Using Hyperpolarized 129Xe MRI to Quantify the Pulmonary Ventilation Distribution

Abstract

Rationale and objectives: Ventilation heterogeneity is impossible to detect with spirometry. Alternatively, pulmonary ventilation can be imaged three-dimensionally using inhaled 129Xe magnetic resonance imaging (MRI). To date, such images have been quantified primarily based on ventilation defects. Here, we introduce a robust means to transform 129Xe MRI scans such that the underlying ventilation distribution and its heterogeneity can be quantified.

Materials and methods: Quantitative 129Xe ventilation MRI was conducted in 12 younger (24.7 ± 5.2 years) and 10 older (62.2 ± 7.2 years) healthy individuals, as well as in 9 younger (25.9 ± 6.4 yrs) and 10 older (63.2 ± 6.1 years) asthmatics. The younger healthy population was used to establish a reference ventilation distribution and thresholds for six intensity bins. These bins were used to display and quantify the ventilation defect region (VDR), the low ventilation region (LVR), and the high ventilation region (HVR).

Results: The ventilation distribution in young subjects was roughly Gaussian with a mean and standard deviation of 0.52 ± 0.18, resulting in VDR = 2.1 ± 1.3%, LVR = 15.6 ± 5.4%, and HVR = 17.4 ± 3.1%. Older healthy volunteers exhibited a significantly right-skewed distribution (0.46 ± 0.20, P = 0.034), resulting in significantly increased VDR (7.0 ± 4.8%, P = 0.008) and LVR (24.5 ± 11.5%, P = 0.025). In the asthmatics, VDR and LVR increased in the older population, and HVR was significantly reduced (13.5 ± 4.6% vs 18.9 ± 4.5%, P = 0.009). Quantitative 129Xe MRI also revealed altered ventilation heterogeneity in response to albuterol in two asthmatics with normal spirometry.

Conclusions: Quantitative 129Xe MRI provides a robust and objective means to display and quantify the pulmonary ventilation distribution, even in subjects who have airway function impairment not appreciated by spirometry.

Keywords: Asthma; aging; albuterol.

Conflict of interest statement

Conflict of Interest: MH, LGQ, YTH have no conflict of interest relevant to the study. BD is founder of Polarean, which is involved in the commercialization of hyperpolarized 129Xe MRI technology.

Copyright © 2016 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

(a) 129Xe MRI as acquired. (b) after correction for B1 inhomogeneity. (c) 1H thoracic cavity image registered to 129Xe MRI. (d) Detection of vascular structures within the 1H image. (e) Segmentation and 1 pixel erosion to create a thoracic cavity mask with vascular structures removed, (f) histogram of 129Xe intensities within the mask, (f) 129Xe histogram after rescaling by the top percentile of all intensities. (h) Binning map generated after applying thresholds.

Figure 2

(A) Rescaled ventilation distribution for the 10 young normal individuals. (B) Average distribution for the 10 subjects has a mean of 0.52 and SD of 0.18. This was used to define the thresholds and widths of 6 bins used to quantify and map the ventilation distribution.

Figure 3

Examples of 129Xe-derived ventilation distributions in a younger and older normal subject. (A) In a 27-year-old healthy subject with FEV1of 86%, the 129Xe ventilation image shows very few ventilation defects. The associated histogram is similar to that of normal young controls, with mean of 0.54 and SD of 0.18. (B) In a 58-year-old healthy subject with FEV1 of 102%, the ventilation image shows areas of ventilation defects (red) and low ventilation (orange). The associated ventilation histogram shifted towards lower values compared to that of normal young controls, with a mean of 0.40 and SD of 0.20.

Figure 4

Comparison of ventilation distribution parameters between the younger and the older control groups. (A) Ventilation distribution histogram from the older control (OC) group (red) is overlaid with that from the young control (YC) group (blue). The ventilation distribution in the older healthy subjects shifted towards lower values compared to the younger group. (B) Comparison of ventilation distribution parameters between the young and the old groups shows significant differences for all parameters except HVR. (*- indicates p ≤ 0.05.)

Figure 5

Representative ventilation distributions in a younger and older asthma subject. (A) The younger asthma subject with FEV1of 76% exhibits a ventilation image with very few ventilation defects. The associated histogram is similar to that of normal young controls. (B) The older asthma subject with FEV1 of 53% has a ventilation image with significant areas of ventilation defects (red) and low ventilation (orange). The associated ventilation histogram was significantly right-skewed compared to that of normal young controls.

Figure 6

Comparison of ventilation distributions between the younger and the older asthma groups. (A) Ventilation distribution histogram from the older asthma (OA) group (red) is overlaid with that from the younger asthma (YA) group (blue). In the older asthma group the distribution is shifted towards lower values compared to the younger asthmatics. (B) Comparison of ventilation distribution parameters between the younger and the older asthma groups shows significant differences in all parameters * p ≤ 0.05.

Figure 7

Comparisons between FEV1 and 129Xe MRI-derived ventilation distribution in patients with asthma. (A) A patient with normal FEV1, but abnormal 129Xe MRI. (B) A patient with abnormal FEV1, but normal 129Xe MRI. (C) and (D) Two patients with moderate asthma with similarly low FEV1, but markedly different ventilation patterns.

Figure 8

Ventilation distribution in a patient with mild intermittent asthma who had positive bronchodilator response to albuterol based on FEV1 criteria. (A) The binning map (top) and histogram (bottom) before albuterol. (B) The binning map (top) and histogram (bottom) after albuterol showed decreased VDR and LVR, and a distribution matching that of young normal controls after bronchodilator.

Figure 9

Ventilation distribution in a patient with mild intermittent asthma who did not respond to albuterol based on FEV1 criteria. (A) The binning map (top) and histogram (bottom) before albuterol. (B) The binning map (top) and histogram (bottom) after albuterol. The ventilation distribution histogram after albuterol showed decreased VDR but increased LVR at the expense of normally ventilated regions. This may be the result of more normal airways dilating and redistributing ventilation.