Abstract
Objectives
To investigate ventilation in mild to moderate asthmatics and age-matched controls using hyperpolarized (HP)129Xenon-MR imaging and correlate findings with pulmonary function tests (PFTs).
Materials and Methods
This single-center, HIPAA-compliant prospective study was approved by our IRB. Thirty subjects (10 young asthmatics, 26±6 years; 3 males, 7 females; 10 older asthmatics, 64±6 years; 3 males, 7 females; 10 healthy controls) were enrolled. After repeated PFTs 1 week apart, subjects underwent 2 MRI-scans within 10 minutes, inhaling 1 liter volumes containing 0.5–1 liter of 129Xe. 129Xe ventilation signal was quantified by linear binning, from which the ventilation defect percentage (VDP) was derived. Differences in VDP among subgroups and variability with age was evaluated using one-tailed t-tests. Correlation of VDP with PFTs was tested using Pearson’s correlation coefficient. Reproducibility of VDP was assessed using Bland-Altman plots, linear regression (R2), intra-class correlation coefficient (ICC) and concordance correlation coefficient (CCC).
Results
VDP was significantly higher in young asthmatics vs. young healthy subjects (8.4±3.2% vs. 5.6±1.7%, P=.031), but not in older asthmatics versus age-matched controls (16.8±10.3% vs. 11.6±6.6%; P=.13). VDP was found to increase significantly with age (healthy: P=.05; asthmatics: P=.033). VDP was highly reproducible (R2=0.976, ICC=0.977, CCC=0.976) and significantly correlated with FEV1% (r=−0.42, P=.025), FEF25–75% (r=−0.45, P=.019), FEV1/FVC (r=−0.71, P<.0001), FeNO (r=0.69, P<.0001), and RV/TLC (r=0.51, P=.0067). Bland-Altman analysis showed a bias for VDP of −0.88±1.52 (FEV1%: −0.33±7.18).
Conclusion
129Xenon-MR imaging is able to depict airway obstructions in mild to moderate asthma and significantly correlates with PFTs.
Keywords: lung MRI, hyperpolarized xenon gas, airway obstruction, asthma, pulmonary function tests
Introduction
Asthma has a prevalence of 6–8% in the American population that peaks in early childhood, declines in late adolescence, and rises again in the late adulthood before decreasing among the elderly (1, 2).
While the diagnosis and clinical monitoring of asthma mainly employ forced expiratory volume in one second (FEV1%), these pulmonary function tests (PFTs) do not completely characterize airway physiology. Even in subjects with normal FEV1%, nitrogen washout shows that significant ventilation heterogeneity can be present (3). Similarly, computed tomography (CT) can reveal heterogeneous regional air trapping and airway thickening in asthmatics (4) with normal PFTs. The most striking shortcoming of PFTs is their broad variation with age, standing height, sex, and ethnic group (5,6). These observations support the need for more precise and reliable measurements that do not require such scaling. In addition to being highly variable, PFTs also cannot depict regional ventilation, nor its heterogeneity. Ventilation defects could represent preclinical stages of disease, offering the opportunity of better monitoring disease progress or treatment efficiency. Another possible application could include preclinical detection of disease-related ventilation defects to further avoid exacerbations (7).
To this end xenon-enhanced dual-energy CT has been able to depict regional ventilation, but confers a considerable radiation dose, which particularly limits use in healthy subjects and young individuals (8). Therefore, MRI techniques are emerging that use short echo time sequences in conjunction with oxygen shortening lung tissue T1 to provide a surrogate of regional ventilation (9). Although the oxygen-enhancement signal depends on more than regional ventilation, it has been shown to significantly correlate with PFT measurements. However, oxygen-enhanced MR imaging is hampered by relatively small signal, long acquisition times, and the need for subtraction that can make it prone to mis-registration artifacts (10).
Even more advantageous are non-proton MR methods that directly image non-endogenous inhaled agents, such as perfluorinated gases, without background signal (11). However, the most thoroughly-investigated and mature non-proton method over the last 2 decades, has been hyperpolarized (HP) 3He MRI (12–15). This technique also readily depicts ventilation heterogeneity and defects and has been used to visualize the effects of both bronchoprovocation and bronchodilation (16). The observed ventilation defects on HP 3He MR imaging significantly correlated with areas of air-trapping resulting from airway remodeling encountered on CT (17). 3He MRI has shown that older asthmatics exhibit significantly more ventilation defects than younger ones (11). Unfortunately, 3He gas is in short supply and faces sharply rising costs, which has increased interest in readily available HP 129Xe. This gas has already been shown to detect more ventilation defects in asthmatics than 3He (18). However, there are no studies further investigating obstructive patterns in asthma with 129Xe MRI or correlating this technique with pulmonary function tests. This gap must be addressed in order for 129Xe MR imaging to expand into clinical practice.
The purpose of our study was to investigate if 129Xe MRI would a) correlate significantly with established lung function metrics; b) accurately depict differences in ventilation defect percentage (VDP) between asthmatics and healthy controls and c) be reproducible at baseline.
