Gastrointestinal Complications of Obesity

Abstract

Obesity usually is associated with morbidity related to diabetes mellitus and cardiovascular diseases. However, there are many gastrointestinal and hepatic diseases for which obesity is the direct cause (eg, nonalcoholic fatty liver disease) or is a significant risk factor, such as reflux esophagitis and gallstones. When obesity is a risk factor, it may interact with other mechanisms and result in earlier presentation or complicated diseases. There are increased odds ratios or relative risks of several gastrointestinal complications of obesity: gastroesophageal reflux disease, erosive esophagitis, Barrett’s esophagus, esophageal adenocarcinoma, erosive gastritis, gastric cancer, diarrhea, colonic diverticular disease, polyps, cancer, liver disease including nonalcoholic fatty liver disease, cirrhosis, hepatocellular carcinoma, gallstones, acute pancreatitis, and pancreatic cancer. Gastroenterologists are uniquely poised to participate in the multidisciplinary management of obesity as physicians caring for people with obesity-related diseases, in addition to their expertise in nutrition and endoscopic interventions.

Obesity usually is associated with morbidity related to diabetes mellitus and cardiovascular diseases. However, there are many gastrointestinal and hepatic diseases for which obesity is the direct cause (eg, nonalcoholic fatty liver diseases [NAFLDs]) or is a significant risk factor such as in reflux esophagitis and gallstones. When obesity is a risk factor, it may interact with other pathogenetic mechanisms and result in earlier presentation of disease or more complicated disease.

The gastrointestinal tract plays a key role in obesity through its contributions to satiation and satiety, production of gut hormones that influence appetite (such as ghrelin, cholecystokinin, and peptide YY), incretins (eg, glucagon-like peptide-1) that impact postprandial glycemia, absorption of nutrients that ultimately determine the positive energy balance that results in obesity, changes in bile acids and the microbiome, and the metabolic products of microbial digestion of nutrients (short-chain fatty acids) that modify some of the metabolic factors that are associated with obesity. Most of these topics are addressed elsewhere in this issue of Gastroenterology. Therefore, this article focuses on the gastrointestinal and hepatobiliary complications of obesity in adults (Figure 1); a separate article addresses the complications of pediatric obesity. Table 1 summarizes the quantified risks (odds ratios [ORs] and relative risks [RRs]) of gastrointestinal complications of obesity in adults.

Figure 1.

Gastrointestinal and hepatic morbidity associated with obesity.

Table 1.

Quantified Risk Ratios of Gastrointestinal Complications of Obesity in Adults

Gastrointestinal disease	Obesity as a risk factor		Reference
	Risk: OR or RR	95% CI
Esophagus
GERD	OR, 1.94	1.46–2.57	14
Erosive esophagitis	OR, 1.87	1.51–2.31	15
Barrett’s esophagus	OR, 4.0	1.4–11.1	23
Esophageal adenocarcinoma	Men: OR, 2.4	1.9–3.2	31
	Women: OR, 2.1	1.4–3.2
	RR, 4.8	3.0–7.7	162
Stomach
Erosive gastritis	OR, 2.23	1.59–3.11	43
Gastric cancer	OR, 1.55	1.31–1.84	45,46
	RR (cardia), 1.8	1.3–2.5	162
Small Intestine
Diarrhea	OR, 2.7	1.10–6.8	38
Colon and rectum
Diverticular disease	RR, 1.78	1.08–2.94	90
Polyps	OR, 1.44	1.23–1.70	163
Colorectal cancer	Men: RR, 1.95	1.59–2.39	104
	Women: RR, 1.15	1.06–1.24
	RR, 1.3	1.3–1.4	162
Clostridium difficile infection	OR, 1.196 per 1 kg/m2 increase in BMI	1.12–1.27	105
Anorectum
Dyssynergic defecation	OR, 1.64	1.09–2.47	162
Liver
NAFLD	RR, 4.6	2.5–110	164
Cirrhosis	RR, 4.1	1.4–11.4	165
Hepatocellular carcinoma	RR, 1.89	1.51–2.36	114
	RR, 1.8	1.6–2.1	162
Gallbladder
Gallstone disease	Men: RR, 2.51	2.16–2.91	148
	Women: RR, 2.32	1.17–4.57
Gallbladder cancer	RR, 1.3	1.2–1.4	162
Pancreas
Acute pancreatitis	RR, 2.20	1.82–2.66	156
Pancreatic cancer	Men: RR, 1.10	1.04–1.22	161
	Women: RR, 1.13	1.05–1.18
	RR, 1.5	1.2–1.8	162

Esophagus

Many esophageal disorders are associated with obesity.

