Epidemiologic data have demonstrated that obesity is an important risk factor for the development of gastroesophageal reflux disease (GERD). There is also accumulating data that obesity is associated with complications related to longstanding reflux such as erosive esophagitis, Barrett esophagus, and esophageal adenocarcinoma. Central obesity, rather than body mass index, appears to be more closely associated with these complications. Surgical data are confounded by the concomitant repair of prevalent hiatal hernias in many patients.
Key points
- •
The prevalence of obesity and gastroesophageal reflux disease (GERD) has increased substantially in the past 30 years.
- •
Central adiposity, measured as the waist-to-hip ratio, is more closely associated with GERD complications than measures of overall obesity such as body mass index.
- •
Visceral adipose tissue is metabolically active and secretes adipokines along with inflammatory cytokines that may predispose to complications of GERD such as Barrett esophagus and esophageal carcinoma.
Introduction
Disease Description
The typical manifestations of gastroesophageal reflux disease (GERD) are heartburn and/or regurgitation. GERD can be further classified into erosive GERD and nonerosive GERD based on endoscopic appearance of esophageal mucosa. The term “atypical GERD” is used in situations where the predominant symptoms are extraesophageal such as cough, laryngitis, and asthma. GERD is a common disorder with a prevalence of approximately 20% in the United States. The recognized sequelae of GERD include Barrett esophagus (BE) and esophageal adenocarcinoma. Obesity, defined as a body mass index (BMI) greater than or equal to 30, is common in the Western world and is increasing in other parts of the world, particularly Asia. Epidemiologic data demonstrate that overall obesity (typically measured as BMI kg/m 2 ) is a risk factor for both GERD and esophageal adenocarcinoma. There is evidence that central abdominal obesity, as opposed to an elevated BMI, is the most important factor associated with BE ( Table 1 ).
| Obesity |
| Caffeine intake |
| Spicy foods |
| Tobacco |
| Pregnancy |
| Alcohol |
| Recumbent position |
| Connective tissue disorders |
| Hiatal hernia |
| Decreased LES tone |
| Zollinger-Ellison syndrome |
| Post-prandial supination |
Introduction
Disease Description
The typical manifestations of gastroesophageal reflux disease (GERD) are heartburn and/or regurgitation. GERD can be further classified into erosive GERD and nonerosive GERD based on endoscopic appearance of esophageal mucosa. The term “atypical GERD” is used in situations where the predominant symptoms are extraesophageal such as cough, laryngitis, and asthma. GERD is a common disorder with a prevalence of approximately 20% in the United States. The recognized sequelae of GERD include Barrett esophagus (BE) and esophageal adenocarcinoma. Obesity, defined as a body mass index (BMI) greater than or equal to 30, is common in the Western world and is increasing in other parts of the world, particularly Asia. Epidemiologic data demonstrate that overall obesity (typically measured as BMI kg/m 2 ) is a risk factor for both GERD and esophageal adenocarcinoma. There is evidence that central abdominal obesity, as opposed to an elevated BMI, is the most important factor associated with BE ( Table 1 ).
| Obesity |
| Caffeine intake |
| Spicy foods |
| Tobacco |
| Pregnancy |
| Alcohol |
| Recumbent position |
| Connective tissue disorders |
| Hiatal hernia |
| Decreased LES tone |
| Zollinger-Ellison syndrome |
| Post-prandial supination |
Prevalence/Incidence
A systematic review estimated the prevalence of GERD in the United States at 18.1% to 27.8%. El-Serag and others in their systematic review divided studies on the prevalence of GERD into 4 temporal categories. Relative to pre-1995, the rate ratio for GERD prevalence was 1.45 for the period 1995 to 1999, 1.46 for 2000 to 2004, and 1.51 for 2005 to 2009. Obesity is an even more common health issue in the United States. Data from the 2009–2010 National Health and Examination Survey estimate a prevalence of 35.5% for men and 35.8% for women, which is not significantly changed compared with the period 2003 to 2008. Previous trends showed that the prevalence of obesity was increasing in America but the trend may be beginning to level.
