Tricuspid incompetence and geometry of the right ventricle as predictors of right ventricular function after implantation of a left ventricular assist device

Abstract

Background: Implantation of a left ventricular assist device (LVAD) is an established treatment for end-stage heart failure. Right ventricular (RV) dysfunction develops in 20% to 50% of patients after LVAD implantation, leading to prolonged ICU stay and elevated mortality. However, the prediction of RV failure remains difficult.

Methods: The pre-operative echocardiographic parameters, tricuspid incompetence (TI), RV end-diastolic diameter (cut-off >35 mm), RV ejection fraction (cut-off <30%), right atrial dimension (cut-off >50 mm) and short/long axis ratio (cut-off >0.6), were analyzed retrospectively in 54 patients. Patients were divided into two groups. One group consisted of patients with RV failure (n = 9), as defined by the presence of two of the following criteria in the first 48 hours after surgery: mean arterial pressure < or =55 mm Hg; central venous pressure > or =16 mm Hg; mixed venous saturation < or =55%; cardiac index <liters/min/m(2); inotropic support score >20 units; or need for an RVAD. The other patients comprised the non-RV-failure group (n = 45).

Results: The RV failure group had a significantly higher short/long axis ratio of the RV (0.63 vs 0.52, p = 0.03; odds ratio 4.4, p = 0.011). For patients with a short/long axis ratio of the RV of <0.6, RV failure occurred in 7% of patients, as compared with 50% for patients with a ratio > or =0.6 (p = 0.013). Among patients with TI Grade III or IV, 75% developed RV failure as compared with 12% in patients with TI Grade I or II (p = 0.054). The odds ratio for RV failure after LVAD implantation for TI Grade III or IV was 4.7 (p = 0.012).

Conclusions: Pre-operative evaluation of tricuspid incompetence and RV geometry may help to select patients who would benefit from biventricular support.