Stress Perfusion Imaging Using Cardiovascular Magnetic Resonance: A Review

[김기환/강원대]

Loader.feature("lp_focus").qCode("ArticleFocus.init('focusButton');");

Heart, Lung and Circulation
Volume 19, Issue 12, December 2010, Pages 697-705

---

doi:10.1016/j.hlc.2010.08.008 | How to Cite or Link Using DOI

  Permissions & Reprints

Review

Stress Perfusion Imaging Using Cardiovascular Magnetic Resonance: A Review

Jonathan P. Christiansen MD, FRACP, FACC, FCSANZ^{a, ,} , Theodoros D. Karamitsos MD, PhD^b, Saul G. Myerson MD, MRCP, FESC^b, Jane M. Francis DCCR^b and Stefan Neubauer MD, FRCP, FACC, FMedSci^b

^a Cardiovascular Division, North Shore Hospital, Waitemata Health, Auckland, New Zealand

^b Centre for Clinical Magnetic Resonance Research, University of Oxford, Dept. of Cardiovascular Medicine, John Radcliffe Hospital, Oxford, UK

Received 24 February 2010; 

revised 9 August 2010; 

accepted 11 August 2010. 

Available online 23 September 2010.

Referred to by:		Corrigendum to “Stress Perfusion Imaging Using Cardiovascular Magnetic Resonance: A Review” [Heart Lung Circ. 19 (2010) 697–705] Heart, Lung and Circulation, Volume 20, Issue 4, April 2011, Page e1, Jonathan P. Christiansen, Theodoros D. Karamitsos, Saul G. Myerson, Jane M. Francis, Stefan Neubauer PDF (39 K)   |

Loader.rt("abs_end");

Stress perfusion CMR can provide both excellent diagnostic and important prognostic information in the context of a comprehensive assessment of cardiac anatomy and function. This coupled with the high spatial resolution, and the lack of both attenuation artefacts and ionising radiation, make CMR stress perfusion imaging a highly attractive stress imaging modality. It is now in routine use in many centres, and shows promise in eval‎uating patients with clinical problems beyond those of epicardial coronary disease.

Key words: Magnetic resonance; Stress; Perfusion

Article Outline

Introduction

Coronary Physiology

Contrast Agents

Imaging Sequences and Artefacts

Imaging Protocol

Image Analysis

Qualitative Interpretation

Quantification and Validation

Diagnostic Performance

Prognostic Information

Beyond Epicardial Coronary Disease

Limitations

Conclusions

Acknowledgements

References

Introduction

Non-invasive assessment of myocardial perfusion is of major importance in the diagnostic eval‎uation and risk stratification of patients with known or suspected coronary disease [1]. Cardiovascular magnetic resonance (CMR) has rapidly become an important imaging modality capable of comprehensive assessment of cardiac structure and function. Furthermore, advances in pulse-sequence development, hardware and reconstruction algorithms have enabled CMR to perform high resolution imaging of first pass myocardial perfusion under vasodilator stress. This has noteworthy advantages over existing techniques for perfusion assessment such as SPECT, including a greater spatial resolution, lack of soft tissue attenuation, the absence of ionising radiation, and avoidance of the need for more than one imaging session. CMR stress perfusion imaging is now a routine investigation in clinical practice. In the EuroCMR Registry, 21% of the 11,040 studies recorded were adenosine stress perfusion scans [2]. We will review the current state of stress myocardial perfusion imaging with CMR, highlighting data regarding its validation, diagnostic accuracy, prognostic importance and incremental benefit over more established modalities.

Coronary Physiology

The coronary circulation consists of the epicardial vessels and the microcirculation, both under constant regulation. For the purposes of understanding the pathophysiology of acute inducible ischaemia as it relates to stress perfusion imaging, we must focus on the behaviour of the microcirculation in the context of varying degrees of stenosis in the epicardial coronary arteries [3]. These principles apply equally to all forms of stress perfusion imaging including myocardial contrast echocardiography, SPECT and PET. It is well understood that whilst resting blood flow may remain normal for up to 85% luminal diameter stenosis of the epicardial artery, flow during maximal hyperaemia is reduced when the luminal diameter falls below 50% [4]. At baseline approximately 8% of LV mass is due to blood in the microcirculation, and importantly 90% of that blood volume resides in capillaries. A constant capillary hydrostatic pressure of 30 mm Hg is maintained through autoregulation and is necessary for homeostasis, particularly the avoidance of interstitial and cellular oedema [5]. If coronary perfusion pressure falls (as in the setting of a significant upstream stenosis), arteriolar vasodilation takes place to maintain normal resting myocardial blood flow. However this capacity for autoregulation is exhausted when the arteriolar bed is maximally vasodilated, which occurs when coronary perfusion pressure is approximately 45 mm Hg (equivalent to an 85% luminal diameter stenosis in an epicardial coronary artery) [6].

If pharmacologic hyperaemia is induced (using adenosine or dipyridamole) in the setting of a significant epicardial stenosis, arteriolar resistance falls still further, and the maintenance of capillary hydrostatic pressure is threatened. The inevitable consequence is that groups of capillaries shut down (de-recruit) to preserve constant hydrostatic pressure, a phenomenon associated with a demonstrable rise in capillary resistance [7]. The myocardial concentration of a perfusion agent such as gadolinium or technetium 99m sestamibi is fundamentally related to the permeability-surface area product, in addition to arterial concentration and flow. As permeability factors are normally constant, the main determinant of tracer extraction is capillary surface area. Therefore the presence, extent and severity of a perfusion defect visualised during exogenous vasodilator stress in the setting of an epicardial coronary stenosis is a direct result of de-recruitment of myocardial capillaries, leading to a decreased capillary surface area and reduced extraction of the perfusion agent [6].

