Emphysema Index Based on Hyperpolarized ³He or ¹²⁹Xe Diffusion MRI: Performance and Comparison with Quantitative CT and Pulmonary Function Tests

Abstract

Background

Apparent diffusion coefficient (ADC) maps of inhaled hyperpolarized gases have shown promise in the characterization of emphysema in patients with chronic obstructive pulmonary disease (COPD), yet an easily interpreted quantitative metric beyond mean and standard deviation has not been established.

Purpose

To introduce a quantitative framework with which to characterize emphysema burden based on hyperpolarized helium 3 (³He) and xenon 129 (¹²⁹Xe) ADC maps and compare its diagnostic performance with CT-based emphysema metrics and pulmonary function tests (PFTs).

Materials and Methods

Twenty-seven patients with mild, moderate, or severe COPD and 13 age-matched healthy control subjects participated in this retrospective study. Participants underwent CT and multiple b value diffusion-weighted ³He and ¹²⁹Xe MRI examinations and standard PFTs between August 2014 and November 2017. ADC-based emphysema index was computed separately for each gas and b value as the fraction of lung voxels with ADC values greater than in the healthy group 99th percentile. The resulting values were compared with quantitative CT results (relative lung area <−950 HU) as the reference standard. Diagnostic performance metrics included area under the receiver operating characteristic curve (AUC). Spearman rank correlations and Wilcoxon rank sum tests were performed between ADC-, CT-, and PFT-based metrics, and intraclass correlation was performed between repeated measurements.

Results

Thirty-six participants were evaluated (mean age, 60 years ± 6 [standard deviation]; 20 women). ADC-based emphysema index was highly repeatable (intraclass correlation coefficient > 0.99) and strongly correlated with quantitative CT (r = 0.86, P < .001 for ³He; r = 0.85, P < .001 for ¹²⁹Xe) with high AUC (≥0.93; 95% confidence interval [CI]: 0.85, 1.00). ADC emphysema indices were also correlated with percentage of predicted diffusing capacity of lung for carbon monoxide (r = −0.81, P < .001 for ³He; r = −0.80, P < .001 for ¹²⁹Xe) and percentage of predicted residual lung volume divided by total lung capacity (r = 0.65, P < .001 for ³He; r = 0.61, P < .001 for ¹²⁹Xe).

Conclusion

Emphysema index based on hyperpolarized helium 3 or xenon 129 diffusion MRI provides a repeatable measure of emphysema burden, independent of gas or b value, with similar diagnostic performance as quantitative CT or pulmonary function metrics.

© RSNA, 2020

Online supplemental material is available for this article.

See also the editorial by Schiebler and Fain in this issue.

Summary

Emphysema index based on hyperpolarized xenon 129 diffusion MRI is introduced as a robust, nonionizing alternative to CT for quantifying emphysema burden in chronic obstructive pulmonary disease.

Key Results

• ■ In 26 patients with chronic obstructive pulmonary disease and 10 age-matched healthy volunteers, emphysema index based on apparent diffusion coefficients (ADCs) obtained with diffusion-weighted MRI with hyperpolarized helium 3 (³He) or xenon 129 (¹²⁹Xe) showed strong correlations with quantitative CT (r = 0.86, P < .001 for ³He; r = 0.85, P < .001 for ¹²⁹Xe).

• ■ The ³He and ¹²⁹Xe ADC-based emphysema indices were correlated with percentage of predicted diffusing capacity of lung for carbon monoxide (r = −0.81, P < .001; r = −0.80, P < .001, respectively) and percentage of predicted residual lung volume divided by total lung capacity (r = 0.65, P < .001; r = 0.61, P < .001, respectively).

• ■ The ³He ADC-based emphysema index was strongly correlated with ¹²⁹Xe ADC-based emphysema index (r = 0.95, P < .001), and both were highly repeatable (intraclass correlation coefficient > 0.99).

