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


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 Arheden, H.
Right arrow Articles by Ståhlberg, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arheden, H.
Right arrow Articles by Ståhlberg, F.
(Radiology. 1999;211:453-458.)
© RSNA, 1999

Cardiac Imaging

Left-to-Right Cardiac Shunts: Comparison of Measurements Obtained with MR Velocity Mapping and with Radionuclide Angiography1

Håkan Arheden, MD, PhD, Catarina Holmqvist, MD, Ulf Thilen, MD, Katarina Hanséus, MD, PhD, Gudrun Björkhem, MD, PhD, Olle Pahlm, MD, PhD, Sven Laurin, MD, PhD and Freddy Ståhlberg, PhD

1 From the Departments of Clinical Physiology (H.A., O.P.), Radiology (C.H., S.L., F.S.), Cardiology (U.T.), and Pediatric Medicine (K.H., G.B.), Lund University Hospital, S-221 85 Lund, Sweden. Received June 2, 1998; revision requested July 22; revision received September 18; accepted November 19. Supported in part by the Swedish Medical Research Council, Stockholm, Sweden; Hellmuth Herz Foundation, Lund, Sweden; Swedish Royal Physiographic Society, Lund, Sweden; and Swedish Heart Lung Foundation, Stockholm, Sweden. Address reprint requests to H.A.


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To investigate the agreement between two noninvasive methods, magnetic resonance (MR) velocity mapping and first-pass radionuclide angiography, to quantify the pulmonary-to-systemic blood flow ratio (QP/QS) in adults, adolescents, and children with left-to-right cardiac shunts.

MATERIALS AND METHODS: The accuracy and precision of MR velocity mapping were studied in 12 control subjects (six men, six women) and in a phantom. MR velocity mapping and radionuclide angiography were performed on the same day in 24 patients (16 adults, two adolescents, six children; five male patients, 19 female patients).

RESULTS: The mean error in QP/QS at MR velocity mapping in phantom experiments was -1% ± 1 (mean ± SD). In control subjects, QP/QS at MR velocity mapping was 1.03 ± 0.03, and the cardiac index was 3.1 L/min/m2 ± 0.2 and 3.2 L/min/m2 ± 0.3 for women and men, respectively. In patients, QP/QS at radionuclide angiography was 14% ± 13, higher than at MR velocity mapping. Interobserver variability was four times higher for radionuclide angiography compared with MR velocity mapping, 0% ± 16 versus 0% ± 4 (n = 12). The difference between repeated MR flow measurements in the same vessel was -1% ± 5 (n = 36).

CONCLUSION: The data suggest that MR velocity mapping is accurate and precise for measurements of shunt size over the whole range of possible QP/QS values.

Index terms: Atrial septal defect, 514.141 • Heart, flow dynamics, 51.12144 • Heart, MR, 51.121411, 51.121412, 51.12144 • Heart, radionuclide studies, 51.12175, 51.12176 • Magnetic resonance (MR), vascular studies, 51.121411, 51.121412, 51.12144 • Pulmonary veins, 565.158 • Ventricular septal defect, 515.142


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Noninvasive assessment of cardiac shunt by quantifying the pulmonary-to-systemic blood flow ratio (QP/QS) may be of value preoperatively to determine the need for surgery and postoperatively to assess the outcome (1). In persons without cardiac shunt, the QP/QS is 1. In left-to-right shunt, this ratio increases. Some of the more commonly used methods to assess QP/QS are oximetry during cardiac catheterization, first-pass radionuclide angiography, and Doppler echocardiography. Cardiac catheterization is invasive and semiquantitative, radionuclide angiography is quantitative but cannot be used to reliably localize the site of the shunt, and quantitative Doppler echocardiography is highly operator dependent (2).

By using magnetic resonance (MR) velocity mapping, blood flow can be measured noninvasively with a high degree of accuracy (3–5), and consequently QP/QS can be determined from measurements of blood flow in the pulmonary trunk and the proximal aorta (6). In adults, QP/QS measured this way has been compared with QP/QS measured with oximetry (7,8), first-pass radionuclide angiography (9), and ventricular volumetric data (10). In children, MR velocity mapping has been compared with cardiac catheterization and ventricular volumetric data in four cases of shunt (11). However, to our knowledge, a numeric analysis of the agreement between MR velocity mapping and first-pass radionuclide angiography has not been performed, and the agreement of these methods in children remains to be established.