Contrast Agents

Gadolinium chelates are the major class of contrast agents used for CMR stress perfusion imaging. They are highly diffusible small molecules, readily distributed to the extravascular space, but remain extracellular, and can be imaged as a first pass bolus following intravenous administration. Gadolinium is an effective paramagnetic agent due to the seven unpaired electrons, and chelation avoids the extreme toxicity of the free molecule [8]. The gadolinium is indirectly detected through a more rapid relaxation rate of those water protons in close proximity to it. The result is that normally perfused tissues have enhanced T1-weighted signal following intravenous administration of gadolinium–DTPA [9]. At low doses there is an approximately linear relationship between contrast concentration in tissues and the signal intensity in the subsequent image [10]. Tissue concentration in the myocardium is a function of multiple factors including the extraction fraction, and the degree of water exchange between the fluid compartments [11]. The doses typically used in stress perfusion imaging do not always result in myocardial concentrations sufficiently low to be in the linear range, but T2 and T2* effects are minimal, except in the LV cavity [12].

Intravascular gadolinium-based agents have also been developed, and one – gadofosveset trisodium (Vasovist™, Schering AG) – is approved for use in Europe [13]. After IV injection, Vasovist™ binds reversibly to human albumin in plasma and results in a macromolecular intravascular contrast agent with high relaxivity. A small unbound portion is eliminated by the kidneys through glomerular filtration. An intravascular contrast agent would enable potentially more accurate quantification of myocardial blood volume and therefore flow, and this has been validated in animal models [14]. Clinical trials are in progress.

Imaging Sequences and Artefacts

First pass stress perfusion imaging with CMR requires pulse sequences with (a) strong T1 weighting to maximise the image contrast intensity in relation to contrast agent concentration, (b) rapid data acquisition to ensure coverage of sufficient myocardial segments for diagnostic purposes, (c) satisfactory spatial resolution and (d) minimal artefacts [15]. Saturation-recovery (SR) pulse sequences are currently widely used, as they provide the best compromise between these inherently competing factors. SR sequences are relatively insensitive to changes in heart rate, have short preparation times, adequate spatial resolution and are suitable for multislice imaging. However they unfortunately have a reduced dynamic range (the number of shades of grey displayed in an image) as compared with earlier inversion-recovery (IR) sequences [15]. Imaging is typically performed in three to five user-defined slices (short or long axis depending on the patient's heart rate) and in-plane resolution is usually in the order of 1.5–3 mm, with a slice thickness of 5–10 mm. Ultra-fast acquisition is therefore essential, and protocols are usually based on gradient echo [16], gradient echo-planar [17] or steady-state free precession (SSFP) sequences [18] W.G. Schreiber, M. Schmitt, P. Kalden, O.K. Mohrs, K.F. Kreitner and M. Thelen, Dynamic contrast-enhanced myocardial perfusion imaging using saturation-prepared True FISP, J Magn Reson Imaging 16 (2002), pp. 641–652. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (0)[18], in combination with parallel imaging techniques [19]. Each of these sequences has advantages and disadvantages and there is no consensus regarding the optimal approach, despite multiple comparison studies [1]. Although the majority of stress perfusion imaging is performed at 1.5 T, trials have suggested that there are advantages in imaging at 3.0, including improved contrast to noise and signal to noise ratios [20]. However there is also a greater potential for artefact, and further work is ongoing to optimise sequences for the 3.0 T environment [21].

All techniques for non-invasive assessment of myocardial perfusion have their own unique artefacts, and CMR is no exception. The most problematic of these is termed the “Dark Rim Artefact” (DRA) [22]. This is seen as a transient rim of reduced signal intensity in the subendocardium (hence “dark rim”), which can give the appearance of hypoperfusion suggesting ischaemia. Characteristically the DRA is transient, lasting only a few heartbeats, whereas a true perfusion defect is more persistent (>5 heartbeats). The DRA is exacerbated by an increased heart rate, a more concentrated contrast bolus, and is most typical in the subendocardium perpendicular to the phase-encoding direction [21]. Studies in ex vivo hearts have suggested that the underlying cause is most likely related to limited spatial resolution [23]. However others have considered the DRA a result of magnetic susceptibility artefact, or even partial volume effects [24]. Gradient-echo sequences have been found to have lower DRA scores than SSFP sequences [25]. Increasing image resolution is the logical way to minimise the DRA, and current constraints on temporal resolution may potentially be circumvented by newer acquisition techniques such as k-t SENSE [26]. Other artefacts such as wrap or those related to parallel imaging are common, but more easily resolved by increasing the phase-encoded field of view, altering the phase-encoding direction or reducing the acceleration factor.

Imaging Protocol

Stress perfusion imaging with CMR generally forms part of a more comprehensive examination, including structural and functional analysis with SSFP sequences, and eval‎uation of myocardial fibrosis with late gadolinium enhancement (LGE) IR sequences. Patient safety in the magnet environment is paramount, given the administration of vasoactive infusions in a difficult environment for patient access. Staff should be certified in basic and advanced cardiac life support, and should have undergone training for dealing with emergencies in the magnet environment. Continuous ECG and regular blood pressure monitoring is required, using MR-compatible equipment. Careful screening of patients for problems that will preclude administration of vasodilator agents should be carried out at the same time as standard screening for contraindications to MRI. A history of active bronchospastic airways disease or high grade conduction abnormalities should be obtained and a baseline 12-lead ECG reviewed. Patients should abstain from caffeine consumption for at least 24 h prior to the study [27]. Life-threatening complications arising from vasodilator stress testing are rare, but the patient should be made aware of the risks and written consent obtained. Patients should also be counselled about the symptoms to expect during vasodilator stress, and given instructions on breathholding. A long breathhold is desirable, but not essential. Patients should be asked to hold their breath as best they can for the more critical first 15 s of the scan, and then to breathe shallowly and gently for the remaining duration of the imaging.