Introduction

Emphysema is a disease characterized by irreversible destruction of alveolar walls that causes loss of lung elastic recoil and impaired gas exchange. Pulmonary function tests (PFTs) provide global measures of lung function that are commonly used to diagnose emphysema and, more broadly, chronic obstructive pulmonary disease (COPD) (1). Parenchymal tissue destruction is chiefly visualized with CT, and its severity can be assessed by using quantitative CT methods including the relative lung area with low attenuation (RA)—that is, with attenuation coefficients lower than a particular Hounsfield unit threshold (eg, RA₉₅₀), which is sometimes referred to as the emphysema index (2–5).

Apparent diffusion coefficients (ADCs) obtained from diffusion-weighted MRI of inhaled hyperpolarized helium 3 (³He) or xenon 129 (¹²⁹Xe) gas provide a relative measure of alveolar size and offer substantial promise for assessing the emphysematous lung. Elevated ADC values have been established to represent airspace enlargement due to emphysematous tissue destruction in animal (6–8) and human studies that used ³He (9–12) and hyperpolarized ¹²⁹Xe (13–20).

This direct correspondence between emphysema and elevated ADC has sparked investigations into whether diffusion-weighted MRI of inhaled hyperpolarized gases might provide useful supplemental information beyond standard PFTs, as well as a nonionizing imaging alternative to CT, for characterizing emphysema in patients with COPD. Numerous studies have compared metrics derived from hyperpolarized gas diffusion MRI with results of quantitative CT (16–18,21,22), PFTs (9,11,15,17–19,21,22), and direct histopathologic assessment (23,24). Recent efforts have focused specifically on establishing the capabilities of ¹²⁹Xe (16,17,19,21,25) owing to the extremely limited availability of ³He and the additional ability of dissolved phase ¹²⁹Xe to directly probe gas exchange (14,20).

Currently, hyperpolarized gas MRI is more commonly used to assess airflow limitation, which is a functional counterpart to emphysema in COPD. Ventilation defect percentage derived from ¹²⁹Xe or ³He ventilation images has recently emerged as a clinically relevant whole-lung metric for quantifying ventilation impairment (16,26). To date, however, an analogous metric has not been developed for ADC measurements of emphysematous tissue destruction. Whole-lung ADC metrics have been limited to means and standard deviations, which are difficult to interpret in the absence of a unifying framework because the measured ADC value depends not only on airspace geometry but also on the b value and timing of the diffusion-sensitizing gradients (27), and the ADC value by itself does not directly reflect emphysema burden.

The purpose of this retrospective study was to introduce and test a quantitative framework for characterizing emphysema burden by using hyperpolarized gas ADC (28). We proposed a definition of emphysema index based on the distribution of ADC values in human lung and assessed its diagnostic performance by comparing ADC-based emphysema index, measured with both ³He and ¹²⁹Xe across a range of different b values, with PFT results and quantitative CT metrics in a study sample composed of patients with varying stages of COPD and age-matched healthy volunteers.

Materials and Methods

Study Participants

This retrospective study complied with the Health Insurance Portability and Accountability Act and was approved by the University of Virginia institutional review board. All study participants provided written informed consent. Twenty-seven patients with clinical diagnosis of COPD and 13 age-matched healthy volunteers underwent imaging between August 2014 and November 2017 with ³He MRI, ¹²⁹Xe MRI, and CT. Exclusion criteria are provided in Appendix E1 (online).

Use of PFTs

Standard PFTs, including spirometry, plethysmography, and diffusing capacity of lung for carbon monoxide (DLco), were performed specifically for research according to American Thoracic Society guidelines in all participants within 1 week of imaging. PFT metrics included forced expiratory volume in 1 second (FEV₁), forced vital capacity (FVC), residual lung volume (RV), and total lung capacity (TLC). COPD was classified as stage I–IV according to Global Initiative for Chronic Obstructive Lung Disease criteria based on FEV₁ percent predicted (1). For group comparisons, stages I and II were grouped together as mild-moderate COPD, and stages III and IV were grouped together as severe COPD.

