Free-breathing cardiovascular magnetic resonance flow quantification can be an alternative to standard breath-holding approach

Abstract

Cardiovascular magnetic resonance (CMR) evaluation of valvular heart disease is an important diagnostic tool when echocardiography is inconclusive. Phase contrast flow quantification is usually performed during breath hold (BH), which can be challenging in patients suffering from dyspnea and heart failure. The purpose of the present study is to compare a free-breathing (FB) with the conventional BH approach for flow quantification in the aortic, pulmonary and tricuspid valves in 20 healthy subjects (HS) and 25 patients with tricuspid regurgitation (TR). Aortic (AoFF) and pulmonary forward flow volume (PuFF), and tricuspid inflow volume (TrIF) were evaluated. Mean, standard deviation (SD) and limits of agreement (LoA) were calculated. There were good agreements between phase contrast flow volumes obtained by FB and BH approach. Mean difference ± SD / LoA for AoFF during BH versus FB were 1 ± 6 / -10 to 13 ml. The corresponding for PuFF were 1 ± 6 / -11 to 13 ml, and for TrIF - 3 ± 6 / -15 to 9 ml, respectively. Thus, free-breathing CMR flow acquisition can be an important alternative in the assessment of stroke volume, valvular regurgitant volume and be useful in all patients with difficulties to hold their breath.

Keywords: Breath-hold; Cardiovascular magnetic resonance; Free-breathing; Phase contrast flow quantification; Valvular heart disease.

Fig. 1

Image acquisition and analysis for CMR flow volume quantification. (A, B) Left and right ventricular outflow tract view in end-diastole. The red line illustrates the slice position for through plane phase contrast velocity imaging. Quantification of aorta and pulmonary flow is depicted in the corresponding phase contrast image below (arrow). (C) Four -chamber view in early diastole with an open tricuspid valve. The red line illustrates the slice position for through plane phase contrast velocity imaging. The corresponding phase contrast image for tricuspid inflow quantification in diastole is shown below (arrow).

Fig. 2

Bland Altman plots depicting the relation between flow volume measurements during free breathing and conventional breath-holding protocol in healthy subjects (n = 20) and patients with tricuspid regurgitation (n = 25).

Fig. 3

Boxplots comparing forward flow volumes in the aortic, pulmonary and tricuspid valve position during breath-holding and free breathing in healthy subjects (n = 20).