Material and Methods
This prospective, single-center Health Insurance Portability and Accountability Act (HIPAA)-compliant study was approved by the Institutional Review Board (IRB) and registered at clinicaltrials.gov. All subjects gave informed, written consent prior to study enrollment.
Patients
Between January 2012 and March 2014, 30 subjects were prospectively recruited for this study (20 asthmatics, 10 age-matched healthy controls). Both study populations were matched for gender, asthma medication use, and percent predicted forced expiratory volume (FEV1%). Asthma was confirmed with a methacholine challenge test according to American Thoracic Society (ATS) guidelines (6). The population was further stratified by age to young (18–35 years; n=9) and older (55–75 years; n=10) asthmatics, and included 5 age-matched healthy subjects for each age group.
The following exclusion criteria were applied: 1) smoking history of more than 5 pack-years or smoked in past 2 years; 2) pre-bronchodilator FEV1% < 60%; 3) diffusing capacity of the lung for carbon monoxide (DLCO) < 80% predicted; 4) asthma exacerbation requiring oral corticosteroids or respiratory tract infection in the past 6 weeks; 5) had been hospitalized for respiratory disease in the past 3 months; 6) were asthmatic and using more than short-acting beta-agonist and/or inhaled corticosteroids (dose equivalent of ≤250 mg Fluticasone). General contraindications for MRI also resulted in exclusion from the study.
All individuals participated in 3 separate visits on different days at the same time of day within ±2 hours (Figure 1). On the first visit, clinical characteristics were recorded and a physical exam was performed. During this session, DLCO and fractional exhaled nitric oxide (FeNO) was measured, and the Asthma Quality of Life Questionnaire (AQLQ), as well as the Asthma Control Questionnaire (ACQ) was administered (19). At the second visit (one week ± 3 days later), subjects returned to the clinic and repeated spirometry, DLCO, AQLQ, FeNO and sequential sputum induction to more accurately measure these parameters (learning/training effect). At the third and final visit, 129Xe ventilation MRI was performed.
Figure 1.
Flowchart of study recruitment and the diagnostic work-up timeline.
Pulmonary Function Testing
Spirometry, plethysmography and DLCO were performed using the Vmax Autobox and SensorMedics (Carefusion, San Diego, CA). Spirometry was used to measure FEV1, FEF25-75 and FVC, while plethysmography was used to assess lung volumes (total lung capacity = TLC, residual volume = RV). Ratios such as FEV1/FVC and RV/TLC were calculated from raw data, but all values are reported as percentage of the predicted values for subjects of comparable characteristics. Fractional exhaled nitric oxide (FeNO), indicating airway inflammation, was measured using the Niox Mino (Aerocrine Inc. USA, Morrisville, NC). Asthmatics were classified according to the guidelines of the U.S. National Asthma Education and Prevention Program [NAEPP].
MR Imaging
HP 129Xe MR scans were performed on a 1.5 Tesla GE Healthcare EXCITE 15M4 MR system. Isotopically enriched 129Xe gas (83% 129Xe), was polarized to 6–10% by spin-exchange optical pumping (20). Subjects underwent two fast gradient echo 129Xe ventilation MRI scans in the supine position, ten minutes apart, using 1-liter of total gas volume, comprised of 0.5–1 liters of HP 129Xe, mixed with N2 buffer gas as described previously (21). In all cases, the subject received a total volume of gas of 1 liter. The xenon volume was determined by the duration of cryogenic accumulation and total gas flow rate. As 129Xe polarization levels became higher, smaller predetermined 129Xe volumes could be used. In case this volume was less than 1 liter, Helium buffer gas was added to make 1 liter.
From each 129Xe MRI scan, VDP was calculated using the linear binning approach (22). Subjects were fitted with a flexible chest coil (Clinical MR solutions, Brookfield, WI) that was tuned to the 17.66 MHz 129Xe frequency and proton-blocked to permit anatomical scans to be acquired using the 1H body coil. After being positioned on the scanner bed, subjects first inhaled a test dose of room air to learn the breathing maneuver. 1H images were acquired in a separate breath hold prior to the 129Xe scan. For the 1H scans, subjects inhaled 1 liter of air from the gas bag in order to achieve a similar level of lung inflation. Individuals were coached to inhale 129Xe from functional residual capacity (FRC). A first bag of gas mixture was used to calibrate the scanner transmit/receive settings in each imaging session. Subjects then underwent two HP 129Xe ventilation MRI scans, ten minutes apart, containing 1-liter of total gas volume, comprised of 0.5–1 liters of HP 129Xe, mixed with N2 buffer gas. The two xenon doses were administered to the patient in the supine position. Specific sequence parameters are summarized in Table 1.
Table 1.