Esophageal Dysmotility

Obesity increases the prevalence of esophageal motility disorders. For example, esophageal transit time was prolonged significantly in subjects with obesity compared with lean subjects,¹ possibly because of increased gastric and gastroesophageal junction resistance.² The typical abnormalities of esophageal motility are nonspecific abnormalities of esophageal peristalsis and, rarely, lower esophageal sphincter (LES) dysfunction, including isolated hypertensive or hypotensive LES pressures. In a recent population-based study, symptoms of dysphagia were more common in patients with obesity compared with lean controls (OR, 1.22; 95% CI, 1.04–1.43).³ A prospective study of 53 Canadian patients with a mean body mass index (BMI) of 46 kg/m² and documented symptoms (heartburn, 66%; regurgitation, 26%; dysphagia, 43%; and chest pain, 6%) reported that almost 50% had esophageal and LES dysmotility, mainly hypomotility.⁴ However, there was no comparator group of lean controls, and the prevalence of dysmotility was similar in the symptomatic and nonsymptomatic participants.⁴ In another study of 116 obese patients with a mean BMI of 42.9 kg/m², abnormal manometric findings were shown in 41% of patients, but these typically were not associated with symptoms.⁵ Diffuse esophageal spasm and achalasia are rare, and a summary of the literature shows that the prevalence and significance of all these dysmotilities are unclear because studies performed did not include lean controls.^5,6

Gastroesophageal Reflux Disease

Excess body weight and, in particular, increased abdominal girth produce higher intra-abdominal pressure and reduces LES pressure. In addition, other factors combine to predispose to gastroesophageal reflux and its complications, including a reduced length of the intra-abdominal portion of the lower esophageal sphincter and peristaltic dysfunction of the esophagus.^7,8 Obesity also results in increased esophageal acid exposure, ^8,9 and this may be related in part to increased estrogen levels, which are higher in obesity than in lean age- and sex-matched controls and are associated strongly with increased acid exposure and gastroesophageal reflux disease (GERD).^10–12 These alterations in functions can result in regurgitation, esophagitis, and GERD, which may progress to Barrett’s esophagus and esophageal adenocarcinoma.

GERD is a chronic disorder characterized by heartburn and regurgitation that occur when gastric acid or bile reflux from the stomach to the esophagus and induce inflammation of the esophageal mucosa. The prevalence of GERD has increased significantly in the past 20 years in parallel with the increased prevalence of obesity. Several meta-analyses have shown a positive association between body weight (BMI) and GERD.^13,14 In addition, central adiposity (apart from BMI) is an independent risk factor of the consequences of GERD, including esophageal inflammation, Barrett’s metaplasia, and esophageal adenocarcinoma, and these effects are mediated by reflux-dependent and reflux-independent mechanisms.¹⁵ The association of BMI with GERD is stronger in women with obesity than in men with obesity; this difference has been attributed to increased estrogen levels in women.¹¹ The role of estrogens in the association of central obesity in men and the higher prevalence of GERD is unclear. The association of BMI and GERD also is stronger in Caucasians than in other ethnicities.¹⁶ The strong association between obesity and GERD is reinforced by improvement of GERD symptoms after weight loss,¹⁷ which was confirmed in a well-designed intervention trial focused on weight loss for GERD.¹⁸

Erosive Esophagitis

Erosive esophagitis results from inflammation of the distal esophageal mucosa, which is secondary to GERD. Obesity is one of the known risk factors for developing erosive esophagitis, in addition to male sex, older age, chronic alcohol intake, chronic smoking, and a long history of GERD.¹⁹ Several meta-analyses have shown the association of a higher BMI, increased waist circumference, or increased waist-to-hip ratio with the presence and severity of erosive esophagitis.^14,15,20 Patients with central adiposity (apple shape) have a 1.87-fold risk of developing erosive esophagitis compared with normal-weight controls, independent of body weight (OR, 1.87; 95% confidence interval [CI], 1.51–2.31).¹⁵ In contrast, obesity with increased hip circumference (pear-shaped) is related inversely to erosive esophagitis and Barrett’s esophagus, analogous to its protective role in progression to type 2 diabetes mellitus and cardiovascular disease.²¹

Barrett’s Esophagus

Barrett’s metaplasia refers to the replacement of the normal squamous epithelium of the distal esophagus by specialized columnar epithelium. Barrett’s esophagus is usually a consequence of chronic GERD and predisposes to adenocarcinoma of the esophagus.²² Several studies have shown an association between obesity, abdominal circumference, and metabolic syndrome with Barrett’s esophagus. ^23,24 Moreover, BMI and abdominal circumference may be indirect risk factors for Barrett’s esophagus through their relationship with GERD.²⁵ However, the association of Barrett’s esophagus with abdominal adiposity is even stronger after adjusting for BMI or GERD, suggesting that abdominal adiposity is an independent risk factor.¹⁵ Potential mechanisms are higher levels of leptin, decreased levels of low-molecular-weight adiponectin, and increased cytokines, which mediate chronic inflammation.^26–28 The relationship of ghrelin and leptin to Barrett’s esophagus is complex. Thus, a recent case–control study of patients with a new diagnosis of Barrett’s esophagus and 2 control groups (GERD and general population) matched for age, sex, and geographic region found that higher levels of ghrelin were associated with an increased risk of Barrett’s esophagus among the general population. In contrast, leptin was associated positively with frequent GERD symptoms, but associated inversely with the risk of Barrett’s esophagus among the GERD controls.²⁹

