Management of Esophageal Chest Pain

G&H What is the prevalence of esophageal chest pain in the general population?

SRA Several studies have examined the prevalence of esophageal chest pain. In the United States, Dr. Locke and associates at the Mayo Clinic in Rochester, Minnesota, found that 23% of the US population suffers from noncardiac chest pain (NCCP). In other countries, similar numbers of patients have been reported to have NCCP, for example 33% of the population in Australia, 24% in Argentina, between 8–28% in Spain, and approximately 21% in south China, making this an international problem.

G&H What are the causes of esophageal chest pain, and how can they be related to gastroesophageal reflux disease and differentiated from heartburn?

SRA Gastroesophageal reflux (GER) is both the best-studied and the most common cause of esophageal pain (here, I have been using the term NCCP to refer to this concept); anywhere between 22–60% of patients with NCCP have GER.

The esophagus has a limited language by which it communicates, and some patients perceive acid reflux as chest pain, whereas other patients perceive it as heartburn. Physicians often ascribe heartburn as a classic symptom of reflux, whereas chest pain may or may not necessarily be associated with reflux. If there is confusion about the origin of the symptoms, an extended pH examination of the esophagus, either with the conventional wire system or the Bravo wireless system, may help clarify the presence or absence of GER. Although pH testing is considered the gold standard for the diagnosis of GER, a negative pH study does not necessarily exclude GER. Alternatively, studies have suggested that a high-dose proton pump inhibitor (PPI) trial for 1 week induces a response rate of close to 80% in patients who suffer from GER-related chest pain.

Other causes of NCCP include esophageal motility disorders and visceral hyperalgesia. Furthermore, patients with NCCP should have other nonesophageal sources of chest pain excluded, such as musculoskeletal conditions, pulmonary and pericardial disorders, and other gastrointestinal problems such as ulcers, pancreatic diseases, and biliary tract diseases.

Finally, clinicians must also keep in mind that a number of these patients may have psychiatric conditions such as panic disorder, depression, and anxiety that may play a role in their chest pain.

G&H What are the presenting symptoms of NCCP, and how can they be differentiated from cardiac chest pain?

SRA The dominant symptom, of course, is that of chest pain, which is so similar to cardiac pain that physicians often find it challenging to distinguish the two. This was illustrated in a study in which board-certified cardiologists were provided with a clinical history of patients presenting for evaluation of chest pain and who had undergone coronary angiography. These cardiologists were asked to predict the results of coronary angiography based on clinical history alone (they were kept unaware of the results of cardiac catheterization). Among these highly skilled physicians, 25% incorrectly diagnosed the presence or absence of coronary artery disease in these patients. This study underscores the challenge that physicians have in attempting to differentiate cardiac pain from NCCP. At the same time, this study also underscores the importance of excluding cardiac sources of chest pain objectively (ie, via cardiac studies as deemed appropriate by a cardiologist) before making the assumption that the chest pain is not cardiac in nature.

There have been at least two publications from insurance carriers about malpractice lawsuits of physicians who erroneously attributed chest pain to the esophagus, neglecting to utilize cardiac examinations and ignoring the possibility that the pain could be caused by the heart. Thus, before assuming that pain is caused by the esophagus, clinicians must objectively establish the absence of cardiac disease.

A few features tend to be more suggestive of esophageal pain than of cardiac pain. For instance, esophageal pain may last hours (whereas cardiac pain tends to last only minutes) and may follow food intake (whereas cardiac pain is precipitated by exercise). Patients with esophageal pain may complain of heartburn, regurgitation, or dysphagia. Nevertheless, there is potential for overlap, and I would remain concerned about excluding heart disease. In addition, some patients may have coexisting heart disease and esophageal disease, making diagnosis even more difficult.

G&H Beyond differentiating it from cardiac chest pain, how is esophageal chest pain diagnosed?

SRA Once we have excluded cardiac sources for the chest pain, we must exclude other sources of pain before attributing it to the esophagus. In addition to life-threatening sources such as heart disease, other conditions should be ruled out appropriately, including pericardial disease, mitral valve prolapse, pulmonary-pleural pericardial diseases, musculoskeletal causes of chest pain, and even other nonesophageal gastrointestinal sources of chest pain such as peptic ulcer disease, pancreatic, or biliary-gallbladder disease. Once we are comfortable that all other sources for the chest pain have been excluded, we can make the diagnosis that the pain is esophageal in nature.

G&H What are the medical therapy options for treatment of esophageal chest pain?

SRA As the most common cause of esophageal pain is GER, a PPI trial is frequently prescribed even as an empirical approach and before pH testing or other gastrointestinal studies are performed. Dr. Fass and his group have shown that a 1-week empirical PPI trial, which they termed “the PPI test,” results in an improvement in close to 80% of patients with gastroesophageal reflux disease–related chest pain. This test also helps to separate out those patients who have reflux as a source of their chest pain from those who do not. It was also found to be a cost-effective strategy. In a longer, 8-week trial, my colleagues and I also showed that PPI therapy was efficacious for the treatment of esophageal chest pain compared with placebo in approximately 80% of the patients. Given the safety profile and wide availability of PPIs, most gastroenterologists and primary care practitioners utilize PPIs as first-line therapy once cardiac and other major noncardiac sources of chest pain have been excluded.