The objectives of the current study were therefore (a) to investigate the precision and accuracy of MR velocity mapping to determine QP/QS and (b) to evaluate how closely radionuclide angiography and MR velocity mapping agree in measurements of QP/QS in adults, adolescents, and children with left-to-right cardiac shunts.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study Population
All subjects examined (n = 36) were questioned about history of cardiac disease and underwent cross-sectional and Doppler echocardiography to exclude or localize the shunt site.

The control group for MR velocity mapping consisted of healthy volunteers (six men, six women; age range, 20–53 years) with no history of cardiovascular disease and no direct or indirect signs of cardiac shunt as determined by means of cross-sectional and Doppler echocardiography.

The patient group (n = 24; five male patients, 19 female patients) (Table 1) consisted of 16 adults (age range, 20–68 years), two adolescents (13 and 17 years), and six children (age range, 2–12 years). Among the adults, there were 11 with atrial septal defect, two with partially anomalous pulmonary venous return, and three with ventricular septal defect. One adolescent had scimitar syndrome, and the other had ventricular septal defect. Among the six children, there were two with atrial septal defect, one with partially anomalous pulmonary venous return, two with atrial septal defect and partially anomalous pulmonary venous return, and one with ventricular septal defect. Nonsinus rhythm was an exclusion criterion.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Studied Population and Results
 
Study Protocol
The protocol and procedures were approved by the research ethics committee at Lund University Hospital, Sweden. Subjects examined were included after written informed consent (by parent or guardian when applicable) was obtained. Patients were examined by means of MR velocity mapping and first-pass radionuclide angiography on the same day. Persons (H.A., C.H., O.P.) evaluating the MR flow data were blinded to the radionuclide angiography data and vice versa. Second observers (H.A., C.H., O.P.) were blinded to the results of the first observers.

MR Imaging
Throughout the study, a 1.5-T system (Magnetom Vision; Siemens, Erlangen, Germany) with a 25-mT/m gradient strength and a 600-µsec gradient ramp time was used. Gradient-echo velocity mapping sequences provided by the manufacturer were used for the determination of blood flow.

In the phantom study, a standard head coil was used, and the flow rate of water doped with 0.13 mmol/L MnCl2 (Merck, Darmstadt, Germany) to the approximate relaxation time of blood was measured in an artificial homemade shunt system by means of MR velocity mapping. This flow rate was compared with the flow rate measured by means of beaker and timer. The artificial shunt system consisted of connected tubes with an inner diameter of 10 mm. The relation between the flow in one tube and the total flow was investigated in the range of a ratio of 1–5.

In control subject and patient studies, a phased-array body coil was used. To localize the ascending aorta, we used a set of coronal spin-echo images (635/30 [repetition time msec/echo time msec]) obtained with electrocardiographic triggering, which resulted in an effective repetition time of one cardiac cycle, a section thickness of 6 mm, a field of view typically of 400 mm, and a matrix typically of 172 x 256 (Fig 1). To localize the pulmonary trunk, after the coronal spin-echo images were obtained, an oblique sagittal turbo spin-echo breath-hold sequence (1,500/85) performed with electrocardiographic triggering was used, which resulted in an effective repetition time of two cardiac cycles, an echo train length of 23, a section thickness of 8 mm, a field of view typically of 400 mm, and a matrix typically of 256 x 512 (Fig 1).



View larger version (162K):
[in this window]
[in a new window]
  		Figure 1a. MR images demonstrate the planes (lines) used to measure blood flow. (a) Coronal spin-echo image (942/30, section thickness of 6 mm, field of view of 380 mm, and matrix of 172 x 256) of the ascending aorta, with the plane used for velocity mapping indicated. (b) Sagittal breath-hold turbo spin-echo image (2,033/85, section thickness of 8 mm, field of view of 400 mm, and matrix of 230 x 512) obtained along the pulmonary trunk. (c) From b, a coronal oblique plane (indicated in b) was used to obtain another breath-hold turbo spin-echo image with the patient in the prone position (1,857/85, section thickness of 8 mm, field of view of 450 mm, and matrix of 276 x 512). Flow was measured in the pulmonary trunk just above the pulmonary valves (velocity mapping plane indicated).

