Inter- and intra-rater reliability of blood and cerebrospinal fluid flow quantification by phase-contrast MRI

Reliability of In Vivo Creatine-Weighted Chemical Exchange Saturation Transfer (CrCEST) MRI in Calf Skeletal Muscle of Healthy Volunteers at 3 T

BACKGROUND

Skeletal muscle mitochondrial oxidative phosphorylation (mtOXPHOS) is important for ATP generation and its dysfunction leads to exercise intolerance. Phosphorus magnetic resonance spectroscopy (31P-MRS) is a useful, noninvasive technique for mtOXPHOS assessment but has limitations. Creatine-weighted chemical exchange saturation transfer (CrCEST) MRI is a potential alternative to assess muscle bioenergetics.

PURPOSE

To evaluate the interscan repeatability, intra- and interobserver reproducibility of CrCEST during mild plantar flexion exercise.

STUDY TYPE

Retrospective.

SUBJECTS

Twenty healthy volunteers (age 37.6 ± 12.4 years, 11 females).

FIELD STRENGTH/SEQUENCE

3 T/CEST imaging using gradient echo readout.

ASSESSMENT

τCrCEST (postexercise Cr recovery time) was assessed in two scans for each participant, following mild plantar flexion exercises targeting the medial gastrocnemius (MG), lateral gastrocnemius (LG), and soleus (Sol) muscles. Three observers measured τCrCEST for interobserver reproducibility. Three readings by one observer were used to measure intraobserver reproducibility. Two scans were used for within-participant interscan repeatability.

STATISTICAL TESTS

Paired t tests, intraclass correlation coefficient (ICC), and Pearson correlation were conducted. Bland-Altman plots were used to analyze the interobserver variability. A P-value of 0.05 was considered statistically significant.

RESULTS

There was excellent intra- (ICC∈ 0.94 - 0.98 $$ \in \left[0.94-0.98\right] $$ ) and interobserver (ICC∈ 0.9 - 0.98 $$ \in \left[0.9-0.98\right] $$ ) reproducibility, with moderate interscan repeatability for τCrCEST in LG and MG (ICC∈ 0.54 - 0.74 $$ \in \left[0.54-0.74\right] $$ ) and poor-to-moderate interscan repeatability in Sol (ICC∈ 0.24 - 0.53 $$ \in \left[0.24-0.53\right] $$ ). Excellent interobserver reproducibility was confirmed by Bland-Altman plots (fixed bias P-value∈ 0.08 - 0.87 $$ \in \left[0.08-0.87\right] $$ ).

DATA CONCLUSION

CrCEST MRI shows promise in assessing muscle bioenergetics by evaluating τCrCEST during mild plantar flexion exercise with reasonable reliability, particularly in LG and MG.

LEVEL OF EVIDENCE

4 TECHNICAL EFFICACY: Stage 1.

Cerebrospinal Fluid Flow in Patients with Huntington's Disease

OBJECTIVE

Investigations of cerebrospinal fluid (CSF) flow aberrations in Huntington's disease (HD) are of growing interest, as impaired CSF flow may contribute to mutant Huntington retention and observed heterogeneous responsiveness to intrathecally administered therapies.

METHOD

We assessed net cerebral aqueduct CSF flow and velocity in 29 HD participants (17 premanifest and 12 manifest) and 51 age- and sex matched non-HD control participants using 3-Tesla magnetic resonance imaging methods. Regression models were applied to test hypotheses regarding: (i) net CSF flow and cohort, (ii) net CSF flow and disease severity (CAP-score), and (iii) CSF volume after correcting for age and sex.

