Dobutamine stress echocardiography: improved endocardial border definition and wall motion analysis with tissue harmonic imaging.

BACKGROUND: We performed a study to determine whether tissue harmonic imaging (THI) facilitates wall motion analysis at rest and whether these benefits extend through the stages of a dobutamine stress echocardiography (DSE) study. We also assessed the impact of THI on the feasibility of DSE in technically difficult patients. Finally we tested the hypothesis that THI by improving endocardial border definition (EBD) could enhance the interobserver agreement between trainees and experienced operators for interpreting DSE studies.
METHODS: Twenty unselected patients underwent DSE by standard protocol. Parasternal and apical views were obtained with the use of fundamental mode (FND) and THI at baseline, low dose, and peak stress. Segmental EBD was characterized as 1 to 4 (1 = excellent) and segmental wall motion was characterized as 1 to 4/x (1 = normal, x = unable to interpret) by a consensus of 2 experienced observers. A trainee in stress echocardiography independently scored all segments, and these results were compared with the consensus of the experienced readers.
RESULTS: EBD improved with THI in 26 +/- 6.7 of 48 segments per patient (54%, 95% confidence interval [CI] 0.40 to 0.68) and deteriorated with THI in only 2 +/- 2.7 (4%, 95% CI 0 to 0.09). Of the total of 48 segments per patient, a mean of 10 +/- 5.7 (21%, 95% CI 0.10 to 0.31) were of inadequate quality to be interpreted for wall motion on FND, and this changed to 4 +/- 3.4 (6%, 95% CI 0.06 to 0.12) on THI (P <.001). EBD improved in a similar degree in all DSE stages 53%, 54%, and 53% for rest, low dose, and peak stress, respectively. Six of the 20 study patients were deemed unsuitable for DSE on FND, and all were changed to suitable subjects on THI. Of the 205 segments deemed unsuitable for interpretation on FND, 140 (68%) were of the anterior and lateral walls of the LV. Improvement with THI was also more prominent on these walls. The mean coefficient of agreement (kappa) for wall motion analysis was 0.82 +/- 0.14 on FND and improved to 0. 92 +/- 0.09 on THI (P <.001).
CONCLUSIONS: THI dramatically improves EBD and the ability to confidently score segmental wall motion. Interobserver agreement is also significantly enhanced. These benefits extend to the peak stage of a DSE study. Routine use of THI may enhance the diagnostic accuracy of DSE and extend its application to technically difficult patients previously deemed unsuitable.