 


View larger version (189K):
[in this window]
[in a new window]
  		Figure 1b. MR images demonstrate the planes (lines) used to measure blood flow. (a) Coronal spin-echo image (942/30, section thickness of 6 mm, field of view of 380 mm, and matrix of 172 x 256) of the ascending aorta, with the plane used for velocity mapping indicated. (b) Sagittal breath-hold turbo spin-echo image (2,033/85, section thickness of 8 mm, field of view of 400 mm, and matrix of 230 x 512) obtained along the pulmonary trunk. (c) From b, a coronal oblique plane (indicated in b) was used to obtain another breath-hold turbo spin-echo image with the patient in the prone position (1,857/85, section thickness of 8 mm, field of view of 450 mm, and matrix of 276 x 512). Flow was measured in the pulmonary trunk just above the pulmonary valves (velocity mapping plane indicated).

 


View larger version (158K):
[in this window]
[in a new window]
  		Figure 1c. MR images demonstrate the planes (lines) used to measure blood flow. (a) Coronal spin-echo image (942/30, section thickness of 6 mm, field of view of 380 mm, and matrix of 172 x 256) of the ascending aorta, with the plane used for velocity mapping indicated. (b) Sagittal breath-hold turbo spin-echo image (2,033/85, section thickness of 8 mm, field of view of 400 mm, and matrix of 230 x 512) obtained along the pulmonary trunk. (c) From b, a coronal oblique plane (indicated in b) was used to obtain another breath-hold turbo spin-echo image with the patient in the prone position (1,857/85, section thickness of 8 mm, field of view of 450 mm, and matrix of 276 x 512). Flow was measured in the pulmonary trunk just above the pulmonary valves (velocity mapping plane indicated).

 
Through-plane flow was measured perpendicular to the flow in the ascending aorta at the level of the pulmonary trunk and in the pulmonary trunk just above the pulmonary valves by using a velocity mapping technique described earlier (12). Two basic pulse sequences (24–50/5) that produce differently shaped gradients in the section-selective direction, one sequence sensitive to motion and the other motion compensated, were executed interleaved with velocity encoding of 150 or 250 cm/sec, a flip angle of 30°, a field of view of 220–440 mm, and a matrix typically of 192 x 256. Using prospective electrocardiographic triggering but no respiratory gating, we obtained 25–40 cinematographic phase image pairs with 24–50-msec time separation, which covered a period that exceeded the R-R interval by approximately 20% to ensure that the measurement included flow during the entire cardiac cycle. The total image information was collected during 512 heartbeats, and the acquisition time varied approximately 5–9 minutes.

MR velocity maps were obtained by means of automatic pixel-by-pixel subtraction of the two phase images, and each study resulted in 25–40 subtracted velocity maps and corresponding modulus information. Measurements were made in vessel 1, vessel 2, and vessel 1 again, where the aorta and the pulmonary trunk were randomly assigned to represent vessel 1 and vessel 2, to avoid effects of physiologic drift in cardiac output (see MR Image Analysis). Subjects with a change in mean heart rate between measurements in the pulmonary trunk and the aorta of more than 10% were reexamined (n = 1).

MR Image Analysis
All data were transferred to a workstation (Sparc 10; Sun Microsystems, Mountain View, Calif) and evaluated by using a specially designed program (RADGOP; Context Vision, Linköping, Sweden). In each velocity map, a region of interest was drawn manually to completely cover the vessel of interest. The region of interest could be changed in size and shape between images, and it was possible to use the vessel contours from modulus images to delineate the vessel in the corresponding velocity maps (Fig 2).



