Right ventricular dysfunction in children supported with pulsatile ventricular assist devices.

OBJECTIVES

To describe the incidence and severity of right ventricular dysfunction (RVD) in pediatric ventricular assist device (VAD) recipients and to identify the preoperative characteristics associated with RVD and their effect on outcomes.

METHODS

Children bridged to transplantation from 2004 to 2011 were included. RVD was defined as the use of a left VAD (LVAD) with an elevated central venous pressure of >16 mm Hg with inotropic therapy and/or inhaled nitric oxide for >96 hours or biventricular assist (BiVAD).

RESULTS

A total of 57 children (median age, 2.97 years; range 35 days to 15.8 years) were supported. Of the 57, 43 (75%) had an LVAD, and of those, 10 developed RVD. The remaining 14 (25%) required BiVAD. Thus, RVD occurred in 24 of 57 patients (42%). Preoperative variables such as younger age (P = .01), use of extracorporeal mechanical support (P = .006), and elevated urea (P = .03), creatinine (P = .02), and bilirubin (P = .001) were associated with RVD. Multiple logistic regression analysis indicated that elevated urea and extracorporeal mechanical support (odds ratio, 26.4; 95% confidence interval, 2.3-307.3; and odds ratio, 27.8; 95% confidence interval, 2.5-312.3, respectively) were risk factors for BiVAD. The patients who developed RVD on LVAD had a complicated postoperative course but excellent survival (100%), comparable to those with preserved right ventricular function (91%). The survival for those requiring BiVAD was reduced (71%).

CONCLUSIONS

RVD occurred in approximately 40% of pediatric VAD recipients and affects their peri-implantation morbidity and bridging outcomes. Preoperative extracorporeal membrane oxygenation and elevated urea were risk factors for BiVAD. Additional studies of the management of RVD in children after VAD implantation are warranted.