RSS Feeds

Email a Friend Printer Friendly

Current diagnostic criteria missing more than 25% of heart attacks

November 18, 2003

The UK's reliance on old criteria to diagnose heart attacks among patients with severe chest pain suggests that more than 25% of cases are being missed, reveals a study in Heart.

In 2000 the European Society of Cardiology and the American College of Cardiology recommended the use of rises in a set of proteins called troponins for the detection of heart attack, on the grounds that these are more reliable.

But the UK and several other countries continue to use ECGs and conventional enzyme tests for diagnosis.

The wisdom of this approach is questioned by a research team from the University of Warwick, who compared the accuracy of heart attack diagnosis in an unselected group of patients with chest pain, who were treated at one teaching hospital, using both the conventional and new criteria.

Over a period of six months, all 401 patients with chest pain for at least 2.5 hours were assessed for heart attack, using current criteria, plus CK-MB (deemed an acceptable alternative to conventional enzyme analysis) and one of the troponin proteins (troponin T). An expert panel then re-assessed the patients, who were subsequently monitored for six months after admission.

The researchers found that a full assessment depended on who admitted the patient. In almost four out of 10 patients, doctors failed to take the full set of samples. This was twice as likely to happen when patients were admitted by doctors other than cardiologists.

The expert panel found that in 4% of cases, patients had been wrongly diagnosed, using the old criteria. Almost all of these cases had been admitted by doctors who were not cardiologists.

But when the new criteria were applied, a further 27% of heart attack cases were identified. And the diagnosis had to be changed in 12% of patients.

Over half of the changed diagnoses were attributable to the diagnostic inaccuracy of the conventional markers. But the rest were down to potentially avoidable causes.

These included inappropriate interpretation of the results and a failure to take the full range of samples. The authors speculate that inadequate training, staffing, supervision, or the patient being moved to different wards might have accounted for these findings.

The researchers found that troponin T was better than any of the other chemical markers for predicting a patient's prognosis six months after discharge, and that CK-MB was as effective, but only if used in conjunction with troponin T.

The authors comment that the old criteria are unreliable, time consuming, and carry a higher risk of error. The new criteria are not only more accurate, they say, but are also simpler and quicker to use. Their prognostic accuracy could help shorten the inpatient stay.

The authors also point out that the differences in care between specialist and general medical treatment reinforce the government's recommendations issued in its national guidance on coronary heart disease (NSF). These stipulate that high risk patients, including those with sudden and severe chest pains, should be managed by heart specialists.

British Medical Journal (BMJ)