Lung diffusion capacity in advanced heart failure: relation to central haemodynamics and outcome

TY - JOUR

T1 - Lung diffusion capacity in advanced heart failure

T2 - relation to central haemodynamics and outcome

AU - Deis, Tania

AU - Balling, Louise

AU - Rossing, Kasper

AU - Wolsk, Emil

AU - Perch, Michael

AU - Gustafsson, Finn

PY - 2019

Y1 - 2019

N2 - Aims: Patients with heart failure (HF) are known to have a reduced pulmonary diffusion capacity for carbon monoxide (D LCO ), but little is known about how lung function relates to central haemodynamics. The aim of this study was to investigate the association between haemodynamic variables and pulmonary diffusion capacity adjusted for alveolar volume in congestive HF patients and to analyse how predicted D LCO /V A affects mortality in relation to the haemodynamic status. Methods and results: We retrospectively studied right heart catheterization (RHC) and lung function data on 262 HF patients (mean age 51 ± 13 years) with a left ventricular ejection fraction < 45% referred non-urgently for evaluation for heart transplantation (HTX) or left ventricular assist device (LVAD). Univariate and multivariate linear regression models were constructed to examine the associations between predicted values of D LCO /V A , forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV 1 ), and haemodynamic parameters [pulmonary capillary wedge pressure (PCWP), central venous pressure, cardiac index, mean pulmonary artery pressure, and mean arterial pressure] as well as other factors known to affect lung function in HF. FEV 1 was reduced to <80% of predicted value in 55% of the population, and D LCO/ V A was reduced in 63% of the population. D LCO /V A correlated positively with pulmonary capillary wedge pressure in both univariate and multivariate analyses for all included patients (P < 0.001 and P = 0.045, respectively) and a restricted population of patients with the shortest time between RHC and lung function testing (P = 0.005, P = 0.015). D LCO /V A predicted mortality in multivariate models [hazard ratio 1.5 (1.1–2.1)] but not the combined endpoint of death, LVAD implantation, or HTX. There was no significant correlation between haemodynamics and predicted FVC or FEV 1 . Conclusions: Pulmonary diffusion capacity correlates positively with left ventricular fillings pressures, and reduced values predict increased mortality in patients with HF. This might be driven by increased lung capillary volume in patients with pulmonary congestion.

AB - Aims: Patients with heart failure (HF) are known to have a reduced pulmonary diffusion capacity for carbon monoxide (D LCO ), but little is known about how lung function relates to central haemodynamics. The aim of this study was to investigate the association between haemodynamic variables and pulmonary diffusion capacity adjusted for alveolar volume in congestive HF patients and to analyse how predicted D LCO /V A affects mortality in relation to the haemodynamic status. Methods and results: We retrospectively studied right heart catheterization (RHC) and lung function data on 262 HF patients (mean age 51 ± 13 years) with a left ventricular ejection fraction < 45% referred non-urgently for evaluation for heart transplantation (HTX) or left ventricular assist device (LVAD). Univariate and multivariate linear regression models were constructed to examine the associations between predicted values of D LCO /V A , forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV 1 ), and haemodynamic parameters [pulmonary capillary wedge pressure (PCWP), central venous pressure, cardiac index, mean pulmonary artery pressure, and mean arterial pressure] as well as other factors known to affect lung function in HF. FEV 1 was reduced to <80% of predicted value in 55% of the population, and D LCO/ V A was reduced in 63% of the population. D LCO /V A correlated positively with pulmonary capillary wedge pressure in both univariate and multivariate analyses for all included patients (P < 0.001 and P = 0.045, respectively) and a restricted population of patients with the shortest time between RHC and lung function testing (P = 0.005, P = 0.015). D LCO /V A predicted mortality in multivariate models [hazard ratio 1.5 (1.1–2.1)] but not the combined endpoint of death, LVAD implantation, or HTX. There was no significant correlation between haemodynamics and predicted FVC or FEV 1 . Conclusions: Pulmonary diffusion capacity correlates positively with left ventricular fillings pressures, and reduced values predict increased mortality in patients with HF. This might be driven by increased lung capillary volume in patients with pulmonary congestion.

KW - Haemodynamics

KW - Heart failure

KW - Pulmonary diffusion capacity

KW - Right heart catheterization

U2 - 10.1002/ehf2.12401

DO - 10.1002/ehf2.12401

M3 - Journal article

C2 - 30784223

AN - SCOPUS:85061787611

SN - 2055-5822

VL - 6

SP - 379

EP - 387

JO - E S C Heart Failure

JF - E S C Heart Failure

IS - 2

ER -