Sequence parameters for thoracic cavity registration and the HP 129Xe ventilation image.
| Parameter | Thoracic cavity 1H image Steady-state free precession sequence |
HP 129Xe ventilation image Spoiled gradient echo sequence |
|---|---|---|
| Repetition time (msec) | 2.8 | 7.9 |
| Echo time (msec) | 1.2 | 1.9 |
| Flip angle (degrees) | 45 | 5 – 7 |
| Slice thickness (mm) | 15 | 15 |
| Field of view (cm2) | 40 × 40 | 40 × 28 – 40 |
| Acquisition matrix | 128 × 128 | 128 × 90 – 128 |
| Receiver bandwidth (kHz) | 125 | 8 |
| Acquisition time (sec) | 15 | 10–14 |
Image Analysis
Images were quantified to determine their ventilation defect percentage (VDP) using a corrected, semi-automated linear-binning approach, first described by He (22). The analysis of the Xe signal was confined to a mask generated by segmenting the proton images to delineate the thoracic cavity. Prior to segmentation, these images had been registered to the HP 129Xe ventilation MRI scans. Both registration and segmentation by region growing used Avizo (Visualization Sciences Group, Burlington, MA). The HP 129Xe images were then corrected for transmit/receive inhomogeneity and rescaled by the 99th percentile of their cumulative intensity distribution, such that all intensities ranged from 0–1. This permitted each 129Xe voxel within the thoracic cavity to be classified into one of four clusters (<0.2 defect, 0.2–0.4 low intensity, 0.4–0.8 medium intensity, and >0.8 high intensity). The volume of the lowest-intensity cluster (<0.2) relative to that of the thoracic cavity was used to quantify ventilation defect percentage (VDP). The 4 clusters representing the HP 129Xe signal intensities within the lungs were color-coded on the ventilation maps: red=defect; yellow=low intensity; green=medium intensity; and blue=high signal intensity. Prior to statistical analysis, the HP 129Xe ventilation images and the corresponding color-coded maps were reviewed in consensus by two readers (blinded to the review process, with 6 and 10 years of experience in chest imaging) to verify that image quality was suitable. Image quality was determined to be inadequate for further analysis if the images exhibited extensive coil shading or insufficient signal to noise to undergo binning analysis.
Statistical Analysis
Statistical analysis was performed using JMP 12 (SAS Institute Inc., Cary, NC) and MedCalc 16.1 (MedCalc Software, Ostend, Belgium). VDP and its change with age was compared between healthy subjects and asthmatics using one-tailed tests; these were justified based on prior knowledge that VDP increases with asthma and age (23). Correlation between VDP and pulmonary function tests was evaluated using the Pearson’s correlation coefficient (r). Reproducibility of VDP over the two consecutive scans was assessed using linear regression, intra-class correlation coefficient (ICC), concordance correlation coefficient (CCC) (21), as well as by Bland-Altman analysis. Results were considered statistically significant with P<.05.
Results
Patients
A total of 30 subjects were enrolled and completed the entire study protocol, with no serious adverse events reported during the imaging sessions. Mild side effects related to xenon inhalation (tingling, euphoria, dizziness) resolved within 3 minutes after exhalation. No further side effects could be observed. Nevertheless, one patient was excluded because the MR images were corrupted by shading artifacts, determined to be caused by a malfunctioning coil (Figure 2). This artifact could not be corrected by retrospective bias field correction due to insufficient signal. This resulted in a final sample size of 29 subjects, of which 19 were mild-to-moderate asthmatics. PFT and VDP results are displayed in Table 2.
Figure 2.
Example of false ventilation defects caused by artifacts. (A): In this healthy young volunteer VDP was 7.28% at baseline and 8.33% at re-scan. VDP clusters <0.2 (red) were slightly visible at baseline and become more severe at rescan. These defects could be attributed to rib cage impressions (white arrows) representing artifacts. These artifacts were identified during the review and were resolved by manual re-segmentation. (B) Young asthmatic subject with FEV1%=80%, that had to be excluded due to extensive coil shading, that caused the binning algorithm (C) to overestimate VDP at 22.4%.
Table 2.
Summary of the subject demographics, results of the pulmonary function tests, and the ventilation defect percentage (VDP). All values are averages of measurements over two time points, and are reported as mean ± standard deviation for the cohort.