³He and ¹²⁹Xe Polarization and Delivery

The ³He gas was polarized to 50%–60% with a homemade polarizer. Isotopically enriched ¹²⁹Xe gas was polarized to 30%–50% with commercial systems (XeBox-10 [Xemed, Durham, NH] or 9820 Hyperpolarizer [Polarean Imaging, Durham, NC]). For each participant, 0.4 L or 1.0 L of the hyperpolarized ³He or ¹²⁹Xe gas, respectively, plus enough medical-grade nitrogen to reach a total volume equal to one-third FVC was dispensed into a plastic bag (Tedlar plastic bag; Jensen Inert Products, Coral Springs, Fla) for each helium or xenon scan. Participants exhaled to RV, then inhaled the contents of the bag through flexible tubing and held their breath for up to 15 seconds during image acquisition.

Diffusion-weighted MRI

Hyperpolarized ³He and ¹²⁹Xe MRI images were obtained in separate breath holds during the same imaging session with a 1.5-T scanner (Avanto; Siemens Healthineers, Erlangen, Germany) and two different transmit-receive vest-shaped radiofrequency coils (Clinical MR Solutions, Brookfield, Wis). Study participants were positioned supine in the scanner and a brief proton scout image was obtained at partial-inspiration breath hold. Following gas inhalation, a two-dimensional spoiled gradient-echo pulse sequence equipped with bipolar diffusion-sensitizing gradients was used to acquire axial image sets at five different b values (0, 1.6, 3.6, 5.6, and 7.6 sec/cm² for ³He and 0, 12.5, 25, 37.5, and 50 sec/cm² for ¹²⁹Xe) in a single breath hold. Rapid (1 second) axial proton scans with spiral readout were obtained at the same slice positions during the same breath hold. Following an identical hyperpolarized gas inhalation and breath hold, a second diffusion-weighted scan was performed at the same slice positions as the multiple b value scan by using a single pair of b values (0, 1.6 sec/cm² for ³He; 0, 10 sec/cm² for ¹²⁹Xe) for repeatability analysis. Pulse sequence parameters for all hyperpolarized gas scans included five 40-mm axial slices with 8-mm interslice gap; 234-mm × 416-mm field of view; 36 × 64 acquisition matrix; readout bandwidth, 200 Hz/pixel; and flip angle, 5°. Parameters for ³He and ¹²⁹Xe included repetition time msec/echo time msec, 11/6.8 and 18/14, respectively.

Image analysis was performed with software (Matlab version 2019b; MathWorks, Natick, Mass). All data analysis for this study was performed by authors (S.T., G.W.M., and W.J.G.; 4, 18, and 3 years of experience, respectively). Four sets of ADC maps were computed from each multiple b value diffusion-weighted scan in the standard manner (9,13) by pairing the b of 0 sec/cm² image set with each of the nonzero b value image sets, and one set of ADC maps was computed from the single ADC scan. Voxels with signal-to-noise ratio less than 15 were excluded from the ADC calculation (29). Fraction of analyzed lung volume was computed as the fraction of available lung voxels included in the ADC analysis, based on lung masks constructed from the same-breath-hold proton images. Additional pulse sequence and analysis details are provided in the Appendix E1 (online).

Quantitative CT

Within 1 week of MRI and with institutional review board approval, nonenhanced chest CT scans were performed for research only in the axial orientation by using a 64-section multi–detector row CT scanner (Somatom Definition Flash; Siemens). Scan parameters included 0.75-mm slice thickness and 0.5-mm slice interval. Participants were coached to exhale to RV, then inhale from a Tedlar plastic bag filled with nitrogen to one-third FVC to achieve the same lung volume as in the hyperpolarized gas MRI scans, as has been done elsewhere (16,18). CT images were also analyzed with Matlab. A three-dimensional nonlung flood-fill method and global threshold were used to segment the lung parenchyma (18). Relative lung areas below −950 HU (RA₉₅₀) and −910 HU (RA₉₁₀) in lung parenchyma were computed from each segmented CT scan. Additional details are provided in Appendix E1 (online).

