Comparison of visual scoring and quantitative planimetry methods for estimation of global infarct size on delayed enhanced cardiac MRI and validation with myocardial enzymes

https://doi.org/10.1016/j.ejrad.2009.09.027Get rights and content

Abstract

Purpose

Although delayed enhanced CMR has become a reference method for infarct size quantification, there is no ideal method to quantify total infarct size in a routine clinical practice. In a prospective study we compared the performance and post-processing time of a global visual scoring method to standard quantitative planimetry and we compared both methods to the peak values of myocardial biomarkers.

Materials and methods

This study had local ethics committee approval; all patients gave written informed consent. One hundred and three patients admitted with reperfused AMI to our intensive care unit had a complete CMR study with gadolinium-contrast injection 4 ± 2 days after admission. A global visual score was defined on a 17-segment model and compared with the quantitative planimetric evaluation of hyperenhancement. The peak values of serum Troponin I (TnI) and creatine kinase (CK) release were measured in each patient.

Results

The mean percentage of total left ventricular myocardium with hyperenhancement determined by the quantitative planimetry method was (20.1 ± 14.6) with a range of 1–68%. There was an excellent correlation between quantitative planimetry and visual global scoring for the hyperenhancement extent's measurement (r = 0.94; y = 1.093x + 0.87; SEE = 1.2; P < 0.001) The Bland–Altman plot showed a good concordance between the two approaches (mean of the differences = 1.9% with a standard deviation of 4.7).

Mean post-processing time for quantitative planimetry was significantly longer than visual scoring post-processing time (23.7 ± 5.7 min vs 5.0 ± 1.1 min respectively, P < 0.001).

Correlation between peak CK and quantitative planimetry was r = 0.82 (P < 0.001) and r = 0.83 (P < 0.001) with visual global scoring. Correlation between peak Troponin I and quantitative planimetry was r = 0.86 (P < 0.001) and r = 0.85 (P < 0.001) with visual global scoring.

Conclusion

A visual approach based on a 17-segment model allows a rapid and accurate assessment of the myocardial global delayed enhancement. This scoring method could be used on a daily practice and useful for the management strategy of post-MI patients.

Introduction

The precise quantification of infarcted myocardium in the setting of ischaemic cardiomyopathy and the distinction between viable and nonviable myocardium are important for the therapeutic management of acute myocardial infarction (AMI) patients [1]. In patients with AMI, infarct size (IS) has been shown to be a very strong factor of prediction of adverse outcome [2].

Delayed enhanced-cardiac magnetic resonance (DE-CMR) imaging has emerged and been confirmed as a reference tool for the assessment of infarcted myocardium with high spatial resolution and high levels of reproducibility [3], [4], [5]. An increasing number of patients are referred to CMR centers after acute myocardial infarction (AMI) for cardiac evaluation of their infarcted myocardium and of its consequences on left ventricular (LV) function [6].

Yet, various arbitrary methods have been employed to quantify IS from DE-CMR from simple quantitative planimetry with different signal intensity thresholdings to more complex computer-assisted methods [7], [8], [9], [10], [11]. All of these methods have various levels of accuracy and are time- and software-consuming.

Until now, no one has validated an ideal and practical method to define infarct size by DE-CMR available for daily clinical practice. A global visual segmental scoring has been described by Comte et al. [9] with good levels of accuracy compared to quantitative planimetry but its accuracy on a wide range of infarct sizes and its correlation with myocardial biomarkers and LV functional parameters remains unknown.

Our principal objective was to assess the visual segmental scoring for infarct size quantification in a group of patients with a wide range of reperfused acute coronary syndromes against quantitative planimetry and our secondary objective was to assess this method's relation with myocardial biomarkers and LV functional parameters.

Section snippets

Study population

A total of 103 patients prospectively admitted to our intensive care unit for acute myocardial infarction with or without ST-elevation were included between May 2005 and December 2008. All patients gave written informed consent according to the ethics committee of our institution standards.