Cross-sectional epidemiologic studies have demonstrated a higher prevalence of GERD in obese individuals compared with the nonobese. Jacobsen and colleagues used a supplemental GERD questionnaire added to the Nurses’ Health Study to show that subjects who reported at least weekly symptoms had a near linear increase in the adjusted odds ratio (OR) for reflux symptoms for each BMI strata. A similar link was seen in the results from the 80,110 insurance members from the Kaiser Permanente MultiPhasic Health Check-Up cohort. The association between BMI and GERD was stronger among whites compared with black members, with ORs of 1.58 and 1.33, respectively. When controlling for abdominal diameter the ORs were 1.39 and 1.15, respectively.
Smaller studies have confirmed the link between obesity and GERD. El-Serag and others interviewed 453 hospital employees and found that 26% had weekly heartburn or regurgitation symptoms. Subjects were offered endoscopy and 196 agreed, and they found that increasing levels of obesity were associated with a greater likelihood of GERD and esophagitis. The proportion of subjects with GERD symptoms were 23.3%, 26.7%, and 50% for BMI groups <25, 25–30, and >30, respectively. Prevalence rates for erosive esophagitis (EE) were 12.5%, 29.8%, and 26.9%. Two small cohort studies from Olmstead County, MN have also evaluated the relationship between obesity and GERD. The first study identified obesity as a risk factor for the initial development of GERD as well as the persistence of symptoms. The second study found that BMI was associated with GERD (OR = 1.9) independent of diet and energy expenditure.
The effect of weight change on GERD symptoms has been studied. Jacobson and colleagues studied select individuals from the Nurses’ Health Study and found that an increase of BMI by more than 3.5 kg/m 2 when compared with no weight change was associated with an increase risk of frequent symptoms of reflux.
World-wide incidence rates
The prevalence of obesity is somewhat lower outside of the United States. The European Prospective Investigation into Cancer and Nutrition study estimated the prevalence of obesity was 17% in 2005, which increased from 13% in 1998. Based on a systematic review, the prevalence rate of GERD in Europe was estimated to be 15% for the period 2005 to 2009. Similar to the trend seen in the United States, this prevalence rate is significantly higher than the rate before 1995. The epidemiologic relationship between obesity and GERD has been observed in Europe as well. The German National Health Interview and Examination Survey found the OR for GERD to be 1.8 for overweight and 2.6 for obese individuals. In England, the Bristol Helicobacter Project found that obese individuals had an OR of 2.91 for heartburn and an OR of 2.23 for regurgitation. A telephone survey in Spain of 2500 subjects revealed that obese individuals had an OR of 1.74 for GERD symptoms. It was also noted that patients with GERD symptoms for more than 10 years were more likely to be obese (OR = 1.92). This group also found that a weight gain of more than 5 kg in the past year demonstrated a 2.7-fold higher risk of new GERD symptoms. In Norway, Nilsson and colleagues conducted nationwide surveys during the periods 1984 to 1986 (N = 74,599) and 1995 to 1997 (N = 65,363). They found that for severely obese men (BMI>35 kg/m 2 ) the OR for GERD was 3.3, whereas the OR for severely obese women was 6.3. A link showing an association between estrogen levels and GERD was observed. Premenopausal women and those who were post-menopausal but taking hormone replacement therapy were at an increased risk for GERD relative to untreated post-menopausal women.
A relationship between obesity and GERD has been seen in Asia. Kang and colleagues studied 2457 subjects who underwent upper endoscopy in Korea. They found a relationship between higher strata of BMI and the presence of EE. In Shanghai, a nested case-control study found an association between obesity and dwelling in an urban environment with GERD.
Studies that have failed to identify a relationship between GERD and obesity have also been reported. A study of 820 subjects from Sweden showed that those who had been overweight or obese had an adjusted OR of 0.99 for GERD. They also found no association between obesity and severity of reflux symptoms. Similarly, a prospective cohort study in Olmsted, MN of 607 individuals surveyed more than 10.5 years did not find an association with GERD symptoms and weight loss of greater than 10 pounds.
In summary, the preponderance of population-based studies supports the association between obesity and GERD reflux. The association has been demonstrated in the United States where obesity rates are the highest and has also been seen in Europe and Eastern Asia ( Fig. 1 ). Shortcomings of these studies are that they primarily relied on self-reported height and weight to calculate BMI and did not look specifically at abdominal obesity. There appears to be a dose response as well with increasing levels of obesity associated with higher prevalence rates. Weight loss has not been consistently associated with amelioration of symptoms at a population level.
