Abstract 13652: Less is More: Use of a Sentinel Segment Approach to Strain Analysis in Stress Echocardiography
Backgrounds: Strain imaging has been applied as an adjunct to subjective interpretation during stress echo (SE). However, it is time consuming to measure all segments, and not all segments have equal feasibility and accuracy. Limited previous work has shown sentinel segments in the apical septum, basal lateral and mid inferior to correlate with LAD, LCx and RCA disease. We sought to compare a sentinel segments to full segmental analysis during SE.
Methods: In this prospective study, 40 patients (32 men, 60±10y) underwent SE and coronary angiography. The change of end-systolic and peak systolic strain and peak systolic strain rate before and peak stress (ΔESS, ΔPSS and ΔPSSR) were analyzed offline using 2DS software (EchoPAC PC, GE Medical) in all segments and sentinel segments. Results were compared with presence of 50% and 70% diameter stenosis at angiography. Areas under the receiver-operating-characteristic curves (AUCs) were used to compare accuracies.
Results: Feasibility of all and sentinel segments were both 97%. There were 16 ischemic segments with >50% stenosis and 6 with >70%. In sentinel segments analysis, the AUC of ΔESS (0.73) exceeded peak ESS (0.63) and ΔPSS (0.73) exceeded peak PSS (0.65). In regional analysis, the AUCs, sensitivity and specificity detecting 70% stenosis were higher than those of >50% (Table). The AUCs for sentinel and all segment analyses using ΔPSS were 0.73 and 0.53 (>70% stenosis) and 0.69 and 0.62 (>50%). Sensitivities and specificities detecting patients with > 70% diameter stenosis were 100% and 53% and those of all segments were 83% and 56%.
Conclusions: Sentinel segment analysis of SE is feasible and more accurate than all segments analysis. The change in deformation indices is more sensitive than the peak values.
- © 2011 by American Heart Association, Inc.