Author + information
- Brian D. Hoit, MD∗ ()
- ↵∗Harrington Heart and Vascular Center, University Hospitals Case Medical Center, 11100 Euclid Avenue, Cleveland, Ohio 44106-5038
I appreciate the interest shown by Dr. Habibi and colleagues, and thank them for their recent contributions to the field of left atrial (LA) function, which as they imply, were neglected in my State-of-the Art review (2). Calibrating the completeness of these reviews can be problematic and requires balancing the novelty of a method with its demonstrated validity and clinical utility. Tissue-tracking cardiac magnetic resonance (FT-CMR) is a new and potentially useful technique for measuring strain from routine cine CMR images using feature-tracking algorithms that were initially designed for echocardiographic strain analysis. FT-CMR tracks motion of a tissue voxel using standard steady-state free precession sequences and is simpler, more practical and available, and less time consuming than other CMR-based strain techniques. However, as Dr. Habibi and colleagues state in their letter, reproducibility and repeatability data for FT-CMR ventricular strain measurements are conflicting (1). More recently, in a study of 145 healthy volunteers, global circumferential but not longitudinal strain showed reasonable agreement with tagged CMR, with acceptable interobserver reproducibility (2). In another study, there was considerable variability between FT-CMR and two-dimensional speckle tracking echo strains of the right ventricle (RV) and left ventricle (LV) in 20 patients with tetralogy of Fallot and 20 control patients (3). Importantly, reproducibility, variability, and validation data for atrial FT-CMR (longitudinal) strain are nonexistent.
Moreover, there are no peer-reviewed studies of FT-CMR atrial strain measurement. Indeed, the studies cited by Dr. Habibi and colleagues are abstracts that have not, to date, undergone a rigorous review process. Those studies concluded that addition of FT-CMR LA strain helps distinguish subjects with atrial fibrillation from controls; that FT-CMR peak systolic atrial strain, but not LA volumes or emptying fraction, was associated with heart failure after adjusting for NT-proBNP (N terminal pro brain natriuretic peptide), traditional heart failure risk factors, and LV mass; and that FT-CMR peak systolic LA strain and diastolic strain rate are associated with significant late gadolinium LV myocardial scars (1). The FT-CMR field is indisputably inchoate; acquisition technique, cine sequence, field strength, and temporal resolution are variables that are likely to affect the accuracy and precision of FT-CMR measurements of atrial strain using CMR. The need for further pre-clinical analysis in addition to clinical trials was trumpeted in a recent editorial (4).
Finally, the focus of my review was the role of LA function in prognosis. Although it is tempting to extrapolate from the experience of echocardiographic LA strain, there are no studies of FT-CMR LA strain that examine outcomes or that compare LA strain measured with that of FT-CMR to echocardiography.
Please note: Dr. Hoit is a speaker for Philips Medical.
- American College of Cardiology Foundation