Acute ST-elevated myocardial infarction (STEMI) was defined as follows: typical symptoms of acute coronary syndrome evolving for less than 6 h; persistent ST-segment elevation on admission 12-lead ECG over two

Patients

The study population comprised 103 patients (77 men, 26 women) with a mean age of 61 ± 13 years; 24 patients were admitted for NSTEMI and 79 were admitted for STEMI. MRI studies were performed 4 ± 2 days after admission.

The mean LVEF was 52.3 ± 11.6%, mean LVEDV was 132 ± 35 mL and mean LVESV was 65 ± 29 mL. The principal infarct location was anterior in 31% (n = 32) of the patients, inferior in 56% (n = 58) and lateral in 13% (n = 13).

Intra- and inter-observer variability

There was an excellent intra-observer variability (r = 0.95; y = 0.96x + 3.81; SEE = 

Discussion

In this study we confirm that the global visual segmental scoring is an accurate, easy-access and quick method to assess IS on DE-CMR. In addition, we show that this method correlates significantly with other direct or indirect markers of IS (myocardial biomarkers and LV functional parameters).

Disclosures

Nathan Mewton was funded by a research grant from the French Federation of Cardiology (Fédération Française de Cardiologie).

References (27)

  • H. Mahrholdt et al.

    Reproducibility of chronic infarct size measurement by contrast-enhanced magnetic resonance imaging

    Circulation

    (2002)
  • D.J. Pennell et al.

    Clinical indications for cardiovascular magnetic resonance (CMR): consensus panel report

    Eur Heart J

    (2004)
  • R.J. Kim et al.

    Performance of delayed-enhancement magnetic resonance imaging with gadoversetamide contrast for the detection and assessment of myocardial infarction: an international, multicenter, double-blinded, randomized trial

    Circulation

    (2008)
  • Cited by (31)

    • Biomechanics of infarcted left Ventricle-A review of experiments

      2020, Journal of the Mechanical Behavior of Biomedical Materials
      Citation Excerpt :

      Currently, this method has been intensively applied in infarct size estimation (Fieno et al., 2000; Schuijf et al., 2004; Bulluck et al., 2018). Numerous automatic algorithms for quantifying MI size accurately have been developed in (Petersen et al., 2003; Amado et al., 2004; Kuhl et al., 2004; Hsu et al., 2006; Heiberg et al., 2008; Kachenoura et al., 2008; Peters et al., 2009; Neizel et al., 2009; Mewton et al., 2011; Baron et al., 2013; Vermes et al., 2013; Biere et al., 2014; Karim et al., 2016; Engblom et al., 2016). Methods for assessment of acute MI have been recommended in (Stillman et al., 2011).

    • Comparison of various methods for quantitative evaluation of myocardial infarct volume from magnetic resonance delayed enhancement data

      2013, International Journal of Cardiology
      Citation Excerpt :

      Finally, our technique complies with clinical requirements in terms of reproducibility, objectivity and processing time (few seconds for each patient). Troponin and CK level increases have been previously reported to be correlated to infarct size visually measured either by means of Tl-201 single photon emission tomography or LGE-CMR [18,19]. Our results are consistent with the previous studies.

    • Association of ST-elevation and non-ST-elevation presentation on ECG with transmurality and size of myocardial infarction as assessed by contrast-enhanced magnetic resonance imaging

      2013, Journal of Electrocardiology
      Citation Excerpt :

      Infarct size in this study was measured by manual delineation of infarct size and it is known, that by measuring the partial volume effect this may lead to an overestimation of infarct size by 5% compared to computer assisted models.25 However the clinical relevance of this overestimation is modest and various visual assessment methods have been evaluated which show good correlation with the computer assisted methods.26 Observational studies such as ours strongly depend on the studied patient population and therapy and selection bias cannot be excluded.

    View all citing articles on Scopus
    View full text