expenditure (TDEE) is comprised of basal or resting metabolic rate (RMR), the thermic effect of physical activity (TEA), and thermic effect of food (TEF); of these, RMR and TEA are profoundly affected by the metabolically active fat-free mass (FFM) comprised of muscles, bones, and organs.^9,10 In addition to the mechanical unloading and loss of neurotrophic influences on muscle and bone after SCI,^11–13 anabolic hormones are significantly reduced after SCI, further diminishing both muscle and bone mass.^14–17 Similarly, diminished sympathetic nervous system activity after SCI decreases heart rate, blood pressure, and metabolic processes,¹⁸ further contributing to lowered RMR.^19,20 While each of the mechanisms listed previously reduce TDEE, satiety after SCI is likely impaired due to diminished vagal sensory processing,²¹ and, in fact, energy intake for persons with SCI is significantly greater than measured energy expenditure.²² There are multiple published studies^23–29 demonstrating neurogenic obesity in SCI by the two most commonly used metrics: BMI ≥22 kg/m^{2 25,30} or excess body fat >22% for men and >35% for women.^{23,29,31–34} With 67% to 97% obesity rates in persons with SCI, it is important to understand the metabolic consequences of neurogenic obesity in this vulnerable population.

Obesity, Inflammation, and Adipokines

As intimated previously, adipose tissue (AT) and its associated macrophages produce a number of proinflammatory adipokines, including TNFα, IL-1b, IL-6, MCP-1, and NFκB.³⁵ TNFα (25 kDa, 233 amino acids) is produced by both adipocytes and AT stromal macrophages and directly impairs insulin signaling by suppressing the expression of insulin receptor substrate-1 (IRS-1) and glucose transporter-4 (GLUT4) within muscle and liver and through upregulation of suppressor of SOCS3.³⁶ TNFα has also been implicated in chronically reducing insulin secretion in later stages of type 2 diabetes mellitus (T2DM) by inducing pancreatic β-cell apoptosis via NFκB.³⁷ IL-1b (17 kDa, 153 amino acids) is a proinflammatory cytokine secreted from adipocytes that, like TNFα, is implicated in the destruction of pancreatic β-cells.³⁸ Like TNFα, IL-6 (20 kDa, 212 amino acids) is a proinflammatory adipokine secreted by adipocytes and AT stromal macrophages that suppresses insulin signaling transduction via suppressor of SOCS3 and down-regulating transcription of IRS-1 and GLUT4.³⁹ MCP-1 (13 kDa, 76 amino acids) is a chemokine produced by adipocytes and AT stromal macrophages after stimulation by TNFα and IL-1b that attracts macrophages, monocytes, and other immune cells to the inflammatory sites of vascular subendothelial space, promoting monocyte migration into the arterial wall to form macrophage-derived foam cells (atherosclerosis).⁴⁰ NFκB is a protein complex that controls DNA transcription, cytokine production, and cell survival within cells when activated by TNFα, IL-1b, IL-6, and other cytokines to block phosphorylation of IRS-1 and IRS-2, inhibiting the phosphoinositide 3-kinase (PI3K)/AKT kinase cascade required within muscle, liver, and fat cells to activate GLUT4 receptor migration to cell membranes.⁴1 Taken together, these proinflammatory adipokines create a chronic, low-grade inflammatory state throughout the vascular tree, while mediating dyslipidemia, insulin resistance, and HTN.⁹

Additional adipokines of importance include leptin, adiponectin, and resistin. Leptin (16-kDa, 167 amino acids) is an adipose-derived cytokine (adipokine) that suppresses appetite, activates early satiety, and decreases insulin resistance under usual conditions.³⁵ Adiponectin (30 kDa, 244 amino acids) is an anti-inflammatory adipokine, exclusively secreted from adipocytes and, under normal conditions of nonadiposity, improves insulin sensitivity, reduces inflammation, and increases energy expenditure.³⁵ Resistin (11 kDa, 114 amino acids) is secreted by both adipocytes and AT stromal macrophages and increases insulin resistance by activating Toll-like receptor 4 (TLR4) and SOCS3, down-regulating transcription of IRS-1 and GLUT4.⁴²

