PROGNOSTIC IMPLICATIONS OF DIFFUSE INTERSTITIAL FIBROSIS IN ASYMPTOMATIC PRIMARY MITRAL REGURGITATION

Chronic volume overload in primary mitral regurgitation (MR) leads to progressive left ventricular (LV) remodeling with extracellular matrix expansion (diffuse interstitial fibrosis)¹. CMR allows for the noninvasive quantification of this via extracellular volume (ECV) mapping, which has demonstrated good correlation with histologic findings of interstitial myocardial fibrosis in several conditions including valvular heart disease². There has been no study reporting a prognostic link between elevated ECV and outcomes in primary MR. We utilized ECV, as a surrogate for diffuse interstitial fibrosis, to evaluate whether elevated ECV represents a prognostic biomarker in asymptomatic patients with primary MR.

Patients undergoing contrast-enhanced CMR for assessment of chronic MR at Houston Methodist Hospital (Houston, TX) from 2012 to 2017 were enrolled into a prospective observational registry. To avoid confounding etiologies of fibrosis, we excluded patients with history of 1)coronary artery disease, 2)cardiomyopathy, 3)infiltrative disease (i.e. amyloidosis/sarcoidosis), 4)prior cardiac surgery or 5)congenital heart disease; and CMR findings of: 1)LV ejection fraction <50%; 2)coexisting valvular disease greater than mild in severity; or 3)greater than mild concentric LV hypertrophy (LV thickness >1.3 cm and relative wall thickness >0.42). These aforementioned criteria also excluded patients with secondary MR. Only asymptomatic patients with at least moderate primary MR (regurgitant fraction ≥30%) and no ACC/AHA class I indications for mitral correction were included for longitudinal follow up³. Patients were followed until they developed 1)referral for mitral valve surgery due to symptoms or other class I indications³ or 2)cardiovascular death. The study was approved by the institutional review board at the Houston Methodist Research Institute.

CMR images were acquired using 1.5-T or 3.0-T scanners (Siemens Avanto, Aera, or Verio). Our protocol and image analysis have been described previously⁴. In brief, cine-CMR was performed for anatomic and functional assessment using a steady-state free-precession sequence. Flow across the aortic valve was ascertained using phase-contrast imaging. An ECG-gated modified Look-Locker inversion recovery sequence (MOLLI) was performed at a mid-LV short axis in a matching position pre-and post-gadolinium (~15 min after 0.15 mmol/kg of extracellular agent)². The pre-and post-contrast MOLLI acquisition utilized a 5(3)3 and a 4(1)3(1)2 sampling scheme, respectively. Quantitative parametric images of myocardial T1 were generated in the mid-LV septum (excluding areas with replacement fibrosis) at the same location for pre-and post-contrast. ECV was calculated per SCMR Guidelines².

Of 144 asymptomatic patients included, 62 (decompensated cohort) had clinical events (59 referred for mitral surgery, successful repair in 95%, and 3 suffered cardiovascular deaths) while 82 remained asymptomatic (compensated cohort). The indication for surgical referral was development of symptoms in 90% of patients. The mean time to clinical event in the decompensated cohort was 12.6±16.8 months and the mean follow-up time in the compensated cohort was 33.8±23.9 months.

Clinical and most of the imaging variables were similar among cohorts (decompensated vs. compensated: age 64.1 vs. 60.6 years, LV ejection fraction 66% vs. 67%, LV end-diastolic volume 100.1 vs. 92.1 ml/m2, prevalence of replacement fibrosis 34% vs. 31%). Patients in the decompensated cohort had more severe MR (regurgitant fraction 47% vs. 38%;p<0.001), larger left atrial size (83.4 vs. 67.9 ml/m2;p<0.001) and higher ECV (median ECV 28.1% vs. 27.1%;p<0.001). Prevalence of elevated ECV (ECV ≥30%, threshold selected a priori based upon normal volunteers and confirmed optimal by ROC curve analysis) was higher in the decompensated group (32% vs. 6%;p<0.001). On multivariate Cox modeling, only MR severity (HR 2.69, 95% CI 1.75-4.14, for 10% increment of regurgitant fraction) and elevated ECV (HR 2.25, 95% CI 1.24-4.11) remain independent predictors of decompensation. For any given MR severity, patients with elevated ECV had worse outcome than those without (Figure).

Figure Legend:

Kaplan Meier curves for survival without mitral valve surgery stratified by mitral regurgitant fraction (RF) and extracellular volume (ECV).

Symptoms related to MR are considered a definitive indication for mitral valve surgery³. However, once they have occurred, the prognosis of patients become unfavorable since operative mortality is higher and postoperative survival is lower even with normal LV size and systolic function⁵. Imaging biomarkers that can detect subclinical changes prior to symptoms or development of LV systolic dysfunction/dilatation may have an important role in management of patient with MR. In our study, the severity of MR, and elevated ECV were the only factors that remained independently predictive with adverse events, while traditional parameters such as LV ejection fraction and LV size did not. Incorporating ECV with quantitative MR severity assessment by CMR may aid in further stratifying patients who were likely to develop future indications for mitral valve surgery. It is possible that less compliant ventricles from greater interstitial fibrosis burden (as measured by elevated ECV) may lead to development of symptoms despite lesser degrees of LV dilation while ventricles with lesser interstitial fibrosis may remain compliant and able to accommodate excessive volume without development of symptoms. Future studies with a larger sample size will be required to confirm our findings, validate the ECV threshold for patient selection, and ultimately demonstrate a potential clinical benefit of early mitral valve surgery.