Esophageal Adenocarcinoma

The incidence of esophageal adenocarcinoma is increasing, and this has been attributed to the increased prevalence of Barrett’s esophagus, erosive esophagitis, and GERD, all of which are associated with obesity and abdominal adiposity. In patients with Barrett’s esophagus, obesity is associated directly with progression to adenocarcinoma, and higher levels of leptin and lower levels of adiponectin have been proposed as markers of progression to adenocarcinoma.³⁰ In a meta-analysis of 2488 cases with esophageal adenocarcinoma, there was a strong association with obesity in both sexes (males: OR, 2.4; 95% CI, 1.9–3.2; females: OR, 2.1; 95% CI, 1.4–3.2).³¹ The risk of esophageal adenocarcinoma also was higher with increased central adiposity (OR, 2.51; 95% CI, 1.56–4.04) when compared with normal body habitus.¹⁵

The molecular mechanisms linking obesity, metabolic syndrome, and esophageal adenocarcinoma have been investigated extensively. These include increased insulin and insulin-like growth factor.^32,33 There is evidence that insulin-like growth factor-1 and insulin-like growth factor-2 induced increased angiogenesis and cell proliferation, decreased apoptosis, and increased cytokines secondary to obesity-induced chronic inflammation with induction of vascular endothelial growth factor, decreased adiponectin, and increased leptin.²⁸ Leptin stimulates cell proliferation by activating epidermal growth factor receptor; leptin also inhibits apoptosis in esophageal cells.³⁴ Histopathologic studies of esophageal adenocarcinoma from obese patients have shown up-regulated expression of leptin and adiponectin receptors in esophageal tumors.³⁵

Stomach

Gastric physiology and its neurohormonal regulation³⁶ are altered in obesity; however, it is unclear whether gastric function abnormalities are the cause or consequence of obesity. Obesity also is associated with symptoms that may arise in the stomach, such as upper abdominal pain, nausea, vomiting, retching, and gastritis.^37,38

Gastric Motor Physiology

Studies have quantified changes in gastric physiological functions using noninvasive approaches. Delgado-Aros et al^39,40 showed that, across a broad spectrum of BMI, there was an association between higher BMI, higher fasting gastric volume, and decreased satiation shown by a reduced fullness sensation and a higher maximum tolerated volume of Ensure (Abbott Nutrition, Chicago, IL) ingested at a constant rate. These studies showed that an increase in the fasting gastric volume of 50 mL was associated with 114 ± 32 kcal more ingested at maximum satiation. A follow-up Mayo Clinic study⁴¹ of 509 people across the BMI spectrum confirmed that obesity was associated with decreased satiation and, for every 5 kg/m² of BMI, there was a 50 kcal higher consumption before sensing fullness. In addition, a higher BMI was associated with a greater fasting gastric volume, and accelerated gastric emptying of solids and liquids (Figure 2).

Figure 2.

Obesity is associated with higher volume to experience fullness during a nutrient drink test (upper left), and faster gastric emptying of solids (upper right). Lower: noninvasive single-photon emission computerized tomography (SPECT) imaging of the stomach, which is used to measure fasting and postprandial gastric volumes. ^*P < .05; T_1/2, time to empty 50% of the ingested meal. Reproduced with permission from Acosta et al.⁴¹

Erosive Gastritis

Erosive gastritis is an inflammation in the mucosa of the stomach that may be acute or chronic and lead to ulceration and bleeding. Obesity is a risk factor for erosive gastritis and gastric and duodenal ulcers.^42,43 An association of low adiponectin with erosive gastritis has been reported to be independent of BMI or Helicobacter pylori infection.⁴⁴

Gastric Cancer

Obesity is considered a proinflammatory and procarcinogenic state and is recognized as an important, potentially modifiable, risk factor for cancer, including gastric cancer. Different meta-analyses (documented in Table 1) have reported associations of obesity (high BMI) with gastric cancer⁴⁵ and with cancer of the gastric cardia.⁴⁶ It is not clear if the association is related to confounding factors, such as an association of obesity with H pylori infection.⁴⁷ Obesity may accelerate H pylori–mediated gastric carcinogenesis. ⁴⁸ In one report of the relationship between BMI and esophageal-gastric cardia cancer, the meta-analysis of case–control and cohort studies confirmed the strong relationship of overweight and obesity with the cancers, but there was no substantial difference across strata of sex and geographic areas in Italy.⁴⁹ Another meta-analysis supported the hypothesis that longer exposure to estrogen effects of either ovarian or exogenous origin may decrease the risk of gastric cancer.⁵⁰ Thus, the analysis of the effect of sex on obesity as a predisposing factor to gastric cancer is complicated by the possible protective role of estrogens in women and the presence of significant environmental risk factors, predominantly cigarette smoking among men or women.⁵¹

