Challenges in echocardiographic assessment of mitral regurgitation in children after repair of atrioventricular septal defect

Abstract

The validity and reproducibility of echocardiographic methods used to quantify mitral regurgitation (MR) in children with congenital heart disease are unknown. We evaluated the usefulness of methods used to quantify MR in children enrolled in a multicenter trial of enalapril 6 months after surgical repair of an atrioventricular septal defect (AVSD). MR severity in this trial was assessed using body surface area (BSA)-adjusted vena contracta lateral (i-VCW(lat)) and anterior-posterior (i-VCW(ap)) dimensions and cross-sectional area (i-VCA), regurgitant volume/BSA, regurgitant fraction, and qualitative MR grade. For each method, association with left ventricular end-diastolic volume (LVEDVz) and end-diastolic dimension (LVEDDz) z-scores and interobserver agreement were assessed. In 149 children (median age 1 year), i-VCW(lat), i-VCW(ap), and i-VCA were best associated with LVEDVz (r (2) = 0.54, r (2) = 0.24, and r (2) = 0.46, respectively; p < 0.001 for all) and showed the highest interobserver agreement (intraclass correlation coefficient = 0.62, 0.73, and 0.68, respectively). Qualitative MR grade was also associated with LVEDVz (r (2) = 0.31, p < 0.001) and showed modest interobserver agreement (kappa 0.56). Regurgitant volume/BSA and regurgitant fraction were associated with LVEDVz (r (2) = 0.45 and r (2) = 0.45, p < 0.001 for both) but showed poor interobserver agreement [ICC = 0.28 (n = 91) and ICC = 0.17 (n = 76), respectively], and their values were negative in 75% of subjects. In conclusion, echocardiographic assessment of MR severity after AVSD remains challenging. Among the quantitative methods used in this trial, i-VCW and i-VCA performed the best but offered little advantage compared with qualitative MR grade. The utility of regurgitant volume and fraction was severely limited by poor interobserver agreement and frequently negative values.

Fig. 1

Measurement of VCW in near-orthogonal planes. a VCW_lat was measured in the apical four-chamber view. b VCW_ap was measured in the parasternal long-axis view

Fig. 2

Distribution of patients by qualitative MR grade

Fig. 3

Box plots showing the relationship between quantitative measures of MR and qualitative MR grade. The middle line in the box represents the median. The lower and upper edges of the box represent the 25th and 75th percentile, respectively. All measures of MR correlated with qualitative MR grade. Values for regurgitant volume and fraction were frequently negative

Fig. 4

Relationship between measures of MR and a LVEDVz and b LVEDDz. Blue lines denote the predicted regression values. All methods were associated with LV size, and most of the relationships were nonlinear. As previously noted, values for i-regurgitant volume (regurgitant volume/BSA) and regurgitant fraction were frequently negative

Fig. 5

Effect of sedation (in patients<2 years of age) on relationship between i-VCA and regurgitant fraction and LVEDVz. Administration of sedation had a significant impact on the association between LVEDVz and the severity of MR as assessed by most measures (all interaction p values ≤ 0.001). Compared with unsedated children, the associations were stronger in children <2 years of age who received sedation

Fig. 6

Relationship between LV stroke and aortic stroke volume (a) scatterplot. The solid line represents y = x. b Bland-Altman plot. The middle solid line represents the mean difference between LV stroke volume and aortic stroke volume. The lower and upper solid lines represent the 95% CIs for the difference. In most subjects, aortic stroke volume was greater than LV stroke volume, resulting in a negative value for regurgitant volume and fraction

Fig. 7

Relationship between Doppler-derived maximum instantaneous LV outflow tract pressure gradient and a regurgitant fraction and b regurgitant volume/BSA. A greater LV outflow tract pressure gradient was associated weakly in an inverse fashion with BSA-indexed regurgitant volume (r = −0.18, p = 0.04) and regurgitant fraction (r = −0.2, p = 0.03), suggesting that greater aortic outflow velocities may have contributed to overestimation of aortic stroke volume. This may have contributed to negatives values frequently observed for regurgitant volume and fraction