Free access
Reviews
2 August 2005

Meta-Analysis: Obesity and the Risk for Gastroesophageal Reflux Disease and Its ComplicationsFREE

Publication: Annals of Internal Medicine
Volume 143, Number 3

Abstract

Background:

The association of body mass index and gastroesophageal reflux disease (GERD), including its complications (esophagitis, Barrett esophagus, and esophageal adenocarcinoma), is unclear.

Purpose:

To conduct a systematic review and meta-analysis to estimate the magnitude and determinants of an association between obesity and GERD symptoms, erosive esophagitis, Barrett esophagus, and adenocarcinoma of the esophagus and of the gastric cardia.

Data Sources:

MEDLINE search between 1966 and October 2004 for published full studies.

Study Selection:

Studies that provided risk estimates and met criteria on defining exposure and reporting outcomes and sample size.

Data Extraction:

Two investigators independently performed standardized search and data abstraction. Unadjusted and adjusted odds ratios for individual outcomes were obtained or calculated for each study and were pooled by using a random-effects model.

Data Synthesis:

Nine studies examined the association of body mass index (BMI) with GERD symptoms. Six of these studies found statistically significant associations. Six of 7 studies found significant associations of BMI with erosive esophagitis, 6 of 7 found significant associations with esophageal adenocarcinoma, and 4 of 6 found significant associations with gastric cardia adenocarcinoma. In data from 8 studies, there was a trend toward a dose–response relationship with an increase in the pooled adjusted odds ratios for GERD symptoms of 1.43 (95% CI, 1.158 to 1.774) for BMI of 25 kg/m2 to 30 kg/m2 and 1.94 (CI, 1.468 to 2.566) for BMI greater than 30 kg/m2. Similarly, the pooled adjusted odds ratios for esophageal adenocarcinoma for BMI of 25 kg/m2 to 30 kg/m2 and BMI greater than 30 kg/m2 were 1.52 (CI, 1.147 to 2.009) and 2.78 (CI, 1.850 to 4.164), respectively.

Limitations:

Heterogeneity in the findings was present, although it was mostly in the magnitude of statistically significant positive associations. No studies in this review examined the association between Barrett esophagus and obesity.

Conclusion:

Obesity is associated with a statistically significant increase in the risk for GERD symptoms, erosive esophagitis, and esophageal adenocarcinoma. The risk for these disorders seems to progressively increase with increasing weight.
Frequent symptoms of gastroesophageal reflux disease (GERD) affect between 10% and 20% of adults in the United States. The prevalence of GERD-related complications, including erosive esophagitis, Barrett esophagus, and esophageal adenocarcinoma, has been steadily increasing in the United States and western Europe. For instance, hospitalizations with GERD among veterans increased 10-fold from the 1970s to the 1990s (1). Similarly, the incidence of esophageal adenocarcinoma increased 4-fold over the past 20 years (2-5). The reasons for the increase in GERD and its complications are not known. Changes in diet, prescription medication use, smoking, and alcohol intake and the declining prevalence of Helicobacter pylori infection have been proposed (6-11). Studies have also hypothesized that the increasing trend of obesity in western populations has paralleled the increase in esophageal adenocarcinoma and may be an important factor in this change (12-14).
The notion of obesity as a cause of GERD is biologically plausible. Obesity has been associated with increased intra-abdominal pressures (15), impaired gastric emptying (16), decreased lower esophageal sphincter pressure, and increased frequency of transient sphincter relaxation (17, 18), thus leading to increased esophageal acid exposure. However, the epidemiologic evidence linking obesity to GERD and its complications has not been critically reviewed. We aimed to evaluate, quantify, and summarize the association of obesity to GERD and its complications.

Methods

Literature Search

Two investigators independently searched the published English-language literature (through October 2004) by using MEDLINE. Search terms included obesity or body mass or anthroprometry searched with reflux or heartburn, (o)esophagitis, Barrett's or Barretts, and (o)esophageal cancer or (o)esophageal adenocarcinoma. We performed a recursive hand search of cited bibliographies to increase completeness.

Study Selection Criteria

The following inclusion criteria had to be fulfilled: 1) cross-sectional, case–control, or cohort study that permitted assessment of a causal association between overweight or obesity and esophageal disease; 2) clear definition of obesity as defined by a body mass index (BMI) in kg/m2 or height-to-weight ratio; and 3) well-defined outcome of interest that included GERD symptoms defined by using validated symptom score, esophageal erosions defined by endoscopy, and Barrett esophagus or esophageal adenocarcinoma validated by pathology review. We excluded case reports and case series, studies with fewer than 50 case-patients, and studies that did not report risk estimates or raw data to allow independent calculation of these estimates. If a study met the selection criteria except for failure to report risk estimates, we contacted the study authors in an attempt to obtain these data.

Statistical Analysis

We abstracted or calculated odds ratios for cross-sectional and case–control studies and risk ratios for cohort studies as the risk estimates for associations between obesity and each outcome of interest. We pooled the results in 2 different ways. All studies either presented or had sufficient information to allow the calculation of unadjusted odds ratios estimates (and 95% CIs). We subsequently pooled these unadjusted estimates. Because of the inconsistency in reporting adjusted risk estimates, we estimated the degree of confounding produced by uncontrolled variables in studies that did not report adjusted risk estimates by using a factorization of the relative risk by external adjustment (19). Subsequently, we pooled the adjusted estimates. In addition, we displayed the adjusted odds ratios in a tabular form, indicating the variables that were adjusted for. We conducted pooling by using a random-effects model. We examined the weighted pooled risk estimates for 2 recognized categories: overweight and obesity (20, 21). We assessed heterogeneity with a chi-square test and by visual inspection of Forest plots. We performed statistical analyses with Comprehensive Meta-Analysis (Biostat, Englewood, New Jersey) and Stata 8.0 (Stata Corp., College Station, Texas). We examined funnel plots to evaluate for publication bias.