Small Intestine

The small intestine is the site of digestion and absorption of most nutrients. In the past, it was thought that the small intestine played a passive role, simply absorbing the excess calories ingested by obese people. Bile acids play a critically important role in the absorption of fats; the role of bile acids in metabolic regulation⁵² or as potential therapeutic approaches for obesity and metabolic syndrome⁵³ are beyond the scope of this article; however, there is no evidence that bile acid synthesis or enterohepatic circulation is altered by obesity. On the other hand, there is evidence that the small intestine is able to adapt its absorptive functions for the 3 macronutrient classes, as follows: (1) lipid absorption capacity adapts to the fat content of the diet, especially through the coordinated induction of lipid binding proteins, which are involved in the intestinal absorption of long-chain fatty acids as well as their uptake, trafficking, and re-assembly into chylomicrons⁵⁴; (2) energy intake from infusion of intraduodenal whey protein hydrolysate tended to be higher in obese nondiabetic men than in lean controls⁵⁵; and (3) in morbid obesity, glucose absorption in the proximal intestine is accelerated and this is related to increased sodium-glucose linked transporter-1 (SGLT-1) expression. The increased glucose absorption in the proximal gut may predispose to obesity and type 2 diabetes.⁵⁶

Although alterations in the physiology of the small intestine may allow greater absorption capacity,⁵⁴ it is likely that small-bowel absorption adapts to the over-consumption of calories associated with obesity, leading to more rapid absorption without an increase in energy absorption.

Diarrhea

The prevalence of diarrhea in obese people is higher compared with normal-weight controls.⁵⁷ A population-based survey study of 2660 people showed that the prevalence of diarrhea in obese individuals was 30% compared with 17% in normal-weight controls (OR, 2.7; 95% CI, 1.1–6.8).³⁸ Similar studies have been replicated in Australia and New Zealand.^58,59 In an epidemiologic study of more than 35,000 persons in France, functional diarrhea was associated with BMI in females (OR, 1.05; 1.03–1.07), but not in males.⁶⁰

Among 1001 Swedes, diarrhea (OR, 2.2; 95% CI, 1.38– 3.46), stool urgency (OR, 1.60; 95% CI, 1.04–2.47), and nocturnal urgency (OR, 2.57; 95% CI, 1.33–4.98) were more prevalent in obese people than in lean controls, adjusting for age, sex, and education.⁶¹

The higher prevalence of diarrhea could be attributed to several potential mechanisms associated with obesity: changes in bile acids resulting in bile acid diarrhea,⁶² accelerated colonic transit,⁶³ increased mucosal permeability, ^64,65 or intestinal inflammation as evidenced by increased levels of fecal calprotectin.⁶⁶ These interesting studies require replication. Medications used by obese individuals, such as metformin for type 2 diabetes mellitus or polycystic ovary syndrome, also may cause diarrhea.⁶⁷

Celiac Disease

Celiac disease is an immune response to gluten in genetically susceptible people and it mainly affects the small intestine. The typical presentation includes weight loss, diarrhea, and malabsorption. Paradoxically, the recognition and prevalence of celiac disease in the obese population is increasing. In patients with newly diagnosed celiac disease, the prevalence of obesity varies from 39% to 44%.^68,69 Obese adult or pediatric patients with celiac disease are more likely to gain more weight on a gluten-free diet.^68,70,71

Inflammatory Bowel Diseases

Crohn’s disease or ulcerative colitis are autoimmune disorders that mainly target the small bowel and colon, respectively.

Epidemiology

In a case-control study, there was a U-shaped association between BMI and Crohn’s disease. Patients who are underweight or overweight were more likely to have Crohn’s disease.⁷² These findings were not reproduced in a European cohort, which showed no association between obesity and IBD in adults,⁷³ and a meta-analysis of 24 studies that included 1442 adult patients with IBD disease and 2059 healthy controls showed that obesity was less prevalent in patients with Crohn’s disease (OR, −1.88; 95% CI, −2.77 to −1.00) and that there was no difference in ulcerative colitis.⁷⁴ On the other hand, in children, the prevalence of obesity in IBD was similar to that in the general population, but obese children with IBD had more severe disease than normal-weight children.⁷⁵ The latter study also identified that treatment with corticosteroids may be a confounder in the interpretation of the relationship between obesity and IBD because the steroid treatment may have predisposed to the development of obesity. In addition, ethnic differences, illustrated by the observation that African Americans (OR, 1.64; 95% CI, 1.10–2.48) and those on Medicaid insurance (OR, 1.67; 95% CI, 1.19–2.34), were associated positively with overweight/obese status, and this may provide an explanation for the absence of this association in European cohorts, which likely were more ethnically homogeneous.

