Long-term pulmonary outcome of children with congenital diaphragmatic hernia: functional lung MRI using matrix-pencil decomposition enables side-specific assessment of lung function

Objectives In patients with congenital diaphragmatic hernia (CDH) the exact functional outcome of the affected lung side is still unknown, mainly due to the lack of spatially resolved diagnostic tools. Functional matrix-pencil decomposition (MP-) lung MRI fills this gap as it measures side-specific ventilation and perfusion. We aimed to assess the overall and side-specific pulmonary long-term outcomes of patients with CDH using lung function tests and MP-MRI. Methods Thirteen school-aged children with CDH (seven with small and six with large defect-sized CDH, defined as > 50% of the chest wall circumference being devoid of diaphragm tissue) and thirteen healthy matched controls underwent spirometry, multiple-breath washout, and MP-MRI. The main outcomes were forced expiratory volume in 1 second (FEV1), lung clearance index (LCI2.5), ventilation defect percentage (VDP), and perfusion defect percentage (QDP). Results Patients with a large CDH showed significantly reduced overall lung function compared to healthy controls (mean difference [95%-CIadjusted]: FEV1 (z-score) −4.26 [−5.61, −2.92], FVC (z-score) −3.97 [−5.68, −2.26], LCI2.5 (TO) 1.12 [0.47, 1.76], VDP (%) 8.59 [3.58, 13.60], QDP (%) 17.22 [13.16, 21.27]) and to patients with a small CDH. Side-specific examination by MP-MRI revealed particularly reduced ipsilateral ventilation and perfusion in patients with a large CDH (mean difference to contralateral side [95%-CIadjusted]: VDP (%) 14.80 [10.50, 19.00], QDP (%) 23.50 [1.75, 45.20]). Conclusions Data indicate impaired overall lung function with particular limitation of the ipsilateral side in patients with a large CDH. MP-MRI is a promising tool to provide valuable side-specific functional information in the follow-up of patients with CDH. Clinical relevance statement In patients with congenital diaphragmatic hernia, easily applicable MP-MRI allows specific examination of the lung side affected by the hernia and provides valuable information on ventilation and perfusion with implications for clinical practice, making it a promising tool for routine follow-up. Key Points • Functional matrix pencil decomposition (MP) MRI data from a small sample indicate reduced ipsilateral pulmonary ventilation and perfusion in children with large congenital diaphragmatic hernia (CDH). • Easily applicable pencil decomposition MRI provides valuable side-specific diagnostic information on lung ventilation and perfusion. This is a clear advantage over conventional lung function tests, helping to comprehensively follow up patients with congenital diaphragmatic hernia and monitor therapy effects. Supplementary Information The online version contains supplementary material available at 10.1007/s00330-023-10395-8.


MP-MRI data acquisition and evaluation
MRI examinations including structural and functional scans were performed on a clinical 1.5T whole-body scanner (MAGNETOM Aera, Siemens Healthineers).Functional scans relied on a multi-slice 2D time-resolved ultra-fast balanced steady-state free precession (uf-bSSFP) pulse sequence (technical parameters: field-of-view (FOV) = 425 mm x 425 mm, 12-mm slice thickness, TE/TR = 0.67 ms/1.52 ms, flip angle a 65°, 2056 Hz/pixel bandwidth, 128 x 128 acquisition matrix (bicubic interpolation to 256 x 256), 150 coronal images, 110 ms per image, 3.33 images/s acquisition rate, 50 s total acquisition time per slice, parallel imaging GRAPPA factor 2) [1].The uf-bSSFP pulse sequence applied used excitation pulses and gradient switching patterns of a conventional Cartesian bSSFP imaging scheme accompanied by partial echo readouts and ramp sampling techniques to shorten echo and repetition time, reduce motion and off-resonance artefacts and improve lung parenchyma signal [1].Accordingly, the entire chest volume was covered from posterior to anterior with around 8 to 14 coronal slices and a voxel size of 3.3mm x 3.3mm x 12mm in supine position during free tidal breathing.At each slice location, 150 images were sequentially acquired during approximately 50 seconds with a frame rate of 3.3 images per second (110ms acquisition time per single image and 190ms interval between consecutive images).The time-resolved uf-bSSFP image series acquired was processed by elastic image registration to a fixed baseline image in mid respiratory state to compensate for respiratory motion [2].Thus, the signal magnitude in each image was preserved, but lung structures (airways, vessels, thoracic walls, etc.) were aligned.On the registered images, the lung parenchyma was segmented automatically as described previously using a deep-learning algorithm [3].Data (motion-corrected and segmented) were further processed with the matrix pencil (MP) algorithm derivated of Fourier decomposition [4]: Voxel-wise spectral analysis of the amplitudes of periodic lung parenchyma signal intensity modulations caused by respiration (frequency corresponding to respiratory rate) and pulsatile blood flow (frequency corresponding to pulse rate) was used to calculate quantitative ventilation and perfusion maps of the lung [4,5].Lung regions with the fractional ventilation or perfusion amplitude below 0.70 of the median of all pixels inside a local region of interest (segmented lung area on the corresponding coronal slice) were considered to show impaired fractional ventilation or impaired perfusion respectively [4,5].Main outcomes were ventilation defect percentage (VDP) and perfusion defect percentage (QDP), which equal the relative amount of lung volume with impaired fractional ventilation resp.relative perfusion [4,5].Homogeneity of defect distribution for ventilation and perfusion was assessed by the defect distribution index DDI (DDIV and DDIQ, resp.)[6].The DDI increases with the defect areas being more clustered as it takes into account how densely and how far away defect voxels are located from each other.