Results

GERD Symptoms

We identified 370 potentially relevant titles. Of these, 9 studies satisfied our inclusion and exclusion criteria (22-30). All studies were cross-sectional examinations of GERD symptoms in randomly selected samples of the general population (Table 1). Only 3 studies (Nilsson and colleagues [(29], Murray and colleagues [28], and Lagergren and colleagues [26]) were conducted for the primary purpose of evaluating the relationship between obesity and GERD.
Table 1. Characteristics of 9 Studies of Body Mass Index and Symptoms of Gastroesophageal Reflux Disease
Table 1. Characteristics of 9 Studies of Body Mass Index and Symptoms of Gastroesophageal Reflux Disease
Of the 9 studies, 6 studies showed a statistically significant association between obesity and GERD (23-25, 28-30), and 3 studies showed no association (22, 26, 27). Eight of the studies presented data to allow for calculation of unadjusted and adjusted odds ratios and 95% CIs for GERD, overweight (BMI, 25 kg/m2 to 30 kg/m2), and obesity (BMI > 30 kg/m2). In these 8 studies, the pooled weighted unadjusted odds ratios for GERD symptoms among overweight and obese persons were 1.5- and 2.0-fold, respectively, greater than normal-weight persons (Figure 1; Appendix Figure 1).
Figure 1. Adjusted odds ratios for the association of overweight (body mass index 25 kg/m2 to 30 kg/m2) and gastroesophageal reflux disease symptoms ( top ) and for the association between obesity (body mass index > 30 kg/m2) and gastroesophageal reflux disease symptoms ( bottom ).
Figure 1. Adjusted odds ratios for the association of overweight (body mass index 25 kg/m2 to 30 kg/m2) and gastroesophageal reflux disease symptoms (
top ) and for the association between obesity (body mass index > 30 kg/m2) and gastroesophageal reflux disease symptoms ( bottom ).
Table 1 presents the adjusted odds ratios. Seven studies adjusted for age and sex. Two studies adjusted for race (24, 27), 3 studies adjusted for nonsteroidal anti-inflammatory drug intake (23, 28, 30), 5 studies adjusted for cigarette smoking (23, 26, 28-30), 1 study adjusted for H. pylori status and socioeconomic status (28), and 1 study adjusted for physical activity (26). None of these adjustments changed the statistical significance or direction of the observed associations. Similarly, adjustments for alcohol (23, 26, 28-30), coffee consumption (23, 28-30), or other dietary factors (such as dietary fiber [29] or total energy intake [26]) resulted in no appreciable changes in the odds ratios. Nilsson and colleagues (29) found that adjustment for postmenopausal hormone therapy in women strengthened the association.
Two studies (Andersen and Jensen [22] and Stanghellini [25]) did not adjust for confounding factors. On the basis of findings from the studies that provided both adjusted and unadjusted odds ratios, we estimated the effect of confounding variables to reduce the unadjusted ratio by a factor of 0.13, which we applied to the 2 studies. The pooled adjusted odds ratios were 1.43 (95% CI, 1.158 to 1.774; P = 0.001) for BMI of 25 kg/m2 to 30 kg/m2 and 1.94 (CI, 1.468 to 2.566; P < 0.001) for BMI greater than 30 kg/m2 (Figure 1). The heterogeneity among the results of these studies was statistically significant (chi-square, 82.1; P < 0.001).

Erosive Esophagitis

Our search identified 111 potentially relevant titles. Of these, 7 studies (4 case–control studies [31-34], 2 cross-sectional studies [35, 36], and 1 cohort study [37]) fulfilled the inclusion and exclusion criteria (Table 2). Three studies (Wilson and colleagues [33], Stene-Larson and colleagues [32], and Nilsson and colleagues [31]) were conducted for the primary purpose of evaluating the relationship between obesity and esophagitis.
Table 2. Characteristics of 5 Studies of Obesity and Erosive Esophagitis
Table 2. Characteristics of 5 Studies of Obesity and Erosive Esophagitis
We calculated a pooled odds ratio for the 4 case–control and 2 cross-sectional studies, excluding the 1 cohort study by Ruhl and Everhart (37). The pooled unadjusted odds ratio of esophagitis related to BMI of 25 kg/m2 or higher was 1.7-fold greater than that of esophagitis related to BMI less than 25 kg/m2 (Appendix Figure 2). When we excluded the 2 studies from Asia (35, 36) from our analysis, the association between obesity and esophagitis did not appreciably change (pooled unadjusted odds ratio, 1.8 [CI, 1.5 to 2.2]) and there was no statistical heterogeneity (chi-square, 0.15; P > 0.2).
Table 2 shows adjusted odds ratios. Four studies adjusted for potential confounding factors (31, 33, 34, 37). Wilson and colleagues (33) found that the association lost statistical significance after adjustment for hiatal hernia. We estimated from these 4 studies that adjustment for confounding variables decreased the odds ratios by 0.16. We applied this factor to the studies by Stene-Larsen and colleagues (32) and Furukawa and colleagues (36). The pooled adjusted odds ratio for 6 studies was 1.76 (CI, 1.156 to 2.677; P = 0.004) (Figure 2). A heterogeneity test among the studies was also significant (chi-square, 18.9; P = 0.002).
Figure 2. Adjusted odds ratios from cross-sectional and case–control studies examining the association between overweight or obesity (body mass index ≥25 kg/m2) and erosive esophagitis.
Figure 2. Adjusted odds ratios from cross-sectional and case–control studies examining the association between overweight or obesity (body mass index ≥25 kg/m2) and erosive esophagitis.