Calculation of ADC Emphysema Index

We defined ADC emphysema index as the fraction of lung voxels having ADC values greater than the 99th percentile of all lung voxel values from all healthy volunteers, for a given gas and b value. Repeatability was assessed by comparing emphysema indices computed from the single-ADC scan with those computed from the first b value pair of the multi-ADC scan for each participant.

Statistical Analysis

To facilitate comparison between different emphysema metrics (RA₉₅₀, ³He-based ADC emphysema index, and ¹²⁹Xe-based ADC emphysema index) across study participants, we selected the top of the 99% confidence interval (CI) of each metric’s healthy group distribution as a threshold to separate apparently healthy from apparently emphysematous ranges for each metric. That is, the participant was classified as emphysematous according to a given metric if its value was above that metric’s emphysematous threshold. These classifications were also used to calculate measures of diagnostic performance: sensitivity, specificity, and area under the receiver operating characteristic curve (AUC), where RA₉₅₀ was the reference standard and ADC emphysema index was the index test.

Intraclass correlation coefficient was computed for repeatability data using statistical software (SPSS version 26.0; IBM, Armonk, NY). All other statistical analyses were performed with Matlab. Spearman rank correlation coefficients and associated P values were calculated between all relevant MRI, CT, and PFT metrics. Healthy, mild-moderate COPD, and severe COPD group distributions were compared by using the Wilcoxon rank sum test. Bland-Altman analysis was used to evaluate systematic bias between ³He-based and ¹²⁹Xe-based ADC emphysema indices and repeatability comparisons. P < .05 was considered to indicate a statistically significant difference.

Results

Data for 36 participants including 10 healthy volunteers (mean age, 58 years ± 6 [standard deviation]; seven women) and seven participants with stage I (59 years ± 4; four women), nine with stage II (59 years ± 6; five women), nine with stage III (62 years ± 8; four women), and one with stage IV (58 years old; male) COPD were analyzed for this report (Fig 1). All healthy volunteers had FEV₁/FVC greater than 0.7 and FEV₁ greater than 80% predicted, and all participants with COPD had FEV₁/FVC less than 0.7 and FEV₁ less than 80% predicted (1). A complete list of participant demographics, disease stage, and PFT results appears in Table 1.

Figure 1:

Study flowchart details exclusion criteria. Among 40 participants who underwent initial imaging, data were excluded from analysis owing to substantial smoking history (healthy participants only) or absence of diffusion-weighted MRI scans with either gas. COPD = chronic obstructive pulmonary disease, ³He = helium 3, PFT = pulmonary function test, ¹²⁹Xe = xenon 129.

Table 1:

Participant Demographics, Disease Stage, and Pulmonary Metrics

Example histograms containing all lung voxel ADC values, computed from the lowest b value pairs for each gas and grouped according to COPD severity, are shown in Figure 2. ADC distributions for the mild-moderate COPD group and severe COPD group were broader and centered at progressively higher ADC values than in the healthy group. Representative ³He and ¹²⁹Xe ADC maps and CT images from three participants are shown in Figure 3. ³He-based and ¹²⁹Xe-based ADC emphysema indices were both highly repeatable (intraclass correlation coefficient, 0.998 and 0.993 for ³He and ¹²⁹Xe, respectively) with no observable bias in Bland-Altman plots (Fig 4).

