Accuracy of multi-slice computed tomography for measurement of left ventricular ejection fraction compared with cardiac magnetic resonance imaging and two-dimensional transthoracic echocardiography: A systematic review and meta-analysis

https://doi.org/10.1016/j.ejrad.2012.02.002Get rights and content

Abstract

Background

Multi-slice computed tomography (MSCT) allows non-invasive assessment of the coronary arteries and simultaneously can provide measurement of left ventricular ejection fraction (LVEF). The accuracy of newer MSCT generations (64-slice or more) for assessment of LVEF compared with magnetic resonance imaging (MRI) and two-dimensional transthoracic echocardiography (TTE) has not been evaluated in a meta-analysis.

Purpose

To evaluate, via a systematic literature review and meta-analysis, whether MSCT can assess LVEF with high accuracy compared with MRI and TTE.

Methods

Electronic databases and reference lists for relevant published studies were searched. Twenty-seven eligible studies provided mean LVEF% with its standard deviation (SD) measured by MSCT versus MRI and TTE. Meta-analysis of weighted mean difference (WMD) and Bland–Altman method were used to quantify the mean difference and agreement between MSCT compared with MRI and TTE.

Results

The results of combining 12 studies showed no significant difference in LVEF% between MSCT and MRI with a WMD of −0.11 (−1.48, 1.26, 95% CI), p = 0.88. Bland–Altman analysis showed excellent agreement between MSCT and MRI with a bias of 0.0 (−3.7, 3.7 ± 1.96SD) with 95% CI. The results of combining 15 studies showed no significant difference in LVEF between MSCT versus TTE measurements with a WMD of 0.19 (−1.13 to 1.50; 95% CI), p = 0.87. Bland–Altman analysis showed excellent agreement between MSCT and TTE with a bias of 0.3 (−4.7, 5.7 ± 1.96SD) with 95% CI.

Conclusion

The newer MSCT generations can provide accurate LVEF measurement compared to MRI and TTE. MSCT represents a valid technique for the combined evaluation of LVEF and coronary artery disease.

Introduction

Measurement of left ventricular ejection fraction (LVEF) is a well-established clinical parameter that has essential diagnostic, therapeutic and prognostic implications, particularly in the settings of coronary artery disease (CAD) and heart failure (HF) [1], [2], [3]. Several non-invasive imaging modalities can provide LVEF measurement; however, transthoracic echocardiography (TTE) is the most feasible, safest and cheapest method that has been used routinely for this purpose for decades [4], [5]. Nevertheless, TTE has some important limitations and uncertainties in case of inadequate acoustic window particularly for measurement of LV dimensions and LVEF. Excellent imaging quality can also be performed by using cardiac magnetic resonance imaging (MRI) which is considered as the gold standard for measurement of ventricular volumes and LVEF [6].

Multi-slice computed tomography (MSCT) has rapidly evolved in the last decades and has become a routinely used non-invasive method for diagnosing CAD with high accuracy [7]. Additionally, MSCT can provide accurate anatomical imaging of the different cardiac structures and measurement of LV systolic and diastolic volumes providing LVEF. Currently, MSCT scanners with 64-slice, dual source, or 320-slice techniques are the most widely used devices allowing lower radiation doses. The accuracy of LVEF determined by 64-MSCT or newer generations compared to MRI and TTE has been assessed in several studies that have included small and different populations with varying results. Some studies have shown an overestimation by MSCT [8], [9], [10] while others have shown no significant difference compared with TTE [11], [12], [13] or MRI [14], [15], [16], [17]. One meta-analysis by van der Vleuten et al. [18] showed that there was a good correlation between LVEF measured by MSCT and MRI. However the MSCT used in this meta-analysis were of four, eight and sixteen slice technique. Furthermore this study did not compare MSCT results with TTE. Accordingly, there is need for clarifying the validity of newer generations MSCT as an established method for accurate determination of LVEF. We have therefore sought a comprehensive systematic literature review and meta-analysis for this purpose.

Section snippets

Search strategy

We searched in the electronic databases: PubMed, EMBASE and Cochrane for all published studies that measured LVEF using MSCT compared with MRI and TTE. The search focused on the new MSCT generations using at least 64-slice or more. We searched without time or language limitation and up to the end of October 2011. The following keywords were used; left ventricle, ejection fraction, echocardiography, computed tomography and magnetic resonance imaging. In-press online published articles were also

Characteristics of the included studies and patients

The results of the search process are shown in Fig. 1. The initial search identified 2141 references. 2089 were excluded after the first screening. The full texts of the 52 remaining studies were evaluated to determine eligibility. 25 studies were excluded due to insufficient data or not meeting the inclusion criteria. In total, 27 eligible studies were included. Of these 27 eligible studies, 12 studies compared MSCT with MRI [14], [15], [16], [17], [19], [20], [21], [22], [23], [24], [25], [26]

Discussion

The results of the present meta-analysis showed that the new MSCT generation (64-slice or DSCT) is a new possible non-invasive method for LVEF estimation. Compared to MRI and TTE, MSCT showed no significant difference in LVEF estimation and the agreement between MSCT and MRI or TTE was excellent. Accordingly, in patients undergoing MSCT for detection of CAD and when a retrospective scanning has already been performed, a measurement of LVEF by MSCT seems to be valid and can be used as an

Conclusions and clinical perspectives

We conclude that MSCT provides similar accurate LVEF estimations compared to MRI and can therefore be considered valid for this purpose. Due to radiation and contrast risks, the currently used techniques cannot be recommended primarily for the purpose of routine functional examination for evaluating LVEF in all patients. TTE still remain the first choice technique for the overall evaluation of cardiac structures and function of patients with suspected or known CAD as well as HF. MSCT may be

Limitations of the meta-analyses

The results of these meta-analyses should be interpreted cautiously due to likely bias. The included populations were small, selected and with different baseline characteristics. None of the studies used multicenter approach limiting the generalizability of the results. The majority of the studies were analysed retrospectively. Several studies used premedication with beta-blockers when LVEF was measured with MSCT, but beta-blockers were not used when LVEF was measured with MRI or TTE.

Conflict of interest

The authors declared no conflicts of interest with respect to the data presented in this manuscript.

References (35)

  • G.J. Taylor et al.

    Predictors of clinical course, coronary anatomy and left ventricular function after recovery from acute myocardial infarction

    Circulation

    (1980)
  • H.D. White et al.

    Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction

    Circulation

    (1987)
  • J.E. Otterstad et al.

    Accuracy and reproducibility of biplane two-dimensional echocardiographic measurements of left ventricular dimensions and function

    Eur Heart J

    (1997)
  • V. Hombach et al.

    Prognostic significance of cardiac magnetic resonance imaging: update 2010

    Cardiol J

    (2010)
  • J. Abdulla et al.

    64-Multislice detector computed tomography coronary angiography as potential alternative to conventional coronary angiography: a systematic review and meta-analysis

    Eur Heart J

    (2007)
  • S.M. Ko et al.

    Assessment of left ventricular ejection fraction and regional wall motion with 64-slice multidetector CT: a comparison with two-dimensional transthoracic echocardiography

    Br J Radiol

    (2010)
  • K. Akram et al.

    Quantification of left ventricular parameters obtained by automated software for 64-slice multidetector computed tomography and comparison with magnetic resonance imaging

    Cardiovasc Intervent Radiol

    (2009)
  • Cited by (0)

    1

    Tel.: +45 60707705.

    2

    Tel.: +45 38113340.

    3

    Tel.: +45 35453545.

    View full text