RESULTS

Group-wise analyses support a decrease in net CSF flow in HD (mean 0.14 ± 0.27 mL/min) relative to control (mean 0.32 ± 0.20 mL/min) participants (p = 0.02), with lowest flow in the manifest HD cohort (mean 0.04 ± 0.25 mL/min). This finding was explained by hyperdynamic CSF movement, manifesting as higher caudal systolic CSF flow velocity and higher diastolic cranial CSF flow velocity across the cardiac cycle, in HD (caudal flow: 0.17 ± 0.07 mL/s, cranial flow: 0.14 ± 0.08 mL/s) compared to control (caudal flow: 0.13 ± 0.06 mL/s, cranial flow: 0.11 ± 0.04 mL/s) participants. A positive correlation between cranial diastolic flow and disease severity was observed (p = 0.02).

INTERPRETATIONS

Findings support aqueductal CSF flow dynamics changing with disease severity in HD. These accelerated changes are consistent with changes observed over the typical adult lifespan, and may have relevance to mutant Huntington retention and intrathecally administered therapeutics responsiveness. ANN NEUROL 2023;94:885-894.

Automated in-depth cerebral arterial labelling using cerebrovascular vasculature reframing and deep neural networks

Identifying the cerebral arterial branches is essential for undertaking a computational approach to cerebrovascular imaging. However, the complexity and inter-individual differences involved in this process have not been thoroughly studied. We used machine learning to examine the anatomical profile of the cerebral arterial tree. The method is less sensitive to inter-subject and cohort-wise anatomical variations and exhibits robust performance with an unprecedented in-depth vessel range. We applied machine learning algorithms to disease-free healthy control subjects (n = 42), patients with stroke with intracranial atherosclerosis (ICAS) (n = 46), and patients with stroke mixed with the existing controls (n = 69). We trained and tested 70% and 30% of each study cohort, respectively, incorporating spatial coordinates and geometric vessel feature vectors. Cerebral arterial images were analyzed based on the 'segmentation-stacking' method using magnetic resonance angiography. We precisely classified the cerebral arteries across the exhaustive scope of vessel components using advanced geometric characterization, redefinition of vessel unit conception, and post-processing algorithms. We verified that the neural network ensemble, with multiple joint models as the combined predictor, classified all vessel component types independent of inter-subject variations in cerebral arterial anatomy. The validity of the categorization performance of the model was tested, considering the control, ICAS, and control-blended stroke cohorts, using the area under the receiver operating characteristic (ROC) curve and precision-recall curve. The classification accuracy rarely fell outside each image's 90-99% scope, independent of cohort-dependent cerebrovascular structural variations. The classification ensemble was calibrated with high overall area rates under the ROC curve of 0.99-1.00 [0.97-1.00] in the test set across various study cohorts. Identifying an all-inclusive range of vessel components across controls, ICAS, and stroke patients, the accuracy rates of the prediction were: internal carotid arteries, 91-100%; middle cerebral arteries, 82-98%; anterior cerebral arteries, 88-100%; posterior cerebral arteries, 87-100%; and collections of superior, anterior inferior, and posterior inferior cerebellar arteries, 90-99% in the chunk-level classification. Using a voting algorithm on the queued classified vessel factors and anatomically post-processing the automatically classified results intensified quantitative prediction performance. We employed stochastic clustering and deep neural network ensembles. Ma-chine intelligence-assisted prediction of vessel structure allowed us to personalize quantitative predictions of various types of cerebral arterial structures, contributing to precise and efficient decisions regarding the cerebrovascular disease.

Imputation models and error analysis for phase contrast MR cerebral blood flow measurements

Cerebral blood flow (CBF) supports brain metabolism. Diseases impair CBF, and pharmacological agents modulate CBF. Many techniques measure CBF, but phase contrast (PC) MR imaging through the four arteries supplying the brain is rapid and robust. However, technician error, patient motion, or tortuous vessels degrade quality of the measurements of the internal carotid (ICA) or vertebral (VA) arteries. We hypothesized that total CBF could be imputed from measurements in subsets of these 4 feeding vessels without excessive penalties in accuracy. We analyzed PC MR imaging from 129 patients, artificially excluded 1 or more vessels to simulate degraded imaging quality, and developed models of imputation for the missing data. Our models performed well when at least one ICA was measured, and resulted in R ² values of 0.998-0.990, normalized root mean squared error values of 0.044-0.105, and intra-class correlation coefficient of 0.982-0.935. Thus, these models were comparable or superior to the test-retest variability in CBF measured by PC MR imaging. Our imputation models allow retrospective correction for corrupted blood vessel measurements when measuring CBF and guide prospective CBF acquisitions.

