The Utility of Left Atrial Strain in Predicting Outcome in Pediatric Restrictive Cardiomyopathy

Background: Prognosis of restrictive cardiomyopathy (RCM) in children is poor. There is variability in their individual clinical course. This can make it difficult to decide timing of heart transplant listing. Since it is a disorder of reduced left ventricular (LV) compliance, abnormalities in left atrial (LA) function often develop. We examined how LA strain abnormalities are associated with outcome and compared them to noninvasive and invasive indices.

Methods: Children with RCM were identified from a heart failure database. Those with associated hypertrophic cardiomyopathy were excluded. The echo at diagnosis and nearest the pre-transplant catheterization were included. Echo analysis was performed by an observer blinded to outcome. Parameters included apical 4-chamber LV strain, LA 4-chamber peak systolic strain (reservoir and contraction), indexed LA volume, mitral E/A, deceleration time, mitral S’ and E’ velocity, mitral E/E’, septal S’ and E’ velocity, septal E/E’, LV isovolumetric relaxation time, peak pulmonary vein S and D velocity, S/D, peak atrial reversal (AR) velocity and duration, LV shortening and ejection fraction, and Simpson’s biplane ejection fraction. Catheter-based values included pulmonary vascular resistance (PVR), Qp, Qs, right ventricular systolic pressure, right ventricular end-diastolic pressure, mean pulmonary artery pressure, pulmonary capillary wedge pressure (PCWP), and LV end-diastolic pressure. Spearman correlations were calculated for each echo variable and PCWP to compare the outcomes of alive and transplant-free, death, and transplanted. Kaplan-Meier curves for outcome of death/transplant were constructed by dichotomizing LA 4-chamber peak systolic strain in reservoir phase at the bottom quartile ( < 25%ile vs >25 %ile).

Results: Twenty-four echo were analyzed in 14 patients (median age at diagnosis 8.8 yrs, 46% female, 1 free of death/transplant). There was no statistical significance between each echo variable and outcome (Table 1), nor between catheter measure and outcome. Spearman correlations compared PCWP with each echo measure resulting in no statistical significance. Kaplan-Meier curves revealed reduced time to death/transplant in patients with LA 4-chamber peak systolic strain reservoir phase < 11%, hazard ratio of 2.70; p=0.17 (Figure 1).

Conclusion: LA strain abnormalities relate to outcome in pediatric patients with RCM and offer promise for a noninvasive echo marker of outcome. Given the paucity of these patients at individual centers, this preliminary observation needs further investigation with a larger multicenter cohort of patients.