Feasibility of CT-derived myocardial strain measurement in patients with advanced cardiac valve disease

Figure 1. Flow chart of the patients in the study. 55 consecutive cardiac CTs, which were obtained for interventional planning of the cardiac valves were screened and 12 CTs were excluded due to either severe artifacts, bad contrast or incomplete coverage of the left ventricle. In the remaining 43 cardiac CTs myocardial strain was assessed. 7 of these 43 cardiac CTs had no analyzable echocardiography. In the remaining 36 examinations the myocardial strain measurements between echocardiography and cardiac CT were compared.

Figure 2. Example of strain measurement in SegmentCT (Medviso, Lund, Sweden). Images (A) and (B) show an example of strain measurement in a four-chamber view of the left ventricle with the corresponding strain curve for radial and longitudinal strain. Images (C) and (D) show an example of strain measurements in a short axis view of the left ventricle with the corresponding strain curves for radial and circumferential strain.

Figure 3. Scatterplot showing the correlation of CT-derived strain in different orientations and CT-derived LVEF. Pearson’s correlation coefficients for the different orientations are as follows: GLS: r = − 0.66 (p < 0.001), CS: r = − 0.53 (p < 0.001), long-axis RS: r = 0.37 (p = 0.02), short-axis RS: r = 0.49 (p < 0.001; GLS, global longitudinal strain; CS, circumferential strain; RS, radial strain; LVEF, left ventricular ejection fraction).

Figure 4. Scatterplots and Bland–Altman plots of global longitudinal strain as well as longitudinal strain in the two-, three- and four-chamber view. The Pearson’s correlation coefficient was as follows: global longitudinal strain: r = 0.60, p < 0.001; two-chamber view: r = 0.28, p = 0.0929; three-chamber view: r = 0.51, p = 0.0014; four-chamber view: r = 0.69, p < 0.001. (GLS, global longitudinal strain; LS, longitudinal strain; TTE, trans-thoracic echocardiography).