In vitro evaluation of cerebrospinal fluid velocity measurement in type I Chiari malformation: repeatability, reproducibility, and agreement using 2D phase contrast and 4D flow MRI

BACKGROUND

Phase contrast magnetic resonance imaging, PC MRI, is a valuable tool allowing for non-invasive quantification of CSF dynamics, but has lacked adoption in clinical practice for Chiari malformation diagnostics. To improve these diagnostic practices, a better understanding of PC MRI based measurement agreement, repeatability, and reproducibility of CSF dynamics is needed.

METHODS

An anatomically realistic in vitro subject specific model of a Chiari malformation patient was scanned three times at five different scanning centers using 2D PC MRI and 4D Flow techniques to quantify intra-scanner repeatability, inter-scanner reproducibility, and agreement between imaging modalities. Peak systolic CSF velocities were measured at nine axial planes using 2D PC MRI, which were then compared to 4D Flow peak systolic velocity measurements extracted at those exact axial positions along the model.

RESULTS

Comparison of measurement results showed good overall agreement of CSF velocity detection between 2D PC MRI and 4D Flow (p = 0.86), fair intra-scanner repeatability (confidence intervals ± 1.5 cm/s), and poor inter-scanner reproducibility. On average, 4D Flow measurements had a larger variability than 2D PC MRI measurements (standard deviations 1.83 and 1.04 cm/s, respectively).

CONCLUSION

Agreement, repeatability, and reproducibility of 2D PC MRI and 4D Flow detection of peak CSF velocities was quantified using a patient-specific in vitro model of Chiari malformation. In combination, the greatest factor leading to measurement inconsistency was determined to be a lack of reproducibility between different MRI centers. Overall, these findings may help lead to better understanding for application of 2D PC MRI and 4D Flow techniques as diagnostic tools for CSF dynamics quantification in Chiari malformation and related diseases.

Role of the spinal canal compliance in regulating posture-related cerebrospinal fluid hydrodynamics in humans

Mechanical compliance of a compartment is defined by the change in its volume with respect to a change in the inside pressure. The compliance of the spinal canal regulates the intracranial pressure (ICP) under postural changes. Understanding how gravity affects ICP is beneficial for poorly understood cerebrospinal fluid (CSF)-related disorders. The aim of this study was to evaluate postural effects on cranial hemo- and hydrodynamics. This was a prospective study, which included 10 healthy volunteers (three males, seven females, mean ± standard deviation age: 29 ± 7 years). Cine gradient-echo phase-contrast sequence acquired at 0.5 T, "GE double-doughnut" scanner was used. Spinal contribution to overall craniospinal compliance (CSC), craniospinal CSF stroke volume (SV), magnetic resonance (MR)-derived ICP (MR-ICP), and total cerebral blood flow (TCBF) were measured in supine and upright postures using automated blood and CSF flows quantification. Statistical tests performed were two-sided Student's t-test, Cohen's d, and Pearson correlation coefficient. MR-ICP and the craniospinal CSF SV were significantly correlated with the spinal contribution to the overall CSC (r = 0.83, p < 0.05) and (r = 0.62, p < 0.05), respectively. Cranial contribution to CSC increased from 44.5% ± 16% in supine to 74.9% ± 8.4% in upright posture. The average MR-ICP dropped from 9.9 ± 3.4 mmHg in supine to -3.5 ± 1.5 mmHg. The CSF SV was over 2.5 times higher in the supine position (0.55 ± 0.14 ml) than in the upright position (0.21 ± 0.13 ml). In contrast, TCBF was slightly higher in the supine posture (822 ± 152 ml/min) than in the upright posture (761 ± 139 ml/min), although not statistically significant (p = 0.16). The spinal-canal compliance contribution to CSC is larger than the cranial contribution in the supine posture and smaller in the upright posture. Thereby, the spinal canal plays a role in modulating ICP upon postural changes. The lower pressure craniospinal CSF system was more affected by postural changes than the higher-pressure cerebral vascular system. Craniospinal hydrodynamics is affected by gravity and is likely to be altered by its absence in space. LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY STAGE: 2.

