**2. Evidence supporting the link between cardiovascular disease and upper gastrointestinal symptoms**

At least a half dozen published studies [8-13] have demonstrated a link between cardiovascular diseases and an increased risk of upper gastrointestinal symptoms. Three of these studies have assessed the association of upper gastrointestinal symptoms with general cardiovascular diagnosis. A recent study created two cohorts of patients derived from health insurance claims data from the Human Capital Management Services research database over a four year period (2001-2004)[9]. The cohorts were based upon the presence or absence of functional dyspepsia diagnosis codes, with the control cohort (n=83,450) being matched to the functional dyspepsia cohort (n=1,669) using a propensity score that included variables such as age, sex, marital status, salary, among others. This study demonstrated that employees with functional dyspepsia were 1.8-fold more likely to suffer from circulatory system disease (preva‐ lence=39.19% in those with functional dyspepsia versus 22.37% in the control group; p<0.05).



The aim of this chapter is to provide a detailed discussion of the evidence suggesting and supporting an increased risk of upper gastrointestinal symptoms in populations suffering from

**2. Evidence supporting the link between cardiovascular disease and upper**

At least a half dozen published studies [8-13] have demonstrated a link between cardiovascular diseases and an increased risk of upper gastrointestinal symptoms. Three of these studies have assessed the association of upper gastrointestinal symptoms with general cardiovascular diagnosis. A recent study created two cohorts of patients derived from health insurance claims data from the Human Capital Management Services research database over a four year period (2001-2004)[9]. The cohorts were based upon the presence or absence of functional dyspepsia diagnosis codes, with the control cohort (n=83,450) being matched to the functional dyspepsia cohort (n=1,669) using a propensity score that included variables such as age, sex, marital status, salary, among others. This study demonstrated that employees with functional dyspepsia were 1.8-fold more likely to suffer from circulatory system disease (preva‐ lence=39.19% in those with functional dyspepsia versus 22.37% in the control group; p<0.05).

**Study Description Key Finding**

Higher prevalence of circulatory system disease in those with functional dyspepsia versus controls (ratio=1.8:1; prevalence=39.19% in those with functional dyspepsia versus 22.37% in the control

Higher odds of cardiovascular condition (OR=2.0), myocardial/endocardial/pericardial/valve condition (OR=2.7) or vascular (extracardiac) condition (OR=2.8) in patients with UGIS diagnosed by a

Higher odds of self-reported cardiovascular symptoms (OR=1.5), or myocardial/endocardial/ pericardial/valve symptoms (OR=4.4) over previous

Higher odds of chest pain (OR: 2.4, 95%CI 2.1-2.7) or angina (OR=1.5, 95%CI=1.2-1.8) comorbidity in dyspepsia cohort in the year prior to index date than

three months in patients with UGIS

group; p<0.05)

doctor

control cohort

Retrospective database analysis of paid health insurance claims within the Human Capital Management Services research database (USA); 275,875 eligible employees, 1,669 with functional dyspepsia diagnosis

Respondents of the Domestic/International Gastroenterology Surveillance Study which surveyed urban, adult populations from 10 countries representing seven geographic areas (Canada, the USA, Switzerland, The Netherlands, Italy, Japan and the Nordic countries) using a study-specific symptom checklist; prevalence rate of upper gastrointestinal symptoms=28%

Analysis UK General Practice Research Database to identify patients with new onset dyspepsia in 1996; overall incidence=15.3

cardiovascular disease.

**Study, year (N=)**

Brook 2012 (N=275,875)

Stanghellini 1999 (N=5,581)

Wallander 2007 (N=17,949)

**gastrointestinal symptoms**

138 Dyspepsia - Advances in Understanding and Management

codes

AF=atrial fibrillation; FD=Functional dyspepsia; HLD=hyperlipidemia; HTN=hypertension; NA=not applicable; NR=not reported; OR=odds ratio; UK=United Kingdom; UGIS=upper gastrointestinal symptoms; USA=United States of America

**Table 2.** Studies Assessing Upper Gastrointestinal Symptoms in Patients with Cardiovascular Disease

A second study, the large Domestic/International Gastroenterology Surveillance Study [8] looked to investigate any association between upper gastrointestinal symptoms (gastroeso‐ phageal-, ulcer- or dysmotility-like) and lifestyle factors (including comorbidities) in a large sample of patients experiencing dyspepsia in the prior 3-months. A sample of urban, adult populations from seven geographic areas (Canada, United States, Switzerland, the Nether‐ lands, Italy, Japan and the Nordic countries) was obtained by door-to-door or telephone recruitment. Subjects were divided into groups depending on whether gastrointestinal symptoms were reported and were analyzed for the association with comorbid conditions. In total, 5,581 subjects were recruited, with 1,566 (28%) reporting relevant upper gastrointestinal symptoms. In the previous three months, subjects reporting gastrointestinal symptoms selfreported more general cardiovascular (odds ratio= 1.5) or vascular myocardial/endocardial/ pericardial and valve (odds ratio=4.4) symptoms or illnesses. Subjects with upper gastroin‐ testinal symptoms also had increased prevalence of clinician-diagnosed cardiovascular (odds ratio=2.0) or myocardial/endocardial/pericardial and valve (odds ratio=2.7) conditions.