View larger version (162K):
[in this window]
[in a new window]
  		Figure 2. MR images show the modulus (upper panels) and corresponding phase (lower panels) used to outline the region of interest for flow measurements in the aorta (left panels: velocity encoding of 150 cm/sec, 40/5, a flip angle of 30°, a field of view of 380 mm, and a matrix of 192 x 256) and the pulmonary trunk just above the pulmonary valves (right panels: velocity encoding of 150 cm/sec, 40/5, a flip angle of 30°, a field of view of 380 mm, and a matrix of 256 x 256). a = aorta, p = pulmonary trunk.

 
Linear phase variations within images were compensated for by placing two background regions of interest at an equal distance from and on a straight line going through the vessel region of interest. The mean background was subtracted image by image. The flow rate was calculated as the product of the area of each region of interest and the net mean velocity within it. Mean flow rate over the cardiac cycle was calculated over the mean R-R interval, which was determined from the imager software. Owing to varying R-R intervals, end-diastolic flow values were substituted with values extrapolated from the last reliable blood velocity data point before velocity started to rise again. Flow was evaluated in vessel 1, vessel 2, and vessel 1, and the flow values in vessel 1 were averaged when the QP/QS was calculated (Fig 3a). Evaluation of data, including outlining of regions of interest and calculation of QP/QS, was performed by either of two observers (H.A., C.H.).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 3a. Graphs show original data from one patient to demonstrate how (a) MR velocity mapping (volume in milliliters per second vs time in milliseconds) and (b) radionuclide angiography (counts per second vs time in seconds) were used to calculate QP/QS. In a, flow was measured in the pulmonary trunk ({bigcirc}), the aorta ({blacktriangleup}), and the pulmonary trunk again (•). QP/QS is calculated as the mean of the two measurements of pulmonary flow in liters per minute divided by the systemic flow in liters per minute to avoid physiologic drift in cardiac output. Note the repeatability of the two pulmonary flow measurements. In b, a gamma variate curve (dotted line a) is fitted to the original time activity data (thick solid line). Subtraction of curve a from the original data provides the recirculation data (thin solid line). A second gamma variate curve (dotted line b) is fitted to the recirculation data, and QP/QS is calculated from the areas under the curves as a/(a - b).

 


View larger version (24K):
[in this window]
[in a new window]
  		Figure 3b. Graphs show original data from one patient to demonstrate how (a) MR velocity mapping (volume in milliliters per second vs time in milliseconds) and (b) radionuclide angiography (counts per second vs time in seconds) were used to calculate QP/QS. In a, flow was measured in the pulmonary trunk ({bigcirc}), the aorta ({blacktriangleup}), and the pulmonary trunk again (•). QP/QS is calculated as the mean of the two measurements of pulmonary flow in liters per minute divided by the systemic flow in liters per minute to avoid physiologic drift in cardiac output. Note the repeatability of the two pulmonary flow measurements. In b, a gamma variate curve (dotted line a) is fitted to the original time activity data (thick solid line). Subtraction of curve a from the original data provides the recirculation data (thin solid line). A second gamma variate curve (dotted line b) is fitted to the recirculation data, and QP/QS is calculated from the areas under the curves as a/(a - b).

 
First-Pass Radionuclide Angiography
A gamma camera system (GCA 901A/ECT; Toshiba, Tokyo, Japan) with a general-purpose collimator and a manufacturer-supplied program to evaluate the data were used. Images were sampled by using the matrix mode (64 x 64, 5 frames per second) after bolus injection of technetium 99m pertechnetate (Mallinckrodt, Petten, the Netherlands; body weight of more than 50 kg, 400 MBq/kg; 25–50 kg, 8 MBq/kg; 0–25 kg, 100 MBq + 4 MBq/kg) in either the superior vena cava directly or in an arm vein during the early reactive hyperemic phase induced by inflating a blood-pressure cuff to a pressure above the arterial pressure for 3 minutes. The mean bolus width of injected 99mTc pertechnetate (full width at half maximum), measured in a region of interest over the superior vena cava, was 0.7 seconds ± 0.3 (n = 24), and the widest bolus was 1.54 seconds. QP/QS was calculated by using the gamma-variate technique (13) (Fig 3b). A QP/QS of more than 3 may not be accurately quantified (13–15), and therefore QP/QS at radionuclide angiography was given a highest value of 3. Evaluation of data, including outlining of regions of interest and calculation of QP/QS, was performed by either of two observers (H.A., O.P.).