Figure 2:

Histograms of all lung voxel apparent diffusion coefficient (ADC) values from all healthy volunteers and patients with chronic obstructive pulmonary disease (COPD) for, A, hyperpolarized helium 3 (³He) and, B, hyperpolarized xenon 129 (¹²⁹Xe) ADC computed from first (smallest) b value. All distributions are self-normalized to have equal areas to facilitate visual comparison. Vertical lines mark 99th percentile of healthy distribution. ³He and ¹²⁹Xe ADC values at 99th percentile of healthy distributions were 0.40 and 0.065 cm²/sec for first b value, 0.32 and 0.055 cm²/sec for second b value, 0.29 and 0.049 cm²/sec for third b value, and 0.25 and 0.047 cm²/sec for fourth b value, respectively.

Figure 3:

Representative axial CT lung images, segmented CT images with voxels having attenuation coefficient levels less than −950 HU and −910 HU displayed in yellow and red, respectively, and corresponding helium 3 (³He) and xenon 129 (¹²⁹Xe) apparent diffusion coefficient (ADC) maps. In top row, CT image in healthy 60-year-old man shows uniformly dense parenchyma throughout lung and uniformly low ADC values in both ³He and ¹²⁹Xe ADC maps. In middle row, CT in 75-year-old man with stage III chronic obstructive pulmonary disease (COPD) depicts large areas of lung parenchyma with low attenuation coefficients. Clear visual concordance is present between these areas in CT image and elevated ADC values in corresponding ³He and ¹²⁹Xe ADC maps. CT RA₉₅₀ and ADC emphysema indices are well outside healthy range (>0.04), indicating that with all techniques participant is identified as emphysematous. In bottom row, clear difference in performance is seen in 68-year-old woman with stage III COPD. Very few pixels in CT image cross −950 HU threshold, whereas both ADC maps show extensive lung regions with elevated values. Accordingly, both ADC-based emphysema indices for this participant lie well outside healthy range, but CT RA₉₅₀ does not (RA₉₅₀ = 0.01; ADC emphysema index = 0.25 and 0.21 based on ³He and ¹²⁹Xe, respectively). RA₉₅₀ = fraction of CT lung voxels with attenuation coefficients less than −950 HU.

Figure 4:

A, Scatterplot of apparent diffusion coefficient (ADC) emphysema indices based on xenon 129 (¹²⁹Xe) versus helium 3 (³He) in all participants using smallest b values. B, Scatterplot of repeatability measurements of ³He-based ADC emphysema indices (y = 1.01× + 0.00, r² > 0.99). C, Scatterplot of repeatability measurements of ¹²⁹Xe-based ADC emphysema indices (y = 1.03× − 0.01, r² = 0.99). D, Bland-Altman plot depicts bias between ³He-based and ¹²⁹Xe-based ADC emphysema indices. Percentage mean ± standard deviation (SD) is 9.8% ± 8.6. Significant correlation is observed between percentage emphysema index difference and mean emphysema index (r = 0.87, P < .001). E, Bland-Altman plot depicts no bias between ³He-based ADC emphysema index repeatability measurements. Percentage mean ± SD is −0.2% ± 1.4. F, Bland-Altman plot depicts no bias between ¹²⁹Xe-based ADC emphysema index repeatability measurements. Percentage mean ± SD is 0.4 ± 3.0. Dashed horizontal lines represent 95% limits of agreement in all Bland-Altman plots. All plots share same legend presented in A.