of dyspepsia for patients with atrial fibrillation was 14.7 events per 100 patient years. At baseline, 62% of patients (n=257,357) had at least one medication which may cause gastroin‐ testinal tolerability issues. The authors conclude that atrial fibrillation was associated with a

Upper Gastrointestinal Symptoms and Cardiovascular Disease

http://dx.doi.org/10.5772/56564

141

Finally in a small case series evaluating the relationship between dyspepsia and congestive heart disease or ischemic heart disease in Italian patients, data showed alterations of motility in the esophagus, stomach and duodenum in every cardiovascular disease patient evaluated

In addition to the aforementioned data suggesting upper gastrointestinal symptoms are more prevalent with patients with cardiovascular diseases; a body of literature suggesting upper gastrointestinal symptoms may in fact induce cardiovascular disease has begun to take shape [15-18]. In 2003, the first signal that gastro esophageal-like symptoms or disease could be linked to the development of atrial fibrillation was published [15]. Clinicians in Australia looked at 18 patients with concomitant diagnoses of lone paroxysmal atrial fibrillation and gastroeso‐ phageal reflux disease and noted that after treatment with a proton pump inhibitor to treat the upper gastrointestinal symptoms, 14 of 18 had a decrease or disappearance of at least one

Since that time, 3 observational studies [16-18] have more thoroughly evaluated this link. In a cohort study of 163,627 patients receiving care from the United States Army National Capitol Area Military Healthcare System between 2001 and 2007 (5% had atrial fibrillation and 29% had gastroesophageal-like symptoms), gastroesophageal symptoms were associated with an increased risk of atrial fibrillation, even after adjusting for age, sex, race and atherosclerotic risk factors (relative risk=1.19, 95% confidence interval=1.13-1.25) or further adjustment for ischemic heart disease, cardiomyopathy, atrial septal defect and being status post-cardiac

The second study [17] similarly sought to assess the relationship between gastroesophageal reflux disease and atrial fibrillation; and the researchers assessed the risk for atrial fibrillation over a follow-up period of greater than 11 years. A self-report survey was sent to 5,288 patients aged 25-74 over the 6 year period of 1988-1994. Of these patients, 741 developed atrial fibrillation. Contrary to the previous study, an inverse relationship with observed between gastroesophageal reflux disease symptoms and atrial fibrillation risk (hazard ratio=0.81, 95% confidence interval=0.68-0.96). However, the frequency of symptoms in those with gastroeso‐ phageal reflux (none, some, weekly, daily) was associated with an increased hazard of atrial fibrillation (p<0.01 for overall association); with daily symptoms associated with the highest hazard (hazard ratio=1.30, 95% Confidence interval=0.98-1.57) of developing atrial fibrillation compared to no gastroesophageal symptoms (p=0.07 unadjusted and p>0.2 after adjustment for confounders). The researchers cite an increase in medical attention in those experiencing gastroesophageal reflux as a possible explanation for the lack of association between the presence of symptoms and atrial fibrillation; hypothesizing that extra physician visits resulting from gastroesophageal symptoms resulted in early and more frequent identification and treatment of known atrial fibrillation risk factors, as well as a higher utilization of proton pump

bypass surgery (relative risk=1.08, 95% confidence interval=1.02-1.13) [16].

40% risk of developing a gastrointestinal event, which was predominantly dyspepsia.

and lesions of the gastric mucous membrane in more than half [14].

paroxysmal atrial fibrillation symptom.

