Determination of cardiac output with dynamic contrast-enhanced computed tomography

Abstract

Cardiac output (CO) is an important diagnostic and prognostic factor in the haemodynamic evaluation of patients. The gold standard for CO measurement, thermodilution, requires an invasive right-heart catheterisation (RHC). In this pilot study we aimed to determine the accuracy of non-invasive CO determination from dynamic contrast-enhanced computed tomography (CT) compared to thermodilution. Patients who underwent diagnostic or follow-up RHC due to suspected or known pulmonary vascular disease at our department and required a thoracic CT between June 2011 and August 2012 were included. CO was determined from CT attenuation-time curves in the pulmonary artery and the ascending aorta using a dynamic contrast-enhanced CT sequence. CO determined in N=18 patients by dynamic CT in the pulmonary artery was in very good agreement with thermodilution data (r=0.84). Bland-Altman analysis showed a systematic overestimation of 0.7±0.6l/min compared to thermodilution. Data from the ascending aorta also showed a good correlation, but with a larger scattering of the values. The average effective dose for the dynamic investigation was 1.2±0.7mSv. CO determined with dynamic contrast-enhanced CT in the main pulmonary artery reliably predicts the values obtained by thermodilution during RHC. This non-invasive technique might provide an alternative for repeated invasive right-heart catheter investigations in the follow-up of pulmonary arterial hypertension patients.