Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print November 5, 2001, 10.1148/radiol.2221010560
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2221010560v1
222/1/127    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Langerak, S. E.
Right arrow Articles by de Roos, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Langerak, S. E.
Right arrow Articles by de Roos, A.
(Radiology 2002;222:127-135.)
© RSNA, 2001

Cardiac Imaging

MR Flow Mapping in Coronary Artery Bypass Grafts: A Validation Study with Doppler Flow Measurements1

Susan E. Langerak, MSc, Patrik Kunz, PhD, Hubert W. Vliegen, MD, J. Wouter Jukema, MD, Aeilko H. Zwinderman, PhD, Paul Steendijk, PhD, Hildo J. Lamb, PhD, Ernst E. van der Wall, MD and Albert de Roos, MD

1 From the Depts of Cardiology (S.E.L., H.W.V., J.W.J., P.S., E.E.v.d.W.), Radiology (S.E.L., P.K., H.J.L., A.d.R.), and Medical Statistics (A.H.Z.), Leiden Univ Medical Center, Albinusdreef 2, 2333 ZA Leiden, the Netherlands; and the Interuniversity Cardiology Inst of the Netherlands, Utrecht (S.E.L, E.E.v.d.W., A.d.R.). Received Mar 2, 2001; revision requested Apr 11; revision received Jun 8; accepted Jul 5. S.E.L. supported by grant 97.173 from the Netherlands Heart Foundation. Address correspondence to A.d.R. (e-mail: a.de_roos@lumc.nl).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To validate fast magnetic resonance (MR) flow mapping with intravascular Doppler flow measurements in vitro and in patients with nonstenotic and stenotic coronary artery bypass grafts.

MATERIALS AND METHODS: MR and Doppler flow measurements were performed in a small-diameter flow phantom with physiologic flow conditions and at baseline and during adenosine stress in 27 grafts in 23 patients, who were scheduled for cardiac catheterization. At invasive analysis, the grafts were divided into those with stenosis of less than 50% (nonstenotic) and those with stenosis greater than or equal to 50% (stenotic). In vitro velocity values and velocity values in nonstenotic and stenotic grafts were compared with linear regression analysis, and the in vitro interstudy variability was determined.

RESULTS: Excellent correlations in average peak velocity (r = 0.99, P < .001) and diastolic peak velocity (r = 0.99, P < .001) were demonstrated in vitro between MR and Doppler flow measurements, with less than 5% interstudy variability. MR and Doppler flow measurements revealed good correlations in peak velocity and velocity reserve both in nonstenotic (n = 20) (average peak velocity: r = 0.81, P < .001; diastolic peak velocity: r = 0.83, P < .001; velocity reserve: r = 0.56, P = .010) and stenotic (n = 7) (average peak velocity: r = 0.83, P < .001; diastolic peak velocity: r = 0.78, P = .001; velocity reserve: r = 0.70, P = .078) grafts.

CONCLUSION: Fast MR flow mapping provides noninvasive measures of peak velocity and velocity reserve, which closely correlate with Doppler values both in vitro and in nonstenotic and stenotic grafts.

Index terms: Arteries, grafts and prostheses, 943.1269, 949.1269 • Arteries, MR, 544.121416, 544.12144, 943.129412, 943.129416, 943.12944 • Arteries, stenosis or obstruction, 943.721, 949.721 • Blood, flow dynamics • Coronary vessels, flow dynamics, 544.731 • Magnetic resonance (MR), vascular studies, 943.12944 • Ultrasound (US), Doppler studies, 943.12985


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Nowadays, the functional assessment of coronary artery and coronary artery bypass graft stenoses is an integrated and accepted part of daily clinical practice in the catheterization laboratory. Sensor-tipped guide wires, such as the Doppler flow wire, are used to determine the average peak velocity (APV), the systolic peak velocity, the diastolic peak velocity (DPV), and the absolute velocity reserve. A reduced capacity to increase the APV during pharmacologic stress has been demonstrated in stenotic vessels and helps the clinician decide whether the lesion is flow limiting and requires an intervention.

The availability of a noninvasive diagnostic tool to identify flow-limiting stenoses in coronary artery bypass grafts or in distal vessels (graft stenosis) would be a major step forward in cardiac care. Magnetic resonance (MR) imaging allows measurements of peak velocities and velocity reserve in grafts, indicating the potential value of this modality. Previous MR angiographic and MR flow mapping studies (1–5) allowed the differentiation between patent and occluded grafts, but the detection of hemodynamically significant graft stenoses has remained difficult. MR flow mapping with a conventional breath-hold flow sequence proved useful in the identification of severe stenoses in the distal left main and proximal and middle left anterior descending coronary artery, whereby a decreased velocity reserve was demonstrated in greater than 70% luminal stenoses (6). Conventional breath-hold MR flow mapping with low temporal resolution has been validated in coronary arteries with Doppler flow measurements and resulted in systematically lower velocities at MR imaging but similar velocity reserve values (6–9).

Recently, a fast MR flow mapping sequence with turbo-field echo-planar imaging that allowed flow quantification in grafts within a breath hold of 20 cardiac cycles and high temporal resolution was described (10). This permits the functional evaluation of multiple grafts in patients during a single 6-minute adenosine infusion.

The purpose of the study was to validate fast MR flow mapping with intravascular Doppler flow measurements in vitro and in patients with nonstenotic and stenotic grafts.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Flow Phantom
A computer-controlled flow pump (UHDC flow simulator; University Hospital Development, London, Ontario, Canada) was used to generate physiologic flow waveforms with user-defined variability in frequency and shape (11). A biphasic flow pattern, which is typical for grafts, was applied and adjusted to yield 10 flow rates ranging from 18.4 to 184.0 mL/min at a frequency of 60 flow cycles per minute, which is comparable to an in vivo heart rate of 60 beats per minute. An electric signal was synchronized to the flow cycle and used for prospective electrocardiographic (ECG) triggering. The phantom (Fig 1) consisted of a cylinder filled with stationary fluid, 0.015 mmol/L of manganese chloride solution in water, which has T1- and T2-weighted relaxation times of 1,000 and 350 msec, respectively. Within this cylinder, a thin 4-mm-diameter glass tube with a length of at least 72 cm was positioned and connected to the pump by means of two acrylic 10-mm-diameter tubes with a length of 5 m and a 3-cm compliant silicon tube for access of the flow wire.