Two more large studies [10,11] have reported on a link between the prevalence of upper gastrointestinal symptoms with angina and chest pain. The first, a cross-sectional study of 6,913 patients aged 20-79 with new diagnoses of dyspepsia and 11,036 age- and sex-matched control patients from the United Kingdom-based General Practice Research Database, demonstrated dyspeptic patients are at increased odds of having a diagnosis for chest pain (odds ratio=2.4, 95% confidence interval=2.1-2.7) or angina (odds ratio=1.5, 95% confidence interval=1.2-1.8) within the previous year. In addition, dyspeptic patients are also more likely to receiving receive a first time diagnosis for chest pain (odds ratio=2.3, 95% confidence interval=2.0-2.8) or angina (odds ratio=2.7, 95% confidence interval=1.8-4.0) [10]. In an older study of 4,962 patients aged 18-61 who took part in the Rand Health Insurance Experiment, a decade-long randomized controlled trial of the effects of alternative methods of financing health care services, about 30% had one chronic illness, with an additional 16% having 2 or more. Ulcer-like symptoms, defined by a previous diagnosis along with taking antacids daily, frequent episodes of stomach pain relieved by milk, occurring one-half hour after eating or at night, was significantly associated with angina (p<0.05) and congestive heart failure (p<0.001) [11].

A single study sought to assess the prevalence of dyspepsia among patients with atrial fibrillation [12]. The population (n=1,297) included a nationwide sample of American adults (from the 2009 National Health and Wellness Survey) with atrial fibrillation divided into two groups: those reporting dyspepsia (defined as any of the following: ulcers, abdominal bloating, abdominal pain, gastroesophageal disease or heartburn) and those who did not. Of these atrial fibrillation patients, 41% reported a diagnosis of a gastrointestinal condition while 34% reported a diagnosis of dyspepsia. Patients with dyspepsia were associated with a significantly higher mean CHADS2 score (1.9 vs. 1.4, p<0.05). Of note, while the CHADS2 score was developed as a tool to determine atrial fibrillation patients' risk for stroke, in this case, it can also serve as a marker of the presence of cardiovascular diseases since 2 of 5 CHADS2 criteria (eg, stroke and congestive heart failure) are in fact cardiovascular diseases and the remaining 3 criteria (eg, age, hypertension, diabetes) are potent risk factors for cardiovascular disease.

A retrospective database study sought to assess the risk of dyspepsia among patients with atrial fibrillation [13]. Analysis of insurance claims from the MarketScan® database from 2005-2009 was conducted. The population (n=413,168) included patients ≥18 years at the date of first atrial fibrillation diagnosis, with 180 days of continuous insurance coverage prior to the index atrial fibrillation diagnosis, and no gastrointestinal event within 180 days of the index atrial fibrillation diagnosis. The risk of dyspepsia was assessed with incidence rates (IRs; new dyspepsia case per patient years of observation). During a mean follow-up of 563 days, the IR of dyspepsia for patients with atrial fibrillation was 14.7 events per 100 patient years. At baseline, 62% of patients (n=257,357) had at least one medication which may cause gastroin‐ testinal tolerability issues. The authors conclude that atrial fibrillation was associated with a 40% risk of developing a gastrointestinal event, which was predominantly dyspepsia.

symptoms were reported and were analyzed for the association with comorbid conditions. In total, 5,581 subjects were recruited, with 1,566 (28%) reporting relevant upper gastrointestinal symptoms. In the previous three months, subjects reporting gastrointestinal symptoms selfreported more general cardiovascular (odds ratio= 1.5) or vascular myocardial/endocardial/ pericardial and valve (odds ratio=4.4) symptoms or illnesses. Subjects with upper gastroin‐ testinal symptoms also had increased prevalence of clinician-diagnosed cardiovascular (odds ratio=2.0) or myocardial/endocardial/pericardial and valve (odds ratio=2.7) conditions.

140 Dyspepsia - Advances in Understanding and Management

Two more large studies [10,11] have reported on a link between the prevalence of upper gastrointestinal symptoms with angina and chest pain. The first, a cross-sectional study of 6,913 patients aged 20-79 with new diagnoses of dyspepsia and 11,036 age- and sex-matched control patients from the United Kingdom-based General Practice Research Database, demonstrated dyspeptic patients are at increased odds of having a diagnosis for chest pain (odds ratio=2.4, 95% confidence interval=2.1-2.7) or angina (odds ratio=1.5, 95% confidence interval=1.2-1.8) within the previous year. In addition, dyspeptic patients are also more likely to receiving receive a first time diagnosis for chest pain (odds ratio=2.3, 95% confidence interval=2.0-2.8) or angina (odds ratio=2.7, 95% confidence interval=1.8-4.0) [10]. In an older study of 4,962 patients aged 18-61 who took part in the Rand Health Insurance Experiment, a decade-long randomized controlled trial of the effects of alternative methods of financing health care services, about 30% had one chronic illness, with an additional 16% having 2 or more. Ulcer-like symptoms, defined by a previous diagnosis along with taking antacids daily, frequent episodes of stomach pain relieved by milk, occurring one-half hour after eating or at night, was significantly associated with angina (p<0.05) and congestive heart failure (p<0.001)