RA₉₅₀ and RA₉₁₀ were both strongly correlated with ³He-based (r = 0.86, P < .001 for RA₉₅₀; r = 0.84, P < .001 for RA₉₁₀) and ¹²⁹Xe-based (r = 0.85, P < .001; r = 0.85, P < .001) ADC emphysema indices computed by using the lowest b value pair for each gas (Figs 5 and E3 [online]). Correlation with ADC emphysema indices computed using higher b values were similarly strong (Table 2). However, a closer look at the data points in Figure 5 suggests important differences between healthy and COPD group distributions along each axis. Nearly all healthy volunteers (10 of 10 with ³He and nine of 10 with ¹²⁹Xe) and 15% of patients with COPD (four of 26) had data points in the lower left quadrant, corresponding to apparently healthy based on both CT and ADC. Most patients with COPD (17 of 26 [65%]) had data points in the upper right quadrant, corresponding to apparently emphysematous based on both CT and ADC. However, a substantial minority of the patients with COPD (five of 26 [19%], four with mild-moderate COPD) had emphysema indices solidly in the lower right quadrant, appearing “healthy” based on quantitative CT but emphysematous based on ADC, whereas no patients appeared emphysematous based on CT but healthy based on ADC (upper left quadrant). Therefore, ADC emphysema index showed near-perfect sensitivity in our study sample (17 of 17 = 100% for the lowest b-value pair with both gases [95% CI: 94%, 100%]) but somewhat lower specificity (14 of 19 = 74% for ³He [95% CI: 49%, 99%]; 13 of 19 = 68% for ¹²⁹Xe [95% CI: 42%, 94%]). The AUC was also relatively high for both gases (≥0.94 for all ³He b values [95% CI: 0.86, 1.00]; ≥0.92 for all ¹²⁹Xe b values [95% CI: 0.83, 1.00]) (Table 3, Fig E4 [online]).

Figure 5:

Superimposed scatterplots of CT RA₉₅₀ versus lowest b-value pair emphysema index based on both helium 3 (³He) apparent diffusion coefficient (ADC) and xenon 129 (¹²⁹Xe) ADC (circles and squares represent ³He-based and ¹²⁹Xe-based ADC emphysema indices, respectively). Solid lines are drawn at upper limit of 99% confidence interval of healthy population to create “apparently healthy” quadrant (lower left) and “apparently emphysematous” quadrant (upper right). Lower right quadrant contains participants whose emphysema indices lie in healthy range based on CT but lie in emphysematous range based on ADC (five ³He-based and five ¹²⁹Xe-based ADC emphysema index data points appear in lower right quadrant). No data points appear in upper left quadrant. RA₉₅₀ = fraction of CT lung voxels with attenuation coefficients less than −950 HU.

Table 2:

Correlation Coefficients between ADC-based Emphysema Indices and CT Metrics

Table 3:

Diagnostic Performance Measures for ADC Emphysema Index with Use of RA₉₅₀

Diagnostic performance measures were similar when RA₉₁₀ replaced RA₉₅₀ as the reference standard (Table E3 [online]). The −950 HU threshold is the current standard for CT emphysema index at full lung inflation (3,5). Because our CT scans were performed at a lower inflation level, the higher (−910 HU) threshold corresponding to a previous standard (2) was analyzed for comparison. However, the use of an RA threshold of −910 HU to calibrate the ADC emphysema index percentile threshold resulted in emphysema indices approaching 20% for healthy participants (Fig E3 [online]). Therefore, calibration against the standard RA threshold of −950 HU, which yielded much lower estimates of emphysema burden in the healthy group, seems more reasonable.

Receiver operating characteristic analysis revealed stronger separation between the healthy group and the mild-moderate COPD group with either ³He-based or ¹²⁹Xe-based ADC emphysema index (AUC, 0.87; 95% CI: 0.71, 1 for both) than with RA₉₅₀ (AUC, 0.76; 95% CI: 0.56, 0.96). The stronger separation persisted regardless of the b value used, for both gases (Figs 6, B, and E5; Table E2 [online]). Because hyperpolarized gas ADC is based on direct probing of airspace size, whereas CT strictly measures tissue density, one possible explanation for these findings is that airway wall thickening accompanies airspace enlargement in the early stages of emphysema, partially offsetting the net decrease in tissue density due to airspace enlargement. This explanation is consistent with recent results involving dissolved phase ¹²⁹Xe, which showed that elevated alveolar wall thickness based on hyperpolarized ¹²⁹Xe MRI may be an indicator of early stage lung disease and emphysema (20).

