Abstract

Nuclear magnetic resonance (NMR) imaging is a new, noninvasive approach for imaging the cardiovascular system. Being a three-dimensional technique, NMR imaging has the capability of measuring volumes without the need for assumptions regarding ventricular geometry. In this study, the technique was validated in 19 excised dog hearts, filled with silicone-rubber, imaged using a multislice spin-echo sequence. The volume of the cavity in each slice was calculated from the number of pixels outlined for each slice multiplied by the pixel volume. Ventricular volumes measured by NMR imaging were highly correlated with cast volumes measured by water displacement: right ventricle (RV):RVNMR = 1.05 RVcast - 1.62; r = 0.99, SEE = 0.96 ml; left ventricle (LV):LVNMR = 0.98 LVcast + 0.35, r = 0.98, SEE = 1.48 ml. After validation in casts, NMR imaging volumes were measured in eight living dogs using a multiphasic gated technique to obtain images at 5, 105, 205, 305 and 405 ms after the QRS complex. Cardiac output (CO) and stroke volume (SV) measured by NMR imaging were significantly correlated with thermodilution (TD) measurements (CONMR = 0.63 COTD + 0.51 liters/min; r = 0.78, SEE = 0.57 liters/min; SVNMR = 0.67 SVTD + 1.95 ml; SEE = 5.58 ml). Right and left stroke volumes were closely related (LVSVNMR = 0.9 RVSVNMR + 1.75; r = 0.94, SEE = 4.32 ml), with the slope and intercept of the regression line showing no difference from 1 and 0, respectively. However, volumes determined by NMR imaging underestimated the thermodilution measurements, presumably reflecting the inability to obtain a true systolic image with the present sampling rate.(ABSTRACT TRUNCATED AT 250 WORDS)

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call