Quantification of biventricular myocardial function using cardiac magnetic resonance feature tracking, endocardial border delineation and echocardiographic speckle tracking in patients with repaired tetralogy of Fallot and healthy controls

J Cardiovasc Magn Reson. 2012 May 31;14(1):32. doi: 10.1186/1532-429X-14-32.

Abstract

Background: Parameters of myocardial deformation have been suggested to be superior to conventional measures of ventricular function in patients with tetralogy of Fallot (ToF), but have required non-routine, tagged cardiovascular magnetic resonance (CMR) techniques. We assessed biventricular myocardial function using CMR cine-based feature tracking (FT) and compared it to speckle tracking echocardiography (STE) and to simple endocardial border delineation (EBD). In addition, the relation between parameters of myocardial deformation and clinical parameters was assessed.

Methods: Overall, 28 consecutive adult patients with repaired ToF (age 40.4 ± 13.3 years) underwent standard steady-state-free precession sequence CMR, echocardiography, and cardiopulmonary exercise testing. In addition, 25 healthy subjects served as controls. Myocardial deformation was assessed by CMR based FT (TomTec Diogenes software), CMR based EBD (using custom written software) and STE (TomTec Cardiac Performance Analysis software).

Results: Feature tracking was feasible in all subjects. A close agreement was found between measures of global left (LV) and right ventricular (RV) global strain. Interobserver agreement for FT and STE was similar for longitudinal LV global strain, but FT showed better inter-observer reproducibility than STE for circumferential or radial LV and longitudinal RV global strain. Reproducibility of regional strain on FT was, however, poor. The relative systolic length change of the endocardial border measured by EBD yielded similar results to FT global strain. Clinically, biventricular longitudinal strain on FT was reduced compared to controls (P < 0.0001) and was related to the number of previous cardiac operations. In addition, FT derived RV strain was related to exercise capacity and VE/VCO2-slope.

Conclusions: Although neither the inter-study reproducibility nor accuracy of FT software were investigated, and its inter-observer reproducibility for regional strain calculation was poor, its calculations of global systolic strain showed similar or better inter-oberver reproducibility than those by STE, and could be applied across RV image regions inaccessible to echo. 'Global strain' calculated by EBD gave similar results to FT. Measurements made using FT related to exercise tolerance in ToF patients suggesting that the approach could have clinical relevance and deserves further study.

Publication types

  • Comparative Study

MeSH terms

  • Adolescent
  • Adult
  • Cardiac Surgical Procedures*
  • Case-Control Studies
  • Chi-Square Distribution
  • Echocardiography / methods*
  • Endocardium / diagnostic imaging
  • Endocardium / physiopathology*
  • Exercise Test
  • Exercise Tolerance
  • Germany
  • Humans
  • Magnetic Resonance Imaging, Cine*
  • Middle Aged
  • Myocardial Contraction*
  • Observer Variation
  • Predictive Value of Tests
  • Reproducibility of Results
  • Tetralogy of Fallot / diagnosis*
  • Tetralogy of Fallot / diagnostic imaging
  • Tetralogy of Fallot / physiopathology
  • Tetralogy of Fallot / surgery*
  • Treatment Outcome
  • Ventricular Function, Left*
  • Ventricular Function, Right
  • Young Adult