RESUMO
OBJECTIVES: To assess the feasibility, accuracy, and reproducibility of tissue-tracking mitral annular displacement (TMAD) compared with other measures of left ventricular systolic function in healthy preterm and term neonates in the transitional period. METHODS: This was a prospective observational study. Two echocardiograms were performed at 24 and 48 hours of life. TMAD, shortening fraction (SF), ejection fraction (EF), s', and global longitudinal strain (GLS) were measured offline. Accuracy to detect impaired GLS was tested by ROC curve analysis. DeLong test was used to compare AUCs. Intra and interobserver reproducibility of the off-line analysis was calculated. RESULTS: Mean ± SD gestational age and weight were 34.2 ± 3.8 weeks and 2162 ± 833 g, respectively. TMAD was feasible in 168/180 scans (93%). At 24 hours the AUC (95% CI) of SF, EF, s', and TMAD (%) was 0.51 (0.36-0.67), 0.68 (0.54-0.82), 0.63 (0.49-0.77), and 0.89 (0.79-0.99) respectively. At 48 hours the AUC (95% CI) of SF, EF, s', and TMAD (%) was 0.64 (0.51-0.77), 0.59 (0.37-0.80), 0.70 (0.54-0.86), and 0.96 (0.91-1.00), respectively. The AUC of TMAD was superior to the AUC of SF, EF, s', at both timepoints (P < .02). Intraclass correlation coefficients (95% CI) of intra and interobserver reproducibility of TMAD were 0.97 (0.95-0.99) and 0.94 (0.88-0.97), respectively. CONCLUSION: TMAD showed improved accuracy and optimal reproducibility in neonates in the first 48 hours of life.