| YH (n=5) | YA (n=9) | OH (n=5) | OA (n=10) | |
|---|---|---|---|---|
| Subject Demographics | ||||
| Age years (± SD) | 23.2 (± 1.3) | 25.9 (± 6.4) | 63.4 (± 5.8) | 63.2 (± 6.1) |
| Male/Female | 2/3 | 2/7 | 2/3 | 3/7 |
| BMI (kg/m2) | 23.7 (± 2.5) | 29.0 (± 8.2) | 29.0 (± 5.2) | 30.3 (± 4.1) |
| ACQ | N/A | 1.3 (± 0.9) | N/A | 1.7 (± 1.2) |
| AQLQ | N/A | 6.1 (± 0.8) | N/A | 5.0 (± 1.9) |
| PFTs | ||||
| FEV1% | 106 (± 14.5) | 84 (± 16.1) | 95.8 (± 14.3) | 81.1 (± 19.8) |
| FVC% | 111.4 (± 12.8) | 98.8 (± 16.2) | 104.0 (± 18.1) | 96.4 (± 12.7) |
| FEV1/FVC | 81.0 (± 8.2) | 72.2 (± 7.1) | 71.7 (± 11.2) | 64.5 (± 11.2) |
| FEF25-75% | 96.2 (± 32.3) | 59.2 (± 21.9) | 87.3 (± 35.5) | 53.1 (± 26.9) |
| RV/TLC | 0.18 (± 0.02) | 0.29 (± 0.12) | 0.31 (± 0.05) | 0.40 (±0.05) |
| DLCO mmol CO | 86.8 (± 10.5) | 97.1 (± 10.8) | 89.5 (± 10.8) | 98.4 (± 17.3) |
| FeNO ppb | 16.8 (± 8.1) | 33.8 (± 21.4) | 15.6(± 10.2) | 36.9 (± 48.2) |
| Eos Sputum % | 4.0 (± 5.0) | 5.1 (± 7.2) | 9.3 (± 3.4) | 5.5 (± 3.8) |
| HP 129Xe MRI | ||||
| VDP | 5.6 (± 1.8) | 8.7 (± 3.2) | 11.6 (± 6.6) | 16.8 (± 10.3) |
YH=young healthy; YA=young asthmatic; OH=old healthy; OA=old asthmatic; ACQ=Asthma Control Questionnaire; AQLQ=Asthma Quality of Life Questionnaire; BMI=body mass index; FEV1%=forced expiratory volume in 1 second; FVC%=forced vital capacity; FEF=forced mid-expiratory flow rate; FeNO=fractional exhaled nitric oxide; RV/TLC=residual volume/total lung capacity; VDP=ventilation defect percentage
HP 129Xe MRI Ventilation Defect Analysis
Among all patient composing the final study cohort, healthy young volunteers yielded the most uniform 129Xe distribution (VDP=5.6±1.7%; Figure 3). By comparison, young asthmatics exhibited significantly more ventilation defects (VDP=8.4±3.2%; P=.031; Figure 4). Similarly, older asthmatics also showed more defects than older healthy subjects (16.8±10.3% vs. 11.6±6.6%; Figures 3 and 4). However, this difference was not significant (P=.13). Older subjects generally exhibit higher VDP (figure 2), and this increase with age was found to be statistically significant (healthy subjects P=.05; asthmatics: P=.02) (Table 3).
Figure 3.
Left: Representative HP 129Xe MRI ventilation scan in a healthy, young volunteer with FEV1%=124%. The 129Xe distribution during the breath hold is largely homogeneous. In the corresponding maps, faint red clusters are primarily attributable to slight mis-registration, yielding VDP=3.83%. Right: 129Xe MRI of an older healthy volunteer with FEV1%=102%. Compared to the young normal, this subject exhibits modest defects in the lung periphery, leading to VDP=14.4 %.
Figure 4.
Left: Young asthmatic subject; FEV1%=59%: Moderate ventilation defects are found mainly in the lung periphery and are triangular in shape. On corresponding binning maps, ventilation defects are seen in red, as well as some high signal intensity pixels (blue). VDP: 14.6%. Right: An older asthmatic is depicted on the right; FEV1%=53%. Scattered ventilation defects are visible throughout the whole lung parenchyma. Both upper and lower lobes, as well as hilar and peripheral lung regions are almost equally affected. The corresponding VDP maps yielded a total VDP percentage of 44.2%.
Table 3.
The acquired parameters were tested for significant differences among the healthy and asthma subjects, and also for change with age. Note: Significant differences are displayed in bold font-type.
| Comparison | Healthy vs. Asthma | Young. vs. Old | ||
|---|---|---|---|---|
|
| ||||
| Young | Old | Healthy | Asthma | |
| PFTs | ||||
| FEV1% | 0.014 | 0.065 | 0.15 | 0.36 |
| FVC% | 0.069 | 0.22 | 0.24 | 0.36 |
| FEV1/FVC | 0.041 | 0.20 | 0.14 | 0.045 |
| FEF25-75% | 0.031 | 0.12 | 0.36 | 0.30 |
| RV/TLC | 0.014 | 0.042 | 0.020 | 0.018 |
| DLCO mmol CO | 0.060 | 0.16 | 0.37 | 0.42 |
| FeNO ppb | 0.036 | 0.10 | 0.42 | 0.57 |
| Eos Sputum % | 0.80 | 0.091 | 0.063 | 0.45 |
| HP 129Xe MRI | ||||
| Average VDP | 0.031 | 0.13 | 0.05 | 0.016 |
FEV1%=forced expiratory volume in 1 second; FVC%=forced vital capacity; FEF=forced mid-expiratory flow rate; FeNO=fractional exhaled nitric oxide; RV/TLC=residual volume/total lung capacity; VDP=ventilation defect percentage
Comparison of VDP and PFTs
Correlation results are reported in Table 4. VDP showed a moderate correlation with FEV1% (r=−0.42, P=.025), FEF25-75% (r=−0.45, P=.019), RV/TLC (r=0.51, P=.0067); and a strong correlation with FEV1/FVC [r=−0.70, P<.0001] (Figure 5). In addition to PFTs, VDP also correlated strongly with the fraction of exhaled nitric oxide (FeNO) (r=0.69, P<.0001). VDP did not correlate significantly with FVC, DLCO, ACQ, and the AQLQ questionnaires.
