Clinical implications of major haemolysis in children supported by a pulsatile ventricular assist device

Haemolysis is developing prominence in the setting of supporting increasingly complex children with heart failure with a ventricular assist device. The goal of this study is to better characterise haemolysis and its implications in children supported with pulsatile ventricular assist devices.

This is a single-centre retrospective review of 44 children who were supported by Berlin Heart EXCOR between January 2006 and June 2020. Patients were divided into major haemolysers and non-major haemolysers. Major haemolysers were defined as patients with lactate dehydrogenase > 500U/L (2.5x the upper limits of normal) with either total bilirubin > 2mg/dL (with predominantly indirect hyperbilirubinemia) or anaemia out of proportion to the clinical scenario more than three days following implantation of the ventricular assist device(s). Patient demographics, ventricular assist device factors, and outcomes, including end-organ function and mortality, were compared between major haemolysers and non-major haemolysers.

Forty-four patients supported by the Berlin EXCOR were included in the analytic cohort of the study: 27 major haemolysers and 17 non-major haemolysers. Major haemolysis was more common in those supported with single-ventricle ventricular assist device (i.e., VAD in the context of functionally univentricular anatomy) compared to those with biventricular hearts, p = 0.01. There were no patients with an isolated left ventricular assist device or isolated right ventricular assist device in our analytic cohort of 44 patients. Of the 19 patients with single-ventricle ventricular assist device, 84% (16/19) were major haemolysers. Of the 25 patients with a biventricular assist device, 44% (11/25) were major haemolysers. Major haemolysers and non-major haemolysers had a body surface area of 0.28 and 0.40, respectively (p = 0.01). Overall, survival to discharge from the hospital was 66% (n = 29/44). Survival to discharge from the hospital was 52% (14/27) in major haemolysers versus 88% (15/17) in non-major haemolysers, p = 0.02. Only 3 of the 27 with major haemolysis had severe haemolysis, that is, lactate dehydrogenase > 2000 and bilirubin above 10. Non-major haemolysers had a better improvement in creatinine clearance during ventricular assist device support, p < 0.0001. (During the same era of this study, 22 patients who were supported with Berlin Heart were excluded from the analytic cohort because they did not have any recorded measurement of lactate dehydrogenase. Seventeen of these 22 patients had no clinical evidence of haemolysis. Survival to discharge from the hospital in this excluded cohort was 86% [19/22].).

Major haemolysis in patients with pulsatile ventricular assist device is more likely with single-ventricle ventricular assist device support and smaller body surface area.