PEDIATRICS Vol. 120 No. 1 July 2007, pp. 78-83 (doi:10.1542/peds.2006-3305)
ARTICLE |
Three-Tesla Cardiac Magnetic Resonance Imaging for Preterm Infants
a Departments of Paediatrics
b Imaging Sciences, Imperial College, London, England
c Medical Research Council Clinical Sciences Centre, Hammersmith Hospital, London, England
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVES. We aimed to establish the feasibility of acquiring 3.0-T cardiac MRIs without sedation, anesthesia, or breath-holding for preterm infants and to obtain preliminary quantitative data on left ventricular function in this population.
METHODS. Twelve preterm infants underwent 3.0-T cardiac MRI without sedation or breath-holding. The median gestational age was 29 weeks (range: 26–33 weeks), the median birth weight was 1240 g (range: 808–2200 g), and the median postconceptional age at the time of cardiac MRI was 33 weeks (range: 31–40 weeks). Anatomic images were acquired with T2-weighted spin-echo sequences, and ventricular function was assessed with balanced steady-state free precession cine sequences. We assessed left ventricular function by using the area-length ejection fraction method on horizontal long-axis images and the volumetric Sergeant's discs method of analysis on short-axis images.
RESULTS. Imaging was successful for 10 of 12 infants. For those 10, the area-length ejection fraction method in the horizontal long-axis plane estimated median stroke volume at 2.9 mL, cardiac output at 0.4 L/minute, end-diastolic volume at 3.8 mL, end-systolic volume at 0.3 mL, and ejection fraction at 74.6%. Short-axis volumetric estimations were made for 4 infants. With this approach, the median stroke volume was 2.4 mL, cardiac output 0.35 L/minute, end-diastolic volume 4.3 mL, end-systolic volume 2.1 mL, and ejection fraction 56%.
CONCLUSIONS. Three-tesla cardiac MRI is feasible for preterm infants without sedation, anesthesia, or breath-holding and has the potential to provide a wide range of precise quantitative data that may be of great value for the investigation of cardiac function in preterm infants.
Key Words: cardiac magnetic resonance imaging • area-length ejection fraction • Sergeant's discs • preterm • newborn • patent ductus arteriosus • left ventricular function
Abbreviations: CMRI—cardiac MRI • ALEF—area-length ejection fraction • PDA—patent ductus arteriosus • b-SSFP—balanced steady-state free precession
Cardiac function in preterm infants is poorly understood. Commonly used bedside tests such as blood pressure and capillary refill time have very low specificity and sensitivity in detecting low blood flow in preterm infants.1 There is emerging evidence that cardiovascular function is an important determinant of outcome. Low superior vena cava flow is common in the first hours after birth and has been associated with subsequent periventricular hemorrhage,2 whereas inflammatory mediators induced by sepsis impair cardiac function3 and low cardiac output is associated with reduced cerebral blood flow.4
Most available data on cardiac function in preterm infants have been acquired by using echocardiography, which is an invaluable tool for observing cardiac morphologic features but is relatively imprecise as a measure of function in individuals.5 Echocardiographic approaches have been unable to provide definitive answers to important practical questions, such as how cardiac output is related to blood pressure and how volume expansion and inotropic therapy affect cardiac output in extremely preterm infants.6,7
Cardiac MRI (CMRI) has been a major advance contributing to the understanding of cardiac function and disease in adults,8 providing quantitative data with high levels of accuracy and reproducibility.9 However, CMRI is technically challenging in children because of small cardiac size, rapid heart rates, and a high level of sensitivity of CMRI to motion-induced degradation of image quality.10 Furthermore, in previous CMRI studies, children were usually sedated or anesthetized and established on mechanical ventilation to overcome movement and to allow breath-holding, a strategy that yields significant improvements in image quality in adults. Unfortunately, sedation or anesthesia may have particular problems in preterm infants.11,12