Finally, plasminogen activator inhibitor-1 (PAI-1) is a 50 kDa protein of 379 amino acids that is secreted from AT adipocytes and macrophages. As its name suggests, PAI-1 inhibits fibrinolysis (i.e., increases the likelihood of blood clots throughout the vascular tree), increasing the likelihood of venous thromboembolism, coronary, and cerebrovascular events.⁴³ Of note, TNFα has also been implicated in chronically reducing insulin secretion in later stages of T2DM by inducing pancreatic β-cell apoptosis via NFκB.³⁷

Obesity and Insulin Resistance

T2DM is characterized by elevated blood glucose concentration, insulin resistance, and hyperinsulinemia, although in the later stages of T2DM, pancreatic β-cell destruction may lead to insulin insufficiency. It is widely recognized that obesity mediates insulin resistance and T2DM through inhibition of the PI3-kinase insulin cascade, which is therefore unable to activate GLUT4 transporters to migrate to the cell membrane for glucose transport; hyperglycemia and hyperinsulinemia ensue.⁴⁴ As a reminder, insulin is secreted from pancreatic β cells in response to elevated blood glucose levels. Activation of tyrosine kinase insulin receptors (INSR) on the cell membranes of muscle, liver, and fat cells promotes glucose uptake and glycogen storage within the cell via the PI3K/AKT cascade, which activates inactive GLUT4 receptors to translocate to the cell membrane and allow the passage of glucose into the cell. As AT accumulates, lipolysis produces NEFAs that are taken up into muscle and liver cells where they are further metabolized into acyl-coA, ceramides, and diacylglycerol. These NEFAs and their metabolites activate, within the cell, protein kinase-c (PKC), jun kinase (JNK), and NFκB-inhibitor kinase B (IKKB), subsequently interfering with the phosphorylation of IRS-1 and IRS-2 and effectively inhibiting the PI3-K/AKT cascade. TNFα, IL-1b, and IL-6 similarly activate PKC, JNK, and IKKB, inhibiting IRS-1 and IRS-2 and the PI3K/AKT kinase cascade. Of note, TNFα, IL-1b, and IL-6 also activate NFκB, which in turn activates SOCS3, down-regulating transcription of IRS-1 and GLUT4. TNFα and IL-1b chronically reduce insulin secretion in later stages of T2DM by inducing pancreatic β-cell apoptosis via NFκB.³⁷

Numerous studies have demonstrated obesity-related hyperglycemia, hyperinsulinemia, and glucose intolerance in persons with SCI,^29,45–51 and recent recommendations encourage increased awareness, screening practices, and treatment strategies of these glycemic disorders.^32,33,52 The comorbidities of T2DM including coronary, cerebral and peripheral artery disease, diabetic retinopathy, autonomic neuropathy, diabetic nephropathy, diabetic gastropathy, peripheral neuropathy, and diabetic ulcers are burdensome to those without SCI but are devastating to persons with SCI.

Obesity and Dyslipidemia

Insulin resistance is the most probable link to obesity-related metabolic dyslipidemia.^53–55 Under normal conditions, insulin suppresses lipolysis in AT by hormone-sensitive lipase, stimulates apo-B degradation, and suppresses very low-density lipoprotein (VLDL-C) secretion from the liver. Insulin also stimulates lipoprotein lipase (LPL) to hydrolyze TG from VLDL particles in the circulation, promoting TG-rich lipoprotein degradation. Under conditions of increasing insulin resistance associated with AT accumulation as described previously, these processes are inhibited and lead to AT lipolysis, hypertriglyceridemia, and the accelerated production of NEFAs.⁵⁴ As NEFAs and associated TG accumulate within the liver, hepatic production of apo-B, LDL-C, and VLDL-C increases, while apolipoprotein-A (apo-A) production is slowed, resulting in reductions of HDL-C, the primary scavenger of peripheral lipids in the vasculature. Additionally, cholesteryl ester transfer protein (CETP) activity promotes the exchange of TG with cholesterol esters between lipoproteins, rendering the circulating HDL-C particles dysfunctional.^54,55 The reduction of functional HDL-C is accompanied by a significant risk of atherosclerosis, such that a 5 mg/dL decrement in HDL-C correlates with a 14% increase incremental risk of cardiovascular events.⁵⁶ This is especially notable in persons with SCI, with multiple cohorts demonstrating more than 60% having HDL-C less than 40 mg/dL.^29,30,48,57