A single study sought to assess the prevalence of dyspepsia among patients with atrial fibrillation [12]. The population (n=1,297) included a nationwide sample of American adults (from the 2009 National Health and Wellness Survey) with atrial fibrillation divided into two groups: those reporting dyspepsia (defined as any of the following: ulcers, abdominal bloating, abdominal pain, gastroesophageal disease or heartburn) and those who did not. Of these atrial fibrillation patients, 41% reported a diagnosis of a gastrointestinal condition while 34% reported a diagnosis of dyspepsia. Patients with dyspepsia were associated with a significantly higher mean CHADS2 score (1.9 vs. 1.4, p<0.05). Of note, while the CHADS2 score was developed as a tool to determine atrial fibrillation patients' risk for stroke, in this case, it can also serve as a marker of the presence of cardiovascular diseases since 2 of 5 CHADS2 criteria (eg, stroke and congestive heart failure) are in fact cardiovascular diseases and the remaining 3 criteria (eg, age, hypertension, diabetes) are potent risk factors for cardiovascular disease.

A retrospective database study sought to assess the risk of dyspepsia among patients with atrial fibrillation [13]. Analysis of insurance claims from the MarketScan® database from 2005-2009 was conducted. The population (n=413,168) included patients ≥18 years at the date of first atrial fibrillation diagnosis, with 180 days of continuous insurance coverage prior to the index atrial fibrillation diagnosis, and no gastrointestinal event within 180 days of the index atrial fibrillation diagnosis. The risk of dyspepsia was assessed with incidence rates (IRs; new dyspepsia case per patient years of observation). During a mean follow-up of 563 days, the IR

[11].

Finally in a small case series evaluating the relationship between dyspepsia and congestive heart disease or ischemic heart disease in Italian patients, data showed alterations of motility in the esophagus, stomach and duodenum in every cardiovascular disease patient evaluated and lesions of the gastric mucous membrane in more than half [14].

In addition to the aforementioned data suggesting upper gastrointestinal symptoms are more prevalent with patients with cardiovascular diseases; a body of literature suggesting upper gastrointestinal symptoms may in fact induce cardiovascular disease has begun to take shape [15-18]. In 2003, the first signal that gastro esophageal-like symptoms or disease could be linked to the development of atrial fibrillation was published [15]. Clinicians in Australia looked at 18 patients with concomitant diagnoses of lone paroxysmal atrial fibrillation and gastroeso‐ phageal reflux disease and noted that after treatment with a proton pump inhibitor to treat the upper gastrointestinal symptoms, 14 of 18 had a decrease or disappearance of at least one paroxysmal atrial fibrillation symptom.

Since that time, 3 observational studies [16-18] have more thoroughly evaluated this link. In a cohort study of 163,627 patients receiving care from the United States Army National Capitol Area Military Healthcare System between 2001 and 2007 (5% had atrial fibrillation and 29% had gastroesophageal-like symptoms), gastroesophageal symptoms were associated with an increased risk of atrial fibrillation, even after adjusting for age, sex, race and atherosclerotic risk factors (relative risk=1.19, 95% confidence interval=1.13-1.25) or further adjustment for ischemic heart disease, cardiomyopathy, atrial septal defect and being status post-cardiac bypass surgery (relative risk=1.08, 95% confidence interval=1.02-1.13) [16].

The second study [17] similarly sought to assess the relationship between gastroesophageal reflux disease and atrial fibrillation; and the researchers assessed the risk for atrial fibrillation over a follow-up period of greater than 11 years. A self-report survey was sent to 5,288 patients aged 25-74 over the 6 year period of 1988-1994. Of these patients, 741 developed atrial fibrillation. Contrary to the previous study, an inverse relationship with observed between gastroesophageal reflux disease symptoms and atrial fibrillation risk (hazard ratio=0.81, 95% confidence interval=0.68-0.96). However, the frequency of symptoms in those with gastroeso‐ phageal reflux (none, some, weekly, daily) was associated with an increased hazard of atrial fibrillation (p<0.01 for overall association); with daily symptoms associated with the highest hazard (hazard ratio=1.30, 95% Confidence interval=0.98-1.57) of developing atrial fibrillation compared to no gastroesophageal symptoms (p=0.07 unadjusted and p>0.2 after adjustment for confounders). The researchers cite an increase in medical attention in those experiencing gastroesophageal reflux as a possible explanation for the lack of association between the presence of symptoms and atrial fibrillation; hypothesizing that extra physician visits resulting from gastroesophageal symptoms resulted in early and more frequent identification and treatment of known atrial fibrillation risk factors, as well as a higher utilization of proton pump


enrolled patients had a diagnosis of atrial fibrillation (n = 86), and while hypertension, dyslipidemia or coronary artery disease were not associated with the prevalence of sympto‐ matic gastroesophageal reflux disease (defined as a total F-scale≥8 points) upon multivariate analysis, atrial fibrillation did show a significant correlation with gastroesophageal reflux disease (p<0.001). In addition, both the dyspeptic sub-score (p=0.018) and the total F-scale score (p=0.019) of atrial fibrillation patients were significantly greater than those in normal sinus