To our knowledge, there have been no previous reports of CMRI in preterm infants undergoing intensive care. The previous studies of CMRI in older infants and children have been at 1.5-T field strength, mirroring the predominance of 1.5-T CMRI for adults and the relative difficulty of developing 3.0-T systems for adult CMRI. However, 3.0 T is becoming widely established as the standard field strength for imaging other parts of the body. Higher field strengths offer the potential advantage of higher signal/noise ratios, allowing improved spatial or temporal resolution once appropriate imaging parameters have been defined. Interestingly, although the small physical size of the preterm heart has been an impediment to development of CMRI in general, it offers a specific advantage in the physical context of higher-field imaging; this suggested to us that development of 3.0-T CMRI might be appropriate for this population.13,14
If appropriate techniques can be developed, CMRI has the potential to provide valuable information on cardiac function in preterm infants and to address long-standing practical problems in neonatal intensive care. The purpose of this study was thus to establish the feasibility of CMRI at 3.0 T in nonsedated preterm infants. We aimed to define imaging parameters and to obtain preliminary quantitative data on left ventricular function by using 2 standard techniques, that is, the area-length ejection fraction (ALEF) approach on the horizontal long-axis view, which is rapid and simple but requires an assumption of ellipsoid ventricular geometry, and the volumetric Sergeant's discs method, which avoids the ellipsoid assumption but requires a contiguous stack of short-axis images.15
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects
Twelve preterm infants underwent CMRI. Clinical details are given in Table 1. The median gestational age was 29 weeks (range: 26–33 weeks), the median birth weight was 1240 g (range: 808–2200 g), and the median postconceptional age at the time of MRI was 33 weeks (range: 31–40 weeks). For 2 infants, patent ductus arteriosus (PDA) had been diagnosed previously through echocardiography; those infants required additional inspired oxygen. The remaining 10 subjects were thought to have structurally and functionally normal hearts and were not oxygen dependent.
|
Infants were fed and allowed to fall into natural sleep in an quiet environment and then were laid in a custom-made, MRI-compatible cradle and placed in the scanner, with MRI-compatible physiologic monitoring and ear protection, as reported previously.16 Infants were monitored with electrocardiography, pulse oximetry, and a video camera, and a pediatrician was present throughout the procedure. Each scan required
45 minutes. The Hammersmith Hospital research ethics committee granted ethical permission for 3.0-T imaging studies. Written informed parental consent was obtained for each infant studied.
Image Acquisition
All scans were conducted with a Philips 3-T Intera system (Philips, Best, Netherlands). A surface receiver coil was placed over the chest wall. Vector electrocardiographic gating was used conventionally, but no motion-correcting pulse sequences or respiratory triggering was used. An axial, electrocardiographically triggered, T2-weighted, black-blood, anatomic sequence was used to give an overview of the heart and great vessels. Automatic shimming was used. Balanced steady-state free precession (b-SSFP) cine sequences were acquired in the horizontal and vertical long-axis planes. These were used to plan a stack of contiguous slices in the left ventricular short-axis plane.
After investigating image acquisition by using varying parameters, we found that, for spin-echo anatomic images, a repetition time of 2 beats and an echo time of 60 milliseconds yielded the highest-quality images. For b-SSFP cine sequences, best tissue contrast was achieved with a flip angle of 45°. Specific absorption rate values, as provided by the scanner, were noted for all examinations, and all infants were monitored for signs of thermal stress. MRI parameters for the reported images are given in Table 2.
|
Quantitation of Cardiac Function
The images were analyzed on a cardiac workstation (Philips View Forum). The first image in each cine sequence was considered to be at end-diastole. The end-systolic phase was determined as the slice with the smallest left ventricular cavity. Endocardial borders were traced manually; papillary muscle was demarcated separately and excluded from the ventricular volume.
Left ventricular volume was measured in 2 ways. First, the ALEF approach was used. A single long-axis view was obtained, and a single-plane ellipsoid model was applied to estimate the ventricular volume, as demonstrated in Fig 1. This technique assumes that the left ventricle has an ellipsoid shape. The left ventricular contour is outlined (green) at end-diastole and end-systole. The ALEF was calculated by using the standard mathematical calculation for volume of an ellipsoid shape, as 
x A2/L, where A is the cavity area and L (yellow line) is the cavity long-axis measurement.
