Rotational deformation of the canine left ventricle measured by magnetic resonance tagging: Effects of catecholamines, ischaemia, and pacing

Objective: The aim was to investigate the generation of rotation of the left ventricular apex with respect to the base by magnetic resonance tagging, a non-invasive method of labelling the myocardium, in a canine model. Methods: 18 dogs were imaged at baseline and during: (1) inotropic stimulation with dobutamine; (2) chronotropic stimulation with atrial pacing; (3) anterior wall ischaemia; (4) posterior wall ischaemia; and (5) varying left ventricular activation site; six dogs underwent each intervention. Apical rotation of the apex (torsion) was quantified. The epicardium and the endocardium were considered separately, as were the anterior and posterior walls. Results: Mean torsion of the epicardium [anterior 3.1(SEM 1.2)°, posterior 9.9(1.0)°] was less than that of the endocardium [anterior 8.1(2.6)°, posterior 14.9(2.0)°, p<0.05 for both]. Anterior torsion was less than posterior torsion for both the epicardium, p<0.05, and the endocardium, p<0.05. Dobutamine increased torsion of both the epicardium [anterior 13.3(2.2)°, posterior 12.6(1.7)°, p<0.05 for both] and the endocardium [anterior 24.6(2.3)°, posterior 16.5(2.1)°, p<0.05 for both]. Atrial pacing at 160% baseline rate increased torsion of both the anterior wall [epicardium 6.6(1.0)°, endocardium 11.3(1.2)°, p<0.05] and the posterior wall [epicardium 13.0(1.3)°, endocardium 19.4(1.9)°, p < 0.05]. Anterior wall ischaemia reduced torsion of the anterior wall only [epicardium -2.0(1.0)°, endocardium 6.7(2.3)°, both p<0.05]. Posterior wall ischaemia reduced torsion of the posterior wall of the epicardium only [7.1(1.2)°, p<0.05] but also reduced torsion of the anterior wall [epicardium 0.7(1.0)°, endocardium 2.4(1.6)°, p<0.05 for both]. Altering the pattern of left ventricular activation by atrioventricular pacing reduced torsion of the posterior wall of the epicardium [6.6(1.2)°, p<0.05] and of the anterior [3.6(1.9)°, p<0.05] and posterior [7.1(1.6)°, p<0.05] walls of the endocardium. Conclusions: Rotational deformation of the left ventricle is dependent on the pattern of left ventricular activation and the contractile state. That a decrease in the contractile state in one area (by ischaemia) can cause a decrease in rotation in another suggests that this rotation depends on the complex fiber arrangement of the whole ventricle.Cardiovascular Research 1994;28:629-635.