Barrett Esophagus

No studies reporting the association of Barrett esophagus and obesity met the selection criteria of our review.

Esophageal Adenocarcinoma and Adenocarcinoma of the Gastric Cardia

Our search strategy identified 93 titles. Of these, 9 case–control studies met the inclusion and exclusion criteria. Eight studies examined population-based samples, and 1 study examined hospital-based samples (38). Four studies (Chow and colleagues [(39], Ji and colleagues [40], Zhang and colleagues [41], and Lagergren and colleagues [42]) were conducted for the primary purpose of evaluating the relationship between obesity and esophageal adenocarcinoma.
The studies of Chow and colleagues (39), Lagergren and colleagues (42), Vaughan and colleagues (43), and Wu and colleagues (44) examined esophageal adenocarcinoma as a separate category from gastric cardia adenocarcinoma. Cheng and colleagues (45) excluded cancer of the cardia but may have included cancer of the gastroesophageal junction. Brown and colleagues (46) and Zhang and colleagues (41) combined data for adenocarcinomas of the esophagus with the gastric cardia. The 2 studies from China included only cases of gastric cardia adenocarcinoma (38, 40).
Because substantial weight loss at the time of cancer diagnosis was a concern, all studies, except 1 study (38), obtained historical weight information. Among the 3 studies that collected several historical heights and weights (39, 42, 44), the magnitude or direction of the association between BMI at different time points and risk for adenocarcinoma did not statistically significantly differ.

Esophageal Adenocarcinoma

Seven studies examined the association between obesity and esophageal adenocarcinoma (Table 3). Weighted pooling of unadjusted odds ratios (Appendix Figure 3) indicates that the risk for esophageal adenocarcinoma is 2.1 times higher in persons with BMI of 25 kg/m2 or greater than in normal-weight persons. We did not include Zhang and colleagues' study (41) in the pooled odds ratio because the study combined cases of both esophageal and gastric cardia cancer.
Table 3. Characteristics of 9 Studies of Body Mass Index and Esophageal Adenocarcinoma or Adenocarcinoma of the Gastric Cardia
Table 3. Characteristics of 9 Studies of Body Mass Index and Esophageal Adenocarcinoma or Adenocarcinoma of the Gastric Cardia
All studies adjusted for age and sex, and these were not found to have any appreciable effect on the association. Further adjustments for race (41, 43, 44), smoking (39-46), alcohol consumption (40-43, 45, 46), caloric intake (40-42, 45, 46), history of reflux symptoms (42, 45), or education level (40-44) did not alter the statistical significance or direction of unadjusted associations. The pooled adjusted odds ratio from 6 studies for BMI of 25 kg/m2 or greater was 2.02 (CI, 1.534 to 2.669; P < 0.001). There was a trend toward a dose–response relationship with an increase in pooled adjusted odds ratio for BMI of 25 kg/m2 to 30 kg/m2 and BMI greater than 30 kg/m2 of 1.52 (CI, 1.147 to 2.009; P = 0.004) Figure 3, top) and 2.78 (CI, 1.850 to 4.164; P < 0.001) Figure 3, bottom).
Figure 3. Adjusted odds ratios from case–control studies examining the association between overweight (body mass index ≥25 kg/m2) and adenocarcinoma of the esophagus ( top ) and the association between obesity (body mass index >30 kg/m2) and esophageal adenocarcinoma ( bottom ).
Figure 3. Adjusted odds ratios from case–control studies examining the association between overweight (body mass index ≥25 kg/m2) and adenocarcinoma of the esophagus (
top ) and the association between obesity (body mass index >30 kg/m2) and esophageal adenocarcinoma ( bottom ).
Appendix Figure 1. Unadjusted odds ratios (95% CIs) for the risk for gastroesophageal reflux disease (GERD) symptoms among overweight ( top ) and obese ( bottom ) versus normal-weight individuals. Normal weight is defined as a body mass index (BMI) less than 25 kg/m2 for all studies, except for the study by Locke and colleagues (BMI < 24 kg/m2) (23). Overweight is defined as a BMI of 25 kg/m2 to 30 kg/m2 for all studies, except for Locke and colleagues' study (BMI, 24 to 30 kg/m2) (23) and Wu and colleagues' study (BMI, 25 kg/m2 to 28 kg/m2) (27). Obese is defined as a BMI greater than 30 kg/m2 for all studies, except Wu and colleagues' study (BMI > 28 kg/m2) (27). Pooled weighted odds ratios are also given for overweight and obesity (that is, BMI ≥ 25 kg/m2 compared with BMI < 25 kg/m2).
Appendix Figure 1. Unadjusted odds ratios (95% CIs) for the risk for gastroesophageal reflux disease (GERD) symptoms among overweight (
top ) and obese ( bottom ) versus normal-weight individuals. Normal weight is defined as a body mass index (BMI) less than 25 kg/m2 for all studies, except for the study by Locke and colleagues (BMI < 24 kg/m2) (23). Overweight is defined as a BMI of 25 kg/m2 to 30 kg/m2 for all studies, except for Locke and colleagues' study (BMI, 24 to 30 kg/m2) (23) and Wu and colleagues' study (BMI, 25 kg/m2 to 28 kg/m2) (27). Obese is defined as a BMI greater than 30 kg/m2 for all studies, except Wu and colleagues' study (BMI > 28 kg/m2) (27). Pooled weighted odds ratios are also given for overweight and obesity (that is, BMI ≥ 25 kg/m2 compared with BMI < 25 kg/m2).
Appendix Figure 2. Unadjusted odds ratios (95% CIs) from cross-sectional and case–control studies examining the risk for erosive esophagitis among overweight (body mass index [ BMI ] ≥ 25 kg/m2) versus normal-weight (BMI < 25 kg/m2) individuals.
Appendix Figure 2. Unadjusted odds ratios (95% CIs) from cross-sectional and case–control studies examining the risk for erosive esophagitis among overweight (body mass index [
BMI ] ≥ 25 kg/m2) versus normal-weight (BMI < 25 kg/m2) individuals.
Appendix Figure 3. Unadjusted odds ratios (95% CIs) from case–control studies examining the association between overweight or obesity (body mass index [ BMI ] ≥25 kg/m2) and adenocarcinoma of the esophagus or gastric cardia.
Appendix Figure 3. Unadjusted odds ratios (95% CIs) from case–control studies examining the association between overweight or obesity (body mass index [
BMI ] ≥25 kg/m2) and adenocarcinoma of the esophagus or gastric cardia.