Interobserver variability of QP/QS for both MR velocity mapping (H.A., C.H.) and radionuclide angiography (H.A., O.P.) was determined in the same group of 12 randomly selected patients with left-to-right shunts.

Statistical Analysis
Values are expressed as mean ± SD. Agreement between the two methods, repeatability of MR flow measurements, and interobserver variability were analyzed according to the method of Bland and Altman (16) and are expressed as percentages because QP/QS is a ratio. In the analysis of agreement or difference, the mean QP/QS at radionuclide angiography and at MR velocity mapping for each patient was considered the best estimate of the true value. The proportional difference in QP/QS was calculated as 100(radionuclide angiography QP/QS - MR velocity mapping QP/QS)/mean of (radionuclide angiography QP/QS + MR velocity mapping QP/QS). Error (bias) was calculated as 100(measurement - true value)/true value. Accuracy and precision were calculated as 100% - error (as a percentage) and 100% - SD (as a percentage), respectively. Repeatability was calculated as 100(measurement 1 - measurement 2)/mean of (measurement 1 + measurement 2).

Because the scatter of the differences between radionuclide angiography and MR velocity mapping in patients increased as QP/QS increased, a paired Student t test of differences between logarithmic measurements was used to test for significance (16). A P value less than .05 was considered to indicate a statistically significant difference.


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Phantom Experiments and Control Subjects
In the phantom experiment, QP/QS at MR imaging showed a mean error of 1% ± 1 versus the QP/QS for the beaker and timer (Fig 4), and the maximum error was less than or equal to 4%. Accuracy and precision were thus both 99%. Absolute linear flow rates had an accuracy of 93% or more and a precision of 99%.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 4. Graph shows findings at MR (MRI) velocity mapping versus findings at beaker and timer analysis to quantitate QP/QS in a phantom and line of equality. The mean error of the MR flow measurements is 1% ± 1.

 
The QP/QS in the control subjects at MR velocity mapping was in the range of 0.98–1.07 (mean, 1.03 ± 0.03) (Table 1). If the mean QP/QS at MR velocity mapping in the control subjects was considered the best estimate of the true QP/QS in healthy people, the maximum error was 5%. Cardiac output at MR velocity mapping in the control subjects was 5.2 L/min ± 0.5 and 6.4 L/min ± 0.9 for women and men, respectively. The cardiac index was 3.1 L/min/m2 ± 0.2 and 3.2 L/min/m2 ± 0.3 for women and men, respectively.

Agreement between Radionuclide Angiography and MR Velocity Mapping
Radionuclide angiography yielded a higher QP/QS than MR velocity mapping in most patients (Fig 5a). In one of these patients, this may have been because radionuclide angiography was limited to an upper value of 3. The mean difference between the methods was 14% ± 13 (P < .001, n = 24) and was essentially proportional to shunt size (Fig 5b). Interobserver variability of QP/QS was four times higher for radionuclide angiography than for MR velocity mapping, 0% ± 16 versus 0% ± 4 (n = 12) (Table 2).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 5a. Graphs show agreement between radionuclide angiography (RNA) and MR velocity mapping measurements of QP/QS in patients with left-to-right cardiac shunt. • = children and adolescents, {bigcirc} = adults. (a) MR velocity mapping results are plotted against those of radionuclide angiography. Dashed line is line of equality. (b) Mean difference between findings at radionuclide angiography and findings at MR velocity mapping is 14% ± 13 (n = 24) and is essentially proportional to shunt size. From top to bottom, the lines represent mean + 2 SD, mean, and mean - 2 SD.

