Isovolumic cardiac contraction on high-temporal-resolution cine MR images: study in heart failure patients and healthy volunteers

• Published in: Radiology Vol. 248; no. 2; p. 458
• Main Authors: Dinh, Howard V, Alvergue, Juan, Sayre, James, Child, John S, Deshpande, Vibhas S, Laub, Gerhard, Finn, J Paul
• Format: Journal Article
• Language: English
• Published: United States 01.08.2008
• Subjects: Adult; Area Under Curve; Case-Control Studies; Female; Heart Failure - physiopathology; Heart Function Tests; Humans; Image Interpretation, Computer-Assisted; Linear Models; Magnetic Resonance Imaging, Cine; Male; Middle Aged; Myocardial Contraction - physiology; Prospective Studies; ROC Curve
• ISSN: 1527-1315
• DOI: 10.1148/radiol.2482071103

To prospectively implement high-temporal-resolution cine magnetic resonance (MR) imaging protocol to compare cardiac preejection contraction (PEC) and prefilling relaxation (PFR) times between heart failure (HF) patients and healthy control subjects and to assess accuracy of PEC times to stratify HF patients, with ejection fraction (EF) and New York Heart Association (NYHA) symptom class as reference standards. Following institutional review board approval of this HIPAA-compliant study and written informed consent, 18 healthy volunteers (10 women, eight men; mean age, 43 years +/- 14 [standard deviation]) and 18 HF patients (five women, 13 men; mean age, 49.8 years +/- 3) were imaged (breath-hold true fast imaging with steady-state precession, with temporal resolution of 5.6 msec at 1.5 T). By using left ventricular (LV) outflow tract acquisition, PEC phase was defined as time at QRS trigger to immediately before aortic valve opening. PFR was defined as time from initial aortic valve closure to immediately before mitral valve opening. Group means were compared (unpaired Student t test). Accuracy of PEC parameters in stratifying participants with severe systolic HF on the basis of EF and NYHA symptom class was assessed (receiver operating characteristic curve analysis). Compared with control subjects, HF patients had prolonged mean PEC time (40.4 msec +/- 11.8 vs 91.3 msec +/- 26, P < .001) and mean PFR time (68.3 msec +/- 26.8 vs 103.7 msec +/- 41.8, P < .01). PEC time correlated with global EF (r = -0.73, P < .001) and LV mass (r = 0.69, P < .001). For identification of patients with severe LV systolic dysfunction (EF <or= 35%), PEC time was highly accurate (area under the curve [AUC], 0.900 [P < .001]). For identification of patients with moderate-to-severe HF symptoms (NYHA class > 2), PEC time had good accuracy (AUC, 0.875 [P < .001]). It is feasible to assess isovolumic PEC and PFR phases of the cardiac cycle with high-frame-rate cine MR images, and PEC time is a surrogate measure of moderate-to-severe systolic HF.