Trial Report Cluster Analysis to Identify Long COVID Phenotypes Using 129Xe Magnetic Resonance Imaging: A Multi-centre Evaluation, 2024, Eddy et al.

[SNT Gatchaman]

Cluster Analysis to Identify Long COVID Phenotypes Using 129Xe Magnetic Resonance Imaging: A Multi-centre Evaluation
Rachel L Eddy; David Mummy; Shuo Zhang; Haoran Dai; Aryil Bechtel; Alexandra Schmidt; Bradie Frizzell; Firoozeh V Gerayeli; Jonathon A Leipsic; Janice M Leung; Bastiaan Driehuys; Loretta G Que; Mario Castro; Don D Sin; Peter J Niedbalski

BACKGROUND
Long COVID impacts ∼10% of people diagnosed with COVID-19, yet the pathophysiology driving ongoing symptoms is poorly understood. We hypothesised that 129Xe magnetic resonance imaging (MRI) could identify unique pulmonary phenotypic subgroups of long COVID, therefore we evaluated ventilation and gas exchange measurements with cluster analysis to generate imaging-based phenotypes.

METHODS
COVID-negative controls and participants who previously tested positive for COVID-19 underwent 129XeMRI ∼14-months post-acute infection across three centres. Long COVID was defined as persistent dyspnea, chest tightness, cough, fatigue, nausea and/or loss of taste/smell at MRI; participants reporting no symptoms were considered fully-recovered. 129XeMRI ventilation defect percent (VDP) and membrane (Mem)/Gas, red blood cell (RBC)/Mem and RBC/Gas ratios were used in k-means clustering for long COVID, and measurements were compared using ANOVA with post-hoc Bonferroni correction.

RESULTS
We evaluated 135 participants across three centres: 28 COVID-negative (40±16yrs), 34 fully-recovered (42±14yrs) and 73 long COVID (49±13yrs). RBC/Mem (p=0.03) and FEV1 (p=0.04) were different between long-and COVID-negative; FEV1 and all other pulmonary function tests (PFTs) were within normal ranges. Four unique long COVID clusters were identified compared with recovered and COVID-negative. Cluster1 was the youngest with normal MRI and mild gas-trapping; Cluster2 was the oldest, characterised by reduced RBC/Mem but normal PFTs; Cluster3 had mildly increased Mem/Gas with normal PFTs; and Cluster4 had markedly increased Mem/Gas with concomitant reduction in RBC/Mem and restrictive PFT pattern.

CONCLUSIONS
We identified four 129XeMRI long COVID phenotypes with distinct characteristics. 129XeMRI can dissect pathophysiologic heterogeneity of long COVID to enable personalised patient care.

TAKE HOME MESSAGE
Cluster analysis of 129 Xe MRI metrics identifies four phenotypes of long COVID with distinct functional MRI and clinical characteristics. MRI-based clusters can be used to dissect long COVID heterogeneity, enabling personalised clinical care and treatment.

Link | PDF (European Respiratory Journal)

 

[SNT Gatchaman]

Summary quotes from introduction —

While it is now clear that long COVID is a multi-organ condition with a wide range of symptoms frequently involving the respiratory system, pulmonary function tests (PFTs) and conventional computed tomography (CT) are typically normal.

One important challenge in identifying novel therapies in long COVID is its heterogeneity. Imaging has the potential to reveal unique phenotypes of long COVID.

129Xe MRI further enables measurement of regional gas exchange as inhaled 129Xe gas uptake in the alveolar membrane and transfer to capillary red blood cells (RBC), and in long COVID has characteristically shown reduced ratio of 129Xe RBC-to-membrane. These findings have been postulated to reflect microvascular abnormalities and have been observed in people with dyspnea or any other persistent symptoms, those who have been hospitalised or not, for one-year or longer after acute infection, and importantly when conventional CT and PFTs have been mostly normal.

We therefore hypothesised that 129Xe MRI could also identify unique pulmonary phenotypic subgroups of long COVID.

 

[SNT Gatchaman]

Without detailing the technology, this section from the materials and methods sets the scene (split and emphasised for legibility) —

Gas exchange was quantified as whole-lung ratios of Mem/Gas, RBC/Mem and RBC/Gas.