 


View larger version (21K):
[in this window]
[in a new window]
  		Figure 5b. Graphs show agreement between radionuclide angiography (RNA) and MR velocity mapping measurements of QP/QS in patients with left-to-right cardiac shunt. • = children and adolescents, {bigcirc} = adults. (a) MR velocity mapping results are plotted against those of radionuclide angiography. Dashed line is line of equality. (b) Mean difference between findings at radionuclide angiography and findings at MR velocity mapping is 14% ± 13 (n = 24) and is essentially proportional to shunt size. From top to bottom, the lines represent mean + 2 SD, mean, and mean - 2 SD.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. QP/QS Measurements: Agreement, Accuracy, and Precision of Radionuclide Angiography and MR Velocity Mapping
 
Flow Measurements in Vivo: Repeatability and Relation to Cardiac Cycle
Repeated MR flow measurements in the same vessel during the same session, as a measure of repeatability, showed a difference of -1% ± 5 (n = 36) (Fig 6, Table 2).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 6. Graph shows that the mean difference between repeated MR flow measurements in liters per minute in the same vessel is -1% ± 5. Dashed line indicates line of equality.

 
If the information in the last 20%–25% of the R-R interval was discarded and the flow value of the last measured time point was extrapolated through the true R-R interval, the difference in calculated QP/QS compared with the QP/QS calculated with the method used in this study was 1% ± 4 (n = 36) (Table 2). This result was not dependent on the heart rate (57–109 beats per minute).


   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
The results of this study show that MR velocity mapping is an accurate and precise method for measuring QP/QS and that radionuclide angiography and MR velocity mapping differed by 14% ± 13.

The error of the QP/QS measurement in the phantom, about 1%, is in agreement with expected theoretic errors in volume flow for vessels of the size examined (17). In control subjects, the mean QP/QS at MR velocity mapping was 1.03 ± 0.03, which is slightly higher than 1. This difference may be a result of the coronary circulation because flow measurements in the aorta were performed distal to the coronary arteries, because of drainage of the thebesian and bronchial veins into the systemic circulation, or because of systematic differences or variability in the measurements of pulmonary and aortic flow rates.

It seems reasonable to assume that the difference in QP/QS between radionuclide angiography and MR velocity mapping may be due to the lower accuracy and precision of the radionuclide angiographic method. The main reasons for this are that the interobserver variability was considerably higher for the radionuclide angiographic method than for the MR velocity mapping method, which is consistent with the findings of earlier studies (7,18), while results from the flow phantom experiment, from repeated flow measurements in vivo, and from the control subjects failed to demonstrate any appreciable bias or imprecision of the MR velocity mapping method (Figs 4, 6; Table 2).

Radionuclide angiography relies on a priori assumptions, becomes mathematically more sensitive when the degree of shunt increases, and cannot be used to accurately measure QP/QS greater than 3 (13–15), which is not the case for MR velocity mapping. Radionuclide angiographic measurements of QP/QS are hampered by valvular regurgitation, slow circulation, or complex anatomy. This is not a major concern with MR velocity mapping, because it measures both forward and backward blood flow in specified vessels, and regurgitation is thus implicitly accounted for. Radionuclide angiography should be expected to show a relatively constant bias (overestimation of QP/QS) because the bronchial arterial circulation gives rise to an early recirculation in the analyzed region of interest (15). This may be one of the explanations for the differences in QP/QS obtained by using radionuclide angiography and by using MR velocity mapping in this study.

Respiratory gating was not used because it might introduce a systematic error in QP/QS measurements. Cardiac output of the left and right ventricles changes slightly in opposite directions during inspiration and expiration (19). QP/QS may thus be underestimated because acquisition of flow data is triggered during the expiration phase when right ventricle output decreases and left ventricle output increases. By not using respiratory gating, the flow data are averaged over the entire respiratory cycle, and the acquisition time is shortened. Acquisition of the radionuclide angiographic data requires 30 seconds while the patient is breathing, which also leads to averaging over a couple of respiratory cycles.

MR flow data were sampled over the entire R-R interval to account for regurgitation and anterograde blood flow during diastole. In this study, however, we found that it was sufficient to sample data during 75%–80% of the R-R interval and extrapolate the last sample value through the true R-R interval because flow in late diastole is close to zero. QP/QS obtained in this way changed by only 1% ± 4 (n = 36) (Table 2). The practical implication of this is that the acquisition time for the flow data can be shortened by 50% without introducing major errors. The acquisition of MR velocity data for QP/QS measurement after anatomic MR mapping would thus require about 3 minutes each for acquisition of pulmonary and systemic flow, provided mean heart rate variation is less than 10% between measurements in the different vessels.

