Luxury lung perfusion in end-stage liver disease during liver transplantation

End-stage liver disease is accompanied by a hyperkinetic circulation sometimes combined with hypoxaemia. Nitric oxide overproduction has been described as a possible cause by dilating the vasculature and decreasing cardiac afterload. The aim of this study was to evaluate haemodynamics, ventilation/perfusion matching, alveolar and alveolar dead space ventilation and resistance of systemic and pulmonary vasculature during liver transplantation.

Ten liver transplantation patients were studied. Cardiac output, CO, was measured with thermodilution technique. Pulmonary shuntflow was calculated from standard formulas. Effective cardiac output, COeff, was defined as the CO in contact with alveolar ventilation, VA. Effective alveolar ventilation, VAeff, was defined as VA in contact with pulmonary circulation. Measurements were performed during dissection, anhepatic and reperfusion phases.

During the dissection phase the shunt was 23 +/- 3%, COeff was 7.9 +/- 0.6 l/min, SVR was 620 +/- 67 dyn.s/cm5, VAeff was 3.4 +/- 0.5 l/min, SaO2 was 98 +/- 1% and SvO2 was 86 +/- 2%. Corresponding values during the anhepatic phase were 16 +/- 2%, 5.6 +/- 0.4 l/min, 931 +/- 78 dyn.s/cm5, 3.1 +/- 0.2 l/min, 99 +/- 1% and 88 +/- 1%. During the reperfusion phase the values returned to levels close to that of the dissection phase. The reduction of COeff between the dissection and the anhepatic phase was significant (P < 0.01).

The low vascular resistance is accompanied by a high cardiac output. In spite of the high shunt fraction, these patients were not hypoxaemic. This is explained by the fact that the increased cardiac output leads to a decrease in arterio-mixed venous oxygen content difference and an increase in mixed venous oxygenation level, SvO2 86-88%, normal value approximately 70%. The VAeff/COeff in this study was approximately 0.5, i.e. the effective cardiac output, COeff is 235, 180 and 197% of the effective alveolar ventilation, VAeff during the three phases. Thus, about twice the amount blood is oxygenated as compared to a normodynamic situation, which compensates for the effect of the shunt flow on oxygenation.