Published in:
01-02-2012 | Original Paper
Clinical feasibility study for detection of myocardial oedema by a cine SSFP sequence in comparison to a conventional T2-weighted sequence
Published in: Clinical Research in Cardiology | Issue 2/2012
Login to get accessAbstract
Background
Cardiovascular magnetic resonance (CMR) T2-imaging is oedema-sensitive and can detect increased myocardial water content to potentially distinguish acute from chronic myocardial infarction (AMI/CMI). Currently applied conventional black-blood T2-weighted-turbo-spin-echo (T2-BB-TSE)-sequences cause various artefacts which limit their image quality and possibly hamper their interpretation. Image contrast of conventional cine steady-state free precession (SSFP)-sequences partly consists of T2 oedema-sensitive information. We therefore sought to prospectively evaluate SSFP cine-imaging to detect myocardial oedema and differentiate AMI from CMI.
Methods
We examined 60 patients with AMI, 30 patients with CMI and 30 healthy volunteers using a 1.5 Tesla-MR whole body scanner. In a blinded fashion, myocardial oedema was assessed with T2-BB-TSE and SSFP-sequences, late gadolinium contrast-enhanced (LGE) CMR images being deemed as the standard reference for identification of infarcted myocardium. Assessment of presence of CMR detectable myocardial oedema was performed visually and quantitatively. P < 0.05 was considered statistically significant.
Results
The contrast-to-noise ratio (CNR) in AMI patients was significantly higher (SSFP-STEMI and SSFP-NSTEMI: 19 ± 12 and 20 ± 14; T2-BB-TSE STEMI and T2-BB-TSE-NSTEMI: 33 ± 16 and 31 ± 13) than in CMI for both MR-sequences (SSFP-STEMI and NSTEMI: 3.5 ± 1.5 and T2-BB-TSE:9.3 ± 9.6, p for all <0.001). By visual analysis, SSFP images achieved a sensitivity of 96%, a specificity of 87%, positive and negative predictive values of 95 and 92% when compared to the existence of gadolinium contrast-enhanced scar imaging. Similarly, for T2-BB-TSE, sensitivity and specificity were 93 and 83% with positive and negative predictive values of 92 and 90%. Inter-observer variability was 0.90 for SSFP and 0.83 for T2-BB-TSE images.
Conclusion
A standard clinical SSFP sequence is not inferior to T2-BB-TSE for the detection of myocardial oedema and can be used to reliably distinguish AMI from CMI. Especially in patients with insufficient T2-BB-TSE image quality, the SSFP sequence may be an alternative for the detection of myocardial oedema.