Vasodilation can be induced with either a continuous intravenous infusion of adenosine, or a bolus of intravenous dipyridamole. Adenosine is a direct agonist of the A_2A receptor in coronary arterioles. Dipyridamole is a pro-drug requiring metabolism in the liver to produce active metabolites that block the cellular reuptake of adenosine and inhibit adenosine deaminase activity—thus leading to increased levels of extracellular adenosine in the myocardium. For the adenosine infusion it is preferable to place two intravenous lines, one in each antecubital fossa. Adenosine is infused at 140 μg/kg/min for 3–6 min [28], and the patient observed for an increase in heart rate (>10 beats/min) accompanied by a fall in blood pressure (>10 mm Hg). In the apparent absence of an appropriate physiologic response, an increased infusion rate up to 210 μg/kg/min has been validated and found safe in patients with prior caffeine consumption [29]. The most common protocol in use for dipyridamole is a slow intravenous bolus injection of 0.56 mg/kg over 4 min, followed by imaging. A higher dose protocol, in which a further bolus of 0.28 mg/kg is administered at 8 min, with imaging commenced at 10 min, has been suggested as providing more predictable and prolonged vasodilation [21]. The side effects of both drugs are similar, and patients typically experience flushing, chest discomfort, breathlessness, headache and nausea. Such side effects are more frequent during adenosine administration, but are more long-lasting when dipyridamole is used due to its longer half-life [30]. Although more expensive, the predictability of adenosine and its short acting properties make it the preferred vasodilator in CMR stress perfusion studies.

First pass imaging is performed during maximal vasodilation using a bolus of gadolinium-based contrast. The dose of this bolus will depend on the plans for subsequent analysis. A higher dose (0.10–0.15 mmol/kg) has been shown to result in improved operating characteristics for qualitative (visual) analysis [31], although the DRA will potentially be exacerbated in a higher dose range. A lower dose (typically 0.03–0.05 mmol/kg) is required for quantitative analysis, especially if a measure of absolute blood flow is to be accurately calculated [32]. These lower doses avoid losses from T2* effects that may result in distorted estimates of the arterial input function, obtained from the signal in the LV cavity [12]. A dual bolus technique has been validated as an alternative means of avoiding this issue, but is not in routine clinical use [33] and [34]. The contrast bolus should be injected at a rate of at least 4 mL/min using a power injector, and followed by a large volume saline flush (20 mL) [21].

Close attention should be paid to the rise in heart rate induced by vasodilation, and the first pass imaging sequence prepared accordingly. Imaging with a temporal repetition rate of every heartbeat is preferred for qualitative analysis, and is essential for quantitative analysis [21]. The patient should be counselled regarding breathholding, and asked to avoid unnecessarily deep breaths when beginning to breathe whilst imaging is completed. The choice of vasodilator may influence the order in which the rest and stress studies are performed. If adenosine is used, the stress imaging should be performed first to avoid contamination of the (more important) stress images with either residual circulating contrast, or that retained in areas of myocardial scar. Some operators have advocated stress imaging only, but we advise against this as the rest images allow comparison for identification of artefacts. An interval of at least 10 min is required following stress scanning before resting perfusion imaging is performed. However the prolonged vasodilatory effect of dipyridamole will necessitate a more lengthy delay in rest imaging if it is performed second, a further factor in recommending adenosine as the vasodilator of choice in CMR perfusion testing. To optimise workflow and efficiency, this time can be valuably used to complete SSFP sequences such as the short axis ‘stack’. In most patients LGE IR imaging for myocardial scar will complete the study, and follows the resting imaging [35]. A further bolus of contrast is usually administered immediately following the resting images, to ensure a total gadolinium dose of 0.15–0.2 mmol/kg, and IR imaging commenced after a further delay of 5–10 min. Fig. 1 summarises one potential protocol for a comprehensive CMR stress perfusion study.

Figure 1. 

Protocol for stress perfusion imaging. A comprehensive assessment of ventricular function, myocardial scar and inducible ischaemia can be obtained in under 45 min. The optimal order of image acquisition, contrast injection and pharmacologic stress is shown in this graphic.

View Within Article

Image Analysis

Qualitative Interpretation

Visual analysis remains the basis of interpretation and reporting of clinical stress perfusion CMR studies. The optimal approach integrates the delayed enhancement IR images with the perfusion data, and an algorithm formalising this methodology has been shown to have a sensitivity, specificity and accuracy for detecting coronary artery disease of 89%, 87% and 88% respectively [36]. Interpretation should start with the LGE images to identify areas of myocardial scar (infarct). In the absence of scar, a defect seen on stress perfusion imaging is considered reversible ischaemia if the defect is absent, or significantly reduced, on the resting images. (Fig. 2) If the resting images show a similar defect, this is either artefact on both sets of images, or an infarct (scar) with a fixed perfusion defect, which should be readily visible on the LGE images (Fig. 3). The potential for DRA should be carefully assessed. Caution is needed in patients with existing scar, as retention of contrast in infarcted myocardium may contaminate the resting perfusion images, leading to over-diagnosis of inducible peri-infarct ischaemia. In this setting the stress images should be compared with the LGE images in the same plane, rather than the rest perfusion images. The high resolution of CMR stress perfusion images enables the detection of three-vessel disease, visible as a circumferential perfusion defect in the endocardium compared to the epicardial myocardial layer. These may be missed on SPECT due to ‘balanced ischaemia’, which the lower resolution of SPECT is unable to visualise [1]. Reporting is based on the 17-segment model for perfusion assessment across different imaging modalities, and the components of a standard report have recently been defined [37].

---

Figure 2. 

Inducible ischaemia in the absence of myocardial scar. Images from two patients with significant coronary disease, inducible ischaemia and no myocardial scar are shown. In patient (A) there is moderate mid-LAD/diagonal bifurcation disease on angiography, and a perfusion defect is shown in the anterior wall and septum at stress, which resolves completely at rest. In patient (B) the proximal LCx is occluded at angiography, and ischaemia is demonstrated in the lateral wall, which again resolves at rest. Additionally a limited area of subendocardial ischaemia is noted in the mid-septum, consistent with extensive atheroma in the LAD and likely ostial narrowing of septal branches.

View Within Article

---

Figure 3. 

Inducible ischaemia in the presence of myocardial scar. Images from two patients with significant coronary disease and prior myocardial infarction are show. In patient (A) there is a partial (50–75%) thickness inferior infarction seen on the delayed enhancement image, and a corresponding fixed perfusion defect is seen at stress and rest. Patient (B) illustrates a small focal area of scar in the mid-inferoseptal segment, but the defect seen on stress perfusion is considerably more extensive, and resolves with rest, consistent with inducible ischaemia in the territory associated with the infarction.