|
Second, by using the Sergeant's disk approach, a series of short-axis views were taken to encompass the left ventricular volume, and the left ventricular volume was calculated as the sum of each short-axis area multiplied by the slice thickness. Slices were considered to be within the left ventricle if
50% of the ventricular myocardium surrounded the blood volume.17 The following left ventricular parameters were then estimated, by using both techniques, for infants for whom appropriate images could be obtained: ventricular ejection fraction (milliliters), ventricular end-diastolic and end-systolic volumes (milliliters), ventricular stroke volume (milliliters), and cardiac output (liters per minute). Test-retest consistency was assessed by calculating the coefficient of variation for 10 repeated determinations for 1 subject with both long-axis and short-axis methods. The mean difference between the long-axis and short-axis methods was estimated through calculation of the mean difference, according to the approach described by Bland and Altman.18
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
For 10 of 12 subjects, images were obtained that allowed long-axis (ALEF) estimations; for 4 of those infants, short-axis (Sergeant's discs) calculations were also performed. For the remaining 2 infants, movement artifacts prevented image analysis. Examples of images at end-systole and end-diastole with these 2 methods of analysis are shown in Figs 1 and 2. A series of images taken from the b-SSFP cine sequence for the cardiac cycle from diastole to systole are given in Fig 3.
|
|
Values for ventricular volume, cardiac output, and ejection fraction, as estimated with both methods, are given in Figs 4 and 5. We also calculated the values related to the infants' weights at the time of scanning. For the ALEF method, the mean values were as follows: stroke volume, 1.6 mL/kg (range: 0.4–3.5 mL/kg); cardiac output, 0.25 L/minute per kg (range: 0.5–0.55 L/minute per kg); end-diastolic volume, 2.2 mL/kg (range: 0.6–4.6 mL/kg); end-systolic volume, 0.6 mL/kg (range: 0.1–1.1 mL/kg). For the Sergeant's discs method, the mean values were as follows: stroke volume, 1.3 mL/kg (range: 0.8–2.0 mL/kg); cardiac output, 0.18 L/minute per kg (range: 0.13–0.29 L/minute per kg); end-diastolic volume, 2.3 mL/kg (range: 1.5–3.5 mL/kg); end-systolic volume, 1.0 mL/kg (range: 0.8–1.5 mL/kg). The 2 infants with previously diagnosed PDA had much larger left ventricles and showed strikingly increased left ventricular output, achieved primarily by an increase in end-diastolic volume, with similar ejection fractions.
|
|
The coefficients of variation for repeated estimations of end-diastolic volume by using long-axis and short-axis methods were 6% and 3%, respectively, and those for end-systolic volume were 7% and 5%, respectively. The 2 methods could be compared for 4 infants (8 measurements). In all except 1 of those comparisons, the short-axis method yielded a slightly higher value, compared with the long-axis approach. The mean ± SD difference (short axis – long axis) for end-diastolic volume was 1.075 ± 0.2 mL, and that for end-systolic volume was 0.3 ± 0.1 mL. Higher end-diastolic volume led to systematically lower estimates for stroke volume and ejection fraction with the short-axis approach (Figs 4 and 5). The indicated specific absorption rate values (3–4 W/kg) were within safety guidelines for all examinations, and none of the infants showed any signs of thermal stress.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study demonstrates that high-quality CMRI at 3.0-T field strength can be performed for preterm infants without sedation, anesthesia, or breath-holding. In the adult population, the excellent cardiac observation offered by CMRI allows a marked improvement in reproducibility for parameters of ventricular function, compared with echocardiography,9 which translates into both more-accurate clinical decision-making and reduced sample sizes for modeling, hemodynamic, and experimental studies.16
Most CMRI experience to date has been gained at 1.5 T, although there is increasing interest in the use of 3.0-T systems, which are becoming widely used for other body MRI applications.17 Imaging at 3.0 T can provide a potential doubling of the signal/noise ratio, compared with conventional 1.5-T platforms.13,19 At present, 3-T CMRI for adults remains challenging, largely because of the technical difficulties inherent with the higher field strength. Although the signal/noise ratio is higher, there are increased problems of signal stability during respiration, difficulty in shimming the static (B0) magnetic field, and inhomogeneity of the radiofrequency (B1) magnetic fields. Susceptibility effects increase linearly with B0 and so double between 1.5 T and 3 T. The air spaces in the lungs pose greater problems at higher field, but this may be less of an issue with neonates, who tend to have higher lung water content and smaller air spaces. The relatively small wavelength, compared with the dimensions of the adult thorax, is associated with B1 inhomogeneity, which is substantially more pronounced at 3 T than at 1.5 T.20 However, the smaller size of neonates greatly reduces this problem. The smaller size of the neonatal heart also puts a premium on the signal/noise ratio, an advantage of 3 T over 1.5 T. Therefore, neonates are likely to be a favorable subject group for 3.0-T imaging. The fact that high-temporal resolution, cardiac, cine imaging can be obtained at 3.0 T is important for accurate measurement of functional parameters.9,13
Because CMRI is highly sensitive to movement artifact, breath-holding protocols are used commonly in adult studies to obtain high-quality images. In pediatric studies, breath-holding is often enforced and movement is prevented through the induction of general anesthesia, with mechanical ventilation. Because many neonatologists now aim to avoid mechanical ventilation for preterm patients whenever possible, this approach is unattractive; if breath-holding were required for successful CMRI, it is unlikely that the technique could be used widely in neonatal care or research. We found that a simple regimen of prefeeding in a quiet environment and good acoustic protection allowed us to obtain high-quality images for 10 of 12 subjects, although gross movement prevented image acquisition suitable for analysis for 2 infants. Scanning was usually completed within 1 hour, including the time needed to ensure that the infant was asleep. These findings show that CMRI is a practical technique for the preterm population.
