Table 2.
CMR Studies Investigating Myocardial Fibrosis in Aortic Stenosis
| Study (Ref. #) | Year | n | Population | CMR | Biopsy | Findings |
|---|---|---|---|---|---|---|
| Native T1Studies | ||||||
| Bull et al. (55) | 2013 | 109 | Severe AS undergoing SAVR Asymptomatic moderate or severe AS |
1.5-T Native T1 shMOLLI |
19 | Native T1 correlated with CVF (r = 0.65; p = 0.002) and increased with disease severity. |
| Lee et al. (56) | 2015 | 80 | Asymptomatic moderate or severe AS | 3-T Native T1 MOLLI |
20 | Native T1 correlated with histology (r = 0.777; p < 0.001) and TTE measures of diastolic dysfunction, and was increased compared with control patients, with overlap. |
| ECV Studies | ||||||
| Flett et al. (62) | 2010 | 18 | Severe AS undergoing SAVR | 1.5-T ECV% EQ-CMR FLASH-IR |
18 | ECV% correlated with CVF (r2 = 0.86; p < 0.001). |
| Fontana et al. (77) | 2012 | 18 | Severe AS undergoing SAVR | 1.5-T ECV% EQ-CMR shMOLLI FLASH-IR |
18 | ECV% correlated with CVF (r2 = 0.685). ShMOLLI was superior to FLASH-IR. |
| White et al. (66) | 2013 | 18 | Severe AS undergoing SAVR | 1.5-T ECV% EQ-CMR DynEQ-CMR shMOLLI |
18 | ECV% by both methods correlated with CVF (r2 = 0.69; p < 0.01 and r2 = 0.71; p < 0.01). |
| Flett et al. (78) | 2012 | 63 | Severe AS undergoing SAVR | 1.5-T ECV% EQ-CMR FLASH-IR |
— | ECV% was increased compared with control subjects, with overlap. At 6 months, LVH had regressed but diffuse fibrosis was unchanged. |
| LGE Studies | ||||||
| Weidemann et al. (27) | 2009 | 46 | Severe AS undergoing AVR | LGE | 46 | LGE appeared to be concordant with histology (88% with severe fibrosis had ≥2 positive segments; 89% with no fibrosis had no positive segments) and did not regress at 9 months post-AVR. |
| Azevedo et al. (28) | 2010 | 28 | Severe AS undergoing AVR | 1.5-T LGE |
28 | LGE was present in 61%. LGE correlated with histology (r = 0.67; p < 0.001). LGE was an independent predictor of all-cause mortality (HR: 1.26; 95% CI: 1.03–1.54; p = 0.02). |
| Debl et al. (79) | 2006 | 22 | Symptomatic AS | 1.5-T LGE |
— | LGE was present in 27%. LGE correlated with more severe AS and LVH. |
| Rudolph et al. (80) | 2009 | 21 | Any AS | 1.5-T LGE |
— | LGE was present in 62%. LGE correlated with increased LV mass and end-diastolic volume index. |
| Dweck et al. (44) | 2011 | 143 | Moderate or severe AS | 1.5-T LGE |
— | LGE present in 66%. Midwall LGE present in 38%. Midwall LGE was an independent predictor of all-cause mortality (HR: 5.35; 95% CI: 1.16–24.56; p = 0.03). |
| Baron-Rochette et al. (45) | 2014 | 154 | Severe AS undergoing AVR | 1.5-T LGE |
— | LGE present in 29%. LGE was an independent predictor of all-cause mortality (HR: 2.8; 95% CI: 1.1 to 6.9; p = 0.025). |
| Rajesh et al. (81) | 2017 | 109 | Severe AS | 1.5-T LGE |
— | LGE present in 43%. Midwall LGE present in 31%. LGE predicted heart failure/hospitalization and a fall in LVEF but did not predict mortality. |
| Musa et al. (46) | 2018 | 674 | Severe AS undergoing AVR | 1.5-T, 3-T LGE |
— | LGE present in 51%. Noninfarct LGE present in 33%. Scar associated with all-cause (26.4% vs 12.9%; p < 0.001) and cardiovascular (15.0% vs 4.8%; p < 0.001) mortality in a dose-dependent fashion (for every 1% increase in scar, HR: 1.11; 95% CI: 1.05–1.17; p < 0.001 for all-cause and HR: 1.08; 95% CI: 1.01–1.17; p < 0.001 for cardiovascular mortality). Infarct and noninfarct scar were both associated with adverse outcomes. |
| de Meester et al. (82) | 2015 | 12 | Severe AS undergoing SAVR | 3-T Native T1 ECV% LGE MOLLI |