If patients do not respond to PPI therapy, the second option that most gastroenterologists explore is pain modulators, particularly tricyclic antidepressants. Two tricyclic antidepressants have been examined objectively in double-blind, randomized, placebo-controlled studies: imipramine, in a study by Dr. Cannon at the National Institutes of Health, and trazodone, in a study by Dr. Clouse. Both medications have been shown to be efficacious. They are administered in low doses (25–50 mg) at nighttime and titrated upward to about 100 mg or even 150 mg. The intent is to not necessarily reach the higher antidepressant doses used by psychiatrists. The perceived mechanism of action of these agents is thought to involve blunting afferent pain transmission, but the intimate details of how this occurs is not well understood. Their side effects may include dry mouth, anticholinergic effects, prostatic retention, and increased intraocular pressure. Trazodone has also been found to cause priapism in male patients.

As tricyclic antidepressants can be regarded as “dirty molecules” that act on a variety of neurotransmitters, we do not know which part of these compounds cause the improvement of pain. Tricyclic antidepressants block different substances such as serotonin, ephinephrine, and histamine receptors, making it difficult to determine which one(s) are responsible for the beneficial effects.

Recently, serotonin receptor antagonists have shown good responses, suggesting that the beneficial effect of tricyclic antidepressants may derive from blocking serotonin. The best support for this comes from a study conducted by Dr. Varia and colleagues at Duke University in 2000, in which sertraline was found to be efficacious in comparison to placebo in a group of patients with chest pain. In addition, an acute study using balloon distention in the laboratory found that another selective serotonin reuptake inhibitor (SSRI), namely citalopram, decreased the perception of pain post–balloon distention, again emphasizing the potential role of serotonin receptor antagonists in blocking pain perception.

If tricyclic antidepressants do not work, gastroenterologists may resort to SSRIs, particularly sertraline, as it has been studied in a controlled trial, albeit of a small sample size. The other SSRI-class drugs may need to be examined, as they may potentially have benefits that have not yet been found.

With regard to muscle relaxants, agents such as nitro-glycerin, calcium channel blockers, and anticholinergic compounds do not work as well as might be expected. Calcium channel blockers and anticholinergic agents, particularly, have shown mixed efficacy, and there have not been any solid, controlled trials showing efficacy versus placebo in the treatment of NCCP. Side effects may also limit the use of muscle relaxants. Nitroglycerin may still have a potential role for some patients who have intermittent severe episodes of squeezing chest pain; they may respond to acute uses of nitroglycerin, particularly in the setting of nonreflux esophageal spasm.

G&H What role do adenosine receptors, particularly theophylline, play as part of the medical treatment options for esophageal chest pain?

SRA Recently, Dr. Rao and collaborators from the University of Iowa published a very interesting, provocative study in the American Journal of Gastroenterology showing that the use of theophylline, a nonspecific adenosine receptor antagonist, is beneficial in NCCP. They noted a decrease in pain perception following acute esophageal balloon distension studies in which theophylline was compared to placebo. Then, during a 4-week crossover trial consisting of randomized assignment of oral theophylline 200 mg twice daily to a small number of patients, they also found significant differences in response when compared to placebo. This study is exciting because for the first time it opens up a new “frontier,” as I called it in an editorial accompanying the article, of potential new medications that may act on the adenosine receptors for the treatment of esophageal chest pain. Theophylline has a very narrow margin of therapeutic benefit. Its side effect profile has been shown to be problematic in the treatment of pulmonary disorders. Although the results of this study will not likely lead to the prescription of theophylline for esophageal chest pain, the study may open the door for further research of newer, redesigned adenosine receptors. Indeed, there are some new agents in the research arena and other similar pain receptors (both agonists and antagonists) that are being examined further.

G&H What are the endoscopic and surgical treatment options for esophageal chest pain?

SRA There have been two small studies focusing on endoscopic therapy with botulinum toxin (Botox, Allergan), a study conducted by Dr. Miller at Temple University examining 29 patients and a European study of 9 patients, both of which have demonstrated some benefit with selected patients with esophageal motility disorders that were not achalasia-related. The problem with these two studies lies not only in the small sample size but also in the fact that there was no placebo control, making it difficult to determine what role, if any, botulinum toxin has in the treatment of NCCP. Larger, controlled trials are sorely needed to determine whether this agent is really of benefit in NCCP.

Similarly, there have been some small studies examining endoscopic techniques for the treatment of atypical forms of gastroesophageal reflux disease such as NCCP and other extraesophageal forms of GER. Most studies have been small, short term, and not placebo-controlled. Thus, these data must be confirmed critically in a double-blind, controlled manner before justifying the use of these techniques for the treatment of esophageal chest pain. The long-term durability of these techniques also requires evaluation.

As for surgery itself, there have been a few studies investigating the impact of antireflux (mainly laparoscopic) surgery on patients with atypical forms of reflux including chest pain. The response rate has been approximately 50%. Thus, I am not convinced that we are ready to embrace these techniques for the widespread management of NCCP yet.

G&H You mentioned earlier that psychiatric disorders may play a role in patients with NCCP. What special considerations do these patients require?

SRA This is another challenging part of the conundrum of NCCP. Patients may have coexisting psychopathology such as depression, anxiety, somatization, or panic disorders. Patients not responding to the therapies mentioned above or those who present with obvious psychiatric symptoms must be referred to a psychiatrist. Patients tend to resist referral, but it is important to empathetically explain to these patients the need to address their psychiatric disorder.