Adenocarcinoma of the Gastric Cardia

Six studies examined the association of obesity and adenocarcinoma of the gastric cardia (Table 3). Four studies showed a statistically significant association (Table 3). Vaughan and colleagues (43) showed an association that did not reach statistical significance. One study from China by Zhang and colleagues (38) showed a statistically significant inverse association. We pooled unadjusted estimates from 5 studies (Figure 3 and Appendix Figure 3) with a modest, weighted pooled odds ratio of 1.5. In general, adjustment resulted in an increase in the odds ratios by a factor of 0.3. We applied this factor to derive adjusted odds ratios in studies that provided only unadjusted estimates. The pooled adjusted odds ratio from all 6 studies was 1.68 (CI, 1.197 to 2.351; P = 0.003). These results had statistically significant heterogeneity (chi-square, 45.3; P < 0.001). However, the 1 study with an inverse association examined BMI 1 year before the cancer diagnosis, whereas other studies used more distant BMI. Excluding that study reduced the heterogeneity (chi-square, 7.6; P = 0.1). Finally, funnel plots were not suggestive of publication bias (not shown).

Discussion

We performed a systematic review of the literature examining the association between BMI and acid-related disorders of the esophagus. The risk for GERD symptoms, erosive esophagitis, or esophageal adenocarcinoma increased with overweight or obesity compared with normal BMI. The association of overweight and obesity with adenocarcinoma of the gastric cardia, a condition less frequently associated with GERD, was weaker and less consistent.
We did not provide a quality score for the individual studies included in our review. However, our inclusion and exclusion criteria ensured a minimum standard of quality. Thus, all studies used acceptable definitions of exposure (BMI) and outcomes of interest (symptoms, endoscopic findings, and pathologic confirmation). Most studies that examined symptoms and cancer were population-based studies with acceptable enrollment rates. However, only 10 of 25 studies stated that the primary purpose of the study was to examine the relationship between obesity and GERD. Another limitation that particularly applies to case–control studies evaluating adenocarcinoma is recall bias, where case-patients may overestimate their historical weight compared with noncancer controls. However, the consistency across studies provides some reassurance against recall bias. To minimize the possibility of recall bias, 3 studies examined patients with squamous-cell carcinoma of the esophagus (39, 42, 43) and 4 studies examined patients with distal gastric adenocarcinoma (38-40, 44) as a second cancer comparison group. These studies found that BMI had either no association or a statistically significant negative association with the cancer controls. Furthermore, in all erosive esophagitis studies, BMI was measured directly.
The heterogeneity test indicated statistically significant degrees of heterogeneity among studies. However, the association between BMI and GERD complications was markedly consistent, and differences in the risk estimates were largely in the magnitude rather than the direction of the association. No studies had a negative association, and the 4 studies with no significant association indicated a trend toward statistical significance. Therefore, the use of pooled odds ratios was appropriate. In the case of esophagitis, the heterogeneity was largely attributed to geographic differences among the study samples. Removing studies that were conducted in Asia eliminated heterogeneity among studies of esophagitis.
Overweight and obesity satisfy several criteria for a causal association with GERD and its complications, including esophagitis and esophageal adenocarcinoma. Our review indicates a consistent modest statistical association, a correct temporal relationship, and possibly a weak dose–response relationship. However, estimates of BMI in most studies were obtained for several years predating the outcomes. The cross-sectional and case–control design of most studies makes it difficult to be emphatic about temporal association, especially for chronic slowly progressive conditions, such as GERD symptoms and erosive esophagitis.
Obese patients may experience extrinsic gastric compression by surrounding adipose tissue, leading to an increase in intragastric pressures and subsequent relaxation of the lower esophageal sphincter (47-49). The findings of manometric studies, however, have been inconsistent, indicating both decreased (50) and normal (17, 51-53) lower esophageal sphincter pressures in obese persons.
Studies have also suggested that rather than obesity, the amount and type of dietary intake, notably fat, are responsible for GERD. Seven studies in our review examined total caloric intake and dietary intake of fiber, fruits and vegetables, or other macronutrients or micronutrients. In all, the effect of BMI on GERD-related disorders was independent of dietary intake, which is further supported by a recent smaller study (54). Moreover, all studies found no consistent association between dietary fat and GERD or esophageal adenocarcinoma.
Obese patients may have an increased risk for hiatal hernia, which has a role in initiating and promoting GERD (55). In our review, 3 studies specifically examined hiatal hernia in relation to obesity and GERD or esophagitis (27, 32, 33). Wilson and colleagues (33) found obesity to be statistically significantly associated with esophagitis. However, after the authors controlled for the presence of hiatal hernia, this association lost statistical significance, implying that hiatal hernia was the mechanism by which obesity leads to esophagitis. Wu and colleagues (44), on the other hand, found no statistically significant association between BMI and hiatal hernia. In a study that examined the association of BMI and grade of esophagitis in persons with esophagitis, El-Serag and Johanson (56) also found that obesity remained an independent risk factor for severe versus mild esophagitis while controlling for the presence of hiatal hernia. The cross-sectional nature of these studies makes it impossible to distinguish temporal associations between obesity and hiatal hernia. In sum, data about hiatal hernia as a cause of increased esophageal reflux in obesity are conflicting.
Humoral factors have also been considered as a mechanism relating obesity to reflux and esophageal adenocarcinoma. Two studies identified in our review conducted by the same group of investigators from Sweden observed that the association of obesity and GERD might be mediated by estrogen (29, 31). The first study reported a statistically significant association between obesity and esophagitis in women, which was potentiated by the use of estrogen in postmenopausal women (31). The second study examined a large population-based cohort and found that obese women had an increased risk for GERD symptoms compared with obese men; the risk was highest in premenopausal women and postmenopausal women receiving estrogen therapy (29). Furukawa and colleagues (36) reported that overweight women older than 70 years of age were the only group to have a statistically significantly increased risk for esophagitis. The authors did not report on estrogen use in these patients. However, several other studies have found that the obesity-related increased risk for esophageal disease was neither confounded nor modified by sex (27, 56).
Some evidence suggests that a change in weight can also affect the risk for GERD and its complications. For example, Nilsson and colleagues (29) found that a weight gain greater than 3.5 kg/m2 was associated with a 2.7-fold (CI, 2.3- to 3.2-fold) increased risk for developing new reflux symptoms. On the other hand, although weight loss is often recommended as a therapeutic measure in reflux disease (57), the studies yielded conflicting results on its efficacy (58, 59). Small nonrandomized studies suggest that weight loss after bariatric surgery for morbid obesity is associated with an improvement in GERD symptoms (60, 61). Finally, no data suggest that weight loss can affect risk for esophageal adenocarcinoma.
The mechanism of obesity-related esophageal adenocarcinoma is unclear. Of interest, no studies satisfying our inclusion and exclusion criteria assessed the association between obesity and Barrett esophagus. Excluded because of the absence of a control group, Caygill and colleagues' study (62) reported a BMI greater than 30 kg/m2 in 24% of patients in their case series of 102 patients with Barrett esophagus, as compared with an obesity prevalence of 13% in the general population of England. Excluded because of few cases, a case–control study by Chak and colleagues (63) and a cross-sectional study by Gerson and colleagues (64) both reported no apparent increase in risk for obesity in 35 and 27 patients with Barrett esophagus, respectively, although Chak and colleagues reported that the duration of obesity was statistically significantly longer in patients with Barrett esophagus.
Our review suggests that overweight and obesity are risk factors for acid-related esophageal disease. Future studies should examine the mechanism by which obesity causes these complications, as well as the potential effects of weight loss. In the meantime, however, it is prudent to counsel all overweight patients who present with GERD-related diseases that weight loss may help improve symptoms.

