Clinical, haemodynamic and echocardiographic features of early cardiac graft dysfunction.

Background: The main cause of early death after heart transplantation (HTx) is so-called early primary or secondary graft failure (GF). The risk of profound GF has not declined in the past decade, as the consequence of the liberalisation of donor acceptance criteria because of the scarcity of donors. It is therefore important to try to diagnose graft failure and recognise the mechanisms of early graft dysfunction.

Aim: To establish haemodynamic and echocardiographic criteria of early GF to define patients who should be considered for assist device support or re-transplantation.

Methods: Between January 2000 and March 2009, 116 HTx patients were studied. On the basis of echocardiography and continuous invasive monitoring, three groups were identified: (1) The true graft failure group (GF) consisted of 46 patients; (2) The latent right ventricular (RV) dysfunction group (RV-D) consisted of 25 patients with small left ventricular (LV) chamber (〈 39 mm) and RV ejection fraction (RVEF) 〈 50%; (3) The control group consisted of 45 consecutive HTx patients without any haemodynamic complications.

Results: Postoperatively, only the GF group required large doses of norepinephrine (〉 0.3 μmg/kg/min) and inhalative NO (40 ppm). Nevertheless, right and left filling pressures were significantly higher than in the controls (right 12 ± 3.6 vs. 9.0 ± 2 and left atrial pressure 13.0 ± 3.2 vs. 9.6 ± 2 mm Hg, both p 〈 0.001). Cardiac index was significantly smaller (2.9 ± 0.7 vs. 3.7 ± 0.9, p 〈 0.001) but neither pulmonary artery pressure (29.5 ± 6 vs. 29.7 ± 7 mm Hg) nor transpulmonary gradient (6 ± 5 vs. 5.1 ± 5 mm Hg) nor pulmonary vascular resistance (273 ± 97 vs. 287 ± 144 dyn × s × cm-5) differed significantly from those of the control group. In the GF group, LV end diastolic dimension (LVEDD) was significantly smaller and function poorer than in controls (39.8 ± 5 vs. 44.4 ± 5 mm, respectively, p = 0.001). RV function was also significantly worse (RVEF 42.2 ± 14% vs. 56.0 ± 9%), respectively, p = 0.001), whereas RV dimension did not differ significantly. Mechanical support after failure of the initial medical treatment was necessary in 37% of patients; 29 (63.0%) patients from the GF group died, the cause of death being sepsis with multi-organ failure. In the RV-D group, remodelling was quite similar but LVEF was excellent and maximal systolic velocity from the posterior wall was significantly higher than in GF. No death occurred.

Conclusions: True early GF represents a grave haemodynamic situation with high mortality. Bedside echocardiography helps to distinguish between latent RV dysfunction and true GF.