Phenotyping of COPD with MRI in comparison to same-day CT in a multi-centre trial

Objectives A prospective, multi-centre study to evaluate concordance of morphologic lung MRI and CT in chronic obstructive pulmonary disease (COPD) phenotyping for airway disease and emphysema. Methods A total of 601 participants with COPD from 15 sites underwent same-day morpho-functional chest MRI and paired inspiratory-expiratory CT. Two readers systematically scored bronchial wall thickening, bronchiectasis, centrilobular nodules, air trapping and lung parenchyma defects in each lung lobe and determined COPD phenotype. A third reader acted as adjudicator to establish consensus. Inter-modality and inter-reader agreement were assessed using Cohen’s kappa (im-κ and ir-κ). Results The mean combined MRI score for bronchiectasis/bronchial wall thickening was 4.5/12 (CT scores, 2.2/12 for bronchiectasis and 6/12 for bronchial wall thickening; im-κ, 0.04–0.3). Expiratory right/left bronchial collapse was observed in 51 and 47/583 on MRI (62 and 57/599 on CT; im-κ, 0.49–0.52). Markers of small airways disease on MRI were 0.15/12 for centrilobular nodules (CT, 0.34/12), 0.94/12 for air trapping (CT, 0.9/12) and 7.6/12 for perfusion deficits (CT, 0.37/12 for mosaic attenuation; im-κ, 0.1–0.41). The mean lung defect score on MRI was 1.3/12 (CT emphysema score, 5.8/24; im-κ, 0.18–0.26). Airway-/emphysema/mixed COPD phenotypes were assigned in 370, 218 and 10 of 583 cases on MRI (347, 218 and 34 of 599 cases on CT; im-κ, 0.63). For all examined features, inter-reader agreement on MRI was lower than on CT. Conclusion Concordance of MRI and CT for phenotyping of COPD in a multi-centre setting was substantial with variable inter-modality and inter-reader concordance for single diagnostic key features. Clinical relevance statement MRI of lung morphology may well serve as a radiation-free imaging modality for COPD in scientific and clinical settings, given that its potential and limitations as shown here are carefully considered. Key Points • In a multi-centre setting, MRI and CT showed substantial concordance for phenotyping of COPD (airway-/emphysema-/mixed-type). • Individual features of COPD demonstrated variable inter-modality concordance with features of pulmonary hypertension showing the highest and bronchiectasis showing the lowest concordance. • For all single features of COPD, inter-reader agreement was lower on MRI than on CT. Supplementary Information The online version contains supplementary material available at 10.1007/s00330-024-10610-0.


Electronic supplementary material Additional exclusion criteria compared to COSYCONET
For the imaging sub-study, the exclusion criteria were complemented excluding pregnant women and subjects with previous lung surgery (e.g.lung volume reduction or lung transplantation), subjects with moderate or severe exacerbations requiring antibiotic treatment within the last four weeks, acute psychosis or other conditions leaving the perception of the participant appear limited, and specific contraindications to MRI (MRIincompatible implants such as pacemakers, lung volume reduction (LVR) coils, claustrophobia, acute and chronic renal insufficiency [GFR <40 ml / min according to

Training of study centres and quality control
Study centres were preselected with the help of a standardized questionnaire, assessing the locally available technical pre-requisites for MRI and CT.For MRI, a field-strength of 1.5-3.0T and gradient systems with a maximum strength of at least 33mT/m were considered to be eligible.Further, power-injectors for contrast application and capabilities for 4D perfusion measurement are required.For CT, multi-detector scanners with ≥40 detector rows were included with an equivalent dose of <3.5 mSv defined as a cut-off.
The MRI and CT protocols were adapted to the locally available scanners/hardware while maintaining comparable image quality.Regular phantom measurements were performed in each study centre.For MRI, in-house developed phantoms designed to mimic the signal properties of lung tissue, blood, muscle tissue and fat were manufactured and distributed to each centre [1].The same MRI protocol used for the participant examinations was used for phantom measurements every three months.For CT, a single Catphan 600 (The Phantom Laboratory) [2] was used for semi-annual CT measurements.Furthermore, standard operating procedures (SOPs) for performing MRI and CT scans, verbal instructions to subjects and pseudonymization of participant data were developed and if necessary, adapted to specific requirements of each study centre.

MRI acquisition
Most of the MRI sequences were to be acquired in inspiratory breath-holds, with several sequences additionally repeated in expiration.Images were acquired in coronal and transversal orientation.The examinations for morphological imaging of airways and parenchyma comprised fast 3D gradient-echo (GE) and fast spin echo (SE) sequences.The coronal GE sequence was repeated in an expiratory breath-hold.After this, a balanced steady state free precession GE sequence was acquired in free breathing followed by a T2weighted fast spin-echo series with periodically rotated overlapping parallel lines with enhanded reconstruction (PROPELLER) in multiple breath holds.An additional transversal ultra-fast SE sequence was acquired with a short T1 inversion recovery (STIR) preparation.
Dynamic perfusion imaging was performed using a T1-weighted keyhole pulse sequence (dynamic contrast enhancement, DCE) [3,4] with a fixed dose of 2 ml gadolinium-based contrast agent (Gadobutrol, 1 mmol/ml, Bayer Vital GmbH) followed by a saline chaser.MR angiography was performed by subtraction of a native spoiled GE image acquired before the TRA measurement and a repeated measurement acquired after i.v.contrast bolus injection of Gadobutrol (0.1 mmol/kg body weight).After the administration of contrast agent, the fast 3D GE acquisitions were repeated in both ex-and inspiration, with additional fat saturation preparation for the transversal images.Breath-hold times were less than 20 seconds each.Detailed information on the applied sequence parameters is listed in

CT protocol
The CT protocol was defined according to the recommendations of the German Radiological Society (DRG) [3] employing a thin slice collimation of 0.6 mm, a pitch of 0.6-1.0, a tube potential of 120 kVp and a tube current of 35 effective mAs for most scanner types.Minor modifications were required to adjust the protocol to the specifics of a broad spectrum of scanner types from different manufacturers used in this study.A detailed CT acquisition protocol is presented in Supplementary Table 2.The low-dose CT protocol provided morphological information of clinical significance on pulmonary and extra-pulmonary structures, comparable to CT examinations in the context of lung cancer screening.The additional end-expiratory images allowed for improved differentiation between air-trapping caused by collapse or obstruction of small airways and emphysema [4].In addition, valuable information on stability of the tracheobronchial system could be obtained.The maximum effective radiation dose of inspiratory and end-expiratory CT scans together was less than 3.5 mSv, which was comparable to CT examinations at that time in the context of lung cancer screening and significantly below the typical values for a standard chest CT with about 6 mSv.CT image reconstructions comprised three-dimensional datasets with thin slice thickness and smooth as well as edge-enhancing algorithms (Supplementary Table 3).All protocols were supported by commercially available CT devices.

Supplementary
Supplementary Table1based on the protocol designed for 1.5T Siemens MAGNETOM Aera (Siemens Healthineers).All sequences are approved for clinical use and commercially available.
The parameters shown are acquisition mode (3D or 2D multi-slice), slice/slab orientation, breathing mode (inspiration, expiration or free breathing and number of breath-holds), status of contrast agent during acquisition, repetition time TR, echo time TE, Field of View, slice thickness, in-plane resolution/voxel size, matrix size, parallelization factor and total scan time.* The slices in the balanced SSFP (TrueFISP) sequence were acquired with 60% overlap.