References

1.
El-Serag HBSonnenberg A. Opposing time trends of peptic ulcer and reflux disease. Gut. 1998;43:327-33. [PMID: 9863476]
2.
Blot WJMcLaughlin JK. The changing epidemiology of esophageal cancer. Semin Oncol. 1999;26:2-8. [PMID: 10566604]
3.
Lagergren J. [Increased incidence of adenocarcinoma of the esophagus and cardia. Reflux and obesity are strong and independent risk factors according to the SECC study]. Lakartidningen. 2000;97:1950-3. [PMID: 10826353]
4.
Polednak AP. Trends in incidence rates for obesity-associated cancers in the US. Cancer Detect Prev. 2003;27:415-21. [PMID: 14642548]
5.
El-Serag HBMason ACPetersen NKey CR. Epidemiological differences between adenocarcinoma of the oesophagus and adenocarcinoma of the gastric cardia in the USA. Gut. 2002;50:368-72. [PMID: 11839716]
6.
Bujanda L. The effects of alcohol consumption upon the gastrointestinal tract. Am J Gastroenterol. 2000;95:3374-82. [PMID: 11151864]
7.
Lagergren JBergström RAdami HONyrén O. Association between medications that relax the lower esophageal sphincter and risk for esophageal adenocarcinoma. Ann Intern Med. 2000;133:165-75. [PMID: 10906830]
8.
Pandolfino JEKahrilas PJ. Smoking and gastro-oesophageal reflux disease. Eur J Gastroenterol Hepatol. 2000;12:837-42. [PMID: 10958210]
9.
Raghunath AHungin APWooff DChilds S. Prevalence of Helicobacter pylori in patients with gastro-oesophageal reflux disease: systematic review. BMJ. 2003;326:737. [PMID: 12676842]
10.
Sharma PVakil N. Review article: Helicobacter pylori and reflux disease. Aliment Pharmacol Ther. 2003;17:297-305. [PMID: 12562442]
11.
Terry PLagergren JWolk ANyrén O. Reflux-inducing dietary factors and risk of adenocarcinoma of the esophagus and gastric cardia. Nutr Cancer. 2000;38:186-91. [PMID: 11525596]
12.
Devesa SSBlot WJFraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer. 1998;83:2049-53. [PMID: 9827707]
13.
Wei JTShaheen N. The changing epidemiology of esophageal adenocarcinoma. Semin Gastrointest Dis. 2003;14:112-27. [PMID: 14653411]
14.
El-Serag HB. The epidemic of esophageal adenocarcinoma. Gastroenterol Clin North Am. 2002;31:421-40, viii. [PMID: 12134611]
15.
Barak NEhrenpreis EDHarrison JRSitrin MD. Gastro-oesophageal reflux disease in obesity: pathophysiological and therapeutic considerations. Obes Rev. 2002;3:9-15. [PMID: 12119661]
16.
Maddox AHorowitz MWishart JCollins P. Gastric and oesophageal emptying in obesity. Scand J Gastroenterol. 1989;24:593-8. [PMID: 2762759]
17.
O'Brien TF Jr. Lower esophageal sphincter pressure (LESP) and esophageal function in obese humans. J Clin Gastroenterol. 1980;2:145-8. [PMID: 7440948]
18.
Orlando RC. Overview of the mechanisms of gastroesophageal reflux. Am J Med. 2001;111 Suppl 8A 174S-177S. [PMID: 11749946]
19.
Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev. 1987;9:1-30. [PMID: 3678409]
20.
Berlin JALongnecker MPGreenland S. Meta-analysis of epidemiologic dose-response data. Epidemiology. 1993;4:218-28. [PMID: 8512986]
21.
Berlin JAColditz GA. A meta-analysis of physical activity in the prevention of coronary heart disease. Am J Epidemiol. 1990;132:612-28. [PMID: 2144946]
22.
Andersen LIJensen G. Risk factors for benign oesophageal disease in a random population sample. J Intern Med. 1991;230:5-10. [PMID: 2066710]
23.
Locke GR 3rdTalley NJFett SLZinsmeister ARMelton LJ 3rd. Risk factors associated with symptoms of gastroesophageal reflux. Am J Med. 1999;106:642-9. [PMID: 10378622]
24.
Oliveria SAChristos PJTalley NJDannenberg AJ. Heartburn risk factors, knowledge, and prevention strategies: a population-based survey of individuals with heartburn. Arch Intern Med. 1999;159:1592-8. [PMID: 10421282]
25.
