Contribution of cross-sectional echocardiography to the diagnosis of right ventricular dysplasia at the asymptomatic stage

The diagnostic accuracy of conventional exercise electrocardiographywas compared with exercise algorithms specifically aimed atapplication in females, with R wave and heart rate adjustedST-segment changes, and with discriminant analysis. A symptom-limitedbicycle exercise test was performed by 189 females without previousinfarction with a normal electrocardiogram (ECG) at rest. Franklead ECG was computer processed. ST-segment amplitudes in leadX were superior to measurements in lead Y, but their sensitivityamounted only to 50% at a specificity of 90%. Correction forR wave amplitude marginally increased sensitivity. It was confirmedthat leads with inferiorly directed vectors such as standardleads II, III, a VF or Frank lead Y should not be used for diagnosticpurposes in females. Exercise-induced increases in R-wave amplitudeof at least 01 mV occurred in only 8% of females with coronarydisease. Previously described discriminant function using attainedheart rate, ST-segment depression and a sex-dependent ST segmentcorrection factor yielded a sensitivity of 60% at a specificityof 90%. Sensitivities of heart rate adjusted ST-segment changesand of a previously proposed discriminant function were greaterthan 70% at a specificity of 90%. It is concluded that the diagnosticyield of exercise testing in females can be improved by useof more sophisticated electrocardio-graphic and exercise variables.