A continuous flow left ventricular assist device (HeartAssist 5; Micromed Cardiovascular Inc. Houston, USA) was implanted in a 66 years old female patient with an end-stage heart failure NYHA IV with a history of dilated cardiomyopathy. Preoperative echocardiography showed a severe impairment of the left ventricular function (EF 10%), a secondary pulmonary hypertension with an estimated systolic pulmonary artery pressure of 48 mmHg and a right ventricular dysfunction with a tricuspid annular plane systolic excursion (TAPSE) of 15 mm. The cardiac output, determined by right heart catheterization, was 3,5 L/min (Cardiac Index 1,92 L/min/m2) with a systemic peripheral vascular resistance of 1350 dyn.sec.cm5 and pulmonary vascular resistance of 383 dyn·s/cm5). After four days intravenous therapy with milrinone (14 μg/kg/hr)., cardiac output improved to 4,6 L/min (Cardiac Index 2,5 L/min/m2) with a systemic vascular resistance of 1000 dyn·s/cm5 and pulmonary vascular resistance decreased to 190 dyn·s/ cm5. After implant of the HeartAssist5, the thorax could not be completely closed, due to severe fluid retention and oedema. Complete closure resulted in severe right ventricular failure. During the first and the second postoperative day the mean real-time flow was 4,8 L/min (range: 4,4 L/min-5,2 L/min) at a fixed pump speed of 8400 rpm (range: 8300–8500 rpm) and the flow waveforms showed very pulsatile flows (2.0 L/min in diastole and 6.5 L/min in systole). In the third postoperative day the mean real-time flow decreased to 4,3 L/min (range: 3,9-5,1 L/min) and the flow curve became flat. Real time flow was correlated to cardiac output measured by thermodilution (4,07 L/min on a systemic peripheral vascular resistance of 766 dyn·s/cm5). Initialy a low-dose inotropic therapy with milrinone (3 μg/kg/hr), noradrenalin (1 μg/kg/hr) and dopamine (200 μg/kg/hr) was enough to restore cardiac output (4,7 L/min with a systemic vascular resistance of 938 dyn·s/cm5). Unfortunately, in the eighth postoperative day, the central venous pressure increased from 9 mmHg to a mean 16 mmHg (range 14–20 mmHg) and the real-time flow decreased (mean 3,9 L/min, range 3,5-4,7 L/min). 24 hours later the patient developed renal failure, necessitating haemodialysis and a high dose inotropic therapy became mandatory (norepinephrine 8 μg/kg/hr, Milrinone 11 μg/kg/hr). Despite maximal inotropic therapy, flows continued to decline and on the tenth postoperative day the patient was placed on venoarterial extracorporeal membrane oxygenation (Levitronix Centrimag, Zurich, Switzerland). The two-stage venous cannula was inserted to the right atrium and the arterial cannula in the main pulmonary artery. Real time flow & pulsatility recovered promptly. Mean flow values over the first 24 hours ranged from 4,8 L/min to 6,2 L/min at a fixed pump speed of 8700 rpm. After four days of ECMO, the inotropes were weaned. At the sixth day of ECMO support, weaning from ECMO was started. Mean real time flow remained stable over the weaning procedure (5 L/min; range 4,3-5,5 L/min). The ECMO circuit was disconnected at the 14th post-operative day. Three days after ECMO removal, the Thorax was closed, and the further course was uneventful.