Fast Phase-Contrast Cine MRI for Assessing Intracranial Hemodynamics and Cerebrospinal Fluid Dynamics

We propose fast phase-contrast cine magnetic resonance imaging (PC-cine MRI) to allow breath-hold acquisition, and we compared intracranial hemo- and hydrodynamic parameters obtained during breath holding between full inspiration and end expiration. On a 3.0 T MRI, using electrocardiogram (ECG)-synchronized fast PC-cine MRI with parallel imaging, rectangular field of view, and segmented k-space, we obtained velocity-mapped phase images at the mid-C2 level with different velocity encoding for transcranial blood flow and cerebrospinal-fluid (CSF) flow. Next, we calculated the peak-to-peak amplitudes of cerebral blood flow (ΔCBF), cerebral venous outflow, intracranial volume change, CSF pressure gradient (ΔPG), and intracranial compliance index. These parameters were compared between the proposed and conventional methods. Moreover, we compared these parameters between different utilized breath-hold maneuvers (inspiration, expiration, and free breathing). All parameters derived from the fast PC method agreed with those from the conventional method. The ΔPG was significantly higher during full inspiration breath holding than at the end of expiration and during free breathing. The proposed fast PC-cine MRI reduced scan time (within 30 s) with good agreement with conventional methods. The use of this method also makes it possible to assess the effects of respiration on intracranial hemo- and hydrodynamics.

Assessing test-retest reliability of phase contrast MRI for measuring cerebrospinal fluid and cerebral blood flow dynamics

PURPOSE

Pathological states occur when cerebrospinal fluid (CSF) and cerebral blood flow (CBF) dynamics become dysregulated in the brain. Phase-contrast MRI (PC-MRI) is a noninvasive imaging technique that enables quantitative measurements of CSF and CBF flow. While studies have validated PC-MRI as an imaging technique for flow, few studies have evaluated its reliability for CSF and CBF flow parameters commonly associated with neurological disease. The purpose of this study was to evaluate test-retest reliability at the cerebral aqueduct (CA) and C2-C3 area using PC-MRI to assess the feasibility of investigating CSF and CBF flow dynamics.

METHODS

This study was performed on 27 cognitively normal young adults (ages 20-35 years). Flow data was acquired on a 3T Siemens Prisma using a 2D cine-PC pulse sequence. Three consecutive flow measurements were acquired at the CA and C2-C3 area. Intraclass correlation coefficient (ICC) and coefficient of variance (CV) were used to evaluate intrarater, inter-rater, and test-retest reliability.

RESULTS

Among the 26 flow parameters analyzed, 22 had excellent reliability (ICC > 0.80), including measurements of CSF stroke volume, flush peak, and fill peak, and 4 parameters had good reliability (ICC 0.60-0.79). 16 flow parameters had a mean CV ≤ 10%, 7 had a CV ≤ 15%, and 3 had a CV ≤ 30%. All CSF and CBF flow measurements had excellent inter-rater and intrarater reliability (ICC > 0.80).

CONCLUSION

This study shows that CSF and CBF flow can be reliably measured at the CA and C2-C3 area using PC-MRI, making it a promising tool for studying flow dynamics in the central nervous system.