Mem/Gas, or membrane tissue uptake, measures 129Xe dissolved in the interstitial membrane normalised to the gas signal and reflects parenchymal integrity; high membrane uptake is typically due to fibrosis or inflammation whereas low membrane uptake is typically due to tissue destruction or emphysema.

RBC/Gas, or RBC transfer, measures 129Xe dissolved in pulmonary capillary RBCs normalised to the gas signal and reflects microvascular integrity; low RBC/Gas is typically due to diffusion/perfusion limitation or vascular destruction.

RBC/Mem measures the ratio of 129Xe dissolved in RBCs to that in interstitial tissue and represents the efficiency of transfer of gas from the interstitium to capillary RBCs as a measure of overall gas exchange; reduced RBC/Mem could be due to reduced RBC transfer and/or increased membrane uptake.

 

[SNT Gatchaman]

Summary quotes from discussion (my emphasis) —

In this multi-centre study, we identified four long COVID clusters with unique 129Xe MRI signatures and corresponding distinct demographic and clinical features.

Previously published 129Xe MRI studies have revealed low RBC/Mem in people with post-acute or long COVID, consistent with Cluster2 phenotype in our work.

RV is Residual Volume (Functional Residual Capacity = Expiratory Reserve Volume + Residual Volume)
TLC is Total Lung Capacity
See Wikipedia

We further identified three other functional pulmonary long COVID groups. Cluster1 exhibited normal 129 Xe MRI with mildly increased RV/TLC and elevated SGRQ. Cluster3 and Cluster4 were uniquely driven by elevated 129Xe membrane uptake that has not been previously noted in long COVID participants.

Thus, gas exchange abnormalities, and not airway or ventilation, were the greatest pathophysiologic drivers of long COVID in these participants. We additionally showed that the characteristic 129Xe gas exchange metrics in Clusters2-4 were different from both COVID-negative controls and fully recovered participants.

There is a cluster of patients (Cluster1, ~30% of participants) who have mildly elevated RV/TLC ratio indicating air trapping despite ‘normal’ MRI and inspiratory CT. This group of long COVID patients may be similar to those from prior work that showed air trapping using expiratory CT.

We were also able to separate participants with low RBC/Mem alone (Cluster2, ~30% of participants) from those with low RBC/Mem concomitant with elevated membrane uptake (Cluster4, ~10% of participants), and those with elevated membrane uptake but preserved RBC/Mem (Cluster3, ~30% of participants).

Low RBC/Mem in Cluster2 in the absence of airway or parenchymal abnormalities adds to the growing evidence of post-COVID microvascular abnormalities such as pulmonary capillary inflammation, vasoconstrictive remodeling or thrombosis that inhibit transfer of gas. Cluster3 was only differentiated by mildly elevated Mem/Gas (with preserved RBC/Mem), which may represent (residual) interstitial inflammation or edema.

Cluster2 and Cluster3 signatures demonstrate the high sensitivity of 129Xe MRI in revealing abnormalities in specific components of pulmonary gas exchange pathophysiology. In contrast, DLCO provides a whole-lung average of the alveolar epithelial-capillary structure, which may explain why participants in Cluster2 and Cluster3 had DLCO values all within normal limits.

Although small, Cluster4 was largely comprised of participants with the greatest severity of acute COVID-19 (based on hospitalisation and in-patient interventions), and were left with persistent post-COVID interstitial lung abnormalities. It should be noted however that ~1/3 of all participants in Cluster4 did not require hospitalisation for acute COVID-19

 

[SNT Gatchaman]

Figure 5. Long COVID Cluster Summary.

(A) 129Xe MR gas, membrane and red blood cell (RBC) images for representative participants in COVID-negative participants, recovered participants, and resulting long COVID clusters. Site location for each participant embedded in images. (B) Qualitative cluster comparison for imaging measurements used to generate clusters and clinical characteristics.
D-defect (no signal); L=low-intensity signal; H=high-intensity signal; RBC-red blood cells; BMI-body mass index; CT=computed tomography; GGO-ground glass opacities; Ret-reticulation.