Table 4.
Correlations of pulmonary functions tests to VDP. Note: significant results are displayed in bold font type.
| Correlation with VDP | r | P-value |
|---|---|---|
| FEV1% | −0.42 | .0249 |
| FVC% | −0.09 | .6284 |
| FEV1/FVC | −0.71 | <.0001 |
| FEF25-75% | −0.45 | .019 |
| RV/TLC | 0.51 | .0067 |
| DLCO mmol CO | 0.34 | .0858 |
| FeNO ppb | 0.69 | <.0001 |
| Eos Sputum % | 0.00 | .9631 |
EV1%=forced expiratory volume in 1 second; FVC%=forced vital capacity; FEF=forced mid-expiratory flow rate; FeNO=fractional exhaled nitric oxide; RV/TLC=residual volume/total lung capacity; VDP=ventilation defect percentage
Figure 5.
Correlation of VDP and FEV1%, FEV25-75%, FEV1/FVC, and RV/TLC.
Repeatability of HP 129Xe VDP and PFTs
The repeatability of VDP proved to be excellent with ICC=0.977 and CCC=0.976 between the two MR scans. Figure 6 additionally shows results of linear regression and Bland-Altman analysis for VDP (A, B). Linear regression yielded an excellent least squares fit for VDP with R2=0.976 (P<.0001). Bland-Altman analysis depicted a small bias (±SD) in VDP of −0.88±1.52, which is largely driven by two subjects who had a difference of ~4% between the two scans. In contrast, the variability in FEV1% at 1 week was ±183 ml (7.18%) across the subject population, which is well illustrated by the Bland-Altman plot for FEV1% (Figure 6D).
Figure 6.
Linear regression and Bland-Altman plots for the two repeated measures of VDP and FEV1%. Linear regression (A, C) showed high agreement between the two scans for VDP (R2 = 0.976), and slightly lower for FEV1% (R2 = 0.879). Solid lines indicate the least squared fit. Bland-Altman analysis (B, D) yielded a bias ±SD of −0.88±1.52 for VDP, and −0.33±7.18 for FEV1. Solid line indicates the mean difference and dotted lines indicate the standard deviation about this mean.
Discussion
Our study results show that the ventilation defect percentage (VDP) derived from HP 129Xe MR imaging is able to detect regional obstruction and significantly correlates with pulmonary function tests.
VDP was moderately to strongly correlated with FEV1%, FEF25-75%, FEV1/FVC and RV/TLC. FEV1% is the predominant metric used for quantitative assessment of airflow limitation in asthma (24,25). Of note, VDP correlated significantly with FEV1%, FEF25-75%, FEV1/FVC and RV/TLC. However, numerous subjects with normal FEV1% exhibiting pronounced ventilation defects, and yet others with abnormal FEV1% exhibiting almost normal lung ventilation on MR. This speaks to the different ways these two techniques are sensitive to obstruction. In order for HP 129Xe to be detectable in a given voxel, the gas must travel largely unobstructed through all 23 generations of airways. This makes HP 129Xe MRI inherently sensitive to the entire bronchial tree, whereas the measurement of FEV1% is known to be dominated by contributions from the larger airways (26).
HP 129Xe VDP further correlated with FEF25-75% (r=−0.45, P=.019), which is thought to be a more sensitive indicator of small airways obstruction. This parameter reflects the most effort-independent segment of the spirometric curve, dominated by flow in the small airways. It is here that diseases of chronic airflow obstruction are thought to start, and thus, the correlation of FEF25-75% with HP 129Xe VDP bolsters the notion that VDP is indeed sensitive to small airways obstruction. Because FEF25-75% is known to be less reproducible than FEV1% (27), HP 129Xe MRI may provide a more reliable metric of small airways obstruction. A finding of potential clinical interest was the strong correlation between VDP and FeNO (r=0.69, P<.0001), which reflects inflammatory activity and may thus aid in the distinction between young asthmatics and young healthy subjects. The sputum eosinophilic count was another recorded measure of airway inflammation; this measure did not correlate with VDP.
The comparison between HP 129Xe VDP and PFTs are relatively virgin ground. To the best of our knowledge there are only two comparable publications addressing the correlations of HP 129Xe VDP and spirometry. In a cohort of COPD subjects, Kirby and co-workers showed a significant correlation between 3He and HP 129Xe MRI VDP, as well as their correlation to FEV1% and FEV1/FVC. Additionally, they found that VDP derived from HP 129Xe MRI was significantly higher than from 3He (28). Kirby and colleagues emphasized that the low density of helium likely makes it less sensitive to obstruction than xenon. To date, the sole study of HP 129Xe MRI in an asthma population was conducted by Svenningsen (16) who focused primarily on bronchodilator response, and did not report on baseline correlations of HP 129Xe VDP with PFTs.