While acquisition of data is a relatively simple and short procedure with radionuclide angiography, it requires skill and takes longer with MR velocity mapping. On the other hand, while evaluation of data requires considerable experience with radionuclide angiography, it is easier with MR velocity mapping.

Study Limitations
MR velocity mapping was validated by using a flow phantom while radionuclide angiography was not. It is, however, an inherent difficulty of some methods, like radionuclide angiography or oximetry, that they cannot be validated by using the "true" values of QP/QS in a phantom. In this context, it may be tempting to regard MR velocity mapping as the standard method. Another limitation is that all subjects examined had sinus rhythm. One of the objectives of the study, however, was to evaluate the accuracy and precision of MR velocity mapping under defined conditions, which is why patients without sinus rhythm were not eligible for the study. This is not the case in the clinical setting. A highly irregular cardiac rhythm, especially with aberrant heartbeats, is likely to decrease the precision and accuracy of the MR velocity mapping method.

MR velocity mapping is an accurate and precise method for measurement of shunt size over the whole range of possible QP/QS values in adults, adolescents, and children with sinus rhythm. If the proportional difference between radionuclide angiography and MR velocity mapping is known, these noninvasive measurements of QP/QS are comparable. MR flow measurements may be the method of choice if MR anatomy mapping is performed anyway or if high accuracy and precision of QP/QS measurements are desirable.


   Acknowledgments
 
We are thankful for the skillful technical assistance of Annmarie Svensson, MLT.


   Footnotes
 
Abbreviation: QP/QS = pulmonary-to-systemic blood flow ratio

Author contributions: Guarantor of integrity of entire study, H.A.; study concepts and design, H.A., C.H., F.S.; definition of intellectual content, H.A., C.H., F.S.; literature research, H.A., C.H., U.T., O.P., S.L., F.S.; clinical studies, H.A., C.H., U.T., K.H., G.B., O.P.; experimental studies, H.A., C.H., F.S.; data acquisition and analysis, H.A., C.H., O.P., F.S.; statistical analysis, H.A., C.H., F.S.; manuscript preparation, H.A., C.H., F.S.; manuscript editing, H.A., C.H., O.P., F.S.; manuscript review, H.A., C.H., U.T., K.H., G.B., O.P., S.L., F.S.