Obesity and Hypertension

The relationship between obesity and HTN is well established, and it is estimated that obesity accounts for 65% to 75% of all cases with primary HTN.^58–62 The pathophysiology of obesity-related HTN includes at least five different mechanisms, including leptin resistance, sympathetic nervous system (SNS) activation, renin-angiotensinaldosterone system activation (RAAS), natriuretic peptide (NP) catabolism, and renal compression.⁵⁸ As adiposity increases, a paradoxical increase in leptin resistance occurs. Further, leptin increases arterial blood pressure by activating aldosterone synthase secretion (via the CYP11B2 gene) from the renal zona glomerulosa^58,60,63 and through activation of the renal SNS via the proopiomelanocortinmelancortin 4 receptor (MC4R) pathway in the central nervous system.^59,60,62 VAT increases both muscular SNS activity (MSNA) and renal SNS activity (RSNA) through adipokine inflammatory effects and influence of the RAAS.^60,64 Adipocytes produce angiotensinogen and subsequently secrete angiotensin II (AngII), a potent vasoconstrictor and RAAS activator.^58,60,64 AngII also interferes with vascular insulin signaling, which is already compromised due to IR, further hampering nitric oxide release; vasodilation is inhibited.^64,65 AngII stimulates the release of aldosterone from the glomerular cortex, increasing sodium and fluid retention.⁶⁴ NPs, consisting of atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and cardiovascular C-type natriuretic peptide (CNP), exert vasodilating, diuretic, natriuretic, and lipolytic actions through their membrane receptors that become resistant with increased AT, and NPs are subsequently degraded through the endopeptidase neprilysin, an adipose-derived adipokine.^58,59 The accretion of perirenal and retroperitoneal fat associated with VAT physically compresses the kidneys, ureters, lymph vessels, renal veins, and renal parenchyma, causing blood pressure (BP) elevations as high as 35 to 40 mm Hg.^58,60 Lastly, the combination of these factors in conjunction with “stiff” vessels associated with arteriosclerosis provides great challenge to the clinician managing obesity-related hypertension.

Neurogenic hypotension associated with sympathetic blunting and circulatory hypokinesis in persons with tetraplegia or paraplegia above T6 often requires mechanical and pharmacological intervention to prevent syncope in the acute and early stages of SCI.1^8,66 However, as AT increases due to a positive energy balance over months and years, the mechanisms for HTN listed previously gradually offset and exceed the neurogenic dysfunction, and persons with SCI subsequently encounter what appear to be normal or, in fact, high blood pressure (21.1%–56.6% prevalence) not associated with autonomic dysreflexia.^26,48,67,68 Of note, venerable work from 2007 described 23% of 7959 veterans with HTN; 68% had BMI >23 kg/m², prompting the question of why these neurologically compromised veterans would have HTN.²⁶ Subsequent studies have demonstrated HTN in 56.6% of 6672 veterans (57% had BMI >25 kg/m²)⁶⁸ and 55.2% of 473 veterans (76.7% had BMI ≥22 kg/m²),⁴⁸ respectively. Thus, despite the risk for neurogenic cardiovascular dysfunction, even persons with SCI appear at high risk for HTN when concomitantly obese.

Obesity and Metabolic Syndrome

The preceding paragraphs have illustrated the pathophysiology of obesity-related systemic inflammation, hyperglycemia, dyslipidemia, and HTN, hallmarks of the “metabolic syndrome.” While several definitions from leading world authorities remain slightly disparate, all have in common the similar elements of obesity, dysglycemia, dyslipidemia, and HTN (Table 1). The appropriately characterized neurogenic obesity seen in persons with SCI has been correlated with metabolic syndrome in multiple studies, suggesting a prevalence of greater than 50%.^29,30,48,69 Of note, a recent investigation reported prevalence of metabolic syndrome in 155 veterans as 31% (IDF), 17% (NCEP), 53% (NHLBI/AHA), or 19% (WHO), depending upon the definition provided by that organization.⁷⁰ The authors of that study reported some degree of the ambiguity was due to the lack of consensus for an SCI-screening tool to accurately classify obesity in this special population.