One cannot overemphasize the need for reproducible markers of airway obstruction. The high degree of variability of FEV1% confounds the clinical value of PFTs and represents one of the major reasons why respiratory clinical trials require large numbers of patients to record an effect. By comparison, the high repeatability observed with HP 129Xe MRI has the potential to mitigate the variability impacting PFTs. We evaluated the repeatability of HP 129Xe MRI by conducting short-term follow up scans as previously reported with 3He MRI (29). In particular, the Bland-Altman analysis yielded only a minimal variation of ±1.52%, whereas the variability of FEV1% was found to be ±7.18%. Although the time between FEV1 measurements was considerably longer than for HP 129Xe MRI, such variability is consistent with the known short-term variability of FEV1% (25,30).
Our study results can have important clinical implications. Measures of VDP with HP 129Xe MR imaging can be clinically leveraged in monitoring asthma-related changes in ventilation in patients with mild to moderate disease. Supported by the remarkable correlation with PFTs and high levels of repeatability, VDP derived from HP 129Xe MR imaging could be valuable for the longitudinal assessment of disease progression and therapy response.
VDP calculation from HP 129Xe MR imaging is, however, not immune to physiologic states that can potentially hinder the image interpretation. Notably, we observed a substantial overlap in VDP values between old asthmatics and age-matched controls. These subjects could have been promptly discriminated by using the RV/TLC, which is a marker of changes occurring in the aging lung (i.e., senile emphysema) (31,32).
Despite these promising results and possible applications, hyperpolarized noble gas MR imaging is not yet used in routine clinical practice. Although the technology (particularly 3He) has been thoroughly studied for two decades, it is still limited to selected academic institutions. This can be partially attributed to the complex and non-traditional workflow of hyperpolarized agents, and the fact that it has not yet been approved by FDA or similar authorities internationally.
Limitations
Our study had notable limitations. Along with the small sample size in the setting of a prospective case-control design, there were differences in the time lapse between repeat measurements of PFTs (i.e., one week) and MR imaging (i.e., ten minutes). Thus, the high repeatability of 129Xe MRI observed on short time scales, should in the future also be verified at longer timescales. In our study, the proton scan and the HP 129Xe ventilation scan were not acquired during the same breath hold owed to the long scan time for each sequence. Consequently, the alignment of the proton images and the ventilation images might not be perfect; potentially leading to artifacts in the binning maps. Future trials could benefit from implementing combined hyperpolarized gas ventilation images and proton images within a single breath hold (33). This approach could facilitate the quantification of ventilation defects. Another limitation is represented by the lack of widespread availability of hyperpolarized noble gas MR imaging, still to date confined to selected academic institutions. Finally, one could argue that findings from HP 129Xe MR imaging would further benefit from correlation with other imaging modalities (e.g., conventional CT imaging).
In conclusion, our data shows that HP 129Xe MR imaging is able to quantify airway obstructions in mild to moderate asthma, and significantly correlates with PFTs. Its high reproducibility can make this technique a surrogate for PFTs in surveillance of ventilation impairment in asthma patients.
Acknowledgments
This study was funded by Chiesi Pharmaceuticals; with additional support from NIH/NHLBI R01 HL105643 and the Duke Center for In Vivo Microscopy, an NIH/NIBIB national Biomedical Technology Resource Center (P41 EB015897).
Lukas Ebner received financial funding by the Swiss National Science Foundation (SNSF; Grant P2SKP3_158645 / 1).
