Ann Thorac Surg 2003;76:1999
© 2003 The Society of Thoracic Surgeons
Invited commentary
John R. Pepper, FRCS
Department of Cardiothoracic Surgery, Royal Brompton Hospital, Sydney Street, London SW3 6NP, United Kingdom
e-mail: j.pepper{at}rbh.nthames.nhs.uk
Chronic unloading of the left ventricle decreases cardiac dimensions, enhances ejection fraction, and lowers left atrial pressure and pulmonary vascular resistance. In the worldwide experience of bridge to transplantation, patients who require only isolated LV assistance rather than biventricular assistance have much better rates of transplantation and discharge. Those who require isolated right heart support or biventricular support with hybrid systems have transplant rates of 30%–40% and hospital discharge rates of 0%–60%, compared to 60%–69% for both in patients supported by LVAD alone. The great advantage of the implantable LVAD is that patients can be safely supported for longer periods allowing complete physical rehabilatation before transplantation.
Despite considerable experience by dedicated teams in specialized centers, no reliable predictors have been found so far that can identify patients unlikely to survive implantable ventricular support. In a nationwide, retrospective survey from Sweden, Granfeldt and colleagues have identified postoperative factors, blood transfusion, and days on a ventilator and in intensive care as significant risk factors for mortality and morbidity. The majority of their patients received first generation pusher-plate pumps save for three patients who received a Jarvik impellor pump. These new impellor pumps are likely to have fewer complications, but their flow capability differs. Some, such as the Jarvik, are LV assist devices while others, like the pusher-plate pumps, are LV replacement devices.
In the bridge to transplant population, the most frequent complications are bleeding, infection, thromboembolism, renal failure, haemolysis, and neurological dysfunction. The most important of these are bleeding and infection. Bleeding rates are as high as 60% because of coagulopathy from chronic hepatic damage, extensive surgery, and the effects of cardiopulmonary bypass and blood pump rheology on platelet function. Although infections can be frequent (30%–40%) and severe, especially beyond three months, they do not preclude successful transplantation. To assess the risk of severe right ventricular failure after insertion of the LVAD can be even more difficult, but Granfeldt found a preoperative elevated C-reactive protein to be highly predictive. The use of inhaled nitric oxide and to a lesser extent the use of prostacyclin, together with phosphodiesterase inhibitors has reduced the occurrence of this problem, but it has not gone away completely.
This treatment is expensive, a fact that concerns all health payers public or private. Robust preoperative risk factors are needed but will only be established by multicenter, prospective studies using shared databases.
