Absence of Myocardial Fibrosis Predicts Favorable Long-Term Survival in New-Onset Heart Failure

TY - JOUR

T1 - Absence of Myocardial Fibrosis Predicts Favorable Long-Term Survival in New-Onset Heart Failure

AU - Gulati, Ankur

AU - Japp, Alan G.

AU - Raza, Sadaf

AU - Halliday, Brian P.

AU - Jones, Daniel A.

AU - Newsome, Simon

AU - Ismail, Nizar A.

AU - Morarji, Kishen

AU - Khwaja, Jahanzaib

AU - Spath, Nick

AU - Shakespeare, Carl

AU - Kalra, Paul R.

AU - Lloyd, Guy

AU - Mathur, Anthony

AU - Cleland, John G.F.

AU - Cowie, Martin R.

AU - Assomull, Ravi G.

AU - Pennell, Dudley J.

AU - Ismail, Tevfik F.

AU - Prasad, Sanjay K.

PY - 2018/9/1

Y1 - 2018/9/1

N2 - BACKGROUND: Myocardial fibrosis, identified by late gadolinium enhancement cardiovascular magnetic resonance, predicts outcomes in chronic heart failure (HF). Its prognostic significance in new-onset HF and reduced left ventricular ejection fraction (LVEF) is unclear. We investigated whether the pattern and extent of fibrosis predict survival in new-onset HF and reduced LVEF of initially uncertain pathogenesis.METHODS AND RESULTS: Of 120 consecutive patients with new-onset (<6 months) HF and reduced LVEF, 31 (26%) had infarct fibrosis, 25 (21%) had midwall fibrosis, and 64 (53%) had no fibrosis. During median follow-up of 8.9 years, 33 (28%) patients died. Patients with infarct fibrosis (hazard ratios [HR], 3.32; 95% CI, 1.46-7.58; P=0.004) or midwall fibrosis (HR, 2.99; 95% CI, 1.24-7.19; P=0.014) were more likely to die compared with those without fibrosis. On multivariable analysis, the pattern and extent of fibrosis were both associated with all-cause mortality (by fibrosis pattern: infarct: HR, 2.60; 95% CI, 1.08-6.27; P=0.033; midwall: HR, 2.64; 95% CI, 1.08-6.47; P=0.034; by fibrosis extent per 1%: HR, 1.07; 95% CI, 1.03-1.12; P<0.001). Fibrosis pattern also predicted composites of cardiovascular mortality or aborted sudden cardiac death (infarct: HR, 3.45; 95% CI, 1.20-9.90; P=0.022; midwall: HR, 6.59; 95% CI, 2.26-19.22; P<0.001), and all-cause mortality, HF hospitalization, or aborted sudden cardiac death (infarct: HR, 2.69; 95% CI, 1.26-5.76; P=0.011; midwall fibrosis: HR, 2.97; 95% CI, 1.37-6.45; P=0.006). Addition of fibrosis pattern to LVEF improved risk prediction for all-cause mortality (LVEF versus LVEF+fibrosis C statistic: 0.66 versus 0.71; P=0.033). Importantly, the absence of fibrosis heralded a favorable prognosis with an 85% survival rate over the duration of follow-up.CONCLUSIONS: The pattern and extent of myocardial fibrosis predict adverse outcomes in new-onset HF and reduced LVEF. In contrast, the absence of fibrosis portends a durable warranty period with a low incidence of adverse events. These findings support a role for late gadolinium enhancement cardiovascular magnetic resonance in the early risk stratification of patients with HF of uncertain pathogenesis.

AB - BACKGROUND: Myocardial fibrosis, identified by late gadolinium enhancement cardiovascular magnetic resonance, predicts outcomes in chronic heart failure (HF). Its prognostic significance in new-onset HF and reduced left ventricular ejection fraction (LVEF) is unclear. We investigated whether the pattern and extent of fibrosis predict survival in new-onset HF and reduced LVEF of initially uncertain pathogenesis.METHODS AND RESULTS: Of 120 consecutive patients with new-onset (<6 months) HF and reduced LVEF, 31 (26%) had infarct fibrosis, 25 (21%) had midwall fibrosis, and 64 (53%) had no fibrosis. During median follow-up of 8.9 years, 33 (28%) patients died. Patients with infarct fibrosis (hazard ratios [HR], 3.32; 95% CI, 1.46-7.58; P=0.004) or midwall fibrosis (HR, 2.99; 95% CI, 1.24-7.19; P=0.014) were more likely to die compared with those without fibrosis. On multivariable analysis, the pattern and extent of fibrosis were both associated with all-cause mortality (by fibrosis pattern: infarct: HR, 2.60; 95% CI, 1.08-6.27; P=0.033; midwall: HR, 2.64; 95% CI, 1.08-6.47; P=0.034; by fibrosis extent per 1%: HR, 1.07; 95% CI, 1.03-1.12; P<0.001). Fibrosis pattern also predicted composites of cardiovascular mortality or aborted sudden cardiac death (infarct: HR, 3.45; 95% CI, 1.20-9.90; P=0.022; midwall: HR, 6.59; 95% CI, 2.26-19.22; P<0.001), and all-cause mortality, HF hospitalization, or aborted sudden cardiac death (infarct: HR, 2.69; 95% CI, 1.26-5.76; P=0.011; midwall fibrosis: HR, 2.97; 95% CI, 1.37-6.45; P=0.006). Addition of fibrosis pattern to LVEF improved risk prediction for all-cause mortality (LVEF versus LVEF+fibrosis C statistic: 0.66 versus 0.71; P=0.033). Importantly, the absence of fibrosis heralded a favorable prognosis with an 85% survival rate over the duration of follow-up.CONCLUSIONS: The pattern and extent of myocardial fibrosis predict adverse outcomes in new-onset HF and reduced LVEF. In contrast, the absence of fibrosis portends a durable warranty period with a low incidence of adverse events. These findings support a role for late gadolinium enhancement cardiovascular magnetic resonance in the early risk stratification of patients with HF of uncertain pathogenesis.

KW - death, sudden, cardiac

KW - fibrosis

KW - gadolinium

KW - heart failure

KW - survival rate

UR - http://www.scopus.com/inward/record.url?scp=85055601195&partnerID=8YFLogxK

U2 - 10.1161/CIRCIMAGING.118.007722

DO - 10.1161/CIRCIMAGING.118.007722

M3 - Article

C2 - 30354674

AN - SCOPUS:85055601195

SN - 1941-9651

VL - 11

SP - e007722

JO - Circulation. Cardiovascular imaging

JF - Circulation. Cardiovascular imaging

IS - 9

ER -