Stanghellini V. Three-month prevalence rates of gastrointestinal symptoms and the influence of demographic factors: results from the Domestic/International Gastroenterology Surveillance Study (DIGEST). Scand J Gastroenterol Suppl. 1999;231:20-8. [PMID: 10565620]
26.
Lagergren JBergström RNyrén O. No relation between body mass and gastro-oesophageal reflux symptoms in a Swedish population based study. Gut. 2000;47:26-9. [PMID: 10861260]
27.
Wu AHTseng CCBernstein L. Hiatal hernia, reflux symptoms, body size, and risk of esophageal and gastric adenocarcinoma. Cancer. 2003;98:940-8. [PMID: 12942560]
28.
Murray LJohnston BLane AHarvey IDonovan JNair Pet al. Relationship between body mass and gastro-oesophageal reflux symptoms: The Bristol Helicobacter Project. Int J Epidemiol. 2003;32:645-50. [PMID: 12913045]
29.
Nilsson MJohnsen RYe WHveem KLagergren J. Obesity and estrogen as risk factors for gastroesophageal reflux symptoms. JAMA. 2003;290:66-72. [PMID: 12837713]
30.
Diaz-Rubio MMoreno-Elola-Olaso CRey ELocke GR 3rd, Rodriguez-Artalejo F. Symptoms of gastro-oesophageal reflux: prevalence, severity, duration and associated factors in a Spanish population. Aliment Pharmacol Ther. 2004;19:95-105. [PMID: 14687171]
31.
Nilsson MLundegårdh GCarling LYe WLagergren J. Body mass and reflux oesophagitis: an oestrogen-dependent association? Scand J Gastroenterol. 2002;37:626-30. [PMID: 12126237]
32.
Stene-Larsen GWeberg RFrøyshov Larsen IBjørtuft OHoel BBerstad A. Relationship of overweight to hiatus hernia and reflux oesophagitis. Scand J Gastroenterol. 1988;23:427-32. [PMID: 3381064]
33.
Wilson LJMa WHirschowitz BI. Association of obesity with hiatal hernia and esophagitis. Am J Gastroenterol. 1999;94:2840-4. [PMID: 10520831]
34.
Labenz JJaspersen DKulig MLeodolter ALind TMeyer-Sabellek Wet al. Risk factors for erosive esophagitis: a multivariate analysis based on the ProGERD study initiative. Am J Gastroenterol. 2004;99:1652-6. [PMID: 15330897]
35.
Chang CSPoon SKLien HCChen GH. The incidence of reflux esophagitis among the Chinese. Am J Gastroenterol. 1997;92:668-71. [PMID: 9128320]
36.
Furukawa NIwakiri RKoyama TOkamoto KYoshida TKashiwagi Yet al. Proportion of reflux esophagitis in 6010 Japanese adults: prospective evaluation by endoscopy. J Gastroenterol. 1999;34:441-4. [PMID: 10452674]
37.
Ruhl CEEverhart JE. Overweight, but not high dietary fat intake, increases risk of gastroesophageal reflux disease hospitalization: the NHANES I Epidemiologic Followup Study. First National Health and Nutrition Examination Survey. Ann Epidemiol. 1999;9:424-35. [PMID: 10501410]
38.
Zhang JSu XQWu XJLiu YHWang HZong XNet al. Effect of body mass index on adenocarcinoma of gastric cardia. World J Gastroenterol. 2003;9:2658-61. [PMID: 14669307]
39.
Chow WHBlot WJVaughan TLRisch HAGammon MDStanford JLet al. Body mass index and risk of adenocarcinomas of the esophagus and gastric cardia. J Natl Cancer Inst. 1998;90:150-5. [PMID: 9450576]
40.
Ji BTChow WHYang GMcLaughlin JKGao RNZheng Wet al. Body mass index and the risk of cancers of the gastric cardia and distal stomach in Shanghai, China. Cancer Epidemiol Biomarkers Prev. 1997;6:481-5. [PMID: 9232333]
41.
Zhang ZFKurtz RCSun MKarpeh M JrYu GPGargon Net al. Adenocarcinomas of the esophagus and gastric cardia: medical conditions, tobacco, alcohol, and socioeconomic factors. Cancer Epidemiol Biomarkers Prev. 1996;5:761-8. [PMID: 8896886]
42.
Lagergren JBergström RNyrén O. Association between body mass and adenocarcinoma of the esophagus and gastric cardia. Ann Intern Med. 1999;130:883-90. [PMID: 10375336]
43.
Vaughan TLDavis SKristal AThomas DB. Obesity, alcohol, and tobacco as risk factors for cancers of the esophagus and gastric cardia: adenocarcinoma versus squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 1995;4:85-92. [PMID: 7742727]