12 | LGE was present in 17 of 31 patients (from total cohort). Only ECV% correlated with histology (r = 0.79; p = 0.011). |
| Kockova et al. (57) | 2016 | 31 | Severe AS undergoing SAVR | 1.5-T Native T1 ECV% MOLLI |
31 | Patient with severe MF (>30%) on histology had higher native T1 times and ECV%. Native T1 ≥1,010 ms and ECV ≥0.32 had AUC of 0.82 and 0.85, respectively, for severe MF. |
| Chin et al. (41) | 2017 | 166 | Any AS | 3-T iECV LGE MOLLI |
11 | Midwall LGE was present in 27%. iECV correlated with histology (r = 0.87; p < 0.001) and was increased compared with control subjects. iECV + LGE predicted unadjusted all-cause mortality (36 vs. 8 deaths/1,000; p = 0.009). |
| Treibel et al. (26) | 2018 | 133 | Severe AS undergoing AVR | 1.5-T ECV% LGE MOLLI |
133 | LGE was present in 60%; noninfarct pattern was more common. Complex MF patterns. LGE, but not ECV%, correlated with CVF in all biopsies (r2 = 0.28; p < 0.001) but more in biopsies with endocardium (r2 = 0.501; p < 0.001). Combined LGE + ECV% best predicted LV remodeling and functional capacity. |
| Child et al. (83) | 2018 | 25 | Severe AS | 3-T Native T1 ECV% LGE MOLLI, shMOLLI, SASHA |
12 | Noninfarct LGE was present in 20%. Sequences differed in discrimination between health and disease as well as association with CVF. Native T1 with MOLLI correlated best (r = 0.582; p = 0.027). |
| Chin et al. (59) | 2014 | 20 | Any AS | 3-T Native T1 ECV% MOLLI |
— | ECV displayed excellent scan-rescan reproducibility and was higher in AS than control subjects. Native T1 was not as reproducible and was not significantly higher in AS than control subjects. |
| Chin et al. (40), Shah et al. (39) | 2014 | 122 | Any AS | 3-T ECV% LGE MOLLI |
— | Midwall LGE was present in 28%. ECV% and LGE were associated with elevated TnI and ECG evidence of strain. |
| Dusenberry et al. (84) | 2014 | 35 | Congenital AS | 1.5-T ECV% LGE Look-Locker |
— | LGE was present in 24%. ECV% was increased compared to control patients and correlated with TTE measures of diastolic dysfunction. |
| Treibel et al. (25) | 2018 | 116 | Severe AS undergoing AVR | 1.5-T iECV LGE MOLLI |
— | At 1 yr, cellular and matrix volume regressed. LGE was unchanged. |
| Everett et al. (42) | 2018 | 99 | 61 asymptomatic AS 38 severe AS undergoing AVR | 1.5-T, 3-T iECV LGE |
— | Midwall LGE was present in 26%. LGE progressed from baseline and was most rapid in patients with more severe stenosis. In patients undergoing AVR, iECV reduced by 11% (4%–16%) but there was no change in LGE. |
| Lee et al. (58) | 2018 | 127 | Moderate or severe AS | 3-T Native T1 LGE MOLLI |
— | LGE was present in 32.3%. Native T1 was increased compared with control patients, with overlap. Native T1 and LGE were independent predictors of poor prognosis. |
AS = aortic stenosis; AUC = area under the curve; CI = confidence interval; CMR = cardiac magnetic resonance; CVF = collagen volume fraction; DynEQ-CMR = dynamic equilibrium contract-cardiac magnetic resonance; ECV% = extra-cellular volume fraction; EQ-CMR = equilibrium contrast cardiac magnetic resonance; FLASH-IR = fast low angle single shot inversion recovery; HR = hazard ratio; iECV = indexed extracellular volume; LGE = late gadolinium enhancement; LVEF = left ventricular ejection fraction; LVH = left ventricular hypertrophy; MOLLI = modified Look-Locker inversion recovery; SASHA = saturation recovery single-shot acquisition; SAVR = surgical aortic valve replacement; shMOLLI = shortened modified Look-Locker inversion recovery; TnI = troponin I; TTE = transthoracic echocardiography.