   References
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Baker E. Congenital heart disease. In: Murray IP, Ell PJ, eds. Nuclear medicine in clinical diagnosis and treatment. New York, NY: Churchill Livingstone, 1994; 1229-1236.
  2. Boehrer JD, Lange RA, Willard JE, Grayburn PA, Hillis LD. Advantages and limitations of methods to detect, localize, and quantitate intracardiac left-to-right shunting. Am Heart J 1992; 124:448-455.[Medline]
  3. 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.[Medline]
  4. Kondo C, Caputo GR, Semelka R, Foster E, Shimakawa A, Higgins CB. Right and left ventricular stroke volume measurements with velocity-encoded cine MR imaging: in vitro and in vivo validation. AJR 1991; 157:9-16.[Abstract/Free Full Text]
  5. Sondergaard L, Stahlberg F, Thomsen C, Stensgaard A, Lindvig K, Henriksen O. Accuracy and precision of MR velocity mapping in measurement of stenotic cross-sectional area, flow rate, and pressure gradient. JMRI 1993; 3:433-437.
  6. Rees S, Firmin D, Mohiaddin R, Underwood R, Longmore D. Application of flow measurements by magnetic resonance velocity mapping to congenital heart disease. Am J Cardiol 1989; 64:953-956.[Medline]
  7. Brenner LD, Caputo GR, Mostbeck G, et al. Quantification of left to right atrial shunts with velocity-encoded cine nuclear magnetic resonance imaging. J Am Coll Cardiol 1992; 20:1246-1250.[Abstract]
  8. Hundley WG, Li HF, Lange RA, et al. Assessment of left-to-right intracardiac shunting by velocity-encoded, phase-difference magnetic resonance imaging: a comparison with oximetric and indicator dilution techniques. Circulation 1995; 91:2955-2960.[Abstract/Free Full Text]
  9. Mohiaddin RH, Underwood R, Romeira L, et al. Comparison between cine magnetic resonance velocity mapping and first-pass radionuclide angiocardiography for quantitating intracardiac shunts. Am J Cardiol 1995; 75:529-532.[Medline]
  10. Rebergen SA, van der Wall EE, Helbing WA, de Roos A, van Voorthuisen AE. Quantification of pulmonary and systemic blood flow by magnetic resonance velocity mapping in the assessment of atrial-level shunts. Int J Card Imaging 1996; 12:143-152.[Medline]
  11. Sieverding L, Jung WI, Klose U, Apitz J. Noninvasive blood flow measurement and quantification of shunt volume by cine magnetic resonance in congenital heart disease: preliminary results. Pediatr Radiol 1992; 22:48-54.[Medline]
  12. Sondergaard L, Lindvig K, Hildebrandt P, et al. Quantification of aortic regurgitation by magnetic resonance velocity mapping. Am Heart J 1993; 125:1081-1090.[Medline]
  13. Maltz DL, Treves S. Quantitative radionuclide angiocardiography: determination of Qp:Qs in children. Circulation 1973; 47:1049-1056.[Abstract/Free Full Text]
  14. Askenazi J, Ahnberg DS, Korngold E, LaFarge CG, Maltz DL, Treves S. Quantitative radionuclide angiocardiography: detection and quantitation of left to right shunts. Am J Cardiol 1976; 37:382-387.[Medline]
  15. Treves S, Parker J. Detection and quantification of intracardiac shunts. In: Strauss H, Pitt B, eds. Cardiovascular nuclear medicine. St Louis, Mo: Mosby, 1979; 148-161.
  16. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307-310.[Medline]
  17. Wolf RL, Ehman RL, Riederer SJ, Rossman PJ. Analysis of systematic and random error in MR volumetric flow measurements. Magn Reson Med 1993; 30:82-91.[Medline]
  18. Villanueva Meyer J, Philippe L, Cordero S, Marcus CS, Mena I. Use of factor analysis in the evaluation of left to right cardiac shunts. J Nucl Med 1986; 27:1442-1448.[Abstract/Free Full Text]
  19. Hoffman JIE, Guz A, Charlier AA, Wilcken DEL. Stroke volume in conscious dogs: effect of respiration, posture, and vascular occlusion. J Appl Physiol 1965; 20:865-877.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Thorac. Cardiovasc. Surg.Home page
R. Petzina, M. Ugander, L. Gustafsson, H. Engblom, J. Sjogren, R. Hetzer, R. Ingemansson, H. Arheden, and M. Malmsjo
Hemodynamic effects of vacuum-assisted closure therapy in cardiac surgery: Assessment using magnetic resonance imaging
J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1154 - 1162.
[Abstract] [Full Text] [PDF]

Home page
 RadioGraphicsHome page
K. I. Norton, C. Tong, R. B. J. Glass, and J. C. Nielsen
Cardiac MR Imaging Assessment Following Tetralogy of Fallot Repair
RadioGraphics, January 1, 2006; 26(1): 197 - 211.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
M. Carlsson, P. Cain, C. Holmqvist, F. Stahlberg, S. Lundback, and H. Arheden
Total heart volume variation throughout the cardiac cycle in humans
Am J Physiol Heart Circ Physiol, July 1, 2004; 287(1): H243 - H250.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
P. Beerbaum, H. Korperich, P. Barth, H. Esdorn, J. Gieseke, and H. Meyer
Noninvasive Quantification of Left-to-Right Shunt in Pediatric Patients : Phase-Contrast Cine Magnetic Resonance Imaging Compared With Invasive Oximetry
Circulation, May 22, 2001; 103(20): 2476 - 2482.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 Arheden, H.
Right arrow Articles by Ståhlberg, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arheden, H.
Right arrow Articles by Ståhlberg, F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE