Incidence and echocardiographic predictors of early postoperative right ventricular dysfunction following left ventricular assist implantation in paediatric patients.

OBJECTIVES: Left ventricular assist device implantation is an established therapy for paediatric patients with end-stage heart failure. Early right ventricular dysfunction (RVD) after implantation still remains a challenge in the postoperative period. This study sought to determine the incidence of RVD and to identify echocardiographic predictors of RVD in paediatric patients, as well describing associated clinical outcome.

METHODS: Prospectively collected preoperative echocardiographic, haemodynamic, demographic and biochemical data from 48 patients scheduled for left ventricular assist device implantation were evaluated. Incidence of high central venous pressure, decreased central venous saturation, high inotropic support requirements or need for mechanical support of the right ventricle during the first 48 h after implantation were used to define RVD. Echocardiographic assessments of right ventricular geometry, function using linear dimensions, areas and tricuspid annular plane systolic excursion (TAPSE) were performed preoperatively and the relative relationships between these parameters were evaluated.

RESULTS: We included 48 consecutive paediatric patients (median age 5 years, range 0-17; median weight 15.9 kg, range 3.6-91). According to our criteria, 24 (50%) patients developed RVD. TAPSE as the parameter for assessment of longitudinal systolic function was significantly lower in this group (P = 0.01). The difference became even more pronounced after normalization to the RV end-diastolic diameter in long axis with P = 0.003. The odds ratio for patients with TAPSE/RV end-diastolic diameter in long axis <17.1% to develop RVD was 7.7 (P = 0.002).

CONCLUSIONS: RVD occurs frequently in paediatric patients after left ventricular assist device. TAPSE, normalized to the RV end-diastolic diameter, may help to identify patients at risk for RVD. The predictive value of this parameter supports decision making to plan for adequate pharmacological support or consider early upgrading to mechanical RV support.