View Within Article

Quantification and Validation

A range of semi-quantitative parameters can be gained from the perfusion images. These include: peak myocardial enhancement, time to peak enhancement, contrast enhancement ratio [38] and [39], and the upslope of the signal intensity curve, which may be used to calculate a flow reserve index [40]. All such measurements are based on regions of interest positioned within the myocardium. Considerable care is needed in the analysis, which may be time consuming. Alignment of all images correcting for respiratory movement is critical, and regions of interest must be placed to avoid contamination from blood pool or surrounding structures. Ventricular ectopy that has resulted in a change in LV cavity size can be particularly problematic. A degree of automation is possible in these processing steps, but manual realignment and contour tracing is the norm. Data should be corrected for in-homogeneities in the field related to the surface coil. This baseline correction is accomplished by dividing the signal intensities after contrast injection with the pre-contrast values. The most sophisticated semi-quantitative analysis is the perfusion reserve index. This is the ratio of the upslopes of the time–signal intensity curves at stress and rest, corrected for the arterial input function by simple division by the upslope of the curve for the ventricular cavity [40]. This method has been validated compared with microsphere-derived data in animal models [41], and has been shown to improve the operating characteristics of the test in clinical trials [42]. Commercial software for image analysis and calculation of this index is available from multiple vendors.

Absolute myocardial blood flow can be calculated, although published methods have considerable complexity and are unsuited to routine clinical use. There are a number of potential pitfalls in accurately calculating absolute flow, including the non-linear relationship of contrast concentration and signal (which is exacerbated if high dose gadolinium contrast is used), and regional variations in the magnetic field [21]. The time course of observed myocardial tissue enhancement is typically corrected for effects of the arterial input function using mathematical deconvolution. A model-independent deconvolution analysis has been validated in animal studies [43], compared with existing modalities such as PET [44], and has been used successfully in clinical research [45] and [46].

Diagnostic Performance

The operating characteristics of CMR stress perfusion imaging are good, and comparable to other stress imaging modalities. A recent meta-analysis in 1183 patients demonstrated a sensitivity of 0.91 (95% CI, 0.88–0.94) and specificity of 0.81 (95% CI, 0.77–0.85) for the detection of coronary artery disease (CAD) [47]. The preval‎ence of coronary disease in this pooled analysis of 14 trials was 57.4%. More recently the multi-centre, multi-vendor MR-IMPACT study demonstrated a sensitivity of 0.85 and specificity of 0.67 (AUC = 0.86 ± 0.06), comparing favourably with SPECT (AUC = 0.75 ± 0.09) [48]. All patients enroled in MR-IMPACT required clinical investigation of possible CAD with either angiography or SPECT, and those with recent infarction or prior CABG were excluded. Non-invasive imaging data was compared with quantitative angiography. The preval‎ence of CAD in the study group was high—77%, and 31% had a history of prior angioplasty. Similar accuracy in women has been reported, with sensitivity, specificity and accuracy for CMR being 84%, 88% and 87% respectively [49]. The 147 women in this study were all investigated for symptoms potentially consistent with coronary ischaemia, but the preval‎ence of CAD at angiography was only 27%. Despite the low rate of significant CAD, the operating characteristics of stress perfusion CMR were very good. Direct comparison has been made with invasive assessment of coronary blood flow, notably fractional flow reserve (FFR) [50] and [51]. Visually assessed stress perfusion CMR accurately reflected findings on FFR in 103 patients, 58% of whom has significant CAD, with a positive and negative predictive values for FFR < 0.75 of 90.9% (95% CI, 84.3–97.5%) and 93.9% (95% CI, 88.9–98.9%) respectively [52].

Prognostic Information

Data demonstrating the prognostic value of CMR stress perfusion are accumulating. The detection of myocardial ischaemia identified patients at high risk of subsequent cardiac death or non-fatal myocardial infarction in a study analysing 461 participants who underwent both stress perfusion imaging and dobutamine stress testing with CMR [53]. The event rate was 16.3% at three years for patients with abnormal stress perfusion imaging (HR = 12.5, 95% CI = 3.64–43.03). A normal CMR perfusion scan was associated with a <1% annual event rate, comparing favourably with prognosis reported for a normal SPECT scan [54]. More recently the potential importance of stress perfusion CMR in assessing patients with epicardial coronary stenosis of intermediate severity was demonstrated [55]. Eighty-one patients with 50–75% diameter stenosis and stable angina were followed for 30 ± 8 months. Patients with a perfusion defect underwent significantly more frequent target vessel revascularisation than those with normal scans (38% vs. 6% respectively, P = 0.005). Similar data are available for dipyridamole CMR stress perfusion imaging [56]. The prognostic value of CMR perfusion has also been confirmed in patients presenting to the emergency department with chest pain but without evidence of acute infarction [57]. A sensitivity of 100% and specificity of 93% for predicting future events associated with CAD was reported in 135 patients at one year follow-up. Recently Steel et al. [58] highlighted the complementary prognostic significance of a reversible perfusion defect and LGE. In a population of 254 patients undergoing CMR stress perfusion imaging, a reversible perfusion defect maintained a 5.5-fold hazards increase in MACE when adjusted for the effects of age, gender, and LGE (P = 0.0004), whilst LGE maintained a 2.7-fold hazards increase in MACE when similarly adjusted (P = 0.04) [58].