Upper Gastrointestinal Symptoms and Cardiovascular Disease

http://dx.doi.org/10.5772/56564

143

Recognizing patients with both cardiovascular diseases and upper gastrointestinal conditions is an important step in their medical care. As demonstrated in available evidence, the links

The World Health Organization, World Heart Federation [1] and the American Heart Associ‐ ation [3] each agree on a set of risk factors for the development of cardiovascular diseases. These risk factors include smoking, being overweight or obese, living a sedentary lifestyle, and poor diet, as well as having pre-existing diagnoses of high cholesterol, hypertension and

In addition to significantly contributing to the risk of developing cardiovascular disease, these same risk factors have also been found in epidemiologic studies to be associated with an increased risk of reporting upper gastrointestinal symptoms. These risk factors are highly

Below we discuss the mechanism behind, and studies supporting, the association between

Over a billion people worldwide are thought to be current smokers. It is estimated that nearly six million people die from tobacco-related deaths annually, and by 2030, this number is projected to surpass 8 million. Smoking is the underlying cause of about 10% of cardiovascular disease [1] and has been consistently found to be a strong and independent risk factor for myocardial infarction and sudden death [2]. Similar findings have been observed with cerebrovascular disease and smoking; with smokers having a 2 to 4 times increased risk of stroke compared with nonsmokers [2]. Consequently, it is not surprising that a large number of studies support the beneficial cardiovascular consequences of smoking cessation [1].

It is theorized that tobacco smoking/use induces upper gastrointestinal symptoms through its effects on the gastric mucosa [19]. The nicotine in tobacco likely causes mucosal injury by augmenting acid and pepsin release, causing duodenogastric reflux and producing free radicals; while at the same time decreasing prostaglandin and mucus production. Addition‐ ally, smoking may reduce lower esophageal sphincter pressure and thus accentuate gastroe‐

between the conditions are strong, and can impact therapeutic decisions.

prevalent both worldwide and in the United States [1,3].

these risk factors and increased rates of upper gastrointestinal symptoms.

rhythm.

diabetes.

**3. Shared risk factors**

**3.1. Current smoking**

sophageal-like dyspeptic symptoms.

\*Widely used questionnaire in Japan to screen for gastroesophageal reflux disease based upon frequency of 12 common symptoms

#Adjusted for age, sex, race, known atherosclerotic risk factors (hypertension, diabetes, hyperlipidemia, and tobacco use)

†Adjusted for strong correlates of AF: ischemic heart disease, cardiomyopathy, atrial septal defect, status post coronary bypass surgery

AF= atrial fibrillation; aRR= adjusted relative risk; GERD= gastroesophageal reflux disorder; HR= hazard ratio; PAF= paroxysmal atrial fibrillation; PPI= proton pump inhibitor; RR= relative risk; USA= United States of America

**Table 3.** Relationship Between Atrial Fibrillation and Gastroesophageal-Like Symptoms

inhibitors (although the researchers did not have data medication use to test this hypothesis). Finally, the most recently published study assessed the relationship between atrial fibrillation and gastroesophageal reflux disease in 188 Japanese patients between 28-91 years of age [18]. Patients' gastroesophageal reflux disease status was classified using the F-scale, a question‐ naire specifically designed to screen for gastroesophageal reflux disease. Almost half of enrolled patients had a diagnosis of atrial fibrillation (n = 86), and while hypertension, dyslipidemia or coronary artery disease were not associated with the prevalence of sympto‐ matic gastroesophageal reflux disease (defined as a total F-scale≥8 points) upon multivariate analysis, atrial fibrillation did show a significant correlation with gastroesophageal reflux disease (p<0.001). In addition, both the dyspeptic sub-score (p=0.018) and the total F-scale score (p=0.019) of atrial fibrillation patients were significantly greater than those in normal sinus rhythm.

Recognizing patients with both cardiovascular diseases and upper gastrointestinal conditions is an important step in their medical care. As demonstrated in available evidence, the links between the conditions are strong, and can impact therapeutic decisions.
