P718Prediction of coronary stenosis during dobutamine stress echocardiography based on three quantitative methods


    loading  Checking for direct PDF access through Ovid

Abstract

Background: Doppler myocardial imaging (DMI), both Pulsed-wave (PW) and Colour-coded (CC) and Speckle tracking imaging (STI) are three methods demonstrated to be useful in determining induced ischaemia during dobutamine stress echocardiography (DSE).Aim: To compare the value of myocardial velocity, strain and strain rate values simultaneously measured by PW-DMI, CC-DMI, STI to predict angiographic coronary stenosis during DSE.Methods: We have investigated 155 patients (age 61.89+/-9.36) without previous MI with mean WMSI at rest 1,06+/-0,12 and mean LV EF 54,08+/-2,37%. Each participant underwent DSE and coronary angiography in ≤ 2 months. Systolic, post-systolic and diastolic velocities, strains and strain rates at rest and during stress were obtained. Quantitative parameters of 9 myocardial segments per patient were analyzed, 1321 segments at all. Results: Normal coronary arteries were found to supply 900 segments, arteries stenosed ≥ 50%, 70% and 90% supplied 421, 257 and 131 segments, respectively. Initial ROC analysis has shown that 45, 52 and 35 quantitative parameters were eligible for predicting of ≥ 50%, 70% and 90% stenoses, respectively (AUC was significantly higher than 0.5). Combining discriminated parameters by quantitative method logistic regression with forward selection was applied and PW-DMI, CC-DMI and STI models were constructed. For either stenosis degree PW-DMI model did not demonstrate the superiority to other techniques but was more time consuming. For predicting stenosis ≥ 50% CC-DMI and STI models showed equal ability (AUC 0.656 vs 0.631, p ns), as well as for predicting stenosis ≥ 70% CC-DMI and STI models did not differ significantly (AUC 0.676 vs 0.695, p ns, respectively). CC-DMI model for ≥ 70% stenosis includes four parameters at stress (early diastolic velocity, systolic velocity, systolic strain and time to post-systolic strain) and one rest parameter: post-systolic velocity. STI model for ≥ 70% stenosis includes early diastolic velocity and systolic strain at rest, and relative changes in post-systolic velocity, post-systolic strain and transverse post-systolic strain rate during stress, as well as absolute change in systolic velocity during stress.Conclusions: Models based on three quantitative methods for prediction of stenoses during DSE are created. Our findings show that PW-DMI method does not demonstrate superior predictive ability for significant stenosis compared to CC-DMI and STI techniques. Suggested CC-DMI and STI models are equally informative for the prediction of angiographic coronary stenosis ≥ 50% and ≥ 70% at particular myocardial segment.

    loading  Loading Related Articles