Beyond Epicardial Coronary Disease

The potential role of stress perfusion CMR has been shown in patients with a range of pathologies in which the epicardial coronary arteries are unobstructed. The most widely studied are patients with Cardiac Syndrome X, characterised by typical exertional angina, abnormal exercise ECG and normal coronary angiography. Global subendocardial hypoperfusion has been demonstrated [59], although this finding has not been universally confirmed and may be difficult to differentiate from “balanced hypoperfusion” in patients with three-vessel coronary artery disease [60]. The role of CMR perfusion assessment in cardiomyopathies has been of interest. In patients with hypertrophic cardiomyopathy, reduced myocardial perfusion in the subendocardium has been shown with CMR (Fig. 4), and the magnitude of the reduction is proportional to the degree of hypertrophy [61]. Reduced resting perfusion has been shown to be reduced in patients with idiopathic dilated cardiomyopathy [62]. A significant impairment in endocardial to epicardial perfusion reserve ratio in patients with a history of aortic coarctation has recently been described [63]. The lack of ionising radiation and the high spatial resolution of CMR stress perfusion imaging also make it an attractive technique for eval‎uating patients with congenital heart disease [64].

---

Figure 4. 

Subendocardial ischaemia in apical hypertrophic cardiomyopathy. The 2-chamber cine image demonstrates typical apical hypertrophic cardiomyopathy in a young woman. The stress image shows significant global subendocardial ischaemia in the apical segments, which resolves with rest.

View Within Article

Limitations

A minority of patients will not be suitable for stress perfusion CMR. The presence of devices such as pacemakers and implantable defibrillators remains a contraindication. Comprehensive lists of other device-related contraindications are widely available, e.g. http://www.MRIsafety.com, and should always be reviewed on an individual basis. Morbid obesity is a limitation due to the relatively small bore diameter of the magnet. This is vendor-specific, but in general a maximal body circumference of greater than 145 cm will preclude CMR. Arrhythmias such as atrial fibrillation or frequent ventricular ectopy may impair image quality and reduce diagnostic accuracy, even when rate control is optimal. More importantly patients with resting tachycardia may be unsuitable for stress perfusion CMR. Firstly it may not be possible to obtain sufficient high resolution imaging data to assess perfusion in all territories, requiring either an increase in voxel size or a reduction in the number of short axis images. Secondly a high resting heart rate may limit the duration of adenosine infusion, potentially compromising the accuracy of the test.

Recently important concerns have been raised regarding the link between the use of gadolinium–DTPA and a rare debilitating disease termed nephrogenic systemic fibrosis (NSF) [65]. In patients with renal insufficiency there is a strong association between NSF and the administration of gadolinium-based contrast agents [66]. Although the pathophysiology of NSF is poorly understood, the dissociation of the toxic gadolinium from its chelating agent in patients with substantially prolonged half-life of contrast agent excretion is thought to be responsible [67]. A strategy of implementation of guidelines restricting the use of gadolinium–DTPA in patients with impaired renal function, and switching to optimally stable forms of the contrast agent has been shown to avert the development of NSF in follow-up, thus making CMR perfusion imaging a safe technique [68].

Conclusions

Stress perfusion CMR can provide both excellent diagnostic and important prognostic information in the context of a comprehensive assessment of cardiac anatomy and function. This coupled with the high spatial resolution, and the lack of both attenuation artefacts and ionising radiation, make CMR stress perfusion imaging a highly attractive stress imaging modality. It is now in routine use in many centres, and shows promise in eval‎uating patients with clinical problems beyond those of epicardial coronary disease.

Acknowledgement

There was no financial assistance for this project.

Loader.rt("lp_refLoad");Loader.feature("lp_ref").load();Loader.rt("ref_start");

References

[1] M. Salerno and G.A. Beller, Noninvasive assessment of myocardial perfusion, Circ Cardiovasc Imaging 2 (2009), pp. 412–424. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (10)

[2] O. Bruder, S. Schneider, D. Nothnagel, T. Dill, V. Hombach and J. Schulz-Menger et al., EuroCMR (European Cardiovascular Magnetic Resonance) registry results of the German Pilot Phase, J Am Coll Cardiol 54 (15) (2009), pp. 1457–1466. Article | PDF (1818 K) | View Record in Scopus | Cited By in Scopus (23)

[3] S. Kaul and H. Ito, Microvasculature in acute myocardial ischemia. Part I. Evolving concepts in pathophysiology, diagnosis, and treatment, Circulation 109 (2004), pp. 146–149. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (48)

[4] K.L. Gould and K. Lipscomb, Effects on coronary stenoses on coronary flow reserve and resistance, Am J Cardiol 34 (1974), pp. 48–55. Abstract | PDF (1036 K) | View Record in Scopus | Cited By in Scopus (268)

[5] J. Jarhult and S. Mellander, Autoregulation of capillary hydrostatic pressure in skeletal muscle during regional arterial hypo- and hypertension, Acta Physiol Scand 91 (1974), pp. 32–41. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (24)

[6] S. Kaul, The role of capillaries in determining coronary blood flow reserve: implications for stress-induced reversible perfusion defects, J Nucl Cardiol 8 (2001), pp. 694–700. Abstract | View Record in Scopus | Cited By in Scopus (13)

[7] A.R. Jayaweera, N. Edwards, W.P. Glasheen, F.S. Villanueva, R.D. Abbott and S. Kaul et al., In vivo myocardial kinetics of air-filled albumin microbubbles during myocardial contrast echocardiography: comparison with radiolabeled red blood cells, Circ Res 74 (1994), pp. 1157–1165. View Record in Scopus | Cited By in Scopus (158)

[8] H.J. Weinmann, R.C. Brasch, W.R. Press and G.E. Wesbey, Characteristics of gadolinium–DTPA complex: a potential NMR contrast agent, Am J Roentgenol 142 (1984), pp. 619–624. View Record in Scopus | Cited By in Scopus (448)

[9] P. Croisille, D. Revel and M. Saeed, Contrast agents and cardiac MR imaging of myocardial ischemia: from bench to bedside, Eur Radiol 16 (2006), pp. 1951–1963. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (10)

[10] K.M. Donahue, R.M. Weisskoff and D. Burstein, Water diffusion and exchange as they influence contrast enhancement, J Magn Reson Imaging 7 (1) (1997), pp. 102–110. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (138)

[11] K.L. Gould, K.O. Kelley and E.L. Bolson, Experimental validation of quantitative coronary arteriography for determining pressure-flow characteristics of coronary stenosis, Circulation 66 (1982), pp. 930–937. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (38)

[12] P. Kellman, A.H. Aletras, L. Hsu, E.R. McVeigh and A.E. Arai, T2* measurement during first-pass contrast-enhanced cardiac perfusion imaging, Magn Reson Med 56 (2006), pp. 1132–1134. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (8)

[13] Gadofosveset: MS 325, MS 32520, Vasovist, ZK 236018. Drugs R D 2004;5(6):339–42.