This is the first study to attempt 3.0-T CMRI for preterm infants, and the preliminary data obtained are not intended to define reference ranges for estimated values. At present, that are few data using CMRI to establish references values for the pediatric population, and no published normal values are available for the preterm population.21 This study provides the correct imaging parameters to allow high-quality CMRI for preterm patients at 3.0 T. The number of images presented is appropriate to achieve this goal. With the small number of infants in this study, we did not look for predicable effects, such as the dependence of cardiac output on age, and we did not formally compare these results with echocardiograph estimations; additional studies should be able to acquire the relevant data.
We obtained images of sufficient quality with both long-axis (ALEF) and short-axis (Sergeant's discs) methods. In this preliminary study, we did not aim for a formal assessment of the comparative values of the ALEF and Sergeant's discs analyses. It seems that the ellipsoid assumption of the ALEF approach may need to be questioned for preterm infants. Calculation of volumes from short-axis images does not require any geometric assumptions to be made about ventricular morphologic features, and this technique estimated systematically higher values for end-systolic volume and consequently lower estimates for stroke volume, cardiac output, and ejection fraction. We acknowledge that there were substantial differences (mean difference: >25%) between the long-axis and short-axis estimations. Despite these differences, both methods detected very high left ventricular output, without obvious alteration of ejection fraction, in the infants with a previous diagnosis of PDA. Although additional work is needed to assess the appropriate approach for volume estimations, our preliminary estimates showed test-retest values that suggested that, as in the adult population, CMRI can provide precise estimations of functional variables.21
Establishing the feasibility of CMRI for preterm infants opens a range of possibilities. CMRI provides additional techniques to define quantitative blood flow, intracardiac flow patterns, wall stress mapping, characterization of normal and damaged tissue, and molecular events in the endothelium.8 Precise quantitative data on cardiac function should also allow studies of therapeutic interventions such as inotropic or antiinflammatory treatments for cardiac dysfunction with acceptable and practical sample sizes. Therefore, CMRI has the potential to make significant contributions to research and the improvement of care for sick preterm infants. In particular, the possibility of acquiring CMRI scans without breath-holding provides the possibility of developing novel MRI sequences that optimize imaging without the time constraints of breath-holding, which may improve image quality significantly.
|
ACKNOWLEDGMENTS |
|---|
We acknowledge the ongoing work and support of Alan Groves, MD.
|
FOOTNOTES |
|---|
Accepted Mar 1, 2007.
Address correspondence to A. David Edwards, FMedSci, Department of Neonatal Medicine, Imperial College London, Hammersmith Hospital, Du Cane Road, London W12 OHS, England. E-mail: david.edwards{at}imperial.ac.uk
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Osborn DA, Evans N, Kluckow M. Clinical detection of low upper body blood flow in very premature infants using blood pressure, capillary refill time and central-peripheral temperature difference.
