Skip to main content


A simple visual algorithm incorporating the components of a routine CMR study improves the determination of infarct age compared with T2-CMR alone


T2-weighted CMR is sensitive but perhaps not specific for detecting acute MI, because T2-hyperintensity can persist for months [1]. Cine and delayed-enhancement (DE)-CMR may help in determining infarct age, since increased end-diastolic-wall-thickness (EDWT) and microvascular obstruction (MO) are frequently found in <1-month-old (acute) MI but not in 1-6-months-old (intermediate-aged) MI [1]. Given that EDWT and MO potentially resolve before T2-hyperintensity, we hypothesized that a simple, visual algorithm incorporating these components with T2-CMR could improve the determination of infarct age.


221 CMR studies were performed at various time points post-MI in 117 STEMI patients enrolled prospectively and consecutively at two centers. True MI age was known given the STEMI date. Images were scored blinded to identity and clinical information. Pre-specified markers of acute MI were: hyperintensity on T2-CMR, MO on DE-CMR, and increased-EDWT (>150% of remote) on cine-CMR. Our algorithm incorporating multiple CMR components was based on: 1) EDWT and MO resolve before T2-hyperintensity, 2) since T2-hyperintensity eventually disappears, T2-size becomes smaller than infarct size over time.


Mean age was 58±11 years. Table 1 shows the diagnostic performance of CMR for discriminating <1 from ≥1-month-old-MI as (a) individual components, (b) basic combinations, and (c) using new algorithm. T2-CMR-alone was sensitive (88%) but not specific (66%) for <1-month-old-MI resulting in only moderate accuracy (77%). Using a later cutpoint for ‘acute' MI (2-months or 3-months) did not improve accuracy since sensitivity decreased with increasing specificity. MO and increased-EDWT were very specific but not sensitive for acute MI. The basic combination of MO-or-increased-EDWT improved sensitivity (73%) while retaining specificity (97%). Basic algebraic combinations including T2-CMR did not improve overall accuracy since ‘OR' function led to low specificity while ‘AND' function led to low sensitivity. The new algorithm resulted in high sensitivity (92%) and specificity (90%). Accuracy (91%) was improved compared with T2-CMR alone (p<0.001) and compared with basic algebraic combinations involving T2-CMR (p<0.05).

Table 1 Diagnostic performance of CMR for discriminating acute (<1-month-old) MI

An additional benefit of the algorithm was the ability to identify intermediate-aged-MI (1-6-month-old). This was based on finding T2-hyperintensity-size < DE-infarct-size, and when present, patients had median infarct age of 110 days (IQR: 96, 115) (Figure 1).

Figure 1

Range infarct age for categories based on the new algorithm


A novel algorithm incorporating components of a routine CMR scan improves the determination of infarct age compared with T2-CMR alone. Certain CMR findings may be specific for intermediate-aged MI.




  1. 1.

    JCMR. 2011, 13 (suppl 1): O68-

Download references

Author information

Correspondence to Martijn W Smulders.

Rights and permissions

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 (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article

Cite this article

Smulders, M.W., Bekkers, S.C., Kim, H.W. et al. A simple visual algorithm incorporating the components of a routine CMR study improves the determination of infarct age compared with T2-CMR alone. J Cardiovasc Magn Reson 15, O73 (2013).

Download citation


  • Individual Component
  • Infarct Size
  • Clinical Information
  • Diagnostic Performance
  • Additional Benefit