Quantitative myocardial perfusion analysis with a dual-bolus contrast-enhanced first-pass MRI technique in humans

PURPOSE: To compare fully quantitative and semiquantitative analysis of rest and stress myocardial blood flow (MBF) and myocardial perfusion reserve (MPR) using a dual-bolus first-pass perfusion MRI method in humans. MATERIALS AND METHODS: Rest and dipyridamole stress perfusion imaging was performed on 10 healthy humans by administering gadolinium contrast using a dual-bolus protocol. Ventricular and myocardial time-signal intensity curves were generated from a series of T1-weighted images and adjusted for surface-coil intensity variations. Corrected signal intensity curves were then fitted using fully quantitative model constrained deconvolution (MCD) to quantify MBF (mL/min/g) and MPR. The results were compared with semiquantitative contrast enhancement ratio (CER) and upslope index (SLP) measurements. RESULTS: MBF (mL/min/g) estimated with MCD averaged 1.02 +/- 0.22 at rest and 3.39 +/- 0.59 for stress with no overlap in measures. MPR was 3.43 +/- 0.71, 1.91 +/- 0.65, and 1.16 +/- 0.19 using MCD, SLP, and CER. Both semiquantitative parameters (SLP and CER) significantly underestimated MPR (P < 0.001) and failed to completely discriminate rest and stress perfusion. CONCLUSION: Rest and stress MBF (mL/min/g) and MPR estimated by dual-bolus perfusion MRI fit within published ranges. Semiquantitative methods (SLP and CER) significantly underestimated MPR.     

Nonlinear myocardial signal intensity correction improves quantification of contrast-enhanced first-pass MR perfusion in humans

PURPOSE: To study the nonlinearity of myocardial signal intensity and gadolinium contrast concentration during first-pass perfusion MRI, and to compare quantitative perfusion estimates using nonlinear myocardial signal intensity correction. MATERIALS AND METHODS: The nonlinearity of signal intensity and contrast concentration was simulated by magnetization modeling and evaluated in phantom measurements. A total of 10 healthy volunteers underwent rest and stress dual-bolus perfusion studies using an echo-planar imaging sequence at both short and long saturation-recovery delay times (TD70 and TD150). Perfusion estimates were compared before and after the correction. RESULTS: The phantom data showed a linear relationship (R(2) = 1.00 and 0.99) of corrected signal intensity vs. contrast concentrations. Peak myocardial contrast concentration averaged 0.64 +/- 0.10 mmol x L(-1) at rest and 0.91 +/- 0.21 mmol x L(-1) during stress for TD70 and were similar for TD150 (P = not significant [NS]). The corrections were larger for stress than rest perfusion and larger for TD150 than TD70 studies (both P < 0.01). Perfusion estimates of TD70 and TD150 stress studies were significantly different before the correction (P < 0.01) but equivalent after the correction (P = NS). CONCLUSION: The nonlinearity between signal intensity and myocardial contrast concentration in perfusion MRI can be corrected through magnetization modeling. A nonlinear correction of myocardial signal intensity is feasible and improves quantitative perfusion analysis.     

Accurate assessment of the arterial input function during high-dose myocardial perfusion cardiovascular magnetic resonance

PURPOSE: To develop a method for accurate measurement of the arterial input function (AIF) during high-dose, single-injection, quantitative T1-weighted myocardial perfusion cardiovascular magnetic resonance (CMR). MATERIALS AND METHODS: Fast injection of high-dose gadolinium with highly T1 sensitive myocardial perfusion imaging is normally incompatible with quantitative perfusion modeling because of distortion of the peak of the AIF caused by full recovery of the blood magnetization. We describe a new method that for each cardiac cycle uses a low-resolution short-axis (SA) image with a short saturation-recovery time immediately after the R-wave in order to measure the left ventricular (LV) blood pool signal, which is followed by a single SA high-resolution image with a long saturation-recovery time in order to measure the myocardial signal with high sensitivity. Fifteen subjects were studied. Using the new method, we compared the myocardial perfusion reserve (MPR) with that obtained from the dual-bolus technique (a low-dose bolus to measure the blood pool signal and a high-dose bolus to measure the myocardial signal). RESULTS: A small significant difference was found between MPRs calculated using the new method and the MPRs calculated using the dual-bolus method. CONCLUSION: This new method for measuring the AIF introduced no major error, while removing the practical difficulties of the dual-bolus approach. This suggests that quantification of the MPR can be achieved using the simple high-dose single-bolus technique, which could also image multiple myocardial slices.     

