Visualization of morphological parenchymal changes in emphysema: Comparison of different MRI sequences to 3D-HRCT

Abstract

Purpose

Thin-section CT is the modality of choice for morphological imaging the lung parenchyma, while proton-MRI might be used for functional assessment. However, the capability of MRI to visualize morphological parenchymal alterations in emphysema is undetermined. Thus, the aim of the study was to compare different MRI sequences with CT.

Materials and methods

22 patients suffering from emphysema underwent thin-section MSCT serving as a reference. MRI (1.5 T) was performed using three different sequences: T2-HASTE in coronal and axial orientation, T1-GRE (VIBE) in axial orientation before and after application of contrast media (ce). All datasets were evaluated by four chest radiologists in consensus for each sequence separately independent from CT. The severity of emphysema, leading type, bronchial wall thickening, fibrotic changes and nodules was analyzed visually on a lobar level.

Results

The sensitivity for correct categorization of emphysema severity was 44%, 48% and 41% and the leading type of emphysema was identical to CT in 68%, 55% and 60%, for T2-HASTE, T1-VIBE and T1-ce-VIBE respectively. A bronchial wall thickening was found in 43 lobes in CT and was correctly seen in MRI in 42%, 33% and 26%. Of those 74 lobes presented with fibrotic changes in CT were correctly identified by MRI in 39%, 35% and 58%. Small nodules were mostly underdiagnosed in MRI.

Conclusion

MRI matched the CT severity classification and leading type of emphysema in half of the cases. All sequences showed a similar diagnostic performance, however a combination of HASTE and ce-VIBE should be recommended.

Introduction

Chronic obstructive pulmonary disease (COPD) is one of the leading causes of morbidity and mortality worldwide. At present it is the 4th most common cause of death among adults [1]. Pulmonary emphysema is a pathological term and is defined by the American Thoracic Society as an abnormal permanent enlargement of the air spaces distal to the terminal bronchiole, accompanied by the destruction of their walls, most often caused by COPD [2], [3].

Spirometry as well as performance test like 6 min walking distance is the clinical reference for COPD measurement. Since compensation mechanisms of the lungs are pronounced, mild pathological affection and minor changes in disease cannot be detected with global analysis. In contrast to spirometry, imaging might allow for regional assessment of the compartments involved. This might be crucial for early detection of COPD and monitoring of its treatment. High-resolution computed tomography (HRCT) is the imaging modality of choice and a long-standing player for non-invasive imaging and quantification of lung parenchyma and airways [4], [5], [6], [7], [8], [9], [10].

Magnetic resonance imaging (MRI) of the chest is more and more recognized as a versatile diagnostic procedure also for imaging the chest with special focus on functional information [11], [12], [13]. As emphysema is associated with extensive functional impairment, which can be examined by proton-MRI (like perfusion), or hyperpolarized gas MRI (e.g. ventilation or measurement of apparent diffusion coefficient), the capability of MR to visualize parenchymal alterations within the same investigation would be highly appreciated [14].

Morphological imaging of the lung parenchyma using proton-MRI has up to now been successfully applied in diseases with an increase of tissue and therefore signal, such as pulmonary nodules, infiltrates and fibrosis [15], [16], [17], [18]. MR signal intensity of the lung parenchyma depends on the proton density of the lung parenchyma per voxel [19] and inhomogeneous B₀-field due to susceptibility artefacts resulting from air-tissue surface. Emphysema, however, is characterized by a loss of parenchyma and reduced blood volume. This decrease of tissue within a voxel results in a low signal [20].

As no study evaluated the potential of morphological MRI in emphysema, the aim of our study was to compare different proton-MRI sequences with modern multislice thin-section-CT of the lung.