Ratio of pulmonary artery diameter to ascending aortic diameter and its association with right ventricular failure after left ventricular assist device implantation

Background: Several invasive hemodynamic parameters help predict right ventricular failure (RVF) after left ventricular assist device (LVAD) implantation. However, prediction using non-invasive parameters alone has not been established. The ratio of the diameters of the pulmonary artery (PAD) to those of the ascending aorta (AoD) may indicate past hemodynamic load and cardiac dysfunction. We aimed to investigate a predictive model for RVF after LVAD implantation using non-invasive parameters including PAD/AoD ratio. Methods: We studied 141 patients who underwent primary LVAD implantation and 117 healthy individuals with computed tomography (CT) data. RVF was defined as the need for a subsequent right ventricular assist device or intravenous inotrope administration for more than 30 days after LVAD implantation. The PAD/AoD ratio was measured at the level of the pulmonary artery bifurcation on the CT transaxial slices. Results: RVF was observed in 29 patients. The correlation between PAD and AoD differed among healthy individuals, patients with and without RVF. Patients with RVF had higher total bilirubin and log brain natriuretic peptide (BNP) levels, a lower left ventricular end-diastolic diameter (LVDd) index, and a higher PAD/AoD ratio than those without RVF. Decision tree analysis indicated that the subgroup with a high PAD/AoD ratio (>= 1.09) >= 1.09) and a small LVDd index (<35.4 mm/m(2)) showed the highest probability of RVF (100 %), while the subgroup with a low PAD/AoD ratio (<1.09) and low log BNP (<2.79) showed the lowest probability of RVF (1 %). Conclusion: Combining non-invasive parameters with the PAD/AoD ratio can predict RVF with high accuracy.