Table 1.

Diagnostic criteria for the metabolic syndrome from World Health Organization, National Cholesterol Education Panel (Adult Treatment Panel III), and International Diabetes Federation

WHO	NCEP (ATP III)	IDF
Required: IR (IGT, IFG, T2DM, or other evidence of IR)	Required: None	Required: Central obesity (WC ≥ 94 cm in men and ≥ 80 cm in women
Criteria: T2DM or impaired glucose or IR (obligatory) plus two or more of the following	Criteria: Any three of five criteria below constitute a diagnosis of metabolic syndrome	Criteria: Central obesity (obligatory) plus any two of the four other criteria constitute a diagnosis of metabolic syndrome
Obesity: BMI >30 kg/m2 or WHR >0.9 in men and >0.85 in women	Obesity: WC >102 cm in men and >88 cm in women	Obesity: Ethnic-specific WC (e.g., South Asians ≥90 cm in men and ≥80 cm in women)
Dyslipidemia: Triglycerides ≥150 mg/dL (1.7 mmol/L)	Dyslipidemia: Triglycerides ≥150 mg/dL (1.7 mmol/L); or, on specific treatment for this lipid disorder	Dyslipidemia: Triglycerides ≥150 mg/dL (1.7 mmol/L); or, on specific treatment for this lipid disorder
Dyslipidemia (second, separate criteria): HDL-C <35 mg/dL (0.9 mmol/L) in men) and <39 mg/dL (1.0 mmol/L) in women), or on specific treatment for this lipid disorder	Dyslipidemia (second, separate criteria): HDL-C <40 mg/dL (1.03 mmol/L) in men and <50 mg/dL (1.29 mmol/L) in women; or, on specific treatment for this lipid disorder	Dyslipidemia (second, separate criteria): HDL-C <40 mg/dL (1.03 mmol/L) in men and <50 mg/dL (1.29 mmol/L) in women; or, on specific treatment for this lipid disorder
Hypertension: BP ≥140/90 mm Hg; or, on treatment for hypertension	Hypertension: Systolic BP ≥130 mm Hg or diastolic BP ≥85 mm Hg; or, on treatment for hypertension	Hypertension: Systolic BP ≥130 mm Hg or diastolic BP ≥85 mm Hg; or, on treatment for hypertension
Other criteria: Microalbuminuria: Albumin excretion ≥20 μg/min or albumin/ creatinine ratio ≥30 mg/g	Hyperglycemia: Fasting Glucose: ≥100 mg/dL (5.6 mmol/L); or previously T2DM	Hyperglycemia: Fasting Glucose: ≥100 mg/dL (5.6 mmol/L); or previously diagnosed T2DM

Note: BP = blood pressure; IFG = impaired fasting glucose; IGT = impaired glucose tolerance; IR = insulin resistance; T2DM = type 2 diabetes mellitus; WC = waist circumference; WHR = waist-hip ratio.

This article is necessarily limited by space and reference restrictions, and not all pertinent and seminal studies may have been identified, fully reviewed, or included. We recognize the need for a transparent and unbiased review using systematic evidence synthesis methodology to fully describe the pathophysiology, prevalence, and incidence of neurogenic obesity in persons with SCI.

Conclusion

Physiological changes in AT following SCI should be characterized as neurogenic obesity due to an obligatory sarcopenia, neurogenic osteoporosis, neurogenic anabolic deficiency, sympathetic dysfunction, and blunted satiety associated with SCI, all of which alter energy balance and subsequently body composition. Because of these findings, recent guidelines recommend characterizing obesity in SCI as BMI ≥22 kg/m² or as %BF >22% for men and >35% for women.^32,33,71 The epidemic of neurogenic obesity associated with SCI puts individuals at high risk for the metabolic syndrome, due to AT-related systemic inflammation, hyperglycemia, dyslipidemia, and HTN. Although future research should focus on the prevention and management of neurogenic obesity through exercise⁷² and nutrition,⁷³ clinicians should optimize their screening practices and management strategies for the accompanying comorbidities.^32,33