Figure 6:

A, Box plots of CT RA₉₅₀ and emphysema index based on hyperpolarized helium 3 (³He) and xenon 129 (¹²⁹Xe) ADCs (first b value pair). B, Box plot of emphysema indices based on different b value pair ¹²⁹Xe ADC. * P < .05, ** P < .01. *** P < .001. Horizontal lines inside boxes are medians; box edges indicate 25th and 75th percentiles of distributions. ADC = apparent diffusion coefficient. RA₉₅₀ = fraction of CT lung voxels with attenuation coefficients less than −950 HU.

³He-based ADC emphysema index showed strong correlation with ¹²⁹Xe-based ADC emphysema index (r = 0.95, P < .001; Fig 4, A). However, ³He-based values were consistently higher than ¹²⁹Xe-based values. Bland-Altman analysis showed an average bias of 9.8% ± 8.6 (standard deviation), and the difference between ³He-based and ¹²⁹Xe-based ADC emphysema indices was correlated with their mean (r = 0.87, P < .001) (Fig 4, D). This bias probably relates to previous findings that ¹²⁹Xe images frequently show larger ventilation defects than ³He images in participants with COPD, presumably because ³He is more diffusible than ¹²⁹Xe and can better penetrate restricted airways or access obstructed lung regions via collateral ventilation (16,21). Regions appearing ventilated with use of ³He but nonventilated with ¹²⁹Xe also tend to be emphysematous, which would increase emphysema burden measured by using ³He ADC. An example of this phenomenon is evident in the ADC maps displayed in the middle row of Figure 3, in which there are highly emphysematous regions in the ³He ADC map that appear nonventilated in the ¹²⁹Xe ADC map. This phenomenon is also reflected in the percentage of available lung used for ADC analysis (Table E1 [online]), which was consistently larger for ³He.

Finally, we note that percentage of predicted DLco, which is often used by pulmonologists as a marker for emphysema, was moderately correlated with each of ³He-based (r = −0.81, P < .001) and ¹²⁹Xe-based (r = −0.80, P < .001) emphysema indices and RA₉₅₀ (r = −0.61, P < .001) in participants with COPD (Fig E6a–E6c [online]). Percentage of predicted RV/TLC, another marker of emphysema, was also moderately correlated with each of ³He-based (r = 0.65, P < .001) and ¹²⁹Xe-based (r = 0.61, P < .001) emphysema indices and RA₉₅₀ (r = 0.57, P < .001) (Fig E6d–E6f [online]). Other relevant correlations among PFT metrics, diffusion-weighted MRI, and CT metrics are reported in Table 4.

Table 4:

Correlation Coefficients between ADC, CT Metrics and Pulmonary, Plethysmography Metrics and Pack-Years

Discussion

Apparent diffusion coefficient (ADC) maps of inhaled hyperpolarized gases have shown promise for characterizing emphysema in patients with chronic obstructive pulmonary disease (COPD), yet an easily interpreted quantitative metric has not been established. Inspired by emphysema index based on CT Hounsfield units, we proposed a similar definition of emphysema index based on ADC. In a study sample of 36 participants, we found that emphysema indices based on helium 3 (³He) and xenon 129 (¹²⁹Xe) ADC were strongly correlated (r = 0.95, P > .001) and both showed strong correlation with relative lung area with low attenuation (RA) on CT images (r ≥ 0.85, P < .001). The ³He-based and ¹²⁹Xe-based ADC emphysema indices were also highly repeatable (intraclass correlation coefficient > 0.99) and showed highly significant differences between healthy, mild-moderate, and severe COPD groups, independent of the b values used (P < .01). Both ³He-based and ¹²⁹Xe-based ADC emphysema indices were also correlated with pulmonary function metrics traditionally used to characterize emphysema, including diffusing capacity of lung for carbon monoxide (r ≥ 0.80, P < .001) and residual lung volume divided by total lung capacity (r ≥ 0.61, P < .002), with a degree of correlation similar to that of quantitative CT (r ≥ 0.57, P < .001).

