Editor,—Yamagishi et al, studying 37 patients within seven weeks of Q wave myocardial infarction (MI), found that exercise induced ST segment elevation was closely associated with the presence of viable myocardium in the infarct territory.1 We also studied this in patients with previous MI and agree with the results2 ,3; however, viable myocardium may persist for a long time after an MI,4 and in these cases ST segment shift is not considered a specific indicator of transmural ischaemia and viability.

To increase the specificity of this sign in patients with an old (> 6 months) MI, we introduced an unconventional, but experimentally validated,5 ECG marker of transmural ischaemia—the stress induced shortening of QTc interval (QT interval corrected for heart rate using Bazett's formula) in Q wave leads—to identify hibernating myocardium in the infarct zone. Experimental studies demonstrated an increase in cellular K⁺ efflux at the onset of myocardial ischaemia accompanied by a progressive shortening of the action potential duration.5

We evaluated 15 consecutive patients (group A) with previous anterior MI presenting with the following: ST segment elevation over Q waves during exercise testing; critical stenosis (⩾ 75%) of the left anterior descending coronary artery (LAD); cross sectional echocardiography and stress–redistribution–reinjection²⁰¹Tl myocardial scintigraphy of viable myocardium in the infarct zone (akinetic segments with normal echoreflectivity plus ⩾ 7mm end diastolic wall thickness and significant²⁰¹Tl redistribution after reinjection (> 50% of the reference myocardium in any scan6)).

The control group (group B) comprised 15 patients with previous anterior MI, critical stenosis of the LAD, and evidence of scar (increased echoreflectivity, associated to < 6 mm end diastolic wall thickness, and no ²⁰¹Tl redistribution) in infarcted areas.

Groups A and B were patients selected at random early or late (> 6 months) after their first anterior MI.

QTc interval was measured at rest and peak stress in leads showing ST segment shift, and the lead by lead fractional difference between the QTc intervals (ΔQTc) was calculated. The ΔQTc was measured again during exercise testing in 11 patients of group A (group A1) who had significant contractility recovery in akinetic areas (83% of akinetic segments) three months after myocardial revascularisation. We considered significant QTc interval shortening as ΔQTc < −10%. Data are presented as mean (SD).

There was no significant difference between patients in group A, B, and A1 (before and after revascularisation) regarding age, sex, number of pathological Q waves in resting ECG, exercise duration, exercise induced maximal workload, maximal heart rate, peak blood pressure, or maximal rate–pressure product.

ST segment elevation over Q waves at rest was higher in group B than in group A (1.8 (0.5) v 0.57 (0.4) mm) (p < 0.001).

All groups had exercise induced ST segment elevation over Q waves, but maximal elevation was significantly higher in group A than group B (2.5 (1.4) v 1.8 (1.1) mm) (p < 0.05) and in patients of group A1 before revascularisation (3 (1.03)v 1.1 (0.1) mm) (p < 0.05).

ΔQTc was significantly shorter in group A than group B (−18.1 (8.5)v −4.2 (7.8)%) (p < 0.0001). Indeed a significant ΔQTc shortening was measured in 14 of 15 patients of group A and only in one of group B (sensitivity 93.3%; specificity 93.3%; p < 0.0001). No group A1 patient had significant ΔQTc shortening in Q wave leads after revascularisation (ΔQTc of group A1 after revascularisation was +6.9 (14.8)%).

ΔQTc shortening in Q wave leads presenting exercise induced ST segment elevation, was a “cheap” ECG marker of transmural ischaemia and, indirectly, of myocardial viability as defined by echocardiographic and radionuclide variables, and confirmed by the results of revascularisation. This sign was no more evident after complete revascularisation and could be helpful in identifying hibernating myocardium even late after an MI.