Mechanisms of IBD related to obesity

There are similarities in some of the pathophysiological features occurring in metabolic syndrome and IBD, including adipose tissue dysregulation, inadequate immune response, dysbiosis, and inflammation. Thus, in metabolic syndrome and IBD, inflammation affects adipose tissue and disturbs adipokine secretion. Both Crohn’s disease and ulcerative colitis are associated with high levels of 2 adipokines (resistin and visfatin) and low levels of a third adipokine (leptin).⁷⁶ Another study showed that active Crohn’s disease was associated with significantly lower adiponectin levels compared with the control group.⁷⁷ Given the contradictory data in these 2 studies, the significance of these adipokine levels still is unclear. Inflammatory disease affecting the ileocolonic mucosa also could impact the synthesis or release of incretins, which may have metabolic effects,⁷⁸ and, paradoxically, increased release of glucagon-like peptide-1 has been reported to retard gastric emptying in patients with active inflammatory bowel disease.⁷⁹

Treatment and Outcomes

Obesity is one of the factors associated with increased risk of surgical site infections in patients with IBD.⁸⁰ In a metropolitan US population, patients with obesity and IBD were significantly less likely to receive anti–tumor necrosis factor-α treatment, undergo surgery, or experience hospitalization for their IBD than their nonobese counterparts.⁸¹ These observations require confirmation. Another retrospective study in 1494 patients with IBD addressed the relationship between BMI and dose of medications for IBD, and reported that obesity was associated with a lower dosage (milligrams per kilogram) of purine analogs and biologics. ⁸² It is interesting that, although the role of obesity in IBD still is unclear, obesity does influence outcome in IBD management. In a study of 124 patients with IBD naive to biologic therapy, who were started on infliximab, it was observed that higher body weight was associated with an earlier time to loss of response to infliximab, this was independent of the dose in patients with Crohn’s disease (adjusted hazard ratio, 3.03; P < .001) or ulcerative colitis (adjusted hazard ratio, 9.68; P = .06).⁸³ Similar findings were reported for adalimumab in obese patients with Crohn’s disease who had a significant loss of response compared with lean controls.⁸⁴

Conversely, in a cohort study of 391 patients who underwent surgery for IBD, obesity did not worsen postoperative complication rates in IBD patients.⁸⁵

Colon and Rectum

Constipation

The association between obesity and constipation is controversial, with a higher prevalence of constipation in obese people in a community-based epidemiologic study in the United States³; this was not reproduced in other large cohort studies.⁵⁹ In children, constipation, but not constipation-predominant irritable bowel syndrome, is more common in obese individuals.^86,87

Diverticular Disease

Obesity is associated with a higher risk of developing diverticulosis,⁸⁸ as well as an increased number of diverticuli⁸⁹ and increased diverticular bleeding and recurrent diverticulitis compared with normal-weight individuals.⁹⁰

Colonic Polyps

There are 3 main types of polyps in the colon: adenomatous, serrated, and hyperplastic polyps. Adenomas and serrated polyps predispose to colon cancer. Several studies have documented an increased prevalence of adenomatous polyps with the highest quartiles of BMI (OR, 2.1; 95% CI, 1.4–2.3) compared with the lowest quartile of BMI.⁹¹ This association was stronger in women (OR, 4.42; 95% CI, 1.53–12.78) than in men (OR, 1.26; 95% CI, 0.52–3.07). Weight gain is another risk factor for adenomas (OR, 2.30; 95% CI, 1.25–4.22).⁹² The association of a higher BMI and colonic adenomatous polyps has been validated in other cohorts and populations.^93–97 Similarly, obesity is associated with an increased risk of adenoma recurrence,⁹⁸ risk of sessile serrated polyps of the colon (OR, 2.57; 95% CI, 1.75– 4.90), and with serrated polyps larger than 1 cm (OR, 3.96; 95% CI, 1.27–12.36).⁹⁹

Colorectal Cancer

Colorectal cancer is the fourth most common cancer in the United States and the second leading cause of cancer deaths. Multiple meta-analyses in more than 70,000 cases of colon cancer show obesity as a risk factor.^100–103 For every increase in BMI of 5 kg/m², the risk of colon cancer increases by 18%.¹⁰¹ The association of obesity and colorectal cancer is stronger in men (RR, 1.24; 95% CI, 1.20– 1.28) than in women (RR, 1.09; 95% CI, 1.04–1.12).¹⁰⁰ The risk of colorectal cancer increases with a higher waist circumference (RR, 1.33; 95% CI, 1.19–1.49 for men; RR, 1.16; 95% CI, 1.09–1.23 for women).¹⁰³ The relationship of obesity, abdominal adiposity, and colon cancer is likely to be multifactorial owing to changes in leptin, adiponectin, the microbiome, secondary bile acids, and insulin resistance, and this relationship has been reviewed extensively elsewhere.¹⁰⁴

Clostridium Difficile Infection

In a retrospective case control study of 6800 hospitalized patients in Israel, 148 cases with C difficile infection were compared with 148 hospitalized controls. A high BMI value (OR, 1.196 per 1-unit increase in kg/m²; 95% CI, 1.12–1.27) was associated significantly with C difficile infection.¹⁰⁵ More studies are needed to verify this relationship.

