Objective
This study sought to demonstrate the potential for myocardial arterial spin labeling (ASL) to identify the ischemic myocardial segments due to stenosis in coronary arteries as detected by X-ray angiography.
Background
Myocardial ASL is a technique for the assessment of myocardial blood flow (MBF) without contrast agents. It can be safely applied to patients with end-stage renal disease who are not candidates for first-pass imaging with contrast agents. Myocardial ASL perfusion imaging performed at rest and during adenosine stress provides perfusion reserve (MBFstress/MBFrest), which is a common indicator for the severity of coronary artery disease. In healthy myocardium, perfusion reserve is known to be approximately four [1].
Methods
Twenty nine patients were recruited from those scheduled for routine cardiac MR (CMR) and X-ray angiography. Myocardial ASL measurements were obtained from a single mid short-axis slice at rest and during adenosine infusion (dosage: 0.14 mg/kg/min) on a GE Signa 3T scanner. The ASL sequence was composed of flow-sensitive alternating inversion recovery (FAIR) tagging and balanced steady-state free precession (SSFP) imaging [2]. Perfusion reserve maps were generated in a standard short-axis view illustration by convolution with a Gaussian filter and resampling onto a polar coordinate [3].
Results
Ten of the twenty-nine patients were found to have significant stenosis on X-ray angiography. Table 1 contains the most ischemic myocardial segments in these ten patients as identified by two cardiologists using either X-ray angiogram or ASL perfusion reserve map independently. Based on McNemar’s test with Bonferroni correction, there was no significant difference between X-ray and ASL MRI in identifying ischemia in all six myocardial segments (p = 1.0000, 0.6170, 0.4795, 0.1336, 0.4795, and 0.4795). Figure 1 contains perfusion reserve maps acquired using myocardial ASL in these patients. The average standard deviation of physiological noise was 0.22 ml/g/min at rest and 0.42 ml/g/min during stress [2].
Table 1.
Most ischemic myocardial segments identified by X-ray angiograms and by ASL perfusion reserve maps
| Pts # | X-ray angiography | ASL MRI | |
|---|---|---|---|
| Worst lesion on angiogram | Ischemic myocardial segments | Ischemic myocardial segments | |
| 1 | Proximal LAD 100% | Anterior | Anterior |
| 2 | RCA 100% | Inferior, inferolateral | Inferolateral |
| 3 | LAD 90% | Anterior | Anterior, anteroseptal |
| 4 | LCS 90% (PDA) | Inferoseptal, inferior, inferolateral | Inferoseptal, inferior |
| 5 | RCA (100%) | Inferoseptal, inferior | Anteroseptal, inferoseptal, inferior |
| 6 | RCA (100%) | Inferoseptal, inferior | Inferior |
| 7 | Distal RCA 80% | Inferior | Anterolateral |
| 8 | Stent to LAD and RCA – now open | Anteroseptal, inferoseptal | Anterior |
| 9 | LCX 100%, RCA 100% | Inferior, inferolateral, anterolateral | Inferolateral |
| 10 | RCA 100% | Inferior, inferolateral | Anteroseptal |
Figure 1.
Perfusion reserve maps acquired using myocardial ASL in patients 1-10. Asterisks denote the most ischemic segments identified based on X-ray angiography.
Conclusion
There was visual agreement (except patients 7, 8, and 10) and no statistically significant difference between ischemic myocardial segments identified by ASL perfusion reserve maps and by X-ray angiograms. This suggests that myocardial ASL with vasodilation may have a potential to identify ischemic myocardial segments in patients with stenosis.