The intracranial pressure curve correlates to the pulsatile component of cerebral blood flow

Current methods to measure cerebral blood flow (CBF) in the neuro critical care setting cannot monitor the CBF continuously. In contrast, continuous measurement of intracranial pressure (ICP) is readily accomplished, and there is a component of ICP that correlates with arterial inflow of blood into the cranial cavity. This property may have utility in using continuous ICP curve analysis to continuously estimate CBF. We examined the data from 13 patients, monitored with an intraventricular ICP device determining the pulsatile amplitude ICP_amp as well as the area under the ICP curve (AUC_ICP). Using an elastance measurement, the ICP curve was converted to craniospinal volume (AUC_ΔV). The patients were examined with Phase Contrast Magnetic Resonance Imaging (MRI), measuring flow in the carotid and vertebral arteries. This made it possible to calculate CBF for one cardiac cycle (ccCBF_MRtot) and divide it into the pulsatile (ccCBF_MRpuls) and non-pulsatile (ccCBF_MRconst) flow. ICP derived data and MRI measurements were compared. Linear regression was used to establish wellness of fit and ANOVA was used to calculate the P value. No correlation was found between ICP_amp and the ccICP_MRpuls (P = 0.067). In contrast there was a correlation between the AUC_ICP and ccCBF_MRpuls (R² = 0.440 P = 0.013). The AUC_ΔV correlated more appropriately with the ccCBF_MRpuls. (R² = 0.688 P < 0.001). Our findings suggests that the pulsatile part of the intracranial pressure curve, especially when transformed into a volume curve, correlates to the pulsatile part of the CBF.

Reproducibility of pulmonary blood flow measurements by phase-contrast MRI using different 1.5 T MR scanners at two institutions

Background

Magnetic resonance imaging (MRI) can be beneficial for diagnosis of disease by offering quantitative information. However, reproducibility can be a major problem when there is a numerical threshold in multi-institution, multi-vendor situations.

Purpose

To measure pulmonary blood flow with phase-contrast (PC) imaging using two different MR scanners (1.5 T) at different institutions in the same participants and to examine the reproducibility of the measurements.

Material and Methods

Participants were 10 healthy volunteers (5 men; age range, 27–36 years). The measurements included the mean and maximal blood velocities, the mean blood flow volume, and the acceleration time and volume (AT and AV), derived from the time-flow curve of the PC-MRI. Simultaneously obtained maximal, minimal, and mean areas from regions of interest set in the pulmonary artery were also calculated. In order to calculate the reproducibility of the quantitative variables, intra-class correlation coefficients (ICCs) were employed. When an adequate ICC was obtained, Bland–Altman analysis was conducted to identify any systematic bias.

Results

The ICCs were almost perfect for the mean blood flow volume and the AV (r = 0.82 and 0.80), and were substantial in the mean and maximal areas, and the AT (r = 0.63, 0.74, and 0.64, respectively). However, there was a fixed bias in the area measurement between the two scanners. Also, the AV had a proportional bias.

Conclusion

Our results reveal that various indices derived from PC-MRI on different MR scanners are promising as common indices for pulmonary flow assessment. Research and clinical use of PC-MRI for the pulmonary artery is expected to extend to multi-institution situations.

Effect of Atlas Vertebrae Realignment in Subjects with Migraine: An Observational Pilot Study

Introduction. In a migraine case study, headache symptoms significantly decreased with an accompanying increase in intracranial compliance index following atlas vertebrae realignment. This observational pilot study followed eleven neurologist diagnosed migraine subjects to determine if the case findings were repeatable at baseline, week four, and week eight, following a National Upper Cervical Chiropractic Association intervention. Secondary outcomes consisted of migraine-specific quality of life measures. Methods. After examination by a neurologist, volunteers signed consent forms and completed baseline migraine-specific outcomes. Presence of atlas misalignment allowed study inclusion, permitting baseline MRI data collection. Chiropractic care continued for eight weeks. Postintervention reimaging occurred at week four and week eight concomitant with migraine-specific outcomes measurement. Results. Five of eleven subjects exhibited an increase in the primary outcome, intracranial compliance; however, mean overall change showed no statistical significance. End of study mean changes in migraine-specific outcome assessments, the secondary outcome, revealed clinically significant improvement in symptoms with a decrease in headache days. Discussion. The lack of robust increase in compliance may be understood by the logarithmic and dynamic nature of intracranial hemodynamic and hydrodynamic flow, allowing individual components comprising compliance to change while overall it did not. Study results suggest that the atlas realignment intervention may be associated with a reduction in migraine frequency and marked improvement in quality of life yielding significant reduction in headache-related disability as observed in this cohort. Future study with controls is necessary, however, to confirm these findings. Clinicaltrials.gov registration number is NCT01980927.