44.
Wu AHWan PBernstein L. A multiethnic population-based study of smoking, alcohol and body size and risk of adenocarcinomas of the stomach and esophagus (United States). Cancer Causes Control. 2001;12:721-32. [PMID: 11562112]
45.
Cheng KKSharp LMcKinney PALogan RFChilvers CECook-Mozaffari Pet al. A case–control study of oesophageal adenocarcinoma in women: a preventable disease. Br J Cancer. 2000;83:127-32. [PMID: 10883680]
46.
Brown LMSwanson CAGridley GSwanson GMSchoenberg JBGreenberg RSet al. Adenocarcinoma of the esophagus: role of obesity and diet. J Natl Cancer Inst. 1995;87:104-9. [PMID: 7707381]
47.
Mercer CDRue CHanelin LHill LD. Effect of obesity on esophageal transit. Am J Surg. 1985;149:177-81. [PMID: 3966634]
48.
Mercer CDWren SFDaCosta LRBeck IT. Lower esophageal sphincter pressure and gastroesophageal pressure gradients in excessively obese patients. J Med. 1987;18:135-46. [PMID: 3480930]
49.
Zacchi PMearin FHumbert PFormiguera XMalagelada JR. Effect of obesity on gastroesophageal resistance to flow in man. Dig Dis Sci. 1991;36:1473-80. [PMID: 1914772]
50.
Hagen JDeitel MKhanna RKIlves R. Gastroesophageal reflux in the massively obese. Int Surg. 1987;72:1-3. [PMID: 3596968]
51.
Backman LGranström LLindahl JMelcher A. Manometric studies of lower esophageal sphincter in extreme obesity. Acta Chir Scand. 1983;149:193-7. [PMID: 6880555]
52.
Fisher BLPennathur AMutnick JLLittle AG. Obesity correlates with gastroesophageal reflux. Dig Dis Sci. 1999;44:2290-4. [PMID: 10573376]
53.
Wajed SAStreets CGBremner CGDeMeester TR. Elevated body mass disrupts the barrier to gastroesophageal reflux; discussion 1018-9. Arch Surg. 2001;136:1014-8. [PMID: 11529823]
54.
Nandurkar SLocke GR 3rdFett SZinsmeister ARCameron AJTalley NJ. Relationship between body mass index, diet, exercise and gastro-oesophageal reflux symptoms in a community. Aliment Pharmacol Ther. 2004;20:497-505. [PMID: 15339321]
55.
Kahrilas PJ. The role of hiatus hernia in GERD. Yale J Biol Med. 1999;72:101-11. [PMID: 10780571]
56.
El-Serag HBJohanson JF. Risk factors for the severity of erosive esophagitis in Helicobacter pylori-negative patients with gastroesophageal reflux disease. Scand J Gastroenterol. 2002;37:899-904. [PMID: 12229963]
57.
Kitchin LICastell DO. Rationale and efficacy of conservative therapy for gastroesophageal reflux disease. Arch Intern Med. 1991;151:448-54. [PMID: 1672062]
58.
Fraser-Moodie CANorton BGornall CMagnago SWeale ARHolmes GK. Weight loss has an independent beneficial effect on symptoms of gastro-oesophageal reflux in patients who are overweight. Scand J Gastroenterol. 1999;34:337-40. [PMID: 10365891]
59.
Kjellin ARamel SRössner SThor K. Gastroesophageal reflux in obese patients is not reduced by weight reduction. Scand J Gastroenterol. 1996;31:1047-51. [PMID: 8938895]
60.
Frezza EEIkramuddin SGourash WRakitt TKingston ALuketich Jet al. Symptomatic improvement in gastroesophageal reflux disease (GERD) following laparoscopic Roux-en-Y gastric bypass. Surg Endosc. 2002;16:1027-31. [PMID: 11984683]
61.
Jones KB Jr. Roux-en-Y gastric bypass: an effective antireflux procedure in the less than morbidly obese. Obes Surg. 1998;8:35-8. [PMID: 9562484]
62.
Caygill CPJohnston DALopez MJohnston BJWatson AReed PIet al. Lifestyle factors and Barrett's esophagus. Am J Gastroenterol. 2002;97:1328-31. [PMID: 12094845]
63.
Chak ALee TKinnard MFBrock WFaulx AWillis Jet al. Familial aggregation of Barrett's oesophagus, oesophageal adenocarcinoma, and oesophagogastric junctional adenocarcinoma in Caucasian adults. Gut. 2002;51:323-8. [PMID: 12171951]
64.
Gerson LBShetler KTriadafilopoulos G. Prevalence of Barrett's esophagus in asymptomatic individuals. Gastroenterology. 2002;123:461-7. [PMID: 12145799]

