Comparison of in vivo and in vitro haemodynamic function in experimental heart failure: use of echocardiography.

OBJECTIVE: To compare in vivo and in vitro haemodynamic performance in two models of experimental cardiac failure. To validate echocardiography as a means of non-invasive assessment of left ventricular dysfunction in rabbits. METHODS: Cardiac failure was induced by doxorubicin injection (1-1.25 mg.kg-1 twice weekly for 8 weeks (n = 16)) or coronary ligation (n = 12), with 12 controls. Left ventricular diastolic dimension and ejection fraction were assessed in vivo by echocardiography. The doxorubicin-treated and ligation hearts were subdivided by ejection fraction > 0.40 or < or = 0.40 into non-failing and failing groups. Thermodilution cardiac output was measured in vivo at baseline and after a fluid load. Basal cardiac output and peak cardiac output achieved by increased preload were measured in vitro in the working heart mode. RESULTS: The mean ejection fractions in the doxorubicin-treated and ligation groups were significantly (P < 0.001) lower than in controls, but there was wide inter-individual variability ranging from normal to severely impaired function [mean +/- s.d. (range) controls 0.65 +/- 0.03 (0.59-0.72), doxorubicin 0.45 +/- 0.11 (0.30-0.67), ligation 0.42 +/- 0.12 (0.25-0.65)]. Basal and peak cardiac outputs in vivo and in vitro were significantly lower in the doxorubicin and coronary ligation groups than in controls, although there was a wider scatter of values in the pathological groups. Among the doxorubicin and coronary ligation groups, hearts with ejection fractions < or = 0.40 demonstrated significantly impaired haemodynamic function compared with those with ejection fractions > 0.40. There were significant correlations between ejection fraction and all indices of haemodynamic function in vivo and in vitro. CONCLUSIONS: Simple non-invasive measurement of ejection fraction allowed improved characterization of haemodynamic responses in vivo and in vitro. Individual assessment of animals by echocardiography will improve interpretation of cellular or molecular studies in experimental heart failure by relating observed abnormalities to the degree of global cardiac dysfunction.