Anal Canal and Pelvic Floor

Dyssynergic Defecation

Female patients have a higher likelihood of experiencing constipation secondary to pelvic floor disorders (83% in 1 large series of 390 female patients)¹⁰⁶ and, particularly, constipation associated with descending perineum syndrome, ¹⁰⁷ which usually is associated with multiparity and is observed almost exclusively in females. In a representative Swedish cohort of 1001 people in the general population, obesity was associated with incomplete rectal evacuation (OR, 1.64; 95% CI, 1.09–2.47), adjusting for age, sex, and education.⁶¹

Fecal Incontinence

Fecal incontinence, defined as the inability to control bowel movements, producing undesired leakage of stool from the rectum, has been associated with a higher BMI, but the association has been weak or not statistically significant. Studies have reported a range of results from no association to up to 69% of obese subjects having fecal incontinence. ^37,108 Obesity is emerging as a potentially modifiable risk factor in fecal incontinence,¹⁰⁹ and obesity (BMI ≥ 30 kg/m²) was a significant risk factor for increased functional difficulty/dependence as a result of fecal incontinence.¹¹⁰

Liver

The liver is central to nutrient regulation, and frequently is involved in obesity-associated NAFLD. NAFLD has surpassed other chronic liver diseases to become the most prevalent chronic liver disease in the United States and the most frequent cause of increased transaminase levels. It affects approximately 30% of the population; with a worldwide prevalence range of 5%–46%.^111–113 Population-based studies have shown a positive correlation between body mass index and NAFLD, suggesting shared pathogenic mechanisms. Furthermore, at-risk populations, such as those with diabetes mellitus or co-existent metabolic syndrome, have significantly higher rates of NAFLD, reaching up to 70%.¹¹¹ Obesity is also a risk factor for hepatocellular carcinoma; this risk may occur independently of NAFLD.¹¹⁴

Nonalcoholic Fatty Liver Disease

NAFLD is a heterogeneous disorder, encompassing 2 broad primarily histologic categories, isolated steatosis and nonalcoholic steatohepatitis (NASH) (Figure 3).¹¹¹ Both show macrovesicular steatosis in more than 5% of hepatocytes; however, NASH shows additional histologic characteristics of liver injury including ballooned hepatocytes, inflammatory foci, and fibrosis. NAFLD pathogenesis involves a complex interplay of nutritional overload, metabolic, microbial, and genetic factors. These diverse pathogenetic factors partly explain the heterogeneous nature of the disease. Fat accumulation in the liver results from caloric overload and the ectopic accumulation of triglycerides in the liver (Figure 3), although other lipid species including sphingolipids and phospholipids also accumulate in hepatocytes. Adipose tissue lipolysis supplies the majority of free fatty acids that subsequently are esterified to form hepatic triglycerides; however, de novo lipogenesis and dietary fat also contribute (Figure 3).^115–117 The abnormal accumulation of lipids contributes to lipotoxicity, broadly defined as the deleterious consequences of accumulated lipids. Target organ damage in this manner leads to insulin resistance, which further exacerbates hepatic steatosis. Radiotracer studies also have shown an increase in hepatic de novo lipogenesis in NAFLD.¹¹⁵ Furthermore, hepatic fatty acid oxidation and very low density lipoprotein secretion rates are increased.^117,118 These may represent compensatory mechanisms for the increase in fatty acid influx to the liver.

Figure 3.

Nonalcoholic fatty liver disease: disease spectrum and sources of triglyceride fatty acids. (A) The spectrum of hepatic manifestations of NAFLD includes isolated steatosis and NASH. Isolated steatosis is characterized by hepatic fat accumulation without any additional pathologic findings or the risk of progression to cirrhosis. NASH is characterized by hepatic inflammation, fibrosis, and risk of progression to cirrhosis. Increased HCC risk is associated with both; although the risk is much greater in NASH. (B) Sources of fatty acids and their respective contribution to hepatic triglycerides are shown. The majority of fatty acids are derived from circulating NEFA, which are derived from adipose tissue lipolysis. Adipose tissue lipolysis is enhanced in insulin resistance. Rates of hepatic de novo lipogenesis are up-regulated in NAFLD. This is the second largest source; newly synthesized fatty acids account for approximately one fourth of the fatty acids in hepatic triglycerides. The remainder is derived from dietary fats. NEFA, nonesterified fatty acid.

