Discriminative ability of conventional echocardiography and tissue Doppler imaging techniques for the detection of subclinical cardiotoxic effects of treatment with anthracyclines.

This study investigated improvement of diagnosing myocardial damage caused by anthracyclines using tissue Doppler imaging (TDI). The optimal set of conventional echocardiographic and/or TDI parameters, needed for the discrimination of survivors from healthy controls, was retrospectively assessed. A total of 60 patients and 99 controls, age range 8.5 to 17.6 years, were studied. The survivors received 50 to 400 mg/m(2) cumulative dose of anthracyclines, with a mean follow-up of 7.3 (+/-2.3) years. The parameters used in the discriminant score (S-score) were selected from a large set of 51 echocardiographic parameters, using logistic regression analysis (stepwise selection). The correct classification probability (C-index) and the generalized distance (d) between the distributions of S-scores were used to measure the overall discriminative performance of each echocardiographic technique separately and in combination. The overall discriminative performance of the conventional echo-Doppler parameters (C = 77.3%, d = 1.04) was lower than that of the TDI (C = 84.2%, d = 1.37); the highest C-index was obtained using both techniques (C = 89.2%, d = 1.66). The set of parameters includes: LV fractional shortening and MV early diastolic flow velocity, two long-axis and five apical 4-CV TDI wall velocities (systolic and diastolic). In the patient group, the S-score was positively associated with cumulative dose of anthracyclines (p = 0.05) and duration of treatment (p = 0.01). The diagnostic index S-score, based on a limited number of variables from both techniques simultaneously, could retrospectively discriminate asymptomatic children with anthracycline-induced cardiomyopathy from healthy controls. The potentials of the S-score for serial and prospective studies are further investigated.