Echocardiography vs Cardiac MRI in Mitral Regurgitation
Echocardiography vs Cardiac MRI in Mitral Regurgitation
All enrolled patients underwent CMR within a period of 1 month after 2DTTE, in 2 patients CMR followed 2DTTE within 2 months but follow-up 2DTTE confirmed stable LV dimensions in these patients. 3 patients had limited image quality on CMR and were excluded from further analysis. The baseline characteristics of the remaining study population are summarised in Table 1. Among the 38 patients included, 79% were males and the origin of MR was predominantly degenerative (92%). Severe MR, as assessed by the PISA method, was found in 45% of patients.
Comparison of measurement by 2D TTE and CMR regarding LV dimensions and parameters of MR severity are represented in Table 2. Measurements of linear LV dimensions by both imaging modalities were statistically similar (LV end-diastolic dimension: 53 ± 6 mm by transthoracic echocardiography vs. 53 ± 8 mm by CMR, p = 0.91; average bias -0.1 mm, 95% confidence interval -8.9 to +8.7 mm and LV end-systolic dimension: 36 ± 5 mm vs. 36 ± 6 mm, p = 0.95; average bias 0 mm, 95% confidence interval -5.7 to +5.7 mm). LV volumes were overall significantly underestimated by modified Simpson's method in comparison with CMR, although LV end-diastolic and end-systolic volumes showed a strong correlation by Pearson correlation analysis (Figure 3A and C). Bland-Altman analysis confirmed general underestimation by 2D TTE of both LV end-diastolic volume (average bias +28 ml, 95% confidence interval -53 to +109 ml) and LV end-systolic volume (average bias +20 ml, 95% confidence interval -25 to + 66 ml) (Figure 3B and D). The LVEF was significantly overestimated by 2D TTE in comparison with CMR and there was no significant correlation between LVEF measurements of the two imaging modalities.
(Enlarge Image)
Figure 3.
Comparison of measurements of LV end-diastolic and end-systolic volumes by 2D TTE and CMR. A. Measurement of LV end-diastolic volumes by both imaging methods showed a strong correlation by Pearson correlation analysis (r = 0.81, p < 0.00001). B. Bland-Altman analysis indicated general underestimation of the LV end-diastolic volume by 2D TTE in comparison with CMR (average bias +28 ml, 95% confidence interval −53 to +109 ml). C. Measurement of LV end-systolic volumes by both imaging methods showed as well a good correlation (Pearson r = 0.7, p < 0.00001). D. Bland-Altman analysis revealed general underestimation of the LV end-systolic volume by 2D TTE compared to CMR (average bias +20 ml, 95 % confidence interval −25 to +66 ml).
Measurement of the ARO was feasible in 21 out of 22 patients (95%). ERO calculated by the PISA method with 2D TTE was similar to ARO measured by planimetry with CMR (47 ± 24 vs. 42 ± 16 mm, p = 0.12). Pearson correlation analysis showed a strong correlation between both imaging techniques of the regurgitant orifice (r = 0.76, p < 0.0001) (Figure 4A). Furthermore Bland-Altman analysis showed good agreement between 2D TTE PISA method and CMR planimetry of the ARO (average bias -5.7 mm, 95% confidence interval -37 to +26 mm), less accordance was observed in some patients with an ERO ≥50 mm (Figure 4B). In addition, the intraclass correlation coefficient (=0.7, p = 0.0001) confirmed strong agreement between both methods to quantify the regurgitant orifice. By contrast, RVol calculated by either of the two 2D TTE methods was significantly higher in comparison with RVol obtained by phase contrast velocity mapping by CMR (PISA method vs. CMR: 69 ± 38 vs. 39 ± 27 ml, p = 0.001; r = 0.45, p = 0.07; Doppler volumetric method vs. CMR: 67 ± 33 vs. 28 ± 16 ml, p = 0.003; r = -0.14, p = 0.6). Assessment of regurgitant volume by PISA and Doppler volumetric method by 2D TTE was similar (74 ± 40 vs. 67 ± 33 ml, p = 0.6). Of note, 20% of patients with moderate MR according to ERO as assessed by the PISA method were reclassified as severe MR according to planimetry of ARO, likewise 20% of patients with severe MR by 2D TTE were reclassified as moderate MR by CMR.
(Enlarge Image)
Figure 4.
Comparison of ERO measured by 2D TTE and ARO measured by CMR. A. Pearson correlation analysis showed a strong correlation between 2D TTE PISA method and CMR planimetry of the ARO (r = 0.76, p < 0.0001). B. Bland-Altman analysis showed good agreement between both imaging techniques of the regurgitant orifice (average bias −5.7 mm, 95% confidence interval −37 to +26 mm), less accordance was observed in some patients with an ERO ≥50 mm².
