Search

Cardiovascular magnetic resonance (CMR) derived ventricular volumes and function guide clinical decision-making for various cardiac pathologies. We aimed to evaluate the efficiency and clinical applicability of a commercially available artificial intelligence (AI) method for performing biventricular volumetric analysis. Three-hundred CMR studies (100 with normal CMR findings, 50 dilated cardiomyopathy, 50 hypertrophic cardiomyopathy, 50 ischaemic heart disease and 50 congenital or valvular heart disease) were randomly selected from database. Manual biventricular volumetric analysis (CMRtools) results were derived from clinical reports and automated volumetric analyses were performed using short axis volumetry AI function of CircleCVI42v5.12 software. For 20 studies, a combined method of manually adjusted AI contours was tested and all three methods were timed. Clinicians` confidence in AI method was assessed using an online survey. Although agreement was better for left ventricle than right ventricle, AI analysis results were comparable to manual method. Manual adjustment of AI contours further improved agreement: within subject coefficient of variation decreased from 5.0% to 4.5% for left ventricular ejection fraction (EF) and from 9.9% to 7.1% for right ventricular EF. Twenty manual analyses were performed in 250 min 12 s whereas same task took 5 min 48 s using AI method. Clinicians were open to adopt AI but concerns about accuracy and validity were raised. The AI method provides clinically valid outcomes and saves significant time. To address concerns raised by survey participants and overcome shortcomings of the automated myocardial segmentation, visual assessment of contours and performing manual corrections where necessary appears to be a practical approach.

AbstractPurposeCardiovascular magnetic resonance assessment of adults late after an atrial redirection operation for transposition is demanding and time consuming. We hypothesised that the relatively fast and standardised 3-dimensional time-resolved contrast-enhanced magnetic resonance angiography, or dynamic angiography, would be valuable in the periodic follow-up of these patients.MethodsWe investigated prospectively 36 adults with transposition using dynamic angiography, comparing our results against a comprehensive but non-contrast cardiovascular magnetic resonance protocol. We acquired 6 dynamic angiographic datasets after injection of contrast. The primary aim was to detect significant obstruction of the pathways for venous flow.ResultsIn 4 patients (11%), we found evidence of moderate-to-severe, and thus clinically important, obstruction of systemic venous channels on standard cardiovascular magnetic resonance. All these patients were correctly identified by dynamic angiography. In 4 additional patients, we found mild and haemodynamically insignificant obstructions in the systemic venous channels. Of the 8 (22%) patients with any obstruction, 6 were detected by angiography. There were no false positives reported, giving sensitivity of 75% and specificity of 100%, a positive predictive value of 100%, and negative predictive value of 93%. In 1 patient, there was a moderate obstruction of the pulmonary venous compartment which was not readily seen by dynamic angiography.Conclusions3-dimensional dynamic angiography is a useful method for detecting anatomically moderate-to-severe, but not mild, obstructions in the systemic venous channels following Mustard repair for transposition. This technique can be used as a single imaging method and/or as complimentary to standard two dimensional cardiovascular magnetic resonance techniques for detection of clinically important obstructions in the systemic venous channels.

Metabolically active mediastinal brown adipose tissue may be mistakenly diagnosed as a malignancy on 18F-fluoro-2-deoxy-D-glucose (FDG)/positron emission tomography (PET). We report the case of a patient with locally recurrent breast carcinoma in which staging PET/CT revealed a suspicious pericardial lesion for which the patient was referred to our centre. The novelty of this case resides in the fact that by tissue characterization, cardiac magnetic resonance imaging allowed the determination that the lesion corresponded to brown fat, a reassuring finding with important impact on management, because the presence of pericardial metastasis would have disqualified this patient for curative resection of her cancer recurrence.

We report a case of primary cardiac lymphoma presenting as myopericarditis and rapidly deteriorating into biventricular heart failure and ventricular arrhythmias. Computed tomography and cardiac magnetic resonance (CMR) imaging showed extensive myocardial infiltration with typical patterns on tissue characterization CMR images, raising clinical suspicion. Diagnosis was confirmed by myocardial histologic examination. Marked regression of tumor burden was apparent after 6 cycles of anthracycline-based chemotherapy. This case illustrates that a high degree of suspicion for this rare entity is mandated to institute timely treatment. Rapid tumor lysis may induce life-threatening acute cardiac decompensation that requires intensive monitoring and support therapy.

Magnetic resonance imaging has been used to study pulmonary arterial anatomy and to measure pulmonary arterial blood flow in patients with single lung transplantation. Nine patients and nine control subjects matched for age and sex were studied. The anatomy of the main pulmonary artery and its main branches, as well as the site of arterial anastomoses, was identified and measurements of their diameters were taken. There were no significant differences in diameters of these vessels between the patients and the control subjects. Arterial blood flows to the transplanted lung were 2.07 ± 0.45 L/min/m2in the group with right lung transplantation and 2.43 ± 0.60 L/min/m2in those with left lung transplantation compared with 1.22 ± 0.22 L/min/m2and 1.27 ± 0.41 L/min/m2in the control group, respectively. The ratio of blood flow in the transplanted and the native lungs in all patients studied was 2.8 ± 0.83:1. The flow profile in the artery of the transplanted lung showed a wide forward flow during systole and most of diastole, whereas that of the native lung showed a narrow early systolic peak and a reverse flow in most of diastole. These differences in the volume and pattern of flow in the transplanted lung are most likely related to the relative resistance in the native and the transplanted lung and could constitute an additional index for monitoring the condition of patients with lung transplantation and help in the understanding of the physiology of the denervated pulmonary vascular bed.

Summary We used magnetic resonance imaging (MRI) velocity mapping to assess the velocity profile of early diastolic mitral inflow in 11 normal subjects. Velocity maps of left ventricular inflow were obtained in the horizontal long axis of the left ventricle at the time of peak early diastolic filling. Velocity profile curves across the mitral inflow were obtained at 1-cm intervals from the mitral ring to 4 cm into the cavity. The jet width was 3.06 ± 0.64cm at the mitral ring level, increasing to 3.6 ± 0.61 cm at 4cm. The peak/mean velocity was 1.2 ± 0.07 at the mitral ring and increased to around 1.4 at 3–4cm from the mitral ring. The point at which the peak velocity was recorded at each level was skewed towards the septal side by 10%–13% of jet width from the center at the mitral ring and 2–4cm from the ring. However, at a depth of 1 cm, corresponding to the mitral tip level, the peak velocity was at the center of the jet. The ratio of vertical and horizontal dimensions of the jet cross section was 1.11 ± 0.05. Thus, the mitral inflow velocity profile is relatively flat at the mitral ring and tip level; the inflow jet cross section is effectively circular.

We describe a noninvasive method for visualization of flow and demonstrate its application in a flow phantom and in the great vessels of healthy volunteers and patients with aortic and pulmonary arterial disease. The technique uses multidirectional MR velocity mapping acquired in selected planes. Maps of orthogonal velocity components were then processed into a graphic form immediately recognizable as flow.

The objective was to investigate how the blood flows in the thoracic aorta, with special emphasis on flow reversal and flow into the coronary arteries. Three‐directional MR velocity mapping was used to map multidirectional flow velocities in the aorta in 14 normal subjects and 14 patients with coronary artery disease. Dynamic flow vector maps and through‐plane velocity maps were used. The flow reversed in all subjects in the upper ascending aorta and usually also in the distal aortic arch. Retrograde flow became antegrade again at various levels in the ascending aorta and in the coronary sinuses. Seven flow characteristics were investigated that, lumped together, were significantly different (P= .0005) in normal subjects compared with patients and in normal subjects 70 years of age and older compared with those younger than 70 years of age.

The authors describe a technique for automatic enhancement, segmentation, and analysis of flow from MR phase shift velocity mapping. This method uses a specific flow related enhancement process to extract and localize the flow field within the image, followed by global flow analysis. By animation on low cost equipment (personal computers), this method can provide various ways of visualizing flow in vivo and its variation in different structures and lesions. Our preliminary results using this method for analyzing flow data in the aortic bifurcation of healthy volunteers are encouraging.

Mitral and pulmonary venous flows are important indexes in the evaluation of left ventricular diastolic function and in the assessment of mitral valve disease. We used MR phase-shift velocity mapping to measure mitral and pulmonary venous flow velocity in 10 healthy volunteers and mitral flow velocity in 5 patients with mitral valve stenosis. Normal mitral flow shows two positive peaks: one during early ventricular diastole and the other during atrial contraction. Peak mitral flow velocity (mean ± SD) in early diastole was 68 ± 12 cm/s and during atrial contraction 39 ± 10 cm/s. The ratio of peak mitral flow velocity in early diastole to that during atrial contraction was 1.9 ± 0.6. In patients with mitral valve stenosis, the initial high flow velocity persisted through diastole. Peak mitral flow velocity of patients with mitral valve stenosis correlated well with values obtained from Doppler echocardiography. Pulmonary venous flow showed two positive peaks: one during ventricular systole and the other in ventricular diastole. A small backflow during atrial contraction was noticed. Peak systolic velocity in the right lower pulmonary vein was 47 ± 11 cm/s, peak diastolic velocity was 40 ± 9 cm/s, and peak backflow velocity was 14 ± 3 cm/s. Magnetic resonance velocity mapping is a noninvasive technique for the evaluation of time-related flow velocity patterns and for quantitative measurement of mitral and pulmonary venous blood flow velocity.

A retrospective analysis of lung signal intensity normalized for fat and lung cross-sectional area normalized for body surface area was obtained from cardiac gated spin-echo (echo time 40 ms) images before and after single-lung transplantation in 12 patients with pulmonary airway disease (seven emphysema, two bronchiectasis and three obliterative bronchiolitis) and 14 with interstitial lung disease (three pulmonary fibrosis, four fibrosing alveolitis and seven lymphangioleiomyomatosis). Nine healthy volunteers were studied for comparison. The native lung in pulmonary airway disease without inflammatory processes has normalised signal intensity (79 ± 12%) similar to that of the control (100 ± 6%) while interstitial lung disease had a higher normalised signal intensity (172 ± 33%; P< 0.05). The transplanted lung had a similar normalised signal intensity (103 ± 12%) to that of the control except when there was a rejection reaction (one patient) or infection (two patients). Compared with the control, the native lung was smaller in pulmonary fibrosis (50 ± 8%) and fibrosing alveolitis (78 ± 8%), while larger in lymphangioleiomyomatosis (136 ± 6%) and pulmonary airway disease (165 ± 8%). The cross-sectional area of the transplanted lung was comparable to that of the control (100 ± 13%). Distinctive features of lung normalised signal intensity and cross-sectional area were demonstrated in patients with pulmonary airway disease and interstitial lung disease before lung transplantation. These measurements could be useful noninvasive indices for assessment of the transplanted lung and for frequent follow-up of patients without exposure to ionizing radiation. Future developments are required to enhance lung signal sufficiently to detect less severe diseases.

Objectives. Nuclear magnetic resonance (MRI) velocity mapping was used to characterize flow waveforms and to measure volume flow in the ascending and descending thoracic aorta in patients with aortic coarctation and in healthy volunteers. We present the method and discuss the relation between these measurements and aortic narrowing assessed by MRI. Finally, we compare coarctation jet velocity measured by MRI velocity mapping with that obtained from continuous wave Doppler echocardiography.

Using an adapted two‐dimensional spatially selective RF excitation scheme, a novel yet practical three‐dimensional (3D) zonal echo‐planar imaging technique for MR coronary angiography has been developed. The robustness of the technique compared with the two‐dimensional (2D) segmented fast low angle shot (FLASH) method was evaluated using the right coronary artery images of 16 asymptomatic volunteers with a 0.5‐T mobile scanner. Each 3D acquisition required multiple breath‐holds, and real‐time navigator echoes were used to ensure consistent breath‐holding. Advantages of the technique include an improved signal‐to‐noise ratio, clearer depiction of tortuous coronary vessels due to decreased partial volume effects, and reduced motion blurring by the use of a short echo‐planar readout.

Noninvasive measurement of blood flow velocity through the cardiac valves has important clinical applications. A wide variety of MR methods are available for flow measurement. The aim of this study was to investigate the ability of cine MR Fourier velocimetry to measure flow through healthy cardiac valves and to compare MR and Doppler peak velocity measurements. Ten healthy volunteers (age mean ± SD, 24 ± 4 years) without history of valvular disease were studied. Four of the subjects were females. In each subject, aortic, pulmonary, mitral, and tricuspid valves were evaluated with MR and Doppler imaging. A whole‐body mobile MR machine was used, operating at .5‐T with actively shielded magnetic field gradient coils on all three axes capable of 20 mT/m at a slew rate of 60 mT/m/msec. The heart rate during MR and Doppler studies was not significantly different. The mean difference between the two studies was 2 beats/min, with a 95% confidence interval of −22 beats/min, +25 beats/min. Peak systolic flow velocity in the aortic and pulmonary valves and peak diastolic flow velocity in the mitral and tricuspid valves measured with MRI and Doppler echocardiography correlated well. The mean difference between the two measurements (MR — Doppler) was 63 mm/sec, with a 95% confidence interval of −180 mm/sec, +310 mm/sec. The agreement between two observers interpreting the same MR velocity maps was close. The mean difference between their two measurements was 23 mm/sec, with a 95% confidence interval of −20 mm/sec, +60 mm/sec. There was no significant difference between MR and Doppler imaging or between the two MR observers. MR Fourier velocimetry has the necessary ease, reliability, and speed to measure blood flow through the cardiac valves, although measurement of late diastolic flow in the atrioventricular valves is limited. Measurement of peak blood velocity through the cardiac valves by this method showed satisfactory agreement with Doppler, but its clinical application for assessing diseased cardiac valves must be established.

We conclude that first-pass radionuclide angio-cardiography has limited value when measuring Qp/Qs in our group of patients, and that magnetic resonance imaging with velocity mapping is an excellent alternative technique that should be considered, particularly in patients with complex congenital disease and right-to-left shunt, or when poor bolus quality is anticipated.

This investigation studied the relation between regional myocardial blood flow and left ventricular function during dobutamine stress in patients with coronary artery disease.

Changes in the dimensions of the thoracic aorta with age were studied in 70 healthy volunteers between the ages of 10 and 83 years. Spin echo images were acquired at end-diastole in three oblique planes through the ascending aorta, aortic arch, and descending aorta. A double oblique image through the whole of the thoracic aorta was also acquired. Measurements of aortic cross-sectional area and length were corrected for body surface area, and normal magnetic resonance standards were established. The direct correlation that aortic dimensions have with age is likely to be due to loss of elasticity. The ratio between areas of the ascending aorta and the aortic arch is directly related to age whereas the ratio between the aortic arch and the descending aorta is inversely related to age.

Objectives. This study was designed to investigate the value of noninvasive imaging modalities for the detection of obstruction in extracardiac ventriculopulmonary conduits.

Shortening the echo times of magnetic resonance (MR) sequences used for phase-shift velocity mapping to 3.6 ms has extended use of the technique to measurement of velocities in turbulent, poststenotic jet flows. We used a 0.5-T MR machine and field even-echo rephasing (FEER) sequences with 3.6 ms echo times for jet velocity mapping.In vitro trials used continuous flow through a phantom with a 6-mm stenosis. Fifteen patients with mitral and/or aortic valve stenosis and 20 patients with repaired aortic coarctation were studied prospectively, with Doppler ultrasonic measurement of peak jet velocity performed independently on the same day. The clinical contribution of MR jet velocity mapping, used during a 3-year period in 306 patients with congenital and acquired disease of heart valves, great vessels, and conduits, was assessed retrospectively. The 3.6-ms sequence allowed accurate measurement of jet velocities up to 6 m s-1in vitro (r=0.996). Prospective studies in patients showed good agreement between MR and Doppler measurements of peak velocity:n=38; range, 1.2–6.1 m s-1; mean, 2.7 m s-1; mean of differences (Doppler-MR), 0.22 ms-1; standard deviation of differences, ±0.38 m s-1 (±14%). MR jet velocity mapping proved particularly valuable for assessment and localization of stenoses at sites where ultrasonic access was limited. The technique represents a diagnostic advance which can obviate the need for catheterization in selected cases.

A case of suspected femoral arteriovenous fistula after cardiac catheterization remained undiagnosed because of inability to obtain access for arteriography. Magnetic resonance imaging with phase-shift velocity mapping allowed a noninvasive diagnosis and qualitative evaluation of the lesion.

Aims Magnetic resonance (MR) conditional pacemakers are increasingly implanted into patients who may need cardiovascular MR (CMR) subsequent to device implantation. We assessed the added value of CMR for diagnosis and management in this population. Methods and results CMR and pacing data from consecutive patients with MR conditional pacemakers were retrospectively reviewed. Images were acquired at 1.5 T (Siemens Magnetom Avanto). The indication for CMR and any resulting change in management was recorded. The quality of CMR was rated by an observer blinded to clinical details, and data on pacemaker and lead parameters were collected pre- and post-CMR. Seventy-two CMR scans on 69 patients performed between 2011 and 2015 were assessed. All scans were completed successfully with no significant change in lead thresholds or pacing parameters. Steady-state free precession (SSFP) cine imaging resulted in a greater frequency of non-diagnostic imaging (22 vs. 1%, P < 0.01) compared with gradient echo sequences (GRE). Right-sided pacemakers were associated with less artefact than left-sided pacemakers. Late gadolinium enhancement imaging was performed in 59 scans with only 2% of segments rated of non-diagnostic quality. The CMR data resulted in a new diagnosis in 27 (38%) of examinations; clinical management was changed in a further 18 (25%). Conclusions CMR in patients with MR conditional pacemakers provided diagnostic or management-changing information in the majority (63%) of our cohort. The use of gradient echo cine sequences can reduce rates of non-diagnostic imaging. Right-sided device implantation may be considered in patients likely to require CMR examination.

Supplemental Digital Content is available in the text.