Variation in right ventricular volumes assessment by real-time three-dimensional echocardiography between dilated and normal right ventricle: Comparison with cardiac magnetic resonance imaging
机构：[a]Institute of Vascular Medicine, Division of Cardiology, and Heart Education And Research Training (HEART) Center, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong[b]Beijing Anzhen Hospital, Capital Medical University, Beijing, China首都医科大学附属安贞医院[c]Department of Cardiovascular Ultrasound and Non-invasive Cardiology, Sichuan Academy of Medical Science, Sichuan Provincial People's Hospital, Sichuan, China非手术科室医技科室超声科心内科四川省人民医院
Real-time three-dimensional echocardiography Right ventricular Volumes Accurate measurement of right ventricular (RV) function and volumes would be important in clinical practice given its predictive value. However, due to its complicated structure and geometry, conventional echocardiography is unable to comprehensively trace RV shape. Currently, real-time three-dimensional echocardiography (RT3DE) is available for measuring RV volumes and function [1–3] , and more reliable than two-dimensional echocardiography. However, there is paucity of data regarding the assessment accuracy between subjects with dilated and normal right ventricles. A recent meta-analysis demonstrated that RT3DE led to an underestimation of RV volumes and systolic function [4] . In this study, we aimed to compare the accuracy and reproducibility of RT3DE in normal and dilated RV groups with cardiac magnetic resonance imaging (MRI) as gold standard. Thirty-eight patients (52 ± 23 years, 20 males) with heart diseases were enrolled. All patients were referred for cardiac MRI and transthoracic echocardiography examination based on clinical indications including systolic heart failure (left ventricular ejection fraction≤50%) and thalasemia (pulmonary hypertension was excluded in this study). RT3DE and cardiac MRI were performed on the same day to minimize differences in loading condition. 16 HZ) were stored digitally and quantitative analysis of RT3DE were performed offline blinded to MRI data by dedicated software (TomTec, RV function, Germany) and RV volumes and ejection fraction were measured with semi-automatical algorithm. Dilated RV was defined as RV basal diameter ≥ 4.2 cm at end-diastole at apical-4 view on two-dimensional echocardiography. [5] Standard echocardiography was done (iE33, Phillips, United States) in sinus rhythm. RV full volume images with clear endocardial border and higher frame rate (> [6] Cardiac MRI was performed with a 1.5-T scanner (Sonata, Siemens, Erlangen, Germany) and RV parameters were assessed with the summation-of-discs method [5] by an experienced investigator who was blinded to echocardiographic data. Inter-observer and intra-observer variability were performed by two independent blinded investigators and by the same observer at 2 different time points. Five patients were excluded in the final analysis due to poor RT3DE image quality. Thus RT3DE and MRI data were obtained in a total of 33 patients. The RV measurements assessed by MRI short-axis views and RT3DE are depicted in . Overall, RV EDV was lower on RT3DE than MRI (p = 0.022) whereas there was no significant difference for RV ESV and EF. Further analysis showed that RV volumes were significantly smaller in normal RV group measured with RT3DE compared to MRI (both p < 0.01); in contrast, all the aforementioned parameters were quite comparable in dilated RV group though EF was almost the same in all groups. Although moderate correlations were found in whole group between RT3DE imaging and MRI (all p < 0.001), there was disparity of the accuracy of RV volumes evaluated with RT3DE between normal and dilated RV groups. Regression and Bland-Altman analysis were performed in both groups. For those with enlarged RV, high correlation was found with the correlation coefficient of 0.99 for EDV, 0.99 for ESV and 0.93 for EF (all p < 0.001). However, in subjects with normal RV, there was only moderate correlation between the two methods with the correlation coefficient of 0.64 for EDV, 0.65 for ESV and 0.64 for EF (all p < 0.001). Mean difference as performed with Bland–Altman was − 13.5 ± 13.5 ml [95% confidence interval (CI): − 39.9 to 12.9 ml] for RVEDV, − 5.9 ± 6.9 ml (95% CI: − 19.5 to 7.7 ml) for RVESV and − 2.0 ± 6.0 (95% CI: − 13.7 to 9.5 %) for RV EF ( ) in dilated RV group. 