Clinical scenario

A 57 year old man presents with a one hour history of central chest pain the character of which he cannot describe. There is no radiation but there is mild sweating and subjective shortness of breath. He has a history of smoking, hypertension, angina, and a hiatus hernia; the pain came on after a curry. He has a normal ECG on admission and an unremarkable examination. You cannot decide whether this is cardiac or oesophageal in origin and wonder whether a single dose of antacid might relieve his pain and therefore clarify the diagnosis

Three part question

In [a patient with chest pain of uncertain aetiology] is [the use of antacids/alginates] useful for [differentiating between cardiac and gastro-oesophogeal causes]?

Search strategy

Medline 1966–12/02 using the OVID interface. [exp chest pain OR exp angina pectoris OR exp angina, unstable OR exp coronary disease OR exp myocardial infarction OR chest pain.af OR angina.af OR coronary.af OR myocardial.af OR cardiac.af OR myocard$.af] AND [exp alginates OR exp antacids OR algin$.af OR antacid.af OR antacids.af OR gaviscon.af OR algicon.af OR gastrocote.af OR peptac.af OR topal.af OR indigestion.af OR heartburn.af OR dyspepsia.af] LIMIT to human AND English language.

Search outcome

Altogether 374 papers were found only two of which directly addressed the three part question (table 2).

Comment(s)

Both studies are small, however in the Henderson paper if the 95% confidence intervals are calculated (81.5% to 100% typical, 9% to 30.2% atypical) there is a distinct difference between the two groups despite all the patients having normal ETTs and angiography. A further paper by Davies et al has shown that the instillation of acid into the stomach decreases the angina threshold on exercise testing. The above tests would suggest that reflux affects angina and vice versa. The vagus nerve has been suggested as the common link between the two.

CLINICAL BOTTOM LINE

Antacids are useful in the relief of pain that is clearly oesophageal in origin but the effect is insufficiently specific to be of value in aiding diagnosis.