Tissue Characterization
Dana C. Peters, PhD, FSCMR
Professor of Radiology and Biomedical Imaging
Yale University, Connecticut, United States
Jérôme Lamy, PhD
Postdoctoral associate
Yale University
New haven, Connecticut, United States
Roy Taoutel, MD
Research Scientist
Yale University, United States
Romy Chamoun
Research Scientist
Yale University, United States
Joseph G. Akar, MD, PhD
Associate Professor
Yale University, United States
Steven A. Niederer, PhD
Professor
King's College London, United Kingdom
Hamid Mojibian, MD
Associate Professor of Radiology and Biomedical Imaging; Director, Cardiac CT/MR Imaging;
Yale University
New Hven, Connecticut, United States
Steffen Huber, MD
Associate Professor
Yale University
New Haven, Connecticut, United States
Lauren A. Baldassarre, MD, FSCMR
Associate Professor, Cardiology Director of Cardiac MR/CT
Yale University
NEW HAVEN, Connecticut, United States
Judith L. Meadows, MD
Associate Professor Term
Yale University
New Haven, Connecticut, United States
A relationship between left atrial (LA) reservoir strain and LV end-diastolic pressure (or similarly mean LA pressure, LAP) has been demonstrated in many studies (1) (2), but not in an atrial fibrillation (AF) cohort. In this study, we hypothesized that tissue fibrosis–common in AF patients--might be a modifier of the LA strain vs. pressure relationship. To understand the impact of LA fibrosis, we investigated the relationship of the LA “stiffness index” (LAP/ LA reservoir strain) to left atrial fibrosis, complemented by a comprehensive study of LA strain.
Methods:
Sixty-seven patients with AF underwent a standard cardiac MR exam including 3D late gadolinium enhancement (LGE) of the atrium and short-axis and long-axis cine views (2 and 4-ch) within 30 days prior to AF ablation, at which procedure invasive mean LAP was measured. Cardiotrack (3) feature tracking software (3) was used to comprehensively analyze LA longitudinal strain values, strain rates and timings (normalized by RR) (Figure 1A). In 41 subjects, 3D LA LGE was performed (1RR between inversions) with fat-saturation and navigator gating, with TR/TE/q= 5ms/1.8ms/15°, and reconstructed voxel size of 0.7 x 0.7x 1.5mm3 with ~5 minutes total acquisition time. Quantitative analysis of the LA LGE was performed by semi-automated segmentation using 3D slicer and a patient-specific threshold equal to an image intensity ratio (IIR) of ~1.4 of mean blood pool signal (Figure 1B) (4). LA LGE was normalized by BSA (LGEi), and compared to Stiffness Index as used in echocardiography, i.e. LAP/LA reservoir strain (5).
Results:
The average age of the AF cohort (n=67) was 61 ± 8 years old, with 45 male patients (67%), with a BMI of 32 ± 6 kg/m2. LAP was not highly elevated (12.2±5.2 mmHg). LVEF was < 50% in 14 of 67 subjects. The median value for LA fibrosis was LGEi = 0.17 ml/m2. The mean min LAVi was 33.0±16.9ml/m2, and the mean LA reservoir strain, conduit strain and active strain were 21.0±10.4%, 14.8±4.4% and 11.7±4.9% respectively.
Figure 2 presents the heat map of all observed correlations. LAP was only correlated to maximal LAVi (R=+0.33), booster LAVi (R=0.28), and time to peak reservoir strain rate (R=+0.32) (all p< 0.05) (Figure 3A). However, in a subgroup of patients with minimal LA LGE< 0.17ml/m2, LAP became correlated with other cardiac timings: shorter length of conduit strain time (R=-0.52, p=0.02), longer time to both peak reservoir strain (R=+0.50, p=0.02) and peak conduit strain rate (R=+0.47, p=0.03). Atrial fibrosis (LGEi) correlated with all LAVis, total and active LAEF, LA reservoir strain, and all three LA strain rates. However, of all tested metrics (shown in Figure 2) its strongest correlate was Stiffness Index (R=+0.59, p< 0.0001) (Figure 3B).
Conclusion: In this AF cohort, there was no correlation between LA strains and LAP but rather a strong link between the Stiffness Index and atrial fibrosis. This is an expression of Hook’s law that stress/strain is proportional to elasticity, e.g. in our study LA LGE.