Results
All enrolled patients underwent CMR within a period of 1 month after 2DTTE, in 2 patients CMR followed 2DTTE within 2 months but follow-up 2DTTE confirmed stable LV dimensions in these patients. 3 patients had limited image quality on CMR and were excluded from further analysis. The baseline characteristics of the remaining study population are summarised in Table 1. Among the 38 patients included, 79% were males and the origin of MR was predominantly degenerative (92%). Severe MR, as assessed by the PISA method, was found in 45% of patients.
Comparison of measurement by 2D TTE and CMR regarding LV dimensions and parameters of MR severity are represented in Table 2. Measurements of linear LV dimensions by both imaging modalities were statistically similar (LV end-diastolic dimension: 53 ± 6 mm by transthoracic echocardiography vs. 53 ± 8 mm by CMR, p = 0.91; average bias -0.1 mm, 95% confidence interval -8.9 to +8.7 mm and LV end-systolic dimension: 36 ± 5 mm vs. 36 ± 6 mm, p = 0.95; average bias 0 mm, 95% confidence interval -5.7 to +5.7 mm). LV volumes were overall significantly underestimated by modified Simpson's method in comparison with CMR, although LV end-diastolic and end-systolic volumes showed a strong correlation by Pearson correlation analysis (Figure 3A and C). Bland-Altman analysis confirmed general underestimation by 2D TTE of both LV end-diastolic volume (average bias +28 ml, 95% confidence interval -53 to +109 ml) and LV end-systolic volume (average bias +20 ml, 95% confidence interval -25 to + 66 ml) (Figure 3B and D). The LVEF was significantly overestimated by 2D TTE in comparison with CMR and there was no significant correlation between LVEF measurements of the two imaging modalities.
(Enlarge Image)
Figure 3.
Comparison of measurements of LV end-diastolic and end-systolic volumes by 2D TTE and CMR. A. Measurement of LV end-diastolic volumes by both imaging methods showed a strong correlation by Pearson correlation analysis (r = 0.81, p < 0.00001). B. Bland-Altman analysis indicated general underestimation of the LV end-diastolic volume by 2D TTE in comparison with CMR (average bias +28 ml, 95% confidence interval −53 to +109 ml). C. Measurement of LV end-systolic volumes by both imaging methods showed as well a good correlation (Pearson r = 0.7, p < 0.00001). D. Bland-Altman analysis revealed general underestimation of the LV end-systolic volume by 2D TTE compared to CMR (average bias +20 ml, 95 % confidence interval −25 to +66 ml).
Measurement of the ARO was feasible in 21 out of 22 patients (95%). ERO calculated by the PISA method with 2D TTE was similar to ARO measured by planimetry with CMR (47 ± 24 vs. 42 ± 16 mm, p = 0.12). Pearson correlation analysis showed a strong correlation between both imaging techniques of the regurgitant orifice (r = 0.76, p < 0.0001) (Figure 4A). Furthermore Bland-Altman analysis showed good agreement between 2D TTE PISA method and CMR planimetry of the ARO (average bias -5.7 mm, 95% confidence interval -37 to +26 mm), less accordance was observed in some patients with an ERO ≥50 mm (Figure 4B). In addition, the intraclass correlation coefficient (=0.7, p = 0.0001) confirmed strong agreement between both methods to quantify the regurgitant orifice. By contrast, RVol calculated by either of the two 2D TTE methods was significantly higher in comparison with RVol obtained by phase contrast velocity mapping by CMR (PISA method vs. CMR: 69 ± 38 vs. 39 ± 27 ml, p = 0.001; r = 0.45, p = 0.07; Doppler volumetric method vs. CMR: 67 ± 33 vs. 28 ± 16 ml, p = 0.003; r = -0.14, p = 0.6). Assessment of regurgitant volume by PISA and Doppler volumetric method by 2D TTE was similar (74 ± 40 vs. 67 ± 33 ml, p = 0.6). Of note, 20% of patients with moderate MR according to ERO as assessed by the PISA method were reclassified as severe MR according to planimetry of ARO, likewise 20% of patients with severe MR by 2D TTE were reclassified as moderate MR by CMR.
(Enlarge Image)
Figure 4.
Comparison of ERO measured by 2D TTE and ARO measured by CMR. A. Pearson correlation analysis showed a strong correlation between 2D TTE PISA method and CMR planimetry of the ARO (r = 0.76, p < 0.0001). B. Bland-Altman analysis showed good agreement between both imaging techniques of the regurgitant orifice (average bias −5.7 mm, 95% confidence interval −37 to +26 mm), less accordance was observed in some patients with an ERO ≥50 mm².
Source...