The excellent repeatability and strong correlation between ³He and ¹²⁹Xe ADC emphysema indices in our study sample echoed previous findings based on mean ADC values (16,18,19), but in a larger study sample containing more patients with early stage disease. In our data set, 25% of the mild-moderate COPD group (four of 16 patients) had ADC emphysema indices outside the healthy range but RA₉₅₀ values inside the healthy range, yet there were no patients with COPD in whom the opposite was true. When RA₉₅₀ and ADC emphysema indices were used to predict mild-moderate COPD, the AUC values for ³He and ¹²⁹Xe were higher than for RA₉₅₀. Taken together, these findings suggest that ADC emphysema index may offer enhanced sensitivity to early stage disease. Similarly intriguing results have been noted in previous ADC studies (11,20). Sensitivity to early disease would be particularly useful for research, especially in pharmaceutical development. Because emphysema is currently a terminal disease with no reversible therapeutic agents commercially available, earlier detection would also facilitate earlier clinical intervention to slow disease progression.

Beyond COPD, which disproportionately affects older smokers, emphysema is frequently found in younger and more radiation-sensitive patients with idiopathic pulmonary fibrosis (30) and survivors of bronchopulmonary dysplasia (31). Dissolved phase ¹²⁹Xe has been investigated for characterizing fibrosis in idiopathic pulmonary fibrosis (32), and ADC emphysema index could provide a valuable adjunct for clinical management in these patients because comorbid emphysema is associated with poor outcomes (30).

An important limitation of our quantitative CT comparisons was that CT scans were performed at a lung inflation level of one-third FVC rather than the recommended inflation level of TLC (33). Measurements at lower inflation levels are known to underestimate emphysema burden (5), and we acknowledge that our quantitative CT results might have been different at full inflation. Limitations of ADC relative to CT include much coarser spatial resolution and longer breath-hold times. However, our results indicate that the limited spatial resolution of ADC does not compromise its ability to quantify emphysema burden, and CT-like spatial resolution is probably not necessary for useful regional analysis. Furthermore, all patients with COPD in our study tolerated the 15-second breath hold required for the ¹²⁹Xe ADC scan, and scan duration would be up to 60% shorter for a single b value nonresearch scan.

Hyperpolarized gas MRI is an expensive procedure, which raises the bar for clinical usefulness. However, it is clear that diffusion-weighted ¹²⁹Xe imaging, especially when combined with ventilation imaging, dissolved phase imaging of gas exchange, and emerging proton MRI methods for parenchyma characterization (34), provides a unique and powerful combination of functional and structural information with segmental or subsegmental regional detail. Such a diagnostic platform could offer a more precise approach to emphysema therapies such as endotracheal valve placement and lung volume reduction surgery (4) in this nascent era of precision medicine (35). In this context, the reduced ability of ¹²⁹Xe to penetrate highly emphysematous regions might not be a serious limitation, as persistent ventilation defects that do not permit ADC measurement would be candidates for regional intervention based on ventilation loss alone.

We propose herein a quantitative framework for characterizing emphysema burden based on hyperpolarized gas diffusion measurements. Within this framework, we found the basic performance of apparent diffusion coefficient (ADC) emphysema index to be largely insensitive to gas type and b value. Widespread application of ADC emphysema index would still require standardization with a particular gas (unequivocally ¹²⁹Xe, due to its abundance), inflation level, and diffusion-sensitizing gradient parameters, although our results indicate that the precise choices are not crucial. Because the clinical definition of chronic obstructive pulmonary disease (COPD) encompasses both airflow limitation and emphysema, ADC emphysema index would provide a useful complement to ventilation defect percentage for staging COPD severity and monitoring disease progression.