RESUMO
OBJECTIVE: Patients with atrial correction of transposition of the great arteries (TGA) may develop right ventricular (RV) failure with time. A reliable non-invasive method for assessment of ventricular function is therefore needed. To evaluate the accuracy of echocardiography in assessment of ventricular volumes and function in these patients we compared echocardiography with magnetic resonance imaging (MRI) in 10 patients late after the Mustard and Senning procedures. DESIGN: Prospective echocardiography and MRI examinations were performed on the same day. Two different echocardiography technicians examined all patients. All echocardiography and MRI examinations were performed at the university hospital outpatient clinic and MRI department respectively. Ten patients, age 14.0+/-2.9 years, who had been operated on with atrial correction of TGA at 8 (2-60) months of age (median and range) were examined. Echocardiography RV and left ventricular (LV) end-systolic volumes (ESVs), end-diastolic volumes (EDVs), stroke volumes (SVs) and ejection fractions (EFs) were calculated, using the modified Simpson method, and compared with the same measurements obtained from MRI. RESULTS: For RV function there was good agreement between echocardiography- and MRI-derived measurements. Both echocardiography and MRI revealed reduced RV function with EFs of 42.6+/-9.1% and 46.4+/-7.2% respectively. For RV volumes there were no significant differences between echocardiography and MRI. LV function was significantly overestimated by echocardiography (EF with echocardiography = 72.7+/-4.4% vs. MRI = 50.5+/-7.6%) while all LV volumes were greatly underestimated. Echocardiography measurements of volumes in repeated examinations by different technicians showed large variations, 13-50%, for different variables in individual patients. CONCLUSION: Echocardiography can provide clinically important information concerning RV function in follow-up of patients late after atrial correction of TGA. It has limited value in assessment of LV function in these patients. Volume measurements by echocardiography are, however, highly user-dependent and interobserver variation is high. MRI may accordingly serve as an important reference method in individual patients.