Combined stress myocardial perfusion and late gadolinium enhancement imaging by cardiac magnetic resonance provides robust prognostic information to cardiac events

  • Otavio Coelho-Filho
  • François-Pierre Mongeon
  • Kevin Steel
  • Ron Blankstein
  • Damien Mandry
  • Bobby Heydari
  • Michael Jerosch-Herold
  • Raymond Y Kwong
Open Access
Oral presentation

Keywords

Myocardial Perfusion Imaging Late Gadolinium Enhancement Major Adverse Cardiovascular Event Late Gadolinium Enhancement Imaging Incremental Prognostic Information 

Background

Accurate non-invasive risk stratification may management and impact survival of CAD patients. Stress perfusion CMR reliably assesses ventricular function, viability and myocardial ischemia in a single examination. While prognostic information may be derived from individual components of a comprehensive CMR exam, evidence that they provide complementary prognostic information is still limited. We sought to determine whether the presence of myocardial ischemia by stress perfusion CMR provides incremental prognostic information for major adverse cardiovascular events (MACE) beyond ventricular function, the presence of myocardial scar and traditional risk factors in a large cohort of patients referred for non-invasive assessment of CAD.

Methods and results

Stress perfusion CMR was performed in 711 consecutive patients (297 females, mean age 56±15 years) referred to assess myocardial ischemia with an intermediate pre-test likelihood of CAD (mean pre-test likelihood of CAD 22±18%). Rest and vasodilator stress perfusion CMR were performed each using a 0.1mmol/Kg bolus infusion of gadolinium, followed by cine function imaging and late gadolinium enhancement (LGE) 10 minutes after a cumulative dose of 0.2mmol/Kg of gadolinium. The presence of myocardial ischemia was defined by a segmental stress-induced perfusion defect without matching segmental LGE. At a median follow-up of 21.4 months (range 2.5 months to 8.2 years), 52 MACE (8%) had occurred (29 cardiac deaths and 28 acute nonfatal MI). By univariable analysis, the presence of ischemia and LGE portended to > 11-fold and > 3-fold increases in MACE (LRχ2, 51.62 and 17.02, both P<0.0001, table1), respectively. Adjusting for age, LVEF, presence LGE and resting ST segment changes, presence of ischemia maintains a strong adjusted association with MACE (adjusted LRχ2 26.1, HR 7.4, P<0.0001). By stepwise forward selection (table 2) considering all pertinent clinical, CMR and ECG variables, presence of ischemia remained the strongest predictor of MACE in the best-overall model. A stress perfusion CMR study without ischemia and LGE predicted a very low negative annual MACE rate (0.6%, figure 1). In addition, the presence of ischemia was strongly associated with a reduced MACE-free survival (figure 2).
Table 1

Univariable prognostic association with MACE

Variable

LRΧ2

HR

P-Value

Age, per decade

21.12

1/06

<0.0001

Gender

0.05

0.94

0.8181

Hypertension

13.62

3.51

0.0002

Diabetes

13.17

2.80

0.0003

Hyperlipidemia

9.75

2.90

0.0018

Hx MI

10/80

2.58

0.0010

Hx PCI

11.60

2.71

0.0007

HX CABG

3.95

2.08

0.0469

Pre-test Probability of CAD

13.12

1.03

0.0003

Left bundle branch block

2.59

2.01

0.1074

Significant Q Waves

10.69

2.73

0.0011

Resting ST changes

27.43

4.48

<0.0001

Resting T wave inversions

10.73

2.57

0.0011

LVEF, per 10%

25.94

0.96

<0.0001

LVEDVi, per 10 ml/m 2

13.54

1.01

0.0002

LVESVi, per 10 ml/m 2

23.02

1.02

<0.0001

Resting RWMA

36.17

5.90

<0.0001

Stress perfusion defect

40.38

8.72

<0.0001

Presence of LGE

17.02

3.36

<0.0001

ISCHEMIA presence

51.62

11.53

<0.0001

ISCH-SCORE

84.06

1.19

<0.0001

Table 2

Best Overall model for MACE

Variable

LRΧ2

P-value

Hazard Ratio

ISCHEMIA presence

14.44

0.0001

5.038

ISCH-Score

7.60

0.0058

1.097

Resting ST changes

16.56

<0.0001

3.621

Figure 1

Cardiac Death/MI

Figure 2

Cardiac Death/MI

Conclusion

The presence of ischemia by stress perfusion CMR provides robust prognostic information for MACE beyond the presence of scar, LVEF, and classical clinical and ECG markers of cardiac prognosis.. The combined absence of ischemia by myocardial perfusion imaging and scar by LGE imaging identifies a very low risk population.

Copyright information

© Coelho-Filho et al; licensee BioMed Central Ltd. 2011

This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Authors and Affiliations

  • Otavio Coelho-Filho
    • 1
  • François-Pierre Mongeon
    • 1
  • Kevin Steel
    • 1
  • Ron Blankstein
    • 1
  • Damien Mandry
    • 1
  • Bobby Heydari
    • 1
  • Michael Jerosch-Herold
    • 1
  • Raymond Y Kwong
    • 1
  1. 1.Brigham and Women'sBostonUSA

Personalised recommendations