Electrocardiogram-gated single-photon emission computed tomography versus cardiac magnetic resonance imaging for the assessment of left ventricular volumes and ejection fraction: a meta-analysis.

OBJECTIVES The purpose of this study was to evaluate the accuracy of electrocardiogram (ECG)-gated single-photon emission computed tomography (SPECT) for assessment of left ventricular (LV) end-diastolic volume (EDV), end-systolic volume (ESV) and ejection fraction (EF) compared with the gold standard of cardiac magnetic resonance imaging (MRI). BACKGROUND Several comparisons of ECG-gated SPECT with cardiac MRI have been performed for evaluation of LV volumes and EF, but each has considered few subjects, thus leaving uncertainty about the frequency of discrepancies between the two methods. METHODS We performed a meta-analysis of data on 164 subjects from nine studies comparing ECG-gated SPECT versus cardiac MRI. Data were pooled in correlation and regression analyses relating ECG-gated SPECT and cardiac MRI measurements. The frequency of discrepancies of at least 30 ml in EDV, 20 ml in ESV and 5% or 10% in EF and concordance for EF < or =40% versus >40% were determined. RESULTS There was an overall excellent correlation between ECG-gated SPECT and cardiac MRI for EDV (r = 0.89), ESV (r = 0.92) and EF (r = 0.87). However, rates of discrepancies for individual subjects were considerable (37% [95% confidence interval [CI], 26% to 50%] for at least 30 ml in EDV; 35% [95% CI, 23% to 49%] for at least 20 ml in ESV; 52% [95% CI, 37% to 63%] for at least 5% in EF; and 23% [95% CI, 11% to 42%] for at least 10% in EF). The misclassification rate for the 40% EF cutoff was 11%. CONCLUSIONS Electrocardiogram-gated SPECT measurements of EDV, ESV and EF show high correlation with cardiac MRI measurements, but substantial errors may occur in individual patients. Electrocardiogram-gated SPECT offers useful functional information, but cardiac MRI should be used when accurate measurement is required.