ASSESSMENT OF MYOCARDIAL PERFUSION RESERVE IN PATIENTS WITH CAD ON 3 TESLA MRI

1
ASSESSMENT OF MYOCARDIAL PERFUSION RESERVE IN
PATIENTSWITH CAD ON 3 TESLA MRI
Kai-Chien Yang1,3, Mao-Yuan Su2, Yen-Wen Wu1,
Wen-Yih Tseng2, Chau-Chung Wu1,3Department of
Internal Medicine and Radiology, National Taiwan
University Hospital, Taipei, Division of
Cardiology, E-Da Hospital, Kaohsiung, Taiwan ,
Republic of China
Background Recent advance in fast MR imaging
(MRI) techniques allows assessment of myocardial
perfusion based on the time course of first-pass
dynamic contrast enhancement. Myocardial
perfusion reserve (MPR) on MRI allows detection
of regional impairment of coronary blood flow
even at the stage of moderate luminal stenosis.
Many studies have reported that MPR derived from
perfusion MR imaging in combination with
pharmacological stress test is sensitive to
perfusion impairment in coronary artery disease
(CAD). Using 3-tesla (3T) MR system to evaluate
myocardial perfusion is potentially advantageous
over lower field systems owing to higher
signal-to-noise ratio (SNR) and better perfusion
contrast. However, reports on the assessment of
myocardial perfusion using 3T system are not
available yet. This study aimed to test the
feasibility of myocardial perfusion imaging at 3T
MR system in detecting coronary artery stenosis.
Figure 1. SI curve of the LV cavity (open
circles) and of the myocardium (closedcircles)
in a normal subject.
Figure 2. Short-axis views of a patient with LAD
stenosis at rest (top row), and stress (middle
row), and the Bulls eye veiws of Upslope and
upslope ratio (MPR) at bottom row.
Method First-pass contrast-enhanced perfusion
imaging was performed in 30 age-matched healthy
subjects and 12 patients with angina in both rest
and dipyridamole-induced stress states. After
bolus injection of contrast agent, Gd-DTPA of
0.025 mmole/kg dose and injection rate of 4
ml/sec, three short-axis images from apex to base
were acquired for 80 cardiac cycles using
saturation recovery turbo FLASH sequence. Four
perfusion indices, maximal upslope (Upslope),
peak value (PV), time to peak (tPeak) and area
under the curve (AUC), and the respective
stress-to-rest ratios were derived from the
signal-time curves of the enhanced myocardium
(Fig. 1). Within 72 hours after MR examination,
all patients received coronary angiography, and
the results were correlated with MR results.
Figure 3. Vasodilatation response of the maximal
upslope (Upslope upper pannel),peak value (PV
middle pannel), and time to peak (tPeak c) in
the ischemic, remote and normal myocardial
segments.
Statistical Analysis Data are presented as
means SD or as numbers and percentages. The
vasodilatation response after dipyridamole
infusion was tested by Wilcoxon matched pairs
test. Group differences in reserve indices were
tested by Kruskal-Wallis ANOVA test and used
Dunns multiple comparison t-test to compare all
pairs of groups. Statistical significance was
considered if p lt 0.05.
Result A total of 42 subjects (30 healthy
volunteers and 12 pts with angina pectoris
completed the MR examination without severe side
effects. In healthy volunteers, perfusion indices
showed no dependence on myocardial segments or
coronary territories. For visual comparison,
spatial distributions of Upslope and Upslope
ratio were rendered in bulls eye views (Fig. 2).
From pts with angina, six out of twelve patients
were found to have significant coronary artery
stenosis in ten vessels. ROC analysis showed that
the sensitivity and specificity of Upslope ratio
and PV ratio ranged from 85 to 92, whereas
those of tPeak ratio and AUC ratio ranged from
70 to 81 (Table 4). The cutoff values for
Upslope ratio, PV ratio, tPeak ratio and AUC
ratio were 1.20, 1.20, 0.79 and 1.16,
respectively. The accuracy areas under the ROC
curves, were 0.90 (95 CI 0.765, 1.031) for MPR,
0.89 (95 CI 0.716, 1.058) for PV, 0.83 (95 CI
0.744, 0.926) for tPeak, and 0.79 (95 CI 0.599,
0.986) for AUC.
Table 1. Clinical Data of Pts Receiving Coronary
Angiography and MRI study
Conclusion We have for the first time established
the clinical feasibility of firstpass
contrast-enhanced myocardial perfusion technique
on a 3 Tesla MR system. The diagnostic power for
ischemic heart disease is comparable or even
superior to that of other myocardial perfusion
imaging modalities and 1.5 Tesla MR system.