View larger version (52K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Schematic representation of the small-diameter flow phantom used for turbo-field echo-planar MR flow mapping and intravascular Doppler flow measurements. The arrows indicate the direction of the flow. PVC = polyvinylchloride.

 
In Vitro MR Flow Mapping
To perform MR flow measurements, the pump was placed outside the MR imaging room, and the system was filled with blood-mimicking fluid (Shelley Medical Imaging Technology, London, Ontario, Canada) and cleared of air bubbles. The phantom was placed parallel to the bore of the magnet. MR imaging was performed with a 1.5-T unit (Gyroscan ACS/NT; Philips Medical Systems, Best, the Netherlands) with a gradient system (25 mT/m, 100 mT/m/msec, PowerTrak 6000; Philips Medical Systems) and a cardiac software patch (CPR 6; Philips Medical Systems). A five-element phased-array cardiac synergy coil was placed around the flow phantom.

MR flow mapping was performed perpendicular to the glass tube under the 10 flow conditions, and measurements were repeated to determine the interstudy variability. A turbo-field echo-planar imaging MR flow sequence involving multiple excitations within one heart phase interval was applied, and each excitation was followed by echo-planar imaging readout. Owing to the use of multiple short echo-planar imaging readouts, this technique allows fast image acquisition while keeping the echo time low (Fig 2). Two alpha pulses, each of which was followed by three echo-planar imaging readouts, were applied per heart phase interval and resulted in a shot length of six k lines and a temporal resolution of 23 msec. Flow-sensitive and flow-compensated images were acquired in 10 shots and resulted in an imaging duration of 20 heartbeats.



View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Schematic representation of the breath-hold turbo-field echo-planar imaging (EPI) pulse sequence used for MR flow mapping. This sequence involves two excitations (two alpha pulses) within one heart phase interval, each of which are followed by three echo-planar imaging readouts. The repetition time (TR) of 11 msec is the time it takes to acquire three k lines after one alpha pulse, and the temporal resolution of 23 msec is the shot length (tshot) of six k lines (two x repetition time). Flow-sensitive and flow-compensated images were acquired in 10 shots and resulted in an imaging duration of 20 heartbeats.

 
The illustration in Figure 2 indicates that the repetition time of 11 msec is the time it takes to acquire three k lines after one alpha pulse, and the temporal resolution of 23 msec is the shot length of six k lines (two x repetition time). A field of view of 200 x 100 mm and a data acquisition matrix of 128 x 60 rendered an in-plane spatial resolution of 1.6 x 1.6 mm, which was reconstructed to 0.8 x 0.8 mm by means of zero filling of k space. Other imaging parameters were a section thickness of 6 mm, flip angle of 20°, echo time of 4.6 msec, and prospective ECG triggering (10). The anterior two elements of the phased-array cardiac synergy coil were used for flow mapping. Velocity encoding (range, 50–100 cm/sec) was adjusted accordingly to prevent aliasing.

In Vitro Doppler Flow Measurements
The flow pump and phantom were subsequently placed in another room, apart from the magnet. The MR blood-mimicking fluid was replaced with ultrasonography blood-mimicking fluid (Shelley Medical Imaging Technology), and the system was cleared of air bubbles. A 7-F sheath was introduced to the flow phantom by means of a puncture through the short compliant tube. Subsequently, a 7-F guiding catheter and 0.014-inch Doppler flow wire (Flo Wire; Cardiometrics, Mountain View, Calif) were advanced into the glass tube, placed at a position similar to that in the MR flow measurement, and adjusted until a stable flow signal was acquired. Doppler flow measurements were carried out twice with the same flow conditions as those used during MR imaging.

Patient Population
Patients with a history of coronary artery bypass graft surgery, who were scheduled to undergo cardiac catheterization for the evaluation of recurrent chest pain, were screened from February to July 1999 for potential inclusion in the study. Patients with chronic obstructive pulmonary disease, sick-sinus syndrome, and/or second- or third-degree atrioventricular block were excluded because of contraindication to adenosine (12,13). Other exclusion criteria were unstable angina, presence of a pacemaker, atrial fibrillation, claustrophobia, and inability to lie flat. Patients were not allowed to have caffeine-containing beverages on the day of MR examination and cardiac catheterization. Inclusion and exclusion criteria were met by 33 patients. In eight patients, no paired MR and Doppler flow data were obtained due to logistic reasons (three patients), intravenous administration of nitrates because of chest pain at the time of Doppler flow mapping (one patient), presence of occluded grafts (two patients), no MR flow mapping during stress (one patient), and inability to perform breath holding (one patient).

In 25 patients, paired MR and Doppler flow data were obtained in 30 grafts. MR imaging was performed before cardiac catheterization (mean, 2.9 days ± 2.6; range, 1–10 days) in 14 patients, the same day in two patients, and after cardiac catheterization (mean, 14.8 days ± 23.6; range, 2–76 days) in nine patients. No change in the clinical status of patients was found between MR examination and cardiac catheterization. We recorded the sex, age, number of years after bypass surgery, number and type of grafts, medical history, and types of medication. The protocol was approved by the medical ethics committee of Leiden University Medical Center, and all patients gave informed consent.