Comments

0 Comments
Sign In to Submit A Comment
Prasanta Padhan25 August 2005
Malnutrition and molecular basis of esophageal adenocarcinoma

Malnutrition either as over and undernutrition can result in deficiency of trace elements such as zinc which can lead to Cyclin D1 overexpression and p53 deficiency,which increases cell proliferation.Further zinc deficiency has been found to dysregulate p16ink4a-cyclin D1/Cdk4-Rb pathway,thereby promoting esophageal tumors.Many other genes and gene products such as TGF-á,EGF,COX- 2,FAS,VEGF,Telomerase are also commonly linked to both esophageal and colonic adenocarcinoma in obese individuals.Whether these can be targeted as a therapeutic option needs further research and large scale studies.Role of aspirin as COX-2 inhibitor in regulating apoptosis appears theoritically as a promising drug as in prevention of colonic adenocarcinoma in patients with familial adenomatous polyposis.

Conflict of Interest:

None declared

Douglas J Sprung15 August 2005
The Obese Refluxer- Nothing New Under the Sun

The meta-analysis by Hampel et al ( 1) was immensely impressive. Yet, it has amazed me over the past 3 decades, how much energy has gone into the debate as to whether or not obesity causes gastroesophageal reflux. As a medical student at Duke in the 1970's I learned that obesity was commonly associated with reflux symptoms by spending time at the Rice Diet program in Durham,NC. Dr Kempner would bring over one obese patient after another and he would discuss their symptoms and medical issues, including hypertension, diabetes,heart failure,gallstones,arthritis and reflux symptoms. After a few weeks on the program,without any antacids or proton- pump inhibitors,the reflux symptoms were resolved in most of the patients ,as their oral intake and weight had dramatically reduced. Those observations were published in the 1960's, and quite frankly the conclusions drawn from this extensive meta-analysis could have been garnered by simply speaking to any of the successful dieter's at the Rice House or other diet centers around the country. It is an academic shame that so much time,effort and grant money had to be spent to this end. As far as Barrett's esophagus(BE) and adenocarcinoma go, we have not found a significant association with obesity in our community private practice. 1/9 patients with incident adenocarcinoma within BE was obese. Very few patients with BE (long segment) were obese.Whereas obesity is correlated with reflux symptoms, Barrett's esophagus and its complications were not in our community.

1.Hampel H,Abraham N,El-Serag HB. Meta-Analysis: Obesity and the Risk for Gastroesophageal Reflux Disease and its Complications. Ann INtern Med. 2005;143:199-211. 2.Sprung DJ,Sprung GM. Barrett's Esophagus In Our Community Over the Past 12 Years. Amer Journal Gastro. Vol 98,No 9,supp 2003; S-1.

Conflict of Interest:

None declared

Information & Authors

Information

Published In

cover image Annals of Internal Medicine
Annals of Internal Medicine
Volume 143Number 32 August 2005
Pages: 199 - 211

History

Published online: 2 August 2005
Published in issue: 2 August 2005

Keywords

Authors

Affiliations

Howard Hampel, MD, PhD
From the Michael E. DeBakey Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Texas.
Neena S. Abraham, MD, MSc(Epi)
From the Michael E. DeBakey Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Texas.
Hashem B. El-Serag, MD, MPH
From the Michael E. DeBakey Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Texas.
Acknowledgments: The authors thank Loralee Capistrano, MPH, and Jennifer R. Kramer, PhD, for their help in using STATA.
Grant Support: Dr. El-Serag is a recipient of the Department of Veterans Affairs Health Services Research and Development Advanced Career Development Award (RCD00-013-2).
Disclosures: None disclosed.
Corresponding Author: Hashem B. El-Serag, MD, MPH, Michael E. DeBakey Veterans Affairs Medical Center, 2002 Holcombe Boulevard (152), Houston, TX 77030; e-mail, [email protected].
Current Author Addresses: Drs. Hampel, Abraham, and El-Serag: Michael E. DeBakey Veterans Affairs Medical Center, 2002 Holcombe Boulevard (152), Houston, TX 77030.

Metrics & Citations

Metrics

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

For more information or tips please see 'Downloading to a citation manager' in the Help menu.

Format





Download article citation data for:
Howard Hampel, Neena S. Abraham, Hashem B. El-Serag. Meta-Analysis: Obesity and the Risk for Gastroesophageal Reflux Disease and Its Complications. Ann Intern Med.2005;143:199-211. [Epub 2 August 2005]. doi:10.7326/0003-4819-143-3-200508020-00006

View More

Get Access

Login Options:
Purchase

You will be redirected to acponline.org to sign-in to Annals to complete your purchase.

Create your Free Account

You will be redirected to acponline.org to create an account that will provide access to Annals.

View options

PDF/ePub

View PDF/ePub

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media