Fast mapping of myocardial blood flow with MR first-pass perfusion imaging.

Accurate and fast quantification of myocardial blood flow (MBF) with MR first-pass perfusion imaging techniques on a pixel-by-pixel basis remains difficult due to relatively long calculation times and noise-sensitive algorithms. In this study, Zierler's central volume principle was used to develop an algorithm for the calculation of MBF with few assumptions on the shapes of residue curves. Simulation was performed to evaluate the accuracy of this algorithm in the determination of MBF. To examine our algorithm in vivo, studies were performed in nine normal dogs. Two first-pass perfusion imaging sessions were performed with the administration of the intravascular contrast agent Gadomer at rest and during dipyridamole-induced vasodilation. Radiolabeled microspheres were injected to measure MBF at the same time. MBF measurements in dogs using MR methods correlated well with the microsphere measurements (R2=0.96, slope=0.9), demonstrating a fair accuracy in the perfusion measurements at rest and during the vasodilation stress. In addition to its accuracy, this method can also be optimized to run relatively fast, providing potential for fast and accurate myocardial perfusion mapping in a clinical setting.     

Optimization of the arterial input function for myocardial perfusion cardiovascular magnetic resonance

PURPOSE: To determine how injection rate, cardiac function, and breathhold influence the arterial input function (AIF), in order to optimize the AIF in the clinical setting for quantitative myocardial perfusion cardiovascular magnetic resonance (CMR). MATERIALS AND METHODS: Gd (0.1 mmol/kg) bolus was injected at 3, 5, or 7 mL/second in 35 patients. In each cardiac cycle during the first-pass, a series of saturation recovery (SR) fast low-angle shot (FLASH) low resolution images with exponentially increasing SR delay times were acquired. Signal intensity (SI) time measurements were made from a region of interest (ROI) drawn in the ascending aorta (AA). The calculation of short T1s and thus peak Gd concentration [Gd] was performed by fitting the mean ROI SI against SR delay times. RESULTS: The mean peak [Gd] in the AA increased as injection rate increased from 3 mL/second (5.0 mM), to 5 mL/second (7.1 mM), to 7 mL/second (4 mM) (P < 0.0001). The peak [Gd] increased as the left ventricular stroke volume (LV SV) increased (P = 0.01). Breath holding was not found to influence peak [Gd]. CONCLUSION: In this study, we found that a high injection rate has advantages over lower injection speeds, although the duration of the AIF was apparently not significantly shortened by faster injection. The choice of expiration or inspiration as breathhold did not have a significant influence upon the AIF. Poor cardiac function was associated with a lower peak [Gd], indicating that first pass perfusion measurements in these patients will be suboptimal.     

Single- or dual-bolus approach for the assessment of myocardial perfusion reserve in quantitative MR perfusion imaging.

A dual-bolus protocol can overcome limitations due to T1-induced MR signal attenuation and hence enables more accurate quantification of myocardial blood flow (MBF) by contrast enhanced MR perfusion imaging. The study explores potential benefits of the dual-bolus technique for the assessment of myocardial perfusion reserve (MPR) over a standard single-bolus protocol. Nineteen patients without obstructive coronary artery disease as assessed by cardiac catheterization underwent a stress-rest MR perfusion study using a dual-bolus protocol. Gd-DTPA dosages of 0.005 and 0.05 mmol/kg of bodyweight were delivered as pre- and main-bolus. For comparison arterial input curves where extracted from left ventricular cavity passage including both, pre-bolus and main-bolus data. Global and segmental MPR were determined from semiquantitative and from full quantitative measures of MBF. As a result good agreement between dual- and single-bolus technique was found with relative differences of maximally 10% in global MPR estimates. For the dual bolus approach a significant relative decrease of 30% (P<0.001) was found for the coefficient of segmental MPR variation, which may allow a more reliable detection of hypoperfused segments in clinical studies.     