38.9 ± 34.2 ml (95% CI: − 106.0 to 28.2), − 22.8 ± 23.2 ml (95% CI: − 68.2 to 22.6) and 1.7 ± 6.9 % (95% CI: − 11.8 to 15.1) for RVEDV, EVESV and RVEF respectively ( 25% and 5% in whole group, normal RV group and dilated RV group respectively between RT3DE and MRI. In the dilated RV group, the inter- and intra-observer variability for RV EDV was 6.7% and 6.0%, respectively, and 6.5 % and 5.8% for ESV, respectively; the corresponding data in normal RV group was 13.2%, 12.5% and 10.1% and 9.8%, respectively. Table 1 Fig. 1 A, B and C Fig. 1 also shows the result of Bland–Altman analysis in normal RV group with mean difference − Fig. 1 E, F,G). RV volumes measured by RT3DE was underestimated when compared to MRI with the degree of 18% for EDV, 17% for ESV in whole group, 37% for EDV and ESV in normal RV group and 9% of EDV and 7% of ESV in dilated group. In respect to RV EF, the difference was 2%, − RV has a complex three-dimensional morphology which is not easy to draw the contour exactly with two-dimensional imaging techniques due to many geometric assumptions, in particular, the irregular shape of RV inflow and outflow > 200 ml) were associated with an increased underestimation of RVEDV and RVESV and older age was associated with overestimation of RV volumes due to due to blurring of the endocardial border, giving rise to wider distance inside the ventricular walls [7] . Data on the discrepancy of RV measurement with RT3DE between normal and dilated RV groups failed to reach the same conclusion [8,9] . It is worth highlighting the findings of a large meta-analysis of 23 studies, which showed that larger volumes (EDV [4] . In current study, we found that those with larger RV not only had high correlation but lower variability compared to MRI. In contrast, the normal RV group had fair correlation of both EDV and ESV. These might be related to easy identification the true RV endocardial border and prominent endocardial trabeculae. Moreover, the dilated RV resembles a round instead of crescentic shape, and contour tracing is readily delineated especially during manual revision. Thus it is highly likely that RV volume analysis is more reliable in those with enlarged RV. Our study is limited by small sample size and the definition of dilated RV based on two-dimensional echocardiography. Full volume acquisition algorithm requiring 4-cardiac-cycle reconstruction instead of one-single beat full volume capture was performed, which might be time-consuming [10] . In conclusion, RT3DE is an effective method to assess RV volume and function, but it is more reliable in dilated RV patients. Therefore, RT3DE should be considered as preferred imaging tool of RV measurements in this subgroup. This study was supported by a research grant from the University Grants Committee of Hong Kong (RGC Collaborative Research Fund 2010/11: CUHK9/CRF/10). We gratefully thank the patients who participated in our study; Tracy Lam and other research staff of the Division of Cardiology, Prince of Wales Hospital for their assistance; Peter Roesch and TomTec Imaging Systems GmbH for training and technical assistance.
第一作者机构：[a]Institute of Vascular Medicine, Division of Cardiology, and Heart Education And Research Training (HEART) Center, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong[b]Beijing Anzhen Hospital, Capital Medical University, Beijing, China
通讯作者：
通讯机构：[a]Institute of Vascular Medicine, Division of Cardiology, and Heart Education And Research Training (HEART) Center, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong
推荐引用方式(GB/T 7714)：
Fang Fang,Anna Chan,Alex Pui-wai Lee,et al.Variation in right ventricular volumes assessment by real-time three-dimensional echocardiography between dilated and normal right ventricle: Comparison with cardiac magnetic resonance imaging[J].International Journal of Cardiology.2013,168(4):4391-4393.doi:10.1016/j.ijcard.2013.05.046.
APA：
Fang Fang,Anna Chan,Alex Pui-wai Lee,John E. Sanderson,Joey S.W. Kwong...&Cheuk-man Yu.(2013).Variation in right ventricular volumes assessment by real-time three-dimensional echocardiography between dilated and normal right ventricle: Comparison with cardiac magnetic resonance imaging.International Journal of Cardiology,168,(4)
MLA：
Fang Fang,et al."Variation in right ventricular volumes assessment by real-time three-dimensional echocardiography between dilated and normal right ventricle: Comparison with cardiac magnetic resonance imaging".International Journal of Cardiology 168..4(2013):4391-4393