Footnotes
Guarantors of integrity of entire study: Lukas Ebner, Bastiaan Driehuys
Study concepts/study design or data acquisition or data analysis/interpretation: all authors
Manuscript drafting or manuscript revision for important intellectual content: all authors
Approval of final version of submitted manuscript: all authors
Literature research: Lukas Ebner, Bastiaan Driehuys
Clinical studies: Timothy Heacock, Monica Kraft
MR experiments: Mu He, Suryanarayanan S. Kaushik, Matthew S. Freeman, Bastiaan Driehuys
Statistical analysis: Mu He, Rohan S. Virgincar
Manuscript editing: Bastiaan Driehuys, H. Page McAdams, Rohan S. Virgincar
References
- 1.Reed CE. The natural history of asthma. J Allergy Clin Immunol. 2006;118(3):543–548. doi: 10.1016/j.jaci.2006.06.020. quiz 549–550. [DOI] [PubMed] [Google Scholar]
- 2.Oraka E, Kim HJ, King ME, Callahan DB. Asthma prevalence among US elderly by age groups: age still matters. J Asthma. 2012;49(6):593–599. doi: 10.3109/02770903.2012.684252. [DOI] [PubMed] [Google Scholar]
- 3.Bourdin A, Paganin F, Prefaut C, Kieseler D, Godard P, Chanez P. Nitrogen washout slope in poorly controlled asthma. Allergy. 2006;61(1):85–89. doi: 10.1111/j.1398-9995.2006.00970.x. [DOI] [PubMed] [Google Scholar]
- 4.Grenier P, Mourey-Gerosa I, Benali K, Brauner MW, Leung AN, Lenoir S, Cordeau MP, Mazoyer B. Abnormalities of the airways and lung parenchyma in asthmatics: CT observations in 50 patients and inter- and intraobserver variability. Eur Radiol. 1996;6(2):199–206. doi: 10.1007/BF00181147. [DOI] [PubMed] [Google Scholar]
- 5.Miller MR, Quanjer PH, Swanney MP, Ruppel G, Enright PL. Interpreting lung function data using 80% predicted and fixed thresholds misclassifies more than 20% of patients. Chest. 2011;139(1):52–59. doi: 10.1378/chest.10-0189. [DOI] [PubMed] [Google Scholar]
- 6.Guidelines for Methacholine and Exercise Challenge Testing—1999. American Journal of Respiratory and Critical Care Medicine. 2000;161(1):309–329. doi: 10.1164/ajrccm.161.1.ats11-99. [DOI] [PubMed] [Google Scholar]
- 7.Roos JE, McAdams HP, Kaushik SS, Driehuys B. Hyperpolarized Gas MR Imaging: Technique and Applications. Magn Reson Med. 2016;75(4):1434–43. doi: 10.1002/mrm.25675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kong X, Sheng HX, Lu GM, et al. Xenon-Enhanced Dual-Energy CT Lung Ventilation Imaging: Techniques and Clinical Applications. AJR Am J Roentgenol. 2014;202(2):309–317. doi: 10.2214/AJR.13.11191. [DOI] [PubMed] [Google Scholar]
- 9.Renne J, Hinrichs J, Schönfeld C, et al. Noninvasive Quantification of Airway Inflammation Following Segmental Allergen Challenge with Functional MR Imaging: A Proof of Concept Study. Radiology. 2015;274(1):267–275. doi: 10.1148/radiol.14132607. [DOI] [PubMed] [Google Scholar]
- 10.Kruger SJ, Fain SB, Johnson KM, et al. Oxygen-enhanced 3D radial ultrashort echo time magnetic resonance imaging in the healthy human lung. NMR Biomed. 2014 Dec;27(12):1535–41. doi: 10.1002/nbm.3158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Halaweish AF, Moon RE, Foster WM, et al. Perfluoropropane Gas as a Magnetic Resonance Lung Imaging Contrast Agent in Humans. CHEST. 2013;144(4):1300–1311. doi: 10.1378/chest.12-2597. [DOI] [PubMed] [Google Scholar]
- 12.Altes TA, Powers PL, Knight-Scott J, Rakes G, Platts-Mills TA, de Lange EE, et al. Hyperpolarized 3He MR lung ventilation imaging in asthmatics: preliminary findings. J Magn Reson Imaging. 2001;13(3):378–384. doi: 10.1002/jmri.1054. [DOI] [PubMed] [Google Scholar]
- 13.de Lange EE, Altes TA, Patrie JT, Gaare JD, Knake JJ, Mugler JP, 3rd, Platts-Mills TA. Evaluation of asthma with hyperpolarized helium-3 MRI: correlation with clinical severity and spirometry. Chest. 2006;130(4):1055–1062. doi: 10.1378/chest.130.4.1055. [DOI] [PubMed] [Google Scholar]
- 14.Tzeng YS, Lutchen K, Albert M. The difference in ventilation heterogeneity between asthmatic and healthy subjects quantified using hyperpolarized 3He MRI. J Appl Physiol (1985) 2009;106(3):813–822. doi: 10.1152/japplphysiol.01133.2007. [DOI] [PubMed] [Google Scholar]
- 15.Costella S, Kirby M, Maksym GN, McCormack DG, Paterson NA, Parraga G. Regional pulmonary response to a methacholine challenge using hyperpolarized (3)He magnetic resonance imaging. Respirology. 2012;17(8):1237–1246. doi: 10.1111/j.1440-1843.2012.02250.x. [DOI] [PubMed] [Google Scholar]