Comparison of different contrast agents and doses for quantitative MR myocardial perfusion imaging

PURPOSE: To investigate three different contrast agents at different injection volumes for repetitive quantitative multislice myocardial perfusion imaging using the prebolus technique. MATERIALS AND METHODS: Two consecutive prebolus perfusion measurements were performed on a 1.5 T scanner using identical injection volumes for the first and second examination to test the reproducibility for possible rest and stress examination in normal volunteers. Either 1-8 mL, 1-12 mL Gd-DTPA, 1-4 mL, 1-6 mL, 1-9 mL Gd-BOPTA, or 1-4 mL, 1-6 mL gadobutrol were applied. RESULTS: In cases where injection volumes were sufficiently small, there was no indication of significant differences in quantitative perfusion values with respect to the different contrast agents. Increasing the bolus volume improved the contrast-to-noise ratio (CNR) but led to saturation effects and underestimation of the true perfusion. The highest CNR was measured for gadobutrol (6 mL, p < 0.0005 compared to 8 mL Gd-DTPA). The smallest difference of perfusion values between the first and the second prebolus examination was found for Gd-BOPTA (p < or = 0.006 compared Gd-DTPA). CONCLUSION: Prebolus examinations for quantitative myocardial perfusion imaging are possible with all three contrast agents for sufficient small injection volumes. Gd-BOPTA was found to be advantageous for a combined quantitative rest and stress examination.     

Noninvasive determination of regional myocardial perfusion with first-pass magnetic resonance (MR) imaging

RATIONALE AND OBJECTIVES: To develop a method to provide absolute values of regional myocardial perfusion by means of color maps, and to determine myocardial perfusion reserve using magnetic resonance imaging during the first pass of gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA). MATERIALS AND METHODS: The study population consisted of five patients with hypertrophic cardiomyopathy, two with dilated cardiomyopathy, four with coronary artery disease, and one with normal coronary arteries who presented with mildly abnormal electrocardiogram findings. For each heartbeat, six continuous slices were acquired during the first pass of Gd-DTPA (0.05 mmol/kg body weight) before and during adenosine triphosphate (ATP) **stress using an electrocardiogram-triggered fast low-angle shot (FLASH) sequence on a 1.5-T magnetic resonance unit. Myocardial perfusion images were created and displayed by means of a color scale. The parameters were calculated pixel by pixel, using the upslope method. Myocardial perfusion reserve was then calculated, as the quotient of myocardial perfusion during ATP stress and perfusion before ATP stress. RESULTS: Myocardial perfusion during ATP stress in patients with normal coronary arteries (n = 1) or after successful percutaneous coronary intervention (n = 2) was increased compared with that before ATP stress. However, the patients with coronary artery disease (n = 2) failed to show increased myocardial perfusion. The patients with hypertrophic cardiomyopathy showed increased myocardial perfusion during ATP stress, although two with dilated cardiomyopathy did not. CONCLUSION: Our new technique can provide absolute values of regional myocardial perfusion by means of color maps, and has potential for widespread use for evaluation of ischemic and other types of heart disease.     

Extended coverage first-pass perfusion imaging using slice-interleaved TSENSE.

Parallel imaging applied to first-pass, contrast-enhanced cardiac MR can yield greater spatial coverage for a fixed temporal resolution. The method combines rate R=2 acceleration using TSENSE with shot-to-shot interleaving of two slices. The square root R SNR loss is largely compensated for by a longer effective repetition time (TR) and increased flip angle associated with slice interleaving. In this manner, increased spatial coverage is achieved while comparable or better image quality is maintained. Single-heartbeat temporal resolution was accomplished with spatial coverage of eight slices at heart rates up to 71 bpm, six slices up to 95 bpm, and four slices up to 143 bpm. Experiments in normal subjects (N=6) were performed to assess signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) values.     

Quantitative myocardial perfusion imaging with a MR cold pressor test

The response of myocardial blood flow to sympathetic stimulation with cold is modulated by endothelium‐related factors. As endothelial dysfunction is an early step in patients with coronary artery disease, the aim of this study was to establish a cold pressor test (CPT) setting for quantitative analysis of myocardial perfusion in a MR scanner. First pass perfusion studies were performed in 10 healthy volunteers using a 1.5 T MR scanner with a multislice steady state free precession perfusion trueFISP sequence in prebolus technique (1 cc/4 cc gadobenate dimeglumine). MR‐CPT was established using an over head ice‐water bath of the left hand. First pass perfusion imaging was started after 1 min to assure an adequate stimulus followed by a second series after 15 min to evaluate the rest perfusion. After motion correction images were segmented with an adapted, automated tool, myocardial contours were determined. Perfusion was quantitatively evaluated after contamination and baseline correction by deconvolution with the arterial input function using an exponential function model as residuum. All data could be evaluated. Mean myocardial perfusion rose from 0.61 ± 0.22 cc/g/min at rest to 1.15 ± 0.34 cc/g/min under CPT. MR myocardial perfusion values show a comparable increase under CPT as published positron emission tomography data. Consequently, CPT for the presence of endothelial dysfunction is feasible in the MR environment. Magn Reson Med, 2011. © 2011 Wiley‐Liss, Inc.     

Multislice first-pass cardiac perfusion MRI: validation in a model of myocardial infarction.

The purpose of this study was to validate a first-pass MRI method for imaging myocardial perfusion with multislice coverage and relatively small analyzable regions of interest (ROIs). A fast gradient-echo (FGRE) sequence with an echo-train (ET) readout was used to achieve multislice coverage, and a high dose of a contrast agent (CA) was used to achieve a high signal-to-noise ratio (SNR). Dogs (N = 6) were studied 1 day after reperfused myocardial infarction, and fluorescent microspheres were used as a standard for perfusion. First-pass MRI correlated well vs. microsphere flow, achieving mean R values of 0.87 (range = 0.82-0.93), 0.71 (range = 0.46-0.85), and 0.72 (range = 0.49-0.95) for subendocardial ROIs, transmural ROIs, and the endocardial-epicardial ratio, respectively. Additionally, analysis of myocardial time-intensity curves (TICs) indicated that 15.8 +/- 6 sectors, corresponding to 260 microl of endocardium, can be analyzed (R(2) > 0.95).     

Multislice first-pass myocardial perfusion imaging on a conventional clinical scanner

A technique is demonstrated for the acquisition and processing of multislice, first-pass contrast-enhanced pelfusion images in the myocardium. The acquisition is a modification of “keyhole” imaging in which time series images are acquired by sampling a limited segment of k-space, corresponding to the low spatial frequencies. In the modification demonstrated here, keyhole samples are divided into two groups that are sampled on alternate cardiac cycles. The alternate “missing” k-space portions are synthesized by Fourier interpolation. Visualization of contrast agent accumulation by image subtraction is demonstrated. A motion artifact reduction process using time domain Fourier filtering is used to reduce artifacts from respiration. Studies were performed on 46 patients at 1.5 T using gadoteridol (0.05–0.1 mmol/kg) injected into the right antecubital vein in conjunction with radionuclide imaging. Fully concordant studies were noted in 27 of these patients. Remaining studies were either partially or completely discordant for reasons relating to the differing natures of radionuclide versus MR contrast agent characteristics.     

Cepstral estimation of arterial input functions in brain perfusion imaging

PURPOSE: To investigate voxel-specific arterial input functions (AIFs) obtained through blind deconvolution using complex cepstrum liftering. Blindly estimated AIFs have the potential of reducing dispersion effects in perfusion maps and are completely user-independent. MATERIALS AND METHODS: The separability of AIFs and tissue residue functions (TRFs) in the cepstrum domain is exemplified using synthetic data, wherein the AIFs are modeled as gamma variate functions and the TRFs are modeled as exponential or linear functions. A novel separation filter is suggested. Initial results of different blind methods are illustrated using data from a stroke patient. RESULTS: The AIFs and the TRFs partly overlap in the complex cepstrum. The AIFs, obtained using the new separation filter, are closer to those obtained using noncepstral blind separation. CONCLUSION: The overlap of AIFs and TRFs in the complex cepstrum makes reliable separation of the two functions more difficult than previously described. Comparison to noncepstral blind methods suggests that similar optima are found using the new separation filter.     

Evaluation of heart perfusion in patients with acute myocardial infarction using dynamic contrast-enhanced magnetic resonance imaging

PURPOSE: To investigate the diagnostic ability of quantitative magnetic resonance imaging (MRI) heart perfusion in acute heart patients, a fast, multislice dynamic contrast-enhanced MRI sequence was applied to patients with acute myocardial infarction. MATERIALS AND METHODS: Seven patients with acute transmural myocardial infarction were studied using a Turbo-fast low angle shot (FLASH) MRI sequence to monitor the first pass of an extravascular contrast agent (CA), gadolinium diethylene triamine pentaacetic acid (Gd-DTPA). Quantitation of perfusion, expressed as Ki (mL/100 g/minute), in five slices, each having 60 sectors, provided an estimation of the severity and extent of the perfusion deficiency. Reperfusion was assessed both by noninvasive criteria and by coronary angiography (CAG). RESULTS: The Ki maps clearly delineated the infarction in all patients. Thrombolytic treatment was clearly beneficial in one case, but had no effect in the two other cases. Over the time-course of the study, normal perfusion values were not reestablished following thrombolytic treatment in all cases investigated. CONCLUSION: This study shows that quantitative MRI perfusion values can be obtained from acutely ill patients following acute myocardial infarction. The technique provides information on both the volume and severity of affected myocardial tissue, enabling the power of treatment regimes to be assessed objectively, and this approach should aid individual patient stratification and prognosis.     

Factor analysis of medical image sequences improves evaluation of first-pass MR imaging acquisitions for myocardial perfusion

RATIONALE AND OBJECTIVES: Factor analysis of medical image sequences (FAMIS) applied to gadolinium chelate-enhanced subsecond magnetic resonance (MR) imaging was evaluated as a postprocessing method for assessing myocardial perfusion in coronary artery disease (CAD). MATERIALS AND METHODS: To assess the accuracy of motion correction, five normal volunteers underwent MR imaging at rest. Thirteen patients with well-documented CAD and no myocardial infarction underwent MR imaging at rest and after dipyridamole administration. After motion correction, a single myocardial tissue factor (FAMISt) image was obtained with FAMIS for each raw MR imaging series acquisition. To evaluate how FAMIS could improve the analysis of these acquisitions, five readers visually assessed myocardial perfusion with FAMISt and raw MR images, and a multicase, multireader receiver operating characteristic analysis was performed. RESULTS: FAMISt images significantly improved detection of the perfusion defects when compared with raw MR images (P = .002). Areas under the receiver operating characteristic curves ranged from 0.84 to 0.93 with FAMISt images and from 0.48 to 0.85 with raw MR images. CONCLUSION: FAMIS applied to first-pass MR imaging series provided myocardial perfusion images that improve the objective assessment of myocardial perfusion in patients with CAD.     

Direct comparison of myocardial perfusion cardiovascular magnetic resonance sequences with parallel acquisition

PURPOSE: To directly compare the three main myocardial perfusion cardiovascular magnetic resonance (CMR) sequences incorporating parallel acquisition methods. MATERIALS AND METHODS: In 15 subjects (12 men, 57 +/- 15.7 years) referred for diagnostic coronary angiography, we acquired first-pass perfusion images (0.1 mmol/kg gadolinium-DTPA) at rest and during adenosine (140 microg/kg/min) on three separate occasions using three sequences incorporating parallel acquisition methods and approximately equivalent spatiotemporal resolution: hybrid echo planar imaging (hEPI), steady-state free precession (SSFP), and gradient echo imaging (GRE). We calculated the contrast-to-noise ratio (CNR) of each scan and blinded observers scored the presence and severity of artifacts (1, worst to 4, best), diagnostic confidence (0, low to 2, high), transmurality, area, and epicardial vessel territory of perfusion defects. RESULTS: CNR was greatest with SSFP and least with hEPI (13.15 vs 7.85 P < 0.001). The most artifacts were recorded with SSFP and least with hEPI (2.00 vs 3.03 P < 0.001). Observers were significantly more confident in reporting hEPI images (1.6 hEPI vs 0.9 SSFP, P < 0.001). Results for GRE were intermediate for all assessments. CONCLUSION: The hEPI sequence scored best for diagnostic performance despite the SSFP sequence having greater CNR. This trial favors hEPI for clinical myocardial perfusion CMR and suggests CNR should not be the sole criterion used to gauge the best candidate sequence.     

T2* effects in the dual-sequence method for high-dose first-pass myocardial perfusion

PURPOSE: To examine whether T2* effects reduce the accuracy of arterial input function (AIF) measurement by the dual-sequence method. MATERIALS AND METHODS: The dual-sequence method obtains a low-resolution AIF image and high-resolution myocardial images in each cycle, with suitable T1 weightings. It was modified to assess T2* effects in the low-resolution AIF image (4.8x4.8x10 mm voxels, TE=0.58 msec) by minimizing T1 weighting in that sequence, while the myocardial sequence remained T1-weighted. In 10 patients who underwent perfusion MRI scans (0.5 M Magnevist, 0.1 mmol/kg, 15-ml flush, 7 mL/second right antecubital) the blood signal in the left ventricle (LV) was measured at the bolus peak and compared with the first cycle's fresh magnetization signal. RESULTS: The bolus peak measured 98%+/-4% (mean+/-SD, N=20) of the value before contrast agent arrival. CONCLUSION: T2* causes insignificant error in the dual-sequence method at the stated parameters.     

Correction for partial volume errors in MR heart perfusion imaging.

Myocardial MR first-pass perfusion time courses are contaminated by signals from the ventricles (spillover) as a consequence of partial volume effects and motion. An early increase in the signal intensity from the myocardium is an indicator of contamination. This contamination leads to under- or overestimation of perfusion, depending on the amount of contamination. In this work a simple method for contamination correction is proposed: curves proportional to the signal intensity time courses in the ventricles are subtracted from the signal intensity time courses in the myocardium to minimize the variance of signal before the arrival of the contrast medium in the myocardium. The proposed correction is easy to apply, removes the contamination, and leads to more precise perfusion values.     

Comparison of first-pass Gd-DOTA and FAIRER MR perfusion imaging in a rabbit model of pulmonary embolism

PURPOSE: To compare the sensitivity of contrast-enhanced magnetic resonance imaging (MRI) and arterial spin labeling to perfusion deficits in the lung. MATERIALS AND METHODS: A rabbit model of pulmonary embolism was imaged with both flow-sensitive alternating inversion recovery with an extra radiofrequency pulse (FAIRER) arterial spin labeling and Gd-DOTA enhanced MRI. The signal-to-noise ratio (SNR) was measured in the area of the perfusion deficit and the normal lung for both techniques. RESULTS: The defect was readily visible in all images. The normal lung had an average of 3.8 +/- 1.2 times the SNR of the unperfused lung with the arterial spin labeling technique, and approximately 13.7 +/- 3.3 times the SNR with the contrast-enhanced technique. CONCLUSION: Gd-DOTA enhanced MRI provides higher SNR in pulmonary perfusion imaging; however, arterial spin labeling is also adequate and may be used when repeated studies are indicated.