[14] M. Jerosch-Herold, X. Hu, N.S. Murthy, C. Rickers and A.E. Stillman, Magnetic resonance imaging of myocardial contrast enhancement with MS-325 and its relation to myocardial blood flow and the perfusion reserve, J Magn Reson Imaging 18 (2003), pp. 544–554. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (31)

[15] P. Kellman and A.E. Arai, Imaging sequences for first pass perfusion: a review, J Cardiovasc Magn Reson 9 (2007), pp. 525–537. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (26)

[16] D.J. Atkinson, D. Burstein and R.R. Edelman, First-pass cardiac perfusion: eval‎uation with ultrafast MR imaging, Radiology 174 (1990), pp. 757–762. View Record in Scopus | Cited By in Scopus (175)

[17] F.H. Epstein and A.E. Arai, Optimization of fast cardiac imaging using an echo-train readout, J Magn Reson Imaging 11 (2000), pp. 75–80. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (18)

[18] W.G. Schreiber, M. Schmitt, P. Kalden, O.K. Mohrs, K.F. Kreitner and M. Thelen, Dynamic contrast-enhanced myocardial perfusion imaging using saturation-prepared True FISP, J Magn Reson Imaging 16 (2002), pp. 641–652. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (0)

[19] P. Kellman, J.A. Derbyshire, K.O. Agyeman, E.R. McVeigh and A.E. Arai, Extended coverage first-pass perfusion imaging using slice-interleaved TSENSE, Magn Reson Med 51 (2004), pp. 200–204. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (41)

[20] A.S. Cheng, T.J. Pegg, T.D. Karamitsos, N. Searle, M. Jerosch-Herold and R.P. Choudhury et al., Cardiovascular magnetic resonance perfusion imaging at 3-tesla for the detection of coronary artery disease: a comparison with 1.5-tesla, J Am Coll Cardiol 49 (25) (2007), pp. 2440–2449. Article | PDF (753 K) | View Record in Scopus | Cited By in Scopus (53)

[21] B.L. Gerber, S.V. Raman, K. Nayak, F.H. Epstein, P. Ferreira and L. Axel et al., Myocardial first-pass perfusion cardiovascular magnetic resonance: history, theory, and current state of the art, J Cardiovasc Magn Res 10 (2008), p. 18. Full Text via CrossRef

[22] A.E. Arai, Magnetic resonance first-pass myocardial perfusion imaging, Top Magn Reson Imaging 11 (6) (2000), pp. 383–398. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (32)

[23] E.V. Di Bella, D.L. Parker and A.J. Sinusas, On the DRA in dynamic contrast-enhanced MRI myocardial perfusion studies, Magn Reson Med 54 (5) (2005), pp. 1295–1299. View Record in Scopus | Cited By in Scopus (68)

[24] W.G. Schreiber, M. Schmitt, P. Kalden, O.K. Mohrs, K.F. Kreitner and M. Thelen, Dynamic contrast-enhanced myocardial perfusion imaging using saturation-prepared TrueFISP, J Magn Reson Imaging 16 (6) (2002), pp. 641–652. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (79)

[25] M. Fenchel, U. Helber, O.P. Simonetti, N.I. Stauder, U. Kramer and C.N. Nguyen et al., Multislice first-pass myocardial perfusion imaging: comparison of saturation recovery (SR)-TrueFISP-two-dimensional (2D) and SR TurboFLASH-2D pulse sequences, J Magn Reson Imaging 19 (5) (2004), pp. 555–563. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (30)

[26] S. Plein, J. Schwitter, D. Suerder, J.P. Greenwood, P. Boesiger and S. Kozerke, k-Space and time sensitivity encoding-accelerated myocardial perfusion MR imaging at 3.0 T: comparison with 1.5 T, Radiology 249 (2) (2008), pp. 493–500. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (21)

[27] M.I. Miyamoto, S.L. Vernotico, H. Majmundar and F.S. Thomas, Pharmacologic stress myocardial perfusion imaging: a practical approach, J Nucl Cardiol 14 (2007), pp. 250–255. Abstract | View Record in Scopus | Cited By in Scopus (13)

[28] J.H. O’Keefe, T.M. Bateman, L.R. Handlin and C.L. Barnhart, Four- versus 6-minute infusion protocol for adenosine thallium-201 single photon emission computed tomography imaging, Am Heart J 129 (1995), pp. 482–487.

[29] E. Reyes, C.Y. Loong, M. Harbinson, J. Donovan, C. Anagnostopoulos and S.R. Underwood, High-dose adenosine overcomes the attenuation of myocardial perfusion reserve caused by caffeine, J Am Coll Cardiol 52 (2008), pp. 2008–2016. Article | PDF (1172 K) | View Record in Scopus | Cited By in Scopus (6)

[30] D.I. Johnston, J.R. Daley, D.O. Hodge, M.R. Hopfenspirger and R.J. Gibbons, Hemodynamic responses and adverse effects associated with adenosine and dipyridamole pharmacologic stress testing: a comparison in 2,000 patients, Mayo Clin Proc 70 (4) (1995), pp. 331–336. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (61)

[31] T.H. Giang, D. Nanz and R. Coulden, Detection of coronary artery disease by magnetic resonance myocardial perfusion imaging with various contrast medium doses: first European multi-centre experience, Eur Heart J 25 (2004), pp. 1657–1665. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (92)

[32] P. Knaapen, P.G. Camici, K.M. Marques, R. Nijveldt, J.J. Bax and N. Westerhof et al., Coronary microvascular resistance: methods for its quantification in humans, Basic Res Cardiol 104 (2009), pp. 485–498. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (8)

[33] T.F. Christian, A.H. Aletras and A.E. Arai, Estimation of absolute myocardial blood flow during first-pass MR perfusion imaging using a dual-bolus injection technique: comparison to single bolus injection method, J Magn Reson Imaging 27 (6) (2008), pp. 1271–1277. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (14)

[34] L.Y. Hsu, K.L. Rhoads, J.E. Holley, P. Kellman, A.H. Aletras and A.E. Arai, Quantitative myocardial perfusion analysis with a dual-bolus contrast-enhanced first-pass MRI technique in humans, J Magn Reson Imaging 23 (2006), pp. 315–322. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (47)

[35] R.J. Kim, D.P. Shah and R.M. Judd, How we perform delayed enhancement imaging, J Cardiovasc Mag Res 5 (3) (2003), pp. 505–514. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (103)

[36] I. Klem, J.F. Heitner, D.J. Shah, M.H. Sketch, V. Behar and J. Weinsaft et al., Improved detection of coronary artery disease by stress perfusion cardiovascular magnetic resonance with the use of delayed enhancement infarction imaging, J Am Coll Cardiol 47 (2006), pp. 1630–1638. Article | PDF (256 K) | View Record in Scopus | Cited By in Scopus (130)

[37] M.D. Cerqueira, N.J. Weissman, V. Dilsizian, A.K. Jacobs, S. Kaul and W.K. Laskey et al., Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association, Circulation 105 (4) (2002), pp. 539–542. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (622)

[38] N. Wilke, C. Simm, J. Zhang, J. Ellermann, X. Ya and H. Merkle et al., Contrast-enhanced first pass myocardial perfusion imaging: correlation between myocardial blood flow in dogs at rest and during hyperemia, Magn Reson Med 29 (1993), pp. 485–497. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (231)

[39] D.L. Kraitchman, N. Wilke, E. Hexeberg, M. Jerosch-Herold, Y. Wang and T.B. Parrish et al., Myocardial perfusion and function in dogs with moderate coronary stenosis, Magn Reson Med 35 (1996), pp. 771–780. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (61)

[40] N. Al-Saadi, E. Nagel, M. Gross, A. Bornstedt, B. Schnackenburg and C. Klein et al., Noninvasive detection of myocardial ischemia from perfusion reserve based on cardiovascular magnetic resonance, Circulation 101 (2000), pp. 1379–1383. View Record in Scopus | Cited By in Scopus (317)

[41] F.J. Klocke, O.P. Simonetti, R.M. Judd, R.J. Kim, K.R. Harris and S. Hedjbeli et al., Limits of detection of regional differences in vasodilated flow in viable myocardium by first-pass magnetic resonance perfusion imaging, Circulation 104 (2001), pp. 2412–2416. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (90)

[42] E. Nagel, C. Klein, I. Paetsch, S. Hettwer, B. Schnackenburg and K. Wegscheider et al., Magnetic resonance perfusion measurements for the noninvasive detection of coronary artery disease, Circulation 108 (4) (2003), pp. 432–437. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (256)

[43] M. Jerosch-Herold, C. Swingen and R.T. Seethamraju, Myocardial blood flow quantification with MRI by model-independent deconvolution, Med Phys 29 (5) (2002), pp. 886–897. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (93)

[44] T. Fritz-Hansen, J.D. Hove, K.F. Kofoed, H. Kelbaek and H.B.W. Larsson, Quantification of MRI measured myocardial perfusion reserve in healthy humans: a comparison with positron emission tomography, J Magn Reson Imaging 27 (2008), pp. 818–824. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (14)

[45] J. Selvanayagam, A.S. Cheng, M. Jerosch-Herold, K. Rahimi, I. Porto and W. van Gaal et al., Effect of distal embolization on myocardial perfusion reserve after percutaneous coronary intervention: a quantitative magnetic resonance perfusion study, Circulation 116 (13) (2007), pp. 1458–1464. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (34)

[46] J.B. Selvanayagam, M. Jerosch-Herold, I. Porto, D. Sheridan, A.S. Cheng and S.E. Petersen et al., Resting myocardial blood flow is impaired in hibernating myocardium: a magnetic resonance study of quantitative perfusion assessment, Circulation 112 (2005), pp. 3289–3296. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (63)

[47] K.R. Nandalur, B.A. Dwamena and A.F. Choudhri, Diagnostic performance of stress cardiac magnetic resonance imaging in the detection of coronary artery disease: a meta-analysis, J Am Coll Cardiol 50 (2007), pp. 1343–1353. Article | PDF (552 K) | View Record in Scopus | Cited By in Scopus (108)

[48] J. Schwitter, C.M. Wacker, A.C. van Rossum, M. Lombardi, N. Al-Saadi and H. Ahlstrom et al., MR IMPACT: comparison of perfusion-cardiac magnetic resonance with single-photon emission computed tomography for the detection of coronary artery disease in a multicentre, multivendor, randomized trial, Eur Heart J 29 (2008), pp. 480–489. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (131)

[49] I. Klem, S. Greulich, J.F. Heitner, H. Kim, H. Vogelsberg and E. Kispert et al., Value of cardiovascular magnetic resonance stress perfusion testing for the detection of coronary artery disease in women, J Am Coll Cardiol Img 1 (2008), pp. 436–445. Article | PDF (421 K) | View Record in Scopus | Cited By in Scopus (17)

[50] M.A. Costa, S. Shoemaker, H. Futamatsu, C. Klassen, D.J. Angiolillo and M. Nguyen, Quantitative magnetic resonance perfusion imaging detects anatomical and physiological coronary artery disease as measured by coronary angiography and fractional flow reserve, J Am Coll Cardiol 50 (2007), pp. 514–522. Article | PDF (3017 K) | View Record in Scopus | Cited By in Scopus (33)

[51] J. Rieber, A. Huber, I. Erhard, S. Mueller, M. Schweyer and A. Koenig et al., Cardiac magnetic resonance perfusion imaging for the functional assessment of coronary artery disease: a comparison with coronary angiography and fractional flow reserve, Eur Heart J 27 (2006), pp. 1465–1471. View Record in Scopus | Cited By in Scopus (46)

[52] S. Watkins, R. McGeoch, J. Lyne, T. Steedman, R. Good and M. McLaughlin et al., Validation of magnetic resonance myocardial perfusion imaging with fractional flow reserve for the detection of significant coronary heart disease, Circulation 120 (2009), pp. 2207–2213. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (17)

[53] C. Jahnke, E. Nagel and R. Gebker, Prognostic value of cardiac magnetic resonance stress tests: adenosine stress perfusion and dobutamine stress wall motion imaging, Circulation 115 (2007), pp. 1769–1776. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (134)

[54] S. Iskander and A.E. Iskandrian, Risk assessment using single-photon emission computed tomographic technetium-99m sestamibi imaging, J Am Coll Cardiol 32 (1998), pp. 57–62. Article | PDF (72 K) | View Record in Scopus | Cited By in Scopus (219)

[55] C. Doesch, A. Seeger and J. Doering, Risk stratification by adenosine stress cardiac magnetic resonance in patients with coronary artery stenoses of intermediate angiographic severity, J Am Coll Cardiol 2 (2009), p. 424. Article | PDF (573 K) | View Record in Scopus | Cited By in Scopus (11)

[56] V Bodi, J. Sanchis, M.P. Lopez-Lereu, J. Nunez, L. Mainar and J.V. Monmeneu et al., Prognostic value of dipyridamole stress cardiovascular magnetic resonance imaging in patients with known or suspected coronary artery disease, J Am Coll Cardiol 50 (12) (2007), pp. 1174–1179. Article | PDF (341 K) | View Record in Scopus | Cited By in Scopus (39)

[57] W.P. Ingkanisorn, R.Y. Kwong and N.S. Bohme, Prognosis of negative adenosine stress magnetic resonance in patients presenting to an emergency department with chest pain, J Am Coll Cardiol 47 (2006), pp. 1427–1432. Article | PDF (228 K) | View Record in Scopus | Cited By in Scopus (104)

[58] K. Steel, R. Broderick, V. Gandla, E. Larose, F. Resnic and M. Jerosch-Herold et al., Complementary prognostic values of stress myocardial perfusion and LGE imaging by cardiac magnetic resonance in patients with known or suspected coronary artery disease, Circulation 120 (2009), pp. 1390–1400. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (25)

[59] J.R. Panting, P.D. Gatehouse, G.Z. Yang, F. Grothues, D.N. Firmin and P. Collins et al., Abnormal subendocardial perfusion in cardiac syndrome X detected by cardiovascular magnetic resonance imaging, N Engl J Med 346 (25) (2002), pp. 1948–1953. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (324)

[60] I.A. Vermeltfoort, O. Bondarenko, P.G. Raijmakers, D.A. Odekerken, A.F. Kuijper and A. Zwijnenburg et al., Is subendocardial ischaemia present in patients with chest pain and normal coronary angiograms? A cardiovascular MR study, Eur Heart J 28 (13) (2007), pp. 1554–1558. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (21)

[61] S.E. Petersen, M. Jerosch-Herold, L.E. Hudsmith, M.D. Robson, J.M. Francis and H.A. Doll et al., Evidence for microvascular dysfunction in hypertrophic cardiomyopathy: new insights from multiparametric magnetic resonance imaging, Circulation 115 (18) (2007), pp. 2418–2425. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (66)

[62] M. Jerosch-Herold, D.C. Sheridan, J.D. Kushner, D. Nauman, D. Burgess and D. Dutton et al., Cardiac magnetic resonance imaging of myocardial contrast uptake and blood flow in patients affected with idiopathic or familial dilated cardiomyopathy, Am J Physiol Heart Circ Physiol 295 (2008), pp. 1234–1242.

[63] S.C. Cook, A.K. Ferketich and S.V. Raman, Myocardial ischemia in asymptomatic adults with repaired aortic coarctation, Int J Cardiol 133 (2009), pp. 95–101. Article | PDF (455 K) | View Record in Scopus | Cited By in Scopus (12)

[64] A. Prakash, A. Powell, R. Krishnamurthy and T. Geva, Magnetic resonance imaging eval‎uation of myocardial perfusion and viability in congenital and acquired pediatric heart disease, Am J Cardiol 93 (2004), pp. 657–661. Article | PDF (218 K) | View Record in Scopus | Cited By in Scopus (31)

[65] S.R. Prasad and J. Jagirdar, Nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy: a primer for radiologists, J Comput Assist Tomogr 32 (2008), pp. 1–3. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (13)

[66] J. Perez-Rodriguez, S. Lai, B.D. Ehst, D.M. Fine and D.A. Bluemke, Nephrogenic systemic fibrosis: incidence, associations, and effect of risk factor assessment—report of 33 cases, Radiology 250 (2) (2009), pp. 371–377. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (41)

[67] T.C. Lauenstein, K. Salman, R. Morreira, S. Tata, D. Tudorascu and G. Baramidze et al., Nephrogenic systemic fibrosis: center case review, J Magn Reson Imaging 26 (2007), pp. 1198–1203. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (51)

[68] E. Altun, D.R. Martin, R. Wertman, A. Lugo-Somolinos, E.R. Fuller and R.C. Semelka, Nephrogenic systemic fibrosis: change in incidence following a switch in gadolinium agents and adoption of a gadolinium policy—report from two U.S. universities, Radiology 253 (3) (2009), pp. 689–696. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (24)

Corresponding author at: Cardiovascular Division, North Shore Hospital, Private Bag 93-503, Shakespeare Rd, Auckland, New Zealand. Tel.: +64 9 4861491.

Copyright © 2010 Australasian Society of Cardiac and Thoracic Surgeons and the Cardiac Society of Australia and New Zealand Published by Elsevier Ltd. on behalf of The Australasian Society of Cardiac and Thoracic Surgeons and The Cardiac Society of Australia and New Zealand

---

Heart, Lung and Circulation
Volume 19, Issue 12, December 2010, Pages 697-705