- 16.Svenningsen S, Kirby M, Starr D, Leary D, Wheatley A, Maksym GN, et al. Hyperpolarized (3) He and (129) Xe MRI: differences in asthma before bronchodilation. J Magn Reson Imaging. 2013;38(6):1521–1530. doi: 10.1002/jmri.24111. [DOI] [PubMed] [Google Scholar]
- 17.Fain SB, Gonzalez-Fernandez G, Peterson ET, Evans MD, Sorkness RL, Jarjour NN, et al. Evaluation of structure-function relationships in asthma using multidetector CT and hyperpolarized He-3 MRI. Acad Radiol. 2008;15(6):753–762. doi: 10.1016/j.acra.2007.10.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kirby M, Svenningsen S, Owrangi A, Wheatley A. Hyperpolarized 3He and 129Xe MR imaging in healthy volunteers and patients with chronic obstructive pulmonary disease. Radiology. 2012 doi: 10.1148/radiol.12120485. [DOI] [PubMed] [Google Scholar]
- 19.Koshak EA. Classification of asthma according to revised 2006 GINA: Evolution from severity to control. Ann Thorac Med. 2007;2(2):45–46. doi: 10.4103/1817-1737.32228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Driehuys B, Cates GD, Miron E, Sauer K, Walter DK, Happer W. High-volume production of laser-polarized Xe-129. Applied Physics Letters. 1996;69(12):1668–1670. [Google Scholar]
- 21.He M, Robertson SH, Kaushik SS, et al. Dose and pulse sequence considerations for hyperpolarized 129Xe ventilation MRI. Magn Reson Imaging. 2015 Sep;33(7):877–85. doi: 10.1016/j.mri.2015.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.He M, Kaushik SS, Robertson SH, Freeman MS, Virgincar RS, McAdams HP, Driehuys B. Extending semiautomatic ventilation defect analysis for hyperpolarized (129)Xe ventilation MRI. Acad Radiol. 2014;21(12):1530–1541. doi: 10.1016/j.acra.2014.07.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Svenningsen S, Kirby M, Starr D, Coxson HO, Paterson NA, McCormack DG, Parraga G. What are ventilation defects in asthma? Thorax. 2014;69(1):63–71. doi: 10.1136/thoraxjnl-2013-203711. [DOI] [PubMed] [Google Scholar]
- 24.Nair P, Dasgupta A, Brightling CE, Chung KF. How to diagnose and phenotype asthma. Clinics in Chest Medicine. 2012;33:445–457. doi: 10.1016/j.ccm.2012.05.003. [DOI] [PubMed] [Google Scholar]
- 25.Miller MR. Standardisation of spirometry. European Respiratory Journal. 2005;26(2):319–338. doi: 10.1183/09031936.05.00034805. [DOI] [PubMed] [Google Scholar]
- 26.van der Wiel E, Hacken NHT, Postma DS, van den Berge M. Small-airways dysfunction associates with respiratory symptoms and clinical features of asthma: A systematic review. Journal of Allergy and Clinical Immunology. 2013;131(3):646–657. doi: 10.1016/j.jaci.2012.12.1567. [DOI] [PubMed] [Google Scholar]
- 27.Ciprandi G, Capasso M, Tosca M, Salpietro C, Salpietro A, Marseglia G, La Rosa M. A forced expiratory flow at 25–75% value <65% of predicted should be considered abnormal: a real-world, cross-sectional study. Allergy Asthma Proc. 2012 Jan-Feb;33(1):5–8. doi: 10.2500/aap.2012.33.3524. [DOI] [PubMed] [Google Scholar]
- 28.Kirby M, Svenningsen S, Owrangi A, Wheatley A. Hyperpolarized 3He and 129Xe MR imaging in healthy volunteers and patients with chronic obstructive pulmonary disease. Radiology. 2012 doi: 10.1148/radiol.12120485. [DOI] [PubMed] [Google Scholar]
- 29.Mathew L, Evans A, Ouriadov A, et al. Hyperpolarized 3He Magnetic Resonance Imaging of Chronic Obstructive Pulmonary Disease: Reproducibility at 3.0 Tesla. Acad Radiol. 2008;15:1298–1311. doi: 10.1016/j.acra.2008.04.019. [DOI] [PubMed] [Google Scholar]
- 30.Reddel HK, Taylor DR, Bateman ED, et al. An Official American Thoracic Society/European Respiratory Society Statement: Asthma Control and Exacerbations. Am J Respir Crit Care Med. 2009;180(1):59–99. doi: 10.1164/rccm.200801-060ST. [DOI] [PubMed] [Google Scholar]
- 31.Janssens JP, Pache JC, Nicod LP. Physiological changes in respiratory function associated with ageing. Eur Respir J. 1999;13:197–205. doi: 10.1034/j.1399-3003.1999.13a36.x. [DOI] [PubMed] [Google Scholar]
- 32.Parraga G, Mathew L, Etemad-Rezai R, McCormack DG, Santyr GE. Hyperpolarized 3He magnetic resonance imaging of ventilation defects in healthy elderly volunteers: initial findings at 3.0 Tesla. Acad Radiol. 2008 Jun;15(6):776–85. doi: 10.1016/j.acra.2008.03.003. [DOI] [PubMed] [Google Scholar]
- 33.Horn FC, Tahir BA, Stewart NJ, et al. Lung ventilation volumetry with same-breath acquisition of hyperpolarized gas and proton MRI. NMR Biomed. 2014 Dec;27(12):1461–7. doi: 10.1002/nbm.3187. [DOI] [PubMed] [Google Scholar]