Showing papers by "Matthias Stuber published in 2016"

Reduction of cortical oxygenation in chronic kidney disease: evidence obtained with a new analysis method of blood oxygenation level-dependent magnetic resonance imaging.

Abstract: Background Determinations of renal oxygenation by blood oxygenation level-dependent magnetic resonance imaging (BOLD-MRI) in chronic kidney disease (CKD) patients have given heterogeneous results, possibly due to the lack of a reproducible method to analyse BOLD-MRI. It therefore remains uncertain whether patients with CKD have a reduced renal tissue oxygenation. We developed a new method to analyse BOLD-MRI signals and applied it to CKD patients and controls. Methods MRI was performed under standardized conditions before and 15 min after IV furosemide in 104 CKD patients, 61 hypertensives and 42 controls. MR images were analysed with the new twelve-layer concentric objects method (TLCO) that divides renal parenchyma in 12 layers of equal thickness. The mean R2* value of each layer was reported, along with the change in R2* between successive layers, as measured by the slope steepness of the relevant curve. Results Inter-observer variability was 2.3 ± 0.9%, 1.9 ± 0.8% and 3.0 ± 2.3% in, respectively, controls, moderate and severe CKD. The mean R2* of the outer (more cortical) layers was significantly higher in CKD, suggesting lower cortical oxygenation as compared with controls. In CKD patients, the response to furosemide was blunted in the inner (more medullary) layers, and the R2* slope was flatter. In multivariable regression analysis, the R2* slope correlated positively with estimated glomerular filtration rate (eGFR) in patients with an eGFR <90 mL/min/1.73 m2 (P < 0.001). Conclusions Using the new TLCO method, we confirm the hypothesis that renal cortical oxygenation is reduced in CKD in humans, and that the level of cortical oxygenation correlates with CKD severity.

An iterative approach to respiratory self-navigated whole-heart coronary MRA significantly improves image quality in a preliminary patient study.

Abstract: Purpose In respiratory self-navigated coronary MRA, the selection of a reference position may have a direct effect on image quality. While end-expiration is commonly used as reference, it may be ill defined in cases of irregular breathing. Here, an iterative self-navigation approach that operates without a reference position was implemented and tested in healthy volunteers and patients. Methods Data were acquired in 15 healthy volunteers and in 23 patients. Images obtained with end-expiratory self-navigation were compared with those obtained with the iterative approach that incorporates cross-correlation to iteratively minimize a global measure of respiratory displacement. Vessel sharpness, length, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were evaluated while differences in breathing patterns between the two sub-groups were assessed, too. Results Vessel sharpness and length were similar for both methods in healthy volunteers. In patients, a significant improvement in vessel sharpness and length was obtained using the iterative approach. SNR and CNR remained constant. While end-expiration was the most frequent respiratory phase in healthy volunteers (57.6 ± 16.2%), intermediate respiratory phases (43.4 ± 30.1%) were predominantly found in patients. Conclusion An iterative approach to respiratory motion correction in self-navigation may lead to significant improvements in coronary MRA image quality in patients with a less consistent end-expiratory respiratory phase. Magn Reson Med, 2015. © 2015 Wiley Periodicals, Inc.

Simultaneous Noninvasive Assessment of Systemic and Coronary Endothelial Function.

Abstract: Background— Normal endothelial function is a measure of vascular health and dysfunction is a predictor of coronary events. Nitric oxide-mediated coronary artery endothelial function, as assessed by vasomotor reactivity during isometric handgrip exercise (IHE), was recently quantified noninvasively with magnetic resonance imaging (MRI). Because the internal mammary artery (IMA) is often visualized during coronary MRI, we propose the strategy of simultaneously assessing systemic and coronary endothelial function noninvasively by MRI during IHE. Methods and Results— Changes in cross-sectional area and blood flow in the right coronary artery and the IMA in 25 patients with coronary artery disease and 26 healthy subjects during IHE were assessed using 3T MRI. In 8 healthy subjects, a nitric oxide synthase inhibitor was infused to evaluate the role of nitric oxide in the IMA-IHE response. Interobserver IMA-IHE reproducibility was good for cross-sectional area ( R =0.91) and blood flow ( R =0.91). In healthy subjects, cross-sectional area and blood flow of the IMA increased during IHE, and these responses were significantly attenuated by monomethyl-l-arginine ( P <0.01 versus placebo). In patients with coronary artery disease, the right coronary artery did not dilate with IHE, and dilation of the IMA was less than that of the healthy subjects ( P =0.01). The blood flow responses of both the right coronary artery and IMA to IHE were also significantly reduced in patients with coronary artery disease. Conclusions— MRI-detected IMA responses to IHE primarily reflect nitric oxide-dependent endothelial function and are reproducible and reduced in patients with coronary artery disease. Endothelial function in both coronary and systemic (IMA) arteries can now be measured noninvasively with the same imaging technique and promises novel insights into systemic and local factors affecting vascular health.

Improved border sharpness of post-infarct scar by a novel self-navigated free-breathing high-resolution 3D whole-heart inversion recovery magnetic resonance approach

Abstract: The border zone of post-infarction myocardial scar as identified by late gadolinium enhancement (LGE) has been identified as a substrate for arrhythmias and consequently, high-resolution 3D scar information is potentially useful for planning of electrophysiological interventions. This study evaluates the performance of a novel high-resolution 3D self-navigated free-breathing inversion recovery magnetic resonance pulse sequence (3D-SN-LGE) vs. conventional 2D breath-hold LGE (2D-LGE) with regard to sharpness of borders (SBorder) of post-infarction scar. Patients with post-infarction scar underwent two magnetic resonance examinations for conventional 2D-LGE and high-resolution 3D-SN-LGE acquisitions (both 15 min after 0.2 mmol/kg Gadobutrol IV) at 1.5T. In the prototype 3D-SN-LGE sequence, each ECG-triggered radial steady-state-free-precession read-out segment is preceded by a non-slice-selective inversion pulse. Scar volume and SBorder were assessed on 2D-LGE and matching reconstructed high-resolution 3D-SN-LGE short-axis slices. In 16 patients (four females, 58 ± 10y) all scars visualized by 2D-LGE could be identified on 3D-SN-LGE (time between 2D-LGE and 3D-SN-LGE 48 ± 53 days). A good agreement of scar volume by 3D-SN-LGE vs. 2D-LGE was found (Bland–Altman: −3.7 ± 3.4 ml, correlation: r = 0.987, p < 0.001) with a small difference in scar volume (20.5 (15.8, 35.2) ml vs. 24.5 (20.0, 41.9)) ml, respectively, p = 0.002] and a good intra- and interobserver variability (1.1 ± 4.1 and −1.1 ± 11.9 ml, respectively). SBorder of border “scar to non-infarcted myocardium” was superior on 3D-SN-LGE vs. 2D-LGE: 0.180 ± 0.044 vs. 0.083 ± 0.038, p < 0.001. Detection and quantification of myocardial scar by 3D-SN-LGE is feasible and accurate in comparison to 2D-LGE. The high spatial resolution of the 3D sequence improves delineation of scar borders.

Coronary endothelial function assessment using self-gated cardiac cine MRI and k-t sparse SENSE

Abstract: Electrocardiogram (ECG)-gated cine MRI, paired with isometric handgrip exercise, can be used to accurately, reproducibly, and noninvasively measure coronary endothelial function (CEF). Obtaining a reliable ECG signal at higher field strengths, however, can be challenging due to rapid gradient switching and an increased heart rate under stress. To address these limitations, we present a self-gated cardiac cine MRI framework for CEF measurements that operates without ECG signal. Cross-sectional slices of the right coronary artery (RCA) were acquired using a two-dimensional golden angle radial trajectory. This sampling approach, combined with the k-t sparse SENSE algorithm, allows for the reconstruction of both real-time images for self-gating signal calculations and retrospectively reordered self-gated cine images. CEF measurements were quantitatively compared using both the self-gated and the standard ECG-gated approach. Self-gated cine images with high-quality, temporal, and spatial resolution were reconstructed for 18 healthy volunteers. CEF as measured in self-gated images was in good agreement (R2 = 0.60) with that measured by its standard ECG-gated counterpart. High spatial and temporal resolution cross-sectional cine images of the RCA can be obtained without ECG signal. The coronary vasomotor response to handgrip exercise compares favorably with that obtained with the standard ECG-gated method. Magn Reson Med 76:1443-1454, 2015. © 2015 International Society for Magnetic Resonance in Medicine.

Is there an optimal respiratory reference position for self‐navigated whole‐heart coronary MR angiography?

Abstract: Purpose To test the direct influence of the reference respiratory position on image quality for self-navigated whole-heart coronary MRI. Methods Self-navigated whole-heart coronary MRI was performed in 11 healthy adult subjects. Respiratory motion was compensated for by using three different respiratory reference positions of the heart: end-inspiratory, end-expiratory, and the mean of the entire respiratory excursion. All datasets were reconstructed without motion compensation for comparison. Image quality was assessed in all reconstructions using signal-to-noise ratio (SNR) and contrst-to-noise ratio (CNR) measurements, as well as percentage vessel sharpness and visible length of the coronary arteries. Results While SNR and CNR remained close to constant in all reconstructions, a clear and significant improvement in vessel sharpness was identified in all motion corrected datasets with respect to their uncorrected counterpart (e.g., percentage sharpness of the proximal right coronary artery (RCA): 61.6 ± 8.2% for end-inspiration, 64.1 ± 10.7% for end-expiration, and 63.3 ± 7.0% for the mean respiratory position versus 55.0 ± 10.4 for the uncorrected datasets; P < 0.05). Among all motion corrected reconstructions, the use of an end-expiratory reference position most consistently provided the highest image quality. In particular, some of the improvements in vessel sharpness and length measured for end-expiration were statistically significant with respect to the reconstructions performed at end-inspiration (e.g., percentage sharpness of the proximal left anterior descending coronary: 58.2 ± 7.4% versus 55.8 ± 8.4%; P < 0.05; and visible length of the RCA: 125.7 ± 25.9 mm versus 114.4 ± 27.4 mm; P < 0.05). Conclusion The use of end-expiration as a reference position for respiratory motion correction in free-breathing self-navigated whole heart coronary MRA significantly improves image quality. J. Magn. Reson. Imaging 2015.

Border sharpness of scar tissue after myocardial infarction as determined by self-navigated free-breathing isotropic 3D whole-heart inversion recovery magnetic resonance

Abstract: Background The border zone of myocardial scar after myocardial infarction (MI) plays an important role for arrhythmia formation. For this reason, high-resolution 3D information of scar tissue for planning of electrophysiological interventions after MI is highly desirable. This study evaluates sharpness of the borders (SB) of scar after MI by a self-navigated isotropic 3D free-breathing wholeheart magnetic resonance with inversion recovery (3DSN-IR) in comparison to a standard 2D inversion recovery sequence.

Renal tissue oxygenation in children with chronic kidney disease due to vesicoureteral reflux.

Abstract: Vesicoureteral reflux (VUR) is a frequent cause of chronic kidney disease (CKD) in children. Using blood oxygenation level-dependent magnetic resonance imaging (BOLD-MRI), we measured cortical and medullary oxygenation in children with CKD due to VUR and compared the results to those obtained on healthy controls. The study population comprised 37 children (19 with CKD due to VUR and 18 healthy age-matched controls). BOLD-MRI was performed before and after furosemide treatment. MR images were analyzed with the region-of-interest (ROI) technique to assess the mean R2* values (=1/T2*) of the cortex and medulla of each kidney and with the concentric object (CO) technique that divides renal parenchyma in 12 equal layers. R2* values were significantly lower (corresponding to higher oxygenation) in the cortex and medulla of kidneys of children with CKD due to VUR than in those of the healthy controls (cortex 16.4 ± 1.4 vs. 17.2 ± 1.6 s−1 , respectively; medulla 28.4 ± 3.2 vs. 30.3 ± 1.9 s−1 , respectively; P < 0.05), and furosemide-induced changes in medullary R2* were smaller in the former than in the latter (−5.7 ± 3.0 vs. −6.9 ± 3.4 s−1, respectively; P < 0.05). Similar results were found with the CO technique. In children with a history of unilateral reflux (n = 9), the non-affected contralateral kidneys presented similar R2* values as the diseased kidneys, but their response to furosemide was significantly larger (−7.4 ± 3.2 vs. −5.7 ± 3.0, respectively; P = 0.05). Chronic kidney disease due to VUR is not associated with kidney tissue hypoxia in children. The significantly larger furosemide-induced decrease in medullary R2* levels in the healthy group and unaffected contralateral kidneys of the VUR group points towards more intense renal sodium transport in these kidneys.

Off-resonance magnetic resonance angiography improves visualization of in-stent lumen in peripheral nitinol stents compared to conventional T1-weighted acquisitions: an in vitro comparison study.

Abstract: To compare the value of inversion recovery with on-resonant water suppression (IRON) to conventional T1-weighted (T1w) MRA and computed tomography angiography (CTA) for visualization of peripheral nitinol stents. We visualized 14 different peripheral nitinol stents in vitro both using Gadolinium (Gd) and ultrasmall superparamagnetic iron nanoparticles (USPIOs) for conventional T1w and IRON-MRA using clinical grade 1.5T MR scanner and iodinated contrast material for CTA using a 256-slice CT scanner. Parameter assessment included signal- and contrast-to-noise ratio (S/CNR), relative in-stent signal and artificial lumen narrowing. X-ray angiography served as gold standard for diameter assessment. Gd-enhanced IRON-MRA exhibited highest in-stent SNR and CNR values compared to conventional T1w MRA (IRON (Gd/USPIO): SNR = 30 ± 3/21 ± 2, CNR = 23 ± 2/14 ± 1; T1w: SNR = 16 ± 1/14 ± 2, CNR = 12 ± 1/10 ± 1, all p < 0.05). Furthermore, IRON-MRA achieved highest relative in-stent signal both using Gd and USPIO (IRON (Gd/USPIO): 121 ± 8 %/103 ± 6 %; T1w: 73 ± 2 %/66 ± 4 %; CTA: 84 ± 6 %, all p < 0.05). However, artificial lumen narrowing appeared similar in all imaging protocols (IRON (Gd/USPIO): 21 ± 3 %/21 ± 2 %; T1w: 16 ± 4 %/17 ± 3 %; CTA: 19 ± 2 %, all p = NS). Finally, IRON-MRA provided improvement of the in-stent lumen visualization with an 'open-close-open' design, which revealed a complete in-stent signal loss in T1w MRA. IRON-MRA improves in-stent visualization in vitro compared to conventional T1w MRA and CTA. In light of the in vitro results with Gd-enhanced IRON-MRA, the clinical implementation of such an approach appears promising.

A Cylindrical, Inner Volume Selecting 2D-T2-Prep Improves GRAPPA-Accelerated Image Quality in MRA of the Right Coronary Artery.

Abstract: BACKGROUND Two-dimensional (2D) spatially selective radiofrequency (RF) pulses may be used to excite restricted volumes. By incorporating a "pencil beam" 2D pulse into a T2-Prep, one may create a "2D-T2-Prep" that combines T2-weighting with an intrinsic outer volume suppression. This may particularly benefit parallel imaging techniques, where artefacts typically originate from residual foldover signal. By suppressing foldover signal with a 2D-T2-Prep, image quality may therefore improve. We present numerical simulations, phantom and in vivo validations to address this hypothesis. METHODS A 2D-T2-Prep and a conventional T2-Prep were used with GRAPPA-accelerated MRI (R = 1.6). The techniques were first compared in numerical phantoms, where per pixel maps of SNR (SNRmulti), noise, and g-factor were predicted for idealized sequences. Physical phantoms, with compartments doped to mimic blood, myocardium, fat, and coronary vasculature, were scanned with both T2-Preparation techniques to determine the actual SNRmulti and vessel sharpness. For in vivo experiments, the right coronary artery (RCA) was imaged in 10 healthy adults, using accelerations of R = 1,3, and 6, and vessel sharpness was measured for each. RESULTS In both simulations and phantom experiments, the 2D-T2-Prep improved SNR relative to the conventional T2-Prep, by an amount that depended on both the acceleration factor and the degree of outer volume suppression. For in vivo images of the RCA, vessel sharpness improved most at higher acceleration factors, demonstrating that the 2D-T2-Prep especially benefits accelerated coronary MRA. CONCLUSION Suppressing outer volume signal with a 2D-T2-Prep improves image quality particularly well in GRAPPA-accelerated acquisitions in simulations, phantoms, and volunteers, demonstrating that it should be considered when performing accelerated coronary MRA.

[op.4d.06] reduction of cortical oxygenation in chronic kidney disease: evidence obtained with bold-mri and a new analytic technique.

Abstract: Objective: The determination of renal tissue oxygenation by BOLD-MRI in chronic kidney disease (CKD) patients has given heterogeneous results, possibly due to the lack of a widely accepted procedure to analyze the R2* maps. We present a new technique for the analysis of renal BOLD-MRI called the twelve-layer concentric objects method (TLCO). We measured its reproducibility and assessed whether differences in the intra-renal distribution of R2* as a measure of oxygenation were detected between CKD patients and controls. Figure. No caption available. Design and method: MR imaging (3 Tesla) was performed under standardized conditions before and 15 min after furosemide injection in 104 CKD patients, 61 hypertensive patients and 42 healthy volunteers. MR images were analyzed with the TLCO technique that divides renal parenchyma in 12 equal layers (see figure). Mean R2* value of each layer was reported, as well as the steepness of the slope of the R2* curve linking the mean R2* from the 3rd to the 7th layer. In a subgroup of 52 individuals (20 controls, 18 moderate CKD, 14 severe CKD) each exam was analyzed by two different observers to assess inter-observer variability. Results: Inter-observer variability was respectively 2.3 ± 0.9%, 1.9 ± 0.8% and 3.0 ± 2.3% in controls, mild to moderate and severe CKD. Mean R2* of the outer (more cortical) layers was significantly higher in CKD suggesting a lower cortical oxygenation. The response to furosemide was smaller in the inner (more medullary) layers, and the R2*slope was flatter in CKD patients than in healthy and hypertensive controls. In multivariable regression analysis, the slope of the R2* profile correlated positively with eGFR in patients with an eGFR < 90 ml/min/1.73m2 (p < 0.001). Conclusions: Our data confirm the hypothesis that renal cortical oxygenation is reduced in CKD and show in humans that the level of cortical oxygenation correlates with the severity of chronic kidney disease. The TLCO-technique is highly reproducible and provides layer-by-layer information on renal R2* values from the cortex to the medulla. Analysis of the slope factor offers information on the intra-renal distribution of R2*, and as such represents a measure of cortico-medullary differentiation.

[op.7b.05] effect of acute hyperglycemia on renal tissue oxygenation as measured with bold-mri in overweight individuals

Abstract: Objective: Diabetic nephropathy affects 30% of diabetics, and its development has been linked to poor glycemic control. Animal studies have suggested that hyperglycemia induces transient renal hypoxia and thus kidney damage, yet this has not been previously tested in humans. In this interventional study we assessed whether hyperglycemia alters renal tissue oxygenation as measured with blood oxygenation level-dependent MRI (BOLD-MRI). Figure. No caption available. Design and method: Nineteen healthy overweight volunteers (age 37 ± 10 years, BMI 28.9 ± 3 kg/m2, HbA1c 5.4 ± 0.3%, 57.9% women) were recruited and underwent an oral glucose tolerance test. Two had impaired glucose tolerance, none had diabetes. On a separate day, BOLD-IRM was performed under standard hydration conditions before, and after the intravenous administration of 0.15 g/kg of glucose in a 20% solution. R2* maps were analyzed using the concentric objects technique, a semi automatic procedure which divides the kidney parenchyma in twelve equal layers at increasing depth. R2* is a measure of local desoxyhemoglobin concentrations, with high R2* values corresponding to low oxygenation. Results: The mean glycemia rose from 4.5 ± 0.3 mmol/l to 9.0 ± 0.9, 8.9 ± 0.7, 7.7 ± 0.6 and 6.8 ± 0.8 mmol/l respectively 1, 10, 20 and 30 minutes after IV glucose administration, whereas circulating insulin levels increased, and no change occurred in HbA1c. The corresponding mean R2* values decreased significantly in all kidney layers (see figure), irrespective of glucose intolerance. In all layers, associations between R2* and insulin levels were stronger than between R2* and glycemia. Conclusions: These findings indicate that glycemia influences the R2* signal and should be measured before each BOLD-MRI. Hyperglycemia leads to an increase, not a decrease, of renal tissue oxygenation as measured with BOLD-MRI in healthy, obese volunteers. Whether this glucose-induced increase in oxygenation is due to alterations in renal perfusion, oxygen consumption or both, and whether this also occurs in patients with diabetes needs further study.

Free-breathing 3D whole-heart coronary mra using respiratory motion-resolved sparse reconstruction

Abstract: Background Navigator gating is commonly used to minimize respiratory motion in free-breathing whole-heart coronary MRA [1]. However, lengthy and unpredictable acquisition times remain a drawback. Respiratory self-navigation (SN) [2-3], conversely, enables 100% scan efficiency, but performs motion correction over a broad range of respiratory displacements, which can result in image artifacts. Here, we propose an alternative respiratory motion-resolved approach based on 3D radial phyllotaxis sampling, respiratory motion sorting and sparse reconstruction.

Quantification of coronary vessel wall thickness using a flexible time-resolved golden angle dual-inversion recovery acquisition for facilitated sequence timing at 3T

Abstract: Background Positive vessel wall remodeling is an early marker of coronary artery diseases. Its early detection could improve prognosis of adverse cardiovascular events. Black blood MRI based on double inversion recovery (DIR) (1) represents a non-invasive technique for the visualization of the coronary vessel wall. However, the need of collecting imaging data at the moment of both optimal blood signal nulling (2) and minimal myocardial motion, still limits the use of the technique in clinic. In order to alleviate difficulties in sequence planning, we propose a continuous acquisition scheme throughout a prolonged acquisition window. Combined with golden angle radial acquisition and k-t sparse SENSE (3), this enables a fully flexible a posteriori selection of imaging parameters.