Patients with NAFLD are at risk for progressive fibrosis and eventual cirrhosis. It is estimated that 20% of NASH patients and less than 5% of patients with isolated steatosis will progress to cirrhosis.¹¹¹ Several studies have sought to identify factors that increase this risk for fibrosis progression. Age, the degree of inflammation, and stage of fibrosis at diagnosis confer fibrosis progression risk.¹¹⁹ In addition, in a meta-analysis, arterial hypertension and a low aspartate aminotransferase:alanine aminotransferase ratio conferred increased risk for fibrosis progression.¹²⁰ Interestingly, it has been calculated that subjects with NASH may progress 1 stage of fibrosis in approximately 7 years; whereas subjects with isolated steatosis may progress 1 stage of fibrosis in approximately 14 years. It is likely that patients diagnosed with isolated steatosis who progressed in fact had low-grade inflammation in the liver, which was insufficient to fulfill NASH diagnostic criteria. Furthermore, a small subset of patients has rapid progression of fibrosis; however, the true estimate of the numbers at risk for rapid progression and the risk factors for rapid progression are incompletely understood. Finally, a substantial proportion (30%–60%) of patients with biopsy-proven NASH can have normal serum transaminase levels, highlighting the need for the development of better disease biomarkers and for a high level of clinical vigilance to detect NAFLD in high-risk populations.

NAFLD confers increased risk of cardiovascular mortality and hepatocellular carcinoma; NASH confers increased risk of liver-related mortality. Therapeutic options for NAFLD target either obesity or hepatic inflammation and fibrosis. The first category includes weight loss through caloric reduction and physical activity and bariatric procedures.^121–123 In NASH subjects with paired liver biopsies before and after weight loss, histologic improvements in steatosis, hepatocyte ballooning, inflammation, and fibrosis were observed at a weight loss of ≥5%; although histologic improvement correlated with the degree of weight loss.¹²¹ The greatest fibrosis resolution occurred in those with 10% or greater body weight loss. Interestingly, all patients who lost 10% or greater of body weight showed histologic improvement. Efficacy of bariatric surgery in improving the histologic features of NASH, including reversal of steatosis, diminished hepatocyte apoptosis, and reversing hepatic fibrosis has been shown in several studies.^124–126 NASH cirrhosis remains the third leading indication for liver transplant and is projected to be the most frequent indication in the near future.¹²⁷

The role of bile acid signaling in improving NASH after bariatric surgery is an area of active investigation. Pioglitazone, vitamin E, and obeticholic acid have shown improvements in liver necroinflammation; however, none has received Food and Drug Administration approval for the treatment of NASH.^128,129 Several other pharmacologic agents targeting steatosis, inflammation, or fibrosis pathways are in clinical trials for the treatment of NASH; more than 50 open phase 2 and 3 clinical trials currently are registered at www.clinicaltrials.gov. A partial list of these agents is provided in Table 2.

Table 2.

Partial List of Drugs in Phase 2 or 3 Clinical Trials for NASH

Drug target	Drug name
Thiazolidinedione	Pioglitazone
Glucagon-like peptide-1 receptor agonists	Liraglutide
PPAR α and δ agonist	Elafibranor
PPAR α and δ agonist	GFT505
Farnesoid X receptor agonist	Obeticholic acid
Farnesoid X receptor agonist	GS-9674
Acetyl-CoA carboxylase inhibitor	GS-0976
Fatty acid synthase inhibitor	3-V Bioscience-2640
Caspase inhibitor	IDN-6556
Immune modulating and antifibrotic	JKB-121
Antioxidant, multiple targets	Vitamin E
Leukotriene-receptor antagonism, inhibition of phosphodiesterases, and inhibition of 5-lipoxygenase	MN-001
CCR2 and CCR5 inhibitor	Cenicriviroc
Anti-LPS antibodies and adjuvants	IMM-124E
Selective aldosterone-receptor antagonist	MT-3995
Liver-directed thyroid hormone receptor-β agonist	MGL-3196
Apical sodium-dependent bile acid transporter	Volixibat

Hepatocellular Carcinoma

Obesity and NAFLD confer increased risk for hepatocellular carcinoma (HCC), the most common primary liver cancer.¹³⁰ A meta-analysis of 11 cohort studies determined that overweight imparted an RR of HCC of 1.17 (95% CI, 1.02–1.34), and obesity an RR of 1.89 (95% CI, 1.51–2.36). In addition, individual studies have shown a greater risk of dying from HCC (RR: 4.5 in men, 1.68 in women) with a BMI of ≥35 kg/m^2.114,130 In a more recent nested case–control study, serum markers of inflammation, including C-reactive protein, interleukin 6, C-peptide, and non–high-molecular-weight adiponectin, were associated with a higher incidence rate ratio of HCC, independent of obesity, likely reflecting the procarcinogenic effects of chronic inflammation.¹³¹ In an analysis of the Surveillance, Epidemiology, and End Results registry in the United States from 2004 to 2009, NAFLD was the third most common underlying cause of HCC after chronic hepatitis C and alcoholic liver disease. This study showed a 9% annual increase in NAFLD-related HCC.¹³² Interestingly however, some studies have shown that some NAFLD patients develop HCC in the absence of cirrhosis, whereas other systematic analyses have shown no increased HCC risk in the absence of cirrhosis.^{111,133–135} The discrepancies likely are owing to patients with isolated steatosis, NASH, and NASH cirrhosis being included in cohorts. Although NASH imparts HCC risk, the absolute risk remains lower than viral hepatitis B and C.¹³⁵ Furthermore, co-existent obesity confers poor prognosis, impacting response to therapy as well as decreased disease-free survival and patient survival.^136,137 NASH confers a significant health care burden; models have predicted annual direct medical costs of $103 billion.¹³⁸ Given the significant health care costs associated with HCC and the increasing rates of NASH-related HCC, this is a considerable public health problem.¹³⁹

