ABSTRACT
BACKGROUND: In patients admitted to outpatient clinics with chest pain, changes in the ST-segments of electrocardiogram (ECG) readings are the most widely used criteria during treadmill ECG tests to determine myocardial ischaemia, despite its poor accuracy. In this study, we evaluated the benefit of combining elongation of P-wave duration (Pdur) and abnormal heart rate recovery (HRR) parameters in addition to changes in the ST-segments for the detection of myocardial ischaemia with treadmill ECG testing. METHODS: Patients (n = 369) with chest pain who underwent both a treadmill ECG test and myocardial perfusion scintigraphy (MPS) were enrolled. P-wave duration was measured at rest and at the end of the first minute of the recovery phase and elongation of the P-wave was calculated. Abnormal HRR was defined as the failure of a decreasing HR at the end of the first minute of the recovery phase >10% of the maximum HR reached during treadmill ECG testing. The sensitivity, specificity, positive and negative predictive values, diagnostic accuracy values, and likelihood ratios (LRs) of changes in the ST-segments, P-wave elongation, abnormal HRR, and the combination of these three variables for predicting myocardial ischaemia detected by MPS, were calculated separately-in patients without previous coronary artery disease (CAD) and in those with CAD. RESULTS: Elongation of Pdur by 20 ms or longer and abnormal HRR during treadmill ECG test were more common in patients with reversible perfusion defects in MPS than in those without perfusion defects (both P < 0.001). When patients were divided into two groups according to the presence or absence of a history of CAD, the addition of elongation of Pdur 20 ms and abnormal HRR to the development of significant changes in the ST-segments detected myocardial ischaemia with 46.7% sensitivity, 97.8% specificity, 67.2% negative predictive value, 88.9% positive predictive value, and 70% diagnostic accuracy in 77 patients with previous CAD. The LR+ of the combination of the three variables was 12.27. The addition of elongation of Pdur 20 ms and abnormal HRR to the development of significant changes in the ST-segments detected myocardial ischaemia with 52.7% sensitivity, 98.7% specificity, 89.9% negative predictive value, 90.6% positive predictive value, and 90% diagnostic accuracy in 292 patients without previous CAD. The LR+ of the combination of the three variables to detect myocardial ischaemia was calculated to be 41.48 in patients without a history of CAD. CONCLUSIONS: The addition of elongation of Pdur and abnormal HRR to the conventionally used criterion of changes in the ST-segments in patients with suspected myocardial ischaemia increased the specificity and positive and negative predictive values of treadmill ECG testing for detecting myocardial ischaemia, which might reduce the need for other expensive noninvasive techniques. The diagnostic utility of adding these two parameters was more obvious in patients with no history of CAD.