Statistical Analysis
Regarding the comparison of the outcomes of the classic lung function tests (FEV1, FVC, FEV1/FVC, TLC, RV/TLC, LCI2.5) and of the MP-MRI examination applied to the lung as a whole (VDP, QDP, DDIV, DDIQ, VQDmatch) between the groups, we chose the method of analysis as required: According to data attributes, for FEV1, FVC, TLC, RV/TLC, LCI2.5, QDP and VQDmatch, Fisher ANOVA, for FEV1/FVC and VDP Welch ANOVA and for DDIV and DDIQ Kruskal-Wallis ANOVA was used.Accordingly, post-hoc analysis was calculated with Student t-test, Welch t-test and Mann-Whitney-U-test and per parameter corrected for multiple testing using Tukey, Games-Howell and Benjamini & Hochberg approaches, respectively.Age as a potential confounder did not differ significantly between the three groups (tested using ANCOVA) and was therefore not implemented as a covariate in the final model.Further, we assessed differences in MP-MRI outcomes (VDP, QDP, DDIV, DDIQ and VQDmatch) between the affected (CDH-) and non-affected lung side.We tested whether these side differences varied between the groups (healthy controls, small CDH, large CDH) using a two-way repeated measures ANOVA including lung side as within-subject factor variable, group as between-subject factor variable and the interaction term between lung side and group.
Per MP-MRI outcome parameter, post-hoc analysis was performed to assess side differences (affected vs. non-affected) in each group (healthy controls, small CDH, large CDH) using paired t-test comparisons with Bonferroni-correction for multiple testing.

Supplemental Tables
Supplemental Table S1.Study population characteristics Small CDH 1  Large CDH

Lung function outcomes of healthy controls and patients with small and large CDH.
Lung function parameters and functional MP-MRI parameters are given as z-scores or absolute values respectively, presented as mean ± standard deviation and compared by post-hoc analysis of one-way ANOVA.Adjustment of CI and p-values for multiple testing using Tukey (FEV1, FVC, RV/TLC, LCI2.5, QDP, VQDmatch), Games-Howell (FEV1/FVC, VDP) and Benjamini & Hochberg (DDIV, DDIQ) approaches.
1 defined as having received a primary closure of the diaphragmatic defect 2 defined as having required a hernia repair with a patch or a muscle flap3diagnosis based on clinical symptoms CDH: congenital diaphragmatic hernia; ECMO: extracorporeal membrane oxygenation; pectus exc: pectus excavatus; NA: non applicable.Supplemental TableS2.

Table S3 . Whole lung function outcomes compared between healthy controls and patients with small and large CDH, results of one-way ANOVA.
CDH: congenital diaphragmatic hernia; FEV1: forced expiratory volume in 1 second; FVC: forced vital capacity; RV: residual volume; TLC: total lung capacity; LCI2.5:Lung clearance index, measured at classical end of nitrogen multiple-breath washout (N2MBW) (2.5% of the normalized nitrogen starting concentration); MP-MRI: Matrix pencil decomposition magnetic resonance imaging; VDP: percentage of the lung volume with impaired fractional ventilation; QDP: percentage of lung volume with impaired relative perfusion; DDIV: defect distribution index of ventilation; DDIQ: defect distribution index of perfusion; VQDmatch: combined ventilation and perfusion defect percentage.Supplemental

Table S5 . Differences in MP-MRI outcomes between the affected (CDH-) and non-affected lung side, variation between the groups (healthy controls, small CDH, large CDH). Results of two-way repeated measures ANOVA.
MP-MRI: Matrix pencil decomposition magnetic resonance imaging; VDP: percentage of the lung volume with impaired fractional ventilation; QDP: percentage of lung volume with impaired relative perfusion; DDIV: defect distribution index of ventilation; DDIQ: defect distribution index of perfusion; VQDmatch: combined ventilation and perfusion defect percentage.