Arch Dis Child Fetal Neonatal Ed. 2004;89
:F168
–F173
[Abstract/Free Full Text] - Hunt RW, Evans N, Rieger I, Kluckow M. Lower superior vena cava flow and neurodevelopment at 3 years in very preterm infants. J Pediatr. 2004;145 :573 –575[CrossRef][Web of Science][Medline]
- Pathan N, Hemingway CA, Alizadeh AA, et al. Role of interleukin 6 in myocardial dysfunction of meningococcal septic shock. Lancet. 2004;363 :203 –209[CrossRef][Web of Science][Medline]
- Kusaka T, Okubo K, Nagano K, Isobe K, Itoh S. Cerebral distribution of cardiac output in newborn infants.
Arch Dis Child Fetal Neonatal Ed. 2005;90
:F77
–F78
[Abstract/Free Full Text] - Otterstad JE, Froeland G, St John Sutton M, Holme I. Accuracy and reproducibility of biplane two-dimensional echocardiographic measurements of left ventricular dimensions and function.
Eur Heart J. 1997;18
:507
–513
[Abstract/Free Full Text] - Osborn DA, Evans N. Early volume expansion for prevention of morbidity and mortality in very preterm infants. Cochrane Database Syst Rev. 2004;(2) :CD002055
- Osborn DA, Evans N. Early volume expansion versus inotropes for prevention of morbidity and mortality in very preterm infants. Cochrane Database Syst Rev. 2001;(2) :CD002056
- Manning WJ, Pennell DJ, eds. Cardiovascular Magnetic Resonance. London, England: Churchill Livingstone; 2002
- Grothues F, Smith GC, Moon JC, et al. Comparison of interstudy reproducibility of cardiovascular magnetic resonance with two-dimensional echocardiography in normal subjects and in patients with heart failure or left ventricular hypertrophy. Am J Cardiol. 2002;90 :29 –34[CrossRef][Web of Science][Medline]
- Didier D, Ratib O, Beghetti M, Oberhaensli I, Friedi B. Morphologic and functional evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging. 1999;10 :639 –655[CrossRef][Web of Science][Medline]
- Schlesinger AE, Hermandez RJ. Magnetic resonance imaging in congenital heart disease in children. Tex Heart Inst J. 1996;23 :128 –143[Web of Science][Medline]
- Tsai-Goodman B, Geva T, Odegard KC, Sena LM, Powell AJ. Clinical role, accuracy and technical aspects of cardiovascular magnetic resonance imaging in infants. Am J Cardiol. 2004;94 :69 –74[CrossRef][Web of Science][Medline]
- Gutberlet M, Spors B, Grothoff M. Comparison of different cardiac MRI sequences at 1.5/3.0T with respect to signal-to-noise and contrast-to-noise ratios: initial experience. Rofo. 2004;176 :801 –808[Web of Science][Medline]
- Hinton DP, Wald LL, Pitts J, Schmitt F. Comparison of cardiac MRI on 1.5 and 3.0 Tesla clinical whole body systems. Invest Radiol. 2003;38 :436 –442[CrossRef][Web of Science][Medline]
- Thiele H, Paetsch I, Schnackenburg B, et al. Improved accuracy of quantitative assessment of left ventricular volume and ejection fraction by geometric models with steady-state free precession. J Cardiovasc Magn Reson. 2002;4 :327 –339[CrossRef][Web of Science][Medline]
- Bellenger NG, Davies LC, Francis JM, Coats AJ, Pennell DJ. Reduction in sample size for studies of remodelling in heart failure by the use of cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2000;2 :271 –278[Web of Science][Medline]
- Schick F. Whole-body MRI at high field: technical limits and clinical potential. Eur Radiol. 2005;15 :946 –959[CrossRef][Web of Science][Medline]
- Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1 :307 –310[CrossRef][Web of Science][Medline]
- Wen H, Denison TJ, Singerman RW, Balaban RS. The intrinsic signal-to-noise ratio in human cardiac imaging at 1.5, 3, and 4 T. J Magn Reson. 1997;125 :65 –71[CrossRef][Web of Science][Medline]
- Collins CM, Liu W, Schreiber W, Yang QX, Smith MB. Central brightening due to constructive interference with, without and despite dielectric resonance. J Magn Reson Imaging. 2005;21 :192 –196[CrossRef][Web of Science][Medline]
- Alfakih K, Plein S, Thiele H, Jones T, Ridgway JP, Sivananthan MU. Normal human left and right ventricular dimensions for MRI as assessed by turbo gradient echo and steady-state free precession imaging sequences. J Magn Reson Imaging. 2000;17 :323 –329[CrossRef]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