MR Flow in Bypass Grafts
All patients were examined while in a supine position after the placement of a respiratory belt to monitor the breath-hold procedure, three ECG leads for cardiac triggering, an ECG patch at the center of the thorax for patient monitoring, and the five-element phased-array cardiac synergy coil. Turbo-field echo-planar imaging MR flow mapping was performed in the proximal part of the graft and perpendicular to the graft segment according to a standardized protocol (10,14). We used a fixed-velocity encoding of 75 cm/sec. The distance from the graft origin to the plane of the MR flow image was measured by one author (S.E.L.). Flow mapping was performed during an end-expiratory breath hold of 20 cardiac cycles. To prevent a Valsalva or Müller maneuver, the patient was instructed to breathe in and out three times and stop breathing after expiration, with the mouth open. After the baseline flow measurement, intravenous adenosine infusion (140-µg/kg/min) was started. When maximal hyperemia was achieved after 2 minutes of infusion, the flow measurement was repeated. The ECG signal and blood pressure at baseline and during adenosine-induced stress were monitored and registered (Millennia 3500; Invivo Research, Orlando, Fla) in all patients. The presence and duration of chest pain, dyspnea, flush, headache, and dizziness during the stress test were registered.

Doppler Flow in Bypass Grafts
Vascular access was obtained by using the femoral approach with the Seldinger technique and a 6- or 7-F guiding catheter. An experienced interventional cardiologist advanced a 0.014-inch Doppler guide wire into the graft and positioned it visually at a level similar to that at MR flow measurement. The position was adjusted until a stable signal was acquired, and the baseline peak velocity was measured over at least three entire cardiac cycles. After a bolus injection of 18-µg adenosine into the graft, the hyperemic peak velocity was recorded. When the peak velocity had returned to baseline, baseline and stress peak velocity measurements were repeated. Peak velocity curves, ECG signal, and blood pressure were displayed and imported into a personal computer. Thereafter, 0.3 mg of nitroglycerine was selectively injected into the graft, and the graft was depicted according to standardized procedures. When visual graft evaluation revealed a diameter of stenosis greater than 20% in the graft or coronary segments beyond the distal anastomosis, subsequent quantitative coronary analysis was performed (Heart Core, Leiden, the Netherlands) to quantify the severity and location of stenosis. On the basis of the most severe stenosis in the graft or coronary segments beyond the distal anastomosis, the grafts were divided into two groups: grafts with luminal stenosis of less than 50% (nonstenotic) and grafts with luminal stenosis greater than or equal to 50% (stenotic).

Data Analysis
Paired modulus and phase MR images were analyzed by using an analytic software package (FLOW; Medis, Leiden, the Netherlands). The quality of the MR flow images was evaluated, and the presence and cause of degraded image quality were registered. A region of interest of 2 x 2 pixels was placed on the phase flow images of either the glass tube or the graft by one author (S.E.L.). They were positioned such that the pixels with the highest velocity over the vessel area were included (8). In each cardiac phase, the mean velocity over the region of interest was determined and defined as the peak velocity over the velocity profile. MR peak velocity (cm/sec) was averaged over the cardiac cycle to determine APV, and the highest peak velocity during diastole was defined as the DPV (cm/sec). Velocity reserve was calculated as the ratio between APV during stress and at baseline. Doppler peak velocity was analyzed off-line by using a computer with a custom-made software program and averaged over three cardiac cycles to calculate APV (cm/sec). Similar to the results of MR data analysis, the highest peak velocity during diastole was the DPV, and the velocity reserve was computed as the ratio of the stress and baseline APV.

Statistical Analysis
In vitro MR and Doppler velocities were compared by means of linear regression analysis and analyzed by means of Bland-Altman analysis (15). Interstudy variability (in percentage) of in vitro MR flow measurements was assessed with linear regression analysis of the two MR measurements and calculated as the absolute difference between the two MR measurements divided by their mean. Similarly, the interstudy variability of in vitro Doppler flow measurements was determined.

Since the distribution of in vivo velocity values was highly skewed, log transformation was performed on APV and DPV values obtained at in vivo MR and Doppler measurements. Data were compared with linear regression analysis results and analyzed by means of Bland-Altman analysis (15). The difference in linear regression between nonstenotic and stenotic grafts was compared by means of covariance analysis. In vivo data are represented as geometric means plus or minus coefficients of variation. Differences between (a) baseline and stress velocity values, (b) MR and Doppler data, and (c) velocity values in nonstenotic and stenotic grafts were tested with a paired or unpaired Student t test. A P value of less than .05 was considered to indicate a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
In Vitro Flow Data
An excellent correlation in APV (r = 0.99, P < .001) and DPV (r = 0.99, P < .001) (Fig 3) was demonstrated between MR and Doppler flow measurements. Bland-Altman analysis revealed good agreement between fast MR flow mapping and the Doppler flow wire but lower APV and DPV at fast MR flow mapping, as compared with the Doppler flow wire (Fig 3). Analyses of the first and second MR flow acquisitions revealed that fast MR flow mapping was very reproducible. This was reflected by an excellent correlation in APV (r = 0.99, P < .001) and DPV (r = 0.99, P < .001) between the two measurements (Fig 3) and a mean interstudy variability of 3.7% ± 3.8 for APV and 5.7% ± 7.2 for DPV. Similar results were obtained by comparing the velocity data of the first and second Doppler flow maps. An excellent correlation coefficient was found between the two Doppler flow measurements for APV (r = 0.99, P < .001) and DPV (r = 0.99, P < .001), and the mean interstudy variabilities were 4.3% ± 4.6 and 4.4% ± 4.8, respectively.



View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. The plot graphs show the correlation of APV (cm/sec) (A) and DPV (cm/sec) (B) with MR flow mapping and the Doppler flow wire in the flow phantom. The limits of agreement between noninvasively and invasively measured peak velocity values are depicted in the Bland-Altman plots (C, D). Dashed horizontal lines in C and D show the mean difference between MR flow mapping and the Doppler flow wire plus or minus 2 SDs.

 
Patients
Breathing artifacts resulted in degraded image quality in three grafts. Flow images of the following were excluded: two single vein grafts to the right coronary artery and one sequential vein graft to the circumflex coronary artery. One of the grafts to the right coronary artery was subtotally occluded. The remaining 27 grafts in 23 patients were used for analysis. Patient characteristics are summarized in Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Baseline Characteristics of Patients

 
The mean baseline heart rate during MR examination was 62.4 beats per minute ± 8.8 and increased to 72.3 beats per minute ± 9.0 (P < .001) during adenosine stress. Similar blood pressures were obtained during MR examination at baseline (108.3 mm Hg ± 20.8) and during stress (113.6 mm Hg ± 18.9) (P = .42). During the invasive procedure, a mean baseline heart rate of 64.5 beats per minute ± 9.8 and mean stress heart rate of 63.7 beats per minute ± 9.7 (P = .019) were found, and the mean aortic pressure (baseline, 94.7 mm Hg ± 15.1; stress, 95.7 mm Hg ± 13.3; P = .33) did not change after adenosine injection. The heart rate during stress (P < .001) and the mean aortic pressure at baseline (P = .012) and during stress (P = .001) were lower during the invasive procedure than during MR examination.

In 17 (74%) of 23 patients, transient side effects such as chest pain (12 patients), dyspnea (12 patients), flush (12 patients), headache (eight patients), and dizziness (one patient) during adenosine infusion were reported. Within 2 minutes after adenosine infusion, all side effects disappeared in all patients, and none of the MR examinations had to be interrupted because of symptoms.

A diameter of stenosis greater than or equal to 50% was considered hemodynamically significant (16–18). Quantitative coronary analyses revealed 20 nonstenotic and seven stenotic grafts (mean luminal stenosis, 59.1% ± 6.9) (Table 1). In four of seven grafts, the stenosis was localized in the graft, and in three patients, the stenosis was present in the coronary artery segments beyond the distal graft anastomosis, whereas the graft itself revealed no significant stenosis. The typical biphasic flow pattern was demonstrated in grafts by using both fast MR flow mapping and the intravascular Doppler flow wire (Fig 4).



View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a. A typical example of MR studies (a-c) and cardiac catheterization (d, e) in a patient with nonstenotic sequential vein graft to the circumflex coronary artery. a, MR angiogram of the graft shows a maximal intensity projection. The plane of the flow image perpendicular to the course of the graft is indicated by an asterisk. Ao = aorta, PA = pulmonary artery, SCV = superior caval vein. The modulus and phase-flow image in middiastole (b) and the resulting baseline and stress velocity curves (c) are depicted. Morphologic and functional information of the same grafts was also obtained during cardiac catheterization by using standardized coronary angiographic techniques (d) and the intravascular Doppler flow wire (e), respectively. Both flow techniques revealed the typical biphasic velocity pattern, with main velocity during diastole. APV and DPV increased during stress.

 


View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b. A typical example of MR studies (a-c) and cardiac catheterization (d, e) in a patient with nonstenotic sequential vein graft to the circumflex coronary artery. a, MR angiogram of the graft shows a maximal intensity projection. The plane of the flow image perpendicular to the course of the graft is indicated by an asterisk. Ao = aorta, PA = pulmonary artery, SCV = superior caval vein. The modulus and phase-flow image in middiastole (b) and the resulting baseline and stress velocity curves (c) are depicted. Morphologic and functional information of the same grafts was also obtained during cardiac catheterization by using standardized coronary angiographic techniques (d) and the intravascular Doppler flow wire (e), respectively. Both flow techniques revealed the typical biphasic velocity pattern, with main velocity during diastole. APV and DPV increased during stress.

 


View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c. A typical example of MR studies (a-c) and cardiac catheterization (d, e) in a patient with nonstenotic sequential vein graft to the circumflex coronary artery. a, MR angiogram of the graft shows a maximal intensity projection. The plane of the flow image perpendicular to the course of the graft is indicated by an asterisk. Ao = aorta, PA = pulmonary artery, SCV = superior caval vein. The modulus and phase-flow image in middiastole (b) and the resulting baseline and stress velocity curves (c) are depicted. Morphologic and functional information of the same grafts was also obtained during cardiac catheterization by using standardized coronary angiographic techniques (d) and the intravascular Doppler flow wire (e), respectively. Both flow techniques revealed the typical biphasic velocity pattern, with main velocity during diastole. APV and DPV increased during stress.

 


View larger version (168K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4d. A typical example of MR studies (a-c) and cardiac catheterization (d, e) in a patient with nonstenotic sequential vein graft to the circumflex coronary artery. a, MR angiogram of the graft shows a maximal intensity projection. The plane of the flow image perpendicular to the course of the graft is indicated by an asterisk. Ao = aorta, PA = pulmonary artery, SCV = superior caval vein. The modulus and phase-flow image in middiastole (b) and the resulting baseline and stress velocity curves (c) are depicted. Morphologic and functional information of the same grafts was also obtained during cardiac catheterization by using standardized coronary angiographic techniques (d) and the intravascular Doppler flow wire (e), respectively. Both flow techniques revealed the typical biphasic velocity pattern, with main velocity during diastole. APV and DPV increased during stress.

 


View larger version (91K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4e. A typical example of MR studies (a-c) and cardiac catheterization (d, e) in a patient with nonstenotic sequential vein graft to the circumflex coronary artery. a, MR angiogram of the graft shows a maximal intensity projection. The plane of the flow image perpendicular to the course of the graft is indicated by an asterisk. Ao = aorta, PA = pulmonary artery, SCV = superior caval vein. The modulus and phase-flow image in middiastole (b) and the resulting baseline and stress velocity curves (c) are depicted. Morphologic and functional information of the same grafts was also obtained during cardiac catheterization by using standardized coronary angiographic techniques (d) and the intravascular Doppler flow wire (e), respectively. Both flow techniques revealed the typical biphasic velocity pattern, with main velocity during diastole. APV and DPV increased during stress.

 
In Vivo Flow Data
An excellent correlation was demonstrated between the APV with fast MR flow mapping and that with Doppler flow wire in all grafts (r = 0.83, P < .001) and in nonstenotic (r = 0.81, P < .001) and stenotic grafts (r = 0.83, P < .001) (Fig 5). A good correlation was also found between DPV at MR and Doppler measurements and velocity reserve at MR and Doppler measurements in nonstenotic (DPV: r = 0.83, P < .001; velocity reserve: r = 0.56, P = .010) and stenotic grafts (DPV: r = 0.78, P = .001; velocity reserve: r = 0.70, P = .078). Regression lines were not significantly different between nonstenotic and stenotic grafts (APV, P = .90; DPV, P = .84; velocity reserve, P = .34).



View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. Correlation between (a) APV (cm/sec), (b) DPV (cm/sec), and (c) velocity reserve (CFR) measured with MR flow mapping and the Doppler flow wire. {circ} = grafts with luminal stenosis of less than 50% (nonstenotic), {bullet} = grafts with luminal stenosis greater than or equal to 50% stenosis (stenotic) in either the graft or coronary segments beyond the distal graft anastomosis. Similar linear regression lines were found in nonstenotic (thin line) and stenotic grafts (thick line). Ln = logarithmic transformation.

 


View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. Correlation between (a) APV (cm/sec), (b) DPV (cm/sec), and (c) velocity reserve (CFR) measured with MR flow mapping and the Doppler flow wire. {circ} = grafts with luminal stenosis of less than 50% (nonstenotic), {bullet} = grafts with luminal stenosis greater than or equal to 50% stenosis (stenotic) in either the graft or coronary segments beyond the distal graft anastomosis. Similar linear regression lines were found in nonstenotic (thin line) and stenotic grafts (thick line). Ln = logarithmic transformation.

 


View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c. Correlation between (a) APV (cm/sec), (b) DPV (cm/sec), and (c) velocity reserve (CFR) measured with MR flow mapping and the Doppler flow wire. {circ} = grafts with luminal stenosis of less than 50% (nonstenotic), {bullet} = grafts with luminal stenosis greater than or equal to 50% stenosis (stenotic) in either the graft or coronary segments beyond the distal graft anastomosis. Similar linear regression lines were found in nonstenotic (thin line) and stenotic grafts (thick line). Ln = logarithmic transformation.

 
A significant increment in APV during pharmacologic stress was demonstrated at fast MR flow mapping in all grafts and resulted in a similar (P = .33) mean velocity reserve of 2.5 ± 0.36 in nonstenotic grafts and 2.1 ± 0.35 in stenotic grafts. Doppler analyses, as compared with MR flow mapping, showed similar mean velocity reserves in nonstenotic (2.5 ± 0.28, P = .87) and stenotic grafts (2.4 ± 0.24, P = .33) (Table 2). In accordance with in vitro flow measurements, systematically lower peak velocities were demonstrated at MR flow mapping as compared with Doppler measurements both at baseline and during stress (Table 2, Fig 6). Furthermore, an impaired stress APV and DPV were found in stenotic grafts compared with those in nonstenotic grafts by using both MR and Doppler flow measurements (Table 2).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Velocity Data in Coronary Artery Bypass Grafts at Fast MR Flow Mapping and Intravascular Doppler Flow Wire

 


View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Bland-Altman plot graph shows the limits of agreement between noninvasively and invasively measured APV (cm/sec) with MR flow mapping and the intravascular Doppler flow wire, respectively. Dashed horizontal lines show the mean difference between MR and Doppler measurements after log transformation plus or minus 2 Sds. Ln = logarithmic transformation.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
The main finding of this study is that noninvasively obtained velocity values at fast MR flow mapping closely correlate with Doppler velocities both in vitro and in nonstenotic and stenotic grafts. First, the reproducibility and accuracy of MR APV and DPV were substantiated when a flow phantom with the Doppler flow wire was used as the standard of reference. Second, a similar comparison was made in patients with nonstenotic and stenotic grafts both at baseline and during stress. Again, an excellent correlation between MR and Doppler APV, DPV, and velocity reserve measurements was demonstrated, although a systematic underestimation of APV and DPV at MR flow mapping was noted, compared with these measurements with the Doppler flow wire.

Recurrent chest pain after coronary artery bypass graft surgery is a frequently seen problem in clinical practice (19,20). The clinician’s interest is in whether there is a flow-limiting stenosis in the graft or its distal vessels that requires coronary angiography and potential intervention. To date, coronary angiography in combination with functional analysis is the technique of choice to identify relevant graft stenosis. This, however, is an invasive diagnostic procedure with inherent risks for the patient (21). Thus, ideally, patients without an ischemic cause of postoperative chest pain should be excluded from invasive analysis. Therefore, noninvasive diagnostic screening tests that allow accurate differentiation between patients with and those without significant stenosis in the graft and/or its distal vessels are needed. The exact location of the stenosis in the epicardial vessel, which consists of the graft and coronary artery segments beyond the distal graft anastomosis, will not influence the management, since further invasive analysis will be indicated when a flow-limiting lesion is suspected.

Fast MR flow mapping may fulfill this clinical requirement, since it is a noninvasive diagnostic tool that allows evaluation of graft function that is similar to that with the intravascular Doppler flow wire. Preliminary clinical data in this study indicate that fast MR flow mapping might be able to aid in distinguishing patients with a relevant stenosis in either the graft or the distal vessels from those with normal graft function by means of a lower stress APV and DPV. However, further studies in which consecutive patients with recurrent chest pain after bypass surgery are examined are needed to determine the diagnostic value of this approach.

Physiology in Coronary Artery Bypass Grafts
Normal baseline values of APV and velocity reserve indicate the presence of a nonstenotic epicardial vessel, including the graft and the coronary arteries in the area supplied by the graft. An epicardial vessel stenosis will increase the resistance to flow, which normally results in dilatation of microcirculation to maintain an adequate baseline flow. Due to a dilated state of the microcirculation at baseline, the capacity to increase flow during stress is reduced, resulting in impaired velocity reserve (22–24). When the distal microcirculation cannot dilate normally, as in conditions like myocardial infarction, small vessel disease, or left ventricular hypertrophy, the maximal flow in a normal epicardial vessel is reduced during stress, leading to an impaired velocity reserve (25,26).

In this study, both MR and Doppler measurements indicate that baseline APV, although not significantly, and stress APV were impaired in the seven stenotic grafts, as compared with the APV in the 20 nonstenotic grafts and lead to a similar velocity reserve in all grafts. This can be explained by a condition in which both the microcirculation and the epicardial vessel are affected, resulting in lower flow at baseline and during stress because the microcirculation cannot compensate for the reduction in baseline flow. To substantiate this hypothesis, more stenotic grafts need to be studied.

MR Flow Mapping
Conventional MR flow mapping (27,28) has been validated and compared with invasive measurements of coronary flow (6–9). In accordance with our results, systematically lower peak velocities but a similar velocity reserve, which is a relative index, were demonstrated at MR imaging (29). Conventional breath-hold MR flow mapping typically includes a low temporal resolution with a large image acquisition window (3,4,6–9,30). Therefore, conventional MR flow mapping may be more susceptible to motion artifacts (31) than breath-hold fast MR flow mapping, which gives flow information every 23 msec during the cardiac cycle.

The fast MR flow sequence used three echo-planar imaging readouts, as opposed to a single-echo phase-contrast sequence. Errors induced at phase evolution over the three echo acquisitions are mainly a problem when there is flow in the readout direction. However, the main component of the velocity vector in the present study was along the section-selection direction, and first-order phase errors (delays during echo-planar imaging readout) were corrected with an echo-planar reference image.

MR versus Doppler Velocities
The lower velocities that were obtained at MR flow mapping, as compared with those obtained with the intravascular Doppler flow wire in the flow phantom and grafts, can be explained by a methodological difference between the two techniques. The region of interest on MR flow images, which consist of 2 x 2 pixels, is much larger than the sample volume of the Doppler flow wire (8). This implies that lower velocities nearer the vessel boundary are included in the MR flow analysis, leading to an underestimation of peak velocities at MR imaging. Another reason for the underestimation of peak velocities at MR imaging may be that peak velocities are missed during MR data acquisition. Although the temporal resolution of the MR sequence is one of the fastest published to date, Doppler measurements are continuous and able to record velocity information throughout the cardiac cycle. However, the averaging effect of MR flow mapping over the 23 msec will be minimal.

In addition, we observed a trend for a larger difference in peak velocity between fast MR flow mapping and the intravascular Doppler flow wire at higher graft flow, and logarithmic transformation corrected for this heteroscedasticity. This phenomenon was only demonstrated at the in vivo measurements, and this indicates that it is not an intrinsic error in the MR measurement. Moreover, errors caused by low signal-to-noise ratio or a high velocity-encoded gradient would result in a higher difference in peak velocity between MR and Doppler measurements at lower flow. A striking difference between grafts and small-diameter glass tubes is that grafts commonly show diffuse atherosclerosis. Vein graft dimensions undergo a variable degree of change after graft placement due to the development of intimal hyperplasia (32,33).

Luminal irregularities, side branches, and curves in the graft result in increased variation in the velocity profile across the vessel area in the in vivo situation, as compared with the in vitro situation. Velocity profiles across the vein graft have indeed shown a skewed pattern in the proximal part of the graft and near the distal anastomosis; even in the middle part of the graft, velocity profiles were not necessarily parabolic (34). Abnormal velocity profiles may influence both MR and Doppler flow measurements and thus result in a relative error between the two flow methods. At higher flow levels, the difference between fast MR flow mapping and the Doppler flow wire has increased. Variations in velocity profiles and turbulent flow, which is becoming more apparent at higher flow, might have contributed to this observation.

Higher flow levels were mainly obtained during the adenosine-induced stress. Adenosine-induced stress was achieved with an intravenous administration during MR imaging and with intracoronary administration during Doppler measurements. Theoretically, the adenosine administration methods might have affected the results in three manners: (a) a less complete coronary vasodilation with intravenous administration, (b) peripheral vasodilation and subsequent reduced aortic pressure with intravenous administration, and (c) an increase in heart rate following the decrease in aortic pressure with intravenous administration.

First, the expected effect from a less complete coronary vasodilation is a reduction in stress peak velocities and velocity reserve at MR imaging. However, it has been demonstrated that both intracoronary and intravenous adenosine induce maximal hyperemia similar to that of intracoronary papaverine (35). Second, the reduced aortic pressure that results from peripheral vasodilation will also reduce stress peak velocities and velocity reserve at MR imaging. In the present study, however, no change in the mean blood pressure was demonstrated during intravenous and intracoronary adenosine administrations, as compared with those at baseline. Third, an increased heart rate will shorten the diastolic phase and result in reduced stress peak velocities and reduced velocity reserve. Indeed, we showed an increased heart rate during the MR stress test, and no change in the heart rate was found during the Doppler stress test. This might be an explanation for the larger difference in peak velocity between MR and Doppler measurements during stress at the higher flow levels, but it does not account for the lower peak velocities at MR imaging, as compared with Doppler measurements at baseline.

In accordance with others (36,37), we demonstrated that variations in heart rate were not associated with substantial changes in velocity reserve. Moreover, the mean MR blood pressure and the mean Doppler aortic pressure did not change during stress, as compared with those at baseline, but the mean Doppler aortic pressure was lower than the mean MR blood pressure. Since there is a linear relationship between the driving pressure and the coronary flow, we expected lower peak velocities during Doppler measurements than at MR imaging.

Study Limitations
We studied one left internal mammary artery graft and 26 vein grafts. Since arterial grafts have a smaller diameter than vein grafts, MR flow mapping might be less accurate in arterial grafts.

Most vein grafts have a considerably larger vessel diameter, as compared with coronary arteries beyond the graft anastomosis, and the overall graft diameter may change over time after bypass surgery as a result of intimal hyperplasia (33). This makes the selection of the reference diameter for quantitative coronary analysis, and thus the reliable assessment of stenosis severity, complicated. Moreover, a 70% stenosis in the native coronary artery will be more likely to impair flow and reach physiologic significance than the same diameter of stenosis in the graft (33). This limits the use of morphologic measures solely for the selection of graft stenoses that require intervention, which underlines the need for functional measures to determine whether lesion specific therapy is required.

In conclusion, fast MR flow mapping provides noninvasive measures of APV, DPV, and velocity reserve, which closely correlate with Doppler values both in vitro and in nonstenotic and stenotic grafts. Since MR flow mapping can be used to determine graft function, it might be a potential noninvasive screening tool for patients who present with recurrent chest pain after bypass surgery. Further studies in which consecutive patients who present with recurrent chest pain after bypass surgery are examined at both MR imaging and cardiac catheterization are needed to determine the diagnostic accuracy of MR flow mapping for the identification and exclusion of relevant lesions.


   ACKNOWLEDGMENTS
 
We thank Henk P. Horree and Nico M. J. Binnendijk from the Division of Medical Instruments of the Department of Radiology, Leiden University Medical Center, for the development and construction of the flow phantom.


   FOOTNOTES
 
Abbreviations: APV = average peak velocity, DPV = diastolic peak velocity, ECG = electrocardiogram

Author contributions: Guarantors of integrity of entire study, H.W.V., E.E.v.d.W., A.d.R.; study concepts, H.W.V., J.W.J., E.E.v.d.W., A.d.R.; study design, S.E.L., P.K., H.W.V.; literature research, S.E.L., P.K.; clinical studies, S.E.L., P.S., H.W.V., J.W.J.; experimental studies, S.E.L., P.K., P.S., H.J.L.; data acquisition, S.E.L., P.K., H.W.V., J.W.J., P.S.; data analysis/interpretation, S.E.L., P.K., H.J.L.; statistical analysis, S.E.L., A.H.Z., P.S.; manuscript preparation, S.E.L.; manuscript definition of intellectual content, P.K., H.W.V., J.W.J., E.E.v.d.W, A.d.R.; manuscript editing, S.E.L., P.K., H.J.L.; manuscript revision/review, H.W.V., J.W.J., A.H.Z., P.S., E.E.v.d.W., A.d.R.; manuscript final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Wintersperger BJ, Engelmann MG, von Smekal A, et al. Patency of coronary bypass grafts: assessment with breath-hold contrast-enhanced MR angiography—value of a non-electrocardiographically triggered technique. Radiology 1998; 208:345-351.
  2. Vrachliotis TG, Bis KG, Aliabadi D, Shetty AN, Safian R, Simonetti O. Contrast-enhanced breath-hold MR angiography for evaluating patency of coronary artery bypass grafts. AJR Am J Roentgenol 1997; 168:1073-1080.
  3. Galjee MA, van Rossum AC, Doesburg T, van Eenige MJ, Visser CA. Value of magnetic resonance imaging in assessing patency and function of coronary artery bypass grafts: an angiographically controlled study. Circulation 1996; 93:660-666.
  4. Hoogendoorn LI, Pattynama PM, Buis B, van der Geest RJ, van der Wall EE, de Roos A. Noninvasive evaluation of aortocoronary bypass grafts with magnetic resonance flow mapping. Am J Cardiol 1995; 75:845-848.
  5. Kalden P, Kreitner KF, Wittlinger T, et al. Assessment of coronary artery bypass grafts: value of different breath-hold MR imaging techniques. AJR Am J Roentgenol 1999; 172:1359-1364.
  6. Hundley WG, Hamilton CA, Clarke GD, et al. Visualization and functional assessment of proximal and middle left anterior descending coronary stenoses in humans with magnetic resonance imaging. Circulation 1999; 99:3248-3254.
  7. Hundley WG, Lange RA, Clarke GD, et al. Assessment of coronary arterial flow and flow reserve in humans with magnetic resonance imaging. Circulation 1996; 93:1502-1508.
  8. Nagel E, Bornstedt A, Hug J, Schnackenburg B, Wellnhofer E, Fleck E. Noninvasive determination of coronary blood flow velocity with magnetic resonance imaging: comparison of breath-hold and navigator techniques with intravascular ultrasound. Magn Reson Med 1999; 41:544-549.
  9. Kessler W, Moshage W, Galland A, et al. Assessment of coronary blood flow in humans using phase difference MR imaging: comparison with intracoronary Doppler flow measurement. Int J Card Imaging 1998; 14:179-186.
  10. Langerak SE, Kunz P, Vliegen HW, et al. Improved MR flow mapping in coronary artery bypass grafts during adenosine stress. Radiology 2001; 218:540-547.
  11. Frayne R, Holdsworth DW, Gowman LM, et al. Computer-controlled flow simulator for MR flow studies. J Magn Reson Imaging 1992; 2:605-612.
  12. Prvulovich E. Resuscitation following arrhythmia in patients undergoing pharmacological/exercise stress for myocardial perfusion imaging. Nucl Med Commun 1996; 17:1003-1005.
  13. Pennell DJ. Pharmacological cardiac stress: when and how?. Nucl Med Commun 1994; 15:578-585.
  14. Langerak SE, Kunz P, de Roos A, Vliegen HW, Van der Wall EE. Evaluation of coronary artery bypass grafts by magnetic resonance imaging. J Magn Reson Imaging 1999; 10:434-441.
  15. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307-310.
  16. van Rugge FP, van der Wall EE, de Roos A, Bruschke AV. Dobutamine stress magnetic resonance imaging for detection of coronary artery disease. J Am Coll Cardiol 1993; 22:431-439.
  17. Elhendy A, van Domburg RT, Bax JJ, et al. Noninvasive diagnosis of coronary artery stenosis in women with limited exercise capacity: comparison of dobutamine stress echocardiography and 99mTc sestamibi single-photon emission CT. Chest 1998; 114:1097-1104.
  18. Rensing BJ, Bongaerts A, van Geuns RJ, van Ooijen P, Oudkerk M, de Feyter PJ. Intravenous coronary angiography by electron beam computed tomography: a clinical evaluation. Circulation 1998; 98:2509-2512.
  19. Bourassa MG, Fisher LD, Campeau L, Gillespie MJ, McConney M, Lesperance J. Long-term fate of bypass grafts: the Coronary Artery Surgery Study (CASS) and Montreal Heart Institute experiences. Circulation 1985; 72:V71-V78.
  20. Campeau L, Enjalbert M, Lesperance J, Vaislic C, Grondin CM, Bourassa MG. Atherosclerosis and late closure of aortocoronary saphenous vein grafts: sequential angiographic studies at 2 weeks, 1 year, 5 to 7 years, and 10 to 12 years after surgery. Circulation 1983; 68(3 pt 2):II1-II7.
  21. Johnson LW, Krone R. Cardiac catheterization 1991: a report of the Registry of the Society for Cardiac Angiography and Interventions (SCA&I). Cathet Cardiovasc Diagn 1993; 28:219-220.
  22. Gould KL, Lipscomb K. Effects of coronary stenoses on coronary flow reserve and resistance. Am J Cardiol 1974; 34:48-55.
  23. Gould KL, Kirkeeide RL, Buchi M. Coronary flow reserve as a physiologic measure of stenosis severity. J Am Coll Cardiol 1990; 15:459-474.
  24. Kern MJ. Coronary physiology revisited: practical insights from the cardiac catheterization laboratory. Circulation 2000; 101:1344-1351.
  25. Brush JE, Jr, Cannon RO, III, Schenke WH, et al. Angina due to coronary microvascular disease in hypertensive patients without left ventricular hypertrophy. N Engl J Med 1988; 319:1302-1307.
  26. Strauer BE. The significance of coronary reserve in clinical heart disease. J Am Coll Cardiol 1990; 15:775-783.
  27. Underwood SR, Firmin DN, Klipstein RH, Rees RS, Longmore DB. Magnetic resonance velocity mapping: clinical application of a new technique. Br Heart J 1987; 57:404-412.
  28. Firmin DN, Nayler GL, Klipstein RH, Underwood SR, Rees RS, Longmore DB. In vivo validation of MR velocity imaging. J Comput Assist Tomogr 1987; 11:751-756.
  29. Shibata M, Sakuma H, Isaka N, Takeda K, Higgins CB, Nakano T. Assessment of coronary flow reserve with fast cine phase contrast magnetic resonance imaging: comparison with measurement by Doppler guide wire. J Magn Reson Imaging 1999; 10:563-568.
  30. Debatin JF, Strong JA, Sostman HD, et al. MR characterization of blood flow in native and grafted internal mammary arteries. J Magn Reson Imaging 1993; 3:443-450.
  31. Hofman MB, Wickline SA, Lorenz CH. Quantification of in-plane motion of the coronary arteries during the cardiac cycle: implications for acquisition window duration for MR flow quantification. J Magn Reson Imaging 1998; 8:568-576.
  32. FitzGibbon GM, Leach AJ, Kafka HP, Keon WJ. Coronary bypass graft fate: long-term angiographic study. J Am Coll Cardiol 1991; 17:1075-1080.
  33. White CW. Clinical applications of Doppler coronary flow reserve measurements. Am J Cardiol 1993; 71:10d-16d.
  34. Fujiwara T, Kajiya F, Kanazawa S, et al. Comparison of blood-flow velocity waveforms in different coronary artery bypass grafts: sequential saphenous vein grafts and internal mammary artery grafts. Circulation 1988; 78:1210-1217.
  35. Wilson RF, Wyche K, Christensen BV, Zimmer S, Laxson DD. Effects of adenosine on human coronary arterial circulation. Circulation 1990; 82:1595-1606.
  36. Rossen JD, Simonetti I, Marcus ML, Winniford MD. Coronary dilation with standard dose dipyridamole and dipyridamole combined with handgrip. Circulation 1989; 79:566-572.
  37. Vassalli G, Hess OM, Krogmann ON, et al. Is atrial pacing needed for determination of coronary flow reserve by parametric imaging?. Am J Cardiol 1993; 71:415-419.



This article has been cited by other articles:


Home page
 J Am Coll CardiolHome page
L. P. Salm, J. J. Bax, H. W. Vliegen, S. E. Langerak, P. Dibbets, J. W. Jukema, H. J. Lamb, E. K.J. Pauwels, A. de Roos, and E. E. van der Wall
Functional significance of stenoses in coronary artery bypass grafts: Evaluation by single-photon emission computed tomography perfusion imaging, cardiovascular magnetic resonance, and angiography
J. Am. Coll. Cardiol., November 2, 2004; 44(9): 1877 - 1882.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
D. J. Pennell, U. P. Sechtem, C. B. Higgins, W. J. Manning, G. M. Pohost, F. E. Rademakers, A. C. van Rossum, L. J. Shaw, and E. K. Yucel
Clinical indications for cardiovascular magnetic resonance (CMR): Consensus Panel report
Eur. Heart J., November 1, 2004; 25(21): 1940 - 1965.
[Full Text] [PDF]

Home page
 RadiologyHome page
L. P. Salm, S. E. Langerak, H. W. Vliegen, J. W. Jukema, J. J. Bax, A. H. Zwinderman, E. E. van der Wall, A. de Roos, and H. J. Lamb
Blood Flow in Coronary Artery Bypass Vein Grafts: Volume versus Velocity at Cardiovascular MR Imaging
Radiology, September 1, 2004; 232(3): 915 - 920.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
S. E. Langerak, H. W. Vliegen, J. W. Jukema, A. H. Zwinderman, H. J. Lamb, A. de Roos, and E. E. van der Wall
Vein Graft Function Improvement after Percutaneous Intervention: Evaluation with MR Flow Mapping
Radiology, September 1, 2003; 228(3): 834 - 841.
[Abstract] [Full Text] [PDF]

Home page
 HeartHome page
C R Peebles
Non-invasive coronary imaging: computed tomography or magnetic resonance imaging?
Heart, June 1, 2003; 89(6): 591 - 594.
[Full Text] [PDF]

Home page
 CirculationHome page
S. E. Langerak, H. W. Vliegen, J. W. Jukema, P. Kunz, A. H. Zwinderman, H. J. Lamb, E. E. van der Wall, and A. de Roos
Value of Magnetic Resonance Imaging for the Noninvasive Detection of Stenosis in Coronary Artery Bypass Grafts and Recipient Coronary Arteries
Circulation, March 25, 2003; 107(11): 1502 - 1508.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2221010560v1
222/1/127    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Langerak, S. E.
Right arrow Articles by de Roos, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Langerak, S. E.
Right arrow Articles by de Roos, A.