Recently, several experimental studies, using differing dietary models with varying degrees of fidelity to human NASH, have added to our understanding of HCC pathogenesis in NASH.^140–143 Notably, several of these studies have linked the immune system with cancer risk. The loss of CD4+ T lymphocytes is associated with increased HCC, implicating the failure of immune surveillance in 1 dietary model of NASH.¹⁴⁰ However, in another study, T helper 17 lymphocytes, which are a subset of CD4+ T lymphocytes, played a pivotal role in insulin resistance, NASH, and HCC.¹⁴² Tumor necrosis factor–driven inflammatory signaling also increased HCC in a mouse model of endoplasmic reticulum stress and fatty liver.¹⁴¹ In human studies, PNPLA3 rs738409 C>G polymorphism imparted increased risk for HCC in NAFLD, although the magnitude of this risk and mechanism remain to be determined.¹⁴⁴ Thus, experimental studies have linked perturbations of both the adaptive and innate immune systems to increased risk of HCC in NASH. The driver mutations in NASH–HCC remain to be defined.

Gallbladder

Obesity has been well recognized for its strong association with gallstone diseases.^145,146 Subjects with obesity have a higher incidence of cholelithiasis, cholecystitis, and cholesterolosis when compared with lean controls.¹⁴⁷ A meta-analysis showed that the risk for gallbladder disease in men was 1.63 (95% CI, 1.42–1.88) for overweight and 2.51 (95% CI, 2.16–2.91) for obesity; in women, the RR was 1.44 (95% CI, 1.05–1.98) for overweight and 2.32 (95% CI, 1.17–4.57) for obesity.¹⁴⁸ Abdominal circumference is also a risk factor for gallbladder diseases, independent of BMI.^149,150 These associations have been attributed to abdominal adiposity, hyperinsulinemia, insulin resistance, hyperleptinemia, hyperlipidemia, and gallbladder dysmotility.^{145,151–153}

Pancreas

Obesity and fat infiltration of the pancreas play a significant role in the endocrine pancreatic dysfunction that leads to the development of type 2 diabetes mellitus (further details are not pertinent for this review).¹⁵⁴ Obesity also has been associated with pancreatitis and pancreatic cancer.

Acute Pancreatitis

Acute pancreatitis is defined as inflammation of the pancreas and can range from mild to fulminant, with mortality up to 20% in severe necrotizing pancreatitis.^155,156 Obesity is associated with more severe acute pancreatitis. In a meta-analysis, obese subjects had an increased risk of developing severe acute pancreatitis (RR, 2.20; 95% CI, 1.82–2.66), a higher risk of local (RR, 2.68; 95% CI, 2.09– 3.43) and systemic complications (RR, 2.14; 95% CI, 1.42– 3.21), and a higher risk of in-hospital mortality (RR, 2.59; 95% CI, 1.66–4.03) when compared with lean subjects.¹⁵⁶ These associations have been attributed to low-grade chronic inflammation and low levels of adiponectin.^157–159

Pancreatic Cancer

Pancreatic cancer is the ninth most common cancer worldwide.¹⁶⁰ Multiple meta-analyses have reported an association between BMI or abdominal obesity and occurrence of adenocarcinoma of the pancreas. On meta-analysis, there is a 10% increased risk in women and a 13% increased risk in men for every 5 kg/m/m² higher BMI (95% CI, 1.04–1.22).¹⁶¹ In addition, for every extra 10 cm of waist circumference, there was an 11% increased risk of pancreatic cancer (RR, 1.11; 95% CI, 1.05–1.18).¹⁶¹

Summary

The increased prevalence of gastrointestinal conditions in the general population may be related to the increased prevalence of obesity in Western countries. Thus, it is important to recognize the role of higher BMI and, particularly, increased abdominal adiposity, in the development of gastrointestinal morbidity and, therefore, to measure BMI and waist circumference in patients presenting with gastrointestinal complaints or abnormal liver function.

Abbreviations used in this paper

BMI

body mass index

CI

confidence interval

GERD

gastroesophageal reflux disease

HCC

hepatocellular carcinoma

LES

lower esophageal sphincter

NAFLD

nonalcoholic fatty liver disease

NASH

nonalcoholic steatohepatitis

OR

odds ratio

RR

relative risk