Impaired lung perfusion in patients with congestive heart failure by quantitative MRI perfusion

Abstract

Introduction: Congestive heart failure (CHF) is associated with reduced end-organ perfusion and can lead to organ failure. Clinical assessment of organ perfusion has relied on measures of organ function, an indirect strategy. Purpose: We sought to assess lung perfusion quantitatively by MRI in patients with CHF and to define factors associated with impaired lung perfusion. Methods: Study subjects were prospectively enrolled and underwent cardiac MRI at 1.5 T. First-pass perfusion was performed with 0.01 mmol/kg gadopentetate in 3 coronal slices covering anterior, mid and posterior lung fields during inspiration using a saturation recovery SSFP sequence. Mean signal intensity of all pixels in pulmonary parenchyma at each time point was evaluated using a custom model-independent deconvolution program for perfusion quantitation. Pulmonary transit time was measured as the time interval between peak signal intensity in the main pulmonary artery and peak signal intensity in the left atrium. Cardiac function was assessed using SSFP standard short axis cine imaging. Pulmonary flow was measured by through plane phase contrast imaging in the main pulmonary artery. Results: Of the 25 subjects enrolled 7 were normal controls and 15 were patients with systolic CHF in I-III NYHA functional class who completed lung perfusion imaging in inspiration. Average lung perfusion was reduced in CHF patients, 81 ± 32 ml/100 ml/min vs 118 ± 48 ml/100 ml/min in controls (p = 0.046). Similar to normal controls patients with CHF maintained a perfusion gradient in the gravity direction with the highest perfusion in posterior and the lowest in anterior lung field. However, the absolute perfusion was reduced in all lung fields with anterior, mid and posterior perfusion 58 ± 26, 85 ± 38, 106 ± 36 ml/100 ml/min in CHF patients and 73 ± 53, 116 ± 53, 165 ± 45 ml/100 ml/min in controls. Reduced lung perfusion was associated with lower pulmonary flow (indexed by BSA) (r = -0.771, p < 0.001), longer pulmonary transit time (r = -0.729, p < 0.001), a marker of total pulmonary resistance, lower LVEF (r = 0.443, p = 0.039) and lower RVEF (r = -0.474, p = 0.026). In a multivariate regression analysis including all variables associated with lung perfusion and an interaction term for pulmonary flow and pulmonary transit time since they were significantly correlated, the predictors of lung perfusion were pulmonary flow and flow/transit time interaction, suggesting the importance of both forward flow and total pulmonary resistance. Conclusion: Lung perfusion is decreased in patients with systolic CHF. The reduced perfusion is associated with both impaired forward flow and increased pulmonary vascular resistance.