Clinical correlation
Complications of GERD
Long-term complications of GERD such as EE, BE, and esophageal adenocarcinoma have been associated with obesity. In a large endoscopic study, El-Serag reported that relative to those with no erosions, those with EE were more likely to be overweight or obese. A similar association was seen in Korea where Lee and colleagues did an endoscopy study in Korea studying 3000 participants. They found that obese individuals compared to normal weight subjects had an OR of 3.3 for EE. A meta-analysis by Hampel and colleagues confirmed the association with increasing levels of obesity and esophageal mucosal injury.
Associations of BE and obesity have been demonstrated by Stein and colleagues who established that for each 5-unit increase in BMI, the risk of BE increased by 35%. Abdominal obesity (“central obesity”) has been shown to be a more specific risk factor for BE. Corley and colleagues, using data from the Kaiser Permanente database, found that a larger abdominal circumference (measured at the iliac crest with the abdomen relaxed), independent of BMI, was associated with BE. Edelstein and colleagues found that for individuals in the highest category of waist-to-hip ratio the adjusted OR for BE was 1.9 and 4.1 for long-segment BE. Rubenstein and colleagues found that abdominal obesity as measured by waist circumference increased the risk of EE and BE, whereas gluteofemoral obesity was protective. Finally, El-Serag used abdominal computed tomographic imaging to demonstrate that greater amounts of visceral adipose tissue but not subcutaneous adipose tissue conferred a significantly increased risk for BE.
Not all studies have demonstrated an association between obesity and BE. An Australian study found that BMI was not an independent risk factor for BE. A study in Canada by Veugelers and colleagues also did not show an association between obesity and BE. They did, however, find an association of BMI with esophageal adenocarcinoma.
The incidence of esophageal adenocarcinoma has been rising in the United States. From 1975 to 2001, the incidence of esophageal adenocarcinoma has increased approximately 6-fold. There are several studies that have examined the relationship between obesity and esophageal adenocarcinoma. In 1998, a National Cancer Institute study by Chow and colleagues found an association between increasing strata of BMI and esophageal cancer, specifically among younger nonsmoking individuals. A Swedish study identified obesity with an OR of 16.2 for the development of adenocarcinoma compared with the leanest individuals (BMI<22 kg/m 2 ). A recently pooled analysis from 12 world-wide epidemiologic studies showed that patients with a BMI greater than or equal to 40 compared with nonoverweight patients had an OR of 4.76 for esophageal adenocarcinoma. Engel and colleagues found that the population attributable risk (proportion of occurrences in the population that may be preventable if a factor were totally eliminated) for body weight (using BMI<23.1 as the control group) increased steadily from 5.4% (BMI = 23.2–25.1) to 21.3% (BMI = 27.3–40.1).
Pathophysiology
Several physiologic abnormalities that could lead to prolonged esophageal acid exposure have been found to occur more frequently in obese compared with normal weight individuals. Many of these disturbances have been identified in the severely obese (BMI>35) before bariatric surgery and may not apply to those with lesser degrees of obesity. For example, esophageal manometry before bariatric surgery has revealed that many patients have a motility disorder. In a study of 345 patients, 25.6% of patients had abnormal manometry. The most common abnormal findings were nutcracker esophagus and nonspecific motility disorder. Other studies in severely obese subjects revealed similar findings, with nonspecific motility disorder, nutcracker esophagus, and hypotensive lower esophageal sphincter (LES) as the most common manometric abnormalities. Interestingly, most of these patients were asymptomatic.
Studies looking specifically at prebariatric surgical patients with symptoms of GERD excluding asymptomatic patients have also been reported. Hong and colleagues studied 61 patients and 32.8% had abnormal manometry, most commonly nonspecific esophageal motor disorder. Another study using manometry, 24-hour pH measurement, and impedance grouped patients into 3 groups. Group 1 (control group) had 10 normal-weight asymptomatic subjects, group 2 had 22 nonobese GERD patients, and group 3 consisted of 22 obese GERD patients. All group 1 patients had normal esophageal acid exposure, motility, and bolus transit. From group 2 there were 5 patients with abnormal manometry, 2 with ineffective esophageal motility, 2 with nutcracker esophagus, and 1 with hypertensive LES (>50 mm Hg). Group 3 also had 5 patients with abnormal manometry, including 2 with ineffective esophageal motility, 2 with nutcracker esophagus, and 1 with diffuse esophageal spasm. The only difference between the obese and nonobese GERD subjects was that obese patients had fewer episodes of complete bolus transit (as measured by impedance) compared with the nonobese, 66% versus 88% P = .01.
A hypotensive LES, defined as basal pressure less than 10 mm Hg, is clearly a predisposing factor for GERD. Studies examining the relationship between LES pressure and BMI have been performed, although the results are inconsistent. One study examined 64 consecutive patients and divided subjects into 3 groups. Group A had 23 subjects with a BMI less than 25, group B had 25 subjects with a BMI between 25 and 30, and group C had 16 subjects with a BMI >30. The investigators observed a strong inverse relationship between BMI and LES pressure ( P <.001).
Transient relaxations of the lower esophageal sphincter (TRLES) have been observed to be more common in patients with obesity. The main stimulus for TRLES is gastric distension, particularly in the fundus. A study by Wu and colleagues divided subjects into 3 groups, 28 obese, 28 overweight, and 28 normal subjects. These individuals were studied with upper endoscopy, manometry, and pH recordings. The overweight and obese groups were found to have significantly higher rates of TRLES during the 2-hour postprandial period (obese group 17.3, overweight 3.8, normal 2.1 episodes per hour; P <.001). Total distal esophageal acid exposure as well as the proportion of TRLES accompanied by acid reflux was also greater in the obese and overweight groups.
The presence of a hiatal hernia has also been associated with obesity. Suter and colleagues studied morbidly obese patients with history of reflux symptoms with upper endoscopy, 24-hour pH monitoring, and manometry. They observed that of 345 subjects approximately half had a hiatal hernia. Furthermore, patients with a hiatal hernia were more likely to have esophagitis compared with those without a hiatal hernia. Pandolfino and colleagues subsequently reported that obese patients have a pressure gradient along the esophagogastric junction that supported the development of a hiatal hernia.
Abdominal obesity likely increases intra-abdominal pressure due to transmission of gravitational force of the adipose tissue to the abdominal cavity. Lambert and colleagues studied morbidly obese patients with a urinary catheter as a surrogate for intra-abdominal pressure and found that obese patients compared with nonobese patients had higher intra-abdominal pressures. This relationship between obesity and elevated intra-abdominal/intragastric pressures has been confirmed by others with use of intragastric manometry.
Gastric volume and motor abnormalities have been proposed as other mechanisms for GERD in obese individuals. Multiple studies have found that the capacitance of gastric contents in obese subjects is larger compared with lean individuals. Whether the greater volume of contents leads to increased GERD is not known. It has also been theorized that obese individuals may have delayed gastric emptying due to neuronal or humoral mechanisms. Buchholz and colleagues using standardized scintigraphic gastric emptying studies showed no difference in gastric emptying in obese and nonobese patients. Retention percentages at 1 hour and 4 hours were 48% and 47% and 1.7% and 1.1%, respectively.
The link between obesity and esophageal neoplasia may be via altered secretion of adipokines such as adiponectin and leptin. Adiponectin is a protein that has antiinflammatory and immunomodulatory functions and stimulates apoptosis. Secretion of adiponectin decreases with obesity. Rubenstein and colleagues found an inverse association between plasma adiponectin levels and the presence of BE in a case-control study. In a separate study, this group found that levels of the low molecular weight subtype of adiponectin were inversely associated with the risk of BE. In contrast to the inverse relationship seen between obesity and adiponectin, leptin levels correlate directly with obesity. Leptin is secreted by adipocytes and gastric chief cells and has been shown to have mitogenic properties and induce proliferation in several human cell lines including esophageal cancer cells. Kendall and colleagues found that male subjects with BE had higher levels of plasma leptin relative to healthy controls. Those with a leptin level in the highest quartile had an OR of 3.3 for the presence of BE. The link between BE and central obesity (rather than BMI) may be partially explained by the fact that leptin reaches very high values in central obesity ( Figs. 2 and 3 ).