Automatic labeling of cerebral arteries in magnetic resonance angiography

OBJECTIVES

In order to introduce 4D flow magnetic resonance imaging (MRI) as a standard clinical instrument for studying the cerebrovascular system, new and faster postprocessing tools are necessary. The objective of this study was to construct and evaluate a method for automatic identification of individual cerebral arteries in a 4D flow MRI angiogram.

MATERIALS AND METHODS

Forty-six elderly individuals were investigated with 4D flow MRI. Fourteen main cerebral arteries were manually labeled and used to create a probabilistic atlas. An automatic atlas-based artery identification method (AAIM) was developed based on vascular-branch extraction and the atlas was used for identification. The method was evaluated by comparing automatic with manual identification in 4D flow MRI angiograms from 67 additional elderly individuals.

RESULTS

Overall accuracy was 93%, and internal carotid artery and middle cerebral artery labeling was 100% accurate. Smaller and more distal arteries had lower accuracy; for posterior communicating arteries and vertebral arteries, accuracy was 70 and 89%, respectively.

CONCLUSION

The AAIM enabled fast and fully automatic labeling of the main cerebral arteries. AAIM functionality provides the basis for creating an automatic and powerful method to analyze arterial cerebral blood flow in clinical routine.

Arterial, venous, and cerebrospinal fluid flow: simultaneous assessment with Bayesian multipoint velocity-encoded MR imaging

PURPOSE

To measure arterial, venous, and cerebrospinal fluid (CSF) velocities simultaneously by using Bayesian multipoint velocity-encoded magnetic resonance (MR) imaging and to compare interacquisition reproducibility relative to that of standard phase-contrast MR imaging for sequential measurements of arterial, venous, and CSF velocities.

MATERIALS AND METHODS

This study was approved by the local ethics committee, and informed consent was obtained from all subjects. Simultaneous measurement of blood and CSF flow was performed at the C1-C2 level in 10 healthy subjects (mean age, 24.4 years ± 2.7; five men, five women) by using accelerated Bayesian multipoint velocity-encoded MR imaging. Data were compared with those obtained from two separate conventional phase-contrast MR imaging acquisitions, one optimized for arterial and venous blood flow (velocity encoding range, ±50 cm/sec) and the other optimized for CSF flow (velocity encoding range, ±10 cm/sec), with an imaging time of approximately 2 minutes each. Data acquisition was repeated six times. Intraclass correlation coefficient (ICC) and linear regression were used to quantify interacquisition reproducibility.

RESULTS

There was no significant difference in arterial blood flow measured with Bayesian multipoint velocity-encoded MR imaging and that measured with phase-contrast MR imaging (mean ICC, 0.96 ± 0.03 vs 0.97 ± 0.02, respectively). Likewise, there was no significant difference between CSF flow measured with Bayesian multipoint velocity-encoded MR imaging and that measured with phase-contrast MR imaging (mean ICC, 0.97 ± 0.02 vs 0.96 ± 0.05, respectively). For venous blood flow, the ICC with Bayesian multipoint MR imaging was significantly larger than that with conventional phase-contrast MR imaging (mean, 0.75 ± 0.23 vs 0.65 ± 0.26, respectively; P = .016).

CONCLUSION

Bayesian multipoint velocity-encoded MR imaging allows for simultaneous assessment of fast and slow flows in arterial, venous, and CSF lumina in a single acquisition. It eliminates the need for vessel-dependent adjustment of the velocity-encoding range, as required for conventional sequential phase-contrast MR imaging measurements.