Persistent brain metabolic impairment in long COVID patients with persistent clinical symptoms: a nine-month follow-up [18F]FDG-PET study

Archive ouverte

Horowitz, Tatiana | Dudouet, Pierre | Campion, Jacques-Yves | Kaphan, Elsa | Radulