Until recent years the negative deflection of the ST-segment of an ECG was almost the only means of interpreting exercise tests in the diagnosis of heart disease. Recent publications have suggested new criteria of positivity. Most of these studies have focused on the absence of a decrease in the amplitude of the R wave in V5 or CM5 derivation, but the results tend to disagree and suffer from lack of reproducibility thus rendering interpretation of this criterion uncertain. Claims have also been made for other criteria with a predictive value at least equal to that of the ST-segment, but very few studies have been devoted to these criteria. The absence of the Q wave in V5 (three studies on a total of 278 patients) has a sensitivity of 82 per cent and a specificity of 76 per cent. Inversion of the U wave in CC5 (one study on a total of 248 patients) has a sensitivity of 21 per cent and a specificity of 99 per cent. Traces of the pressure in abnormally increased recovery (a study on a total of 87 patients) has a sensitivity of 95 per cent and a specificity of 90 per cent. These results cannot replace those obtained with the ST-segment which relate to hundreds of studies on tens of thousands of patients. For this reason it is important to specify better the value of these new criteria and to assess their contribution in association with changes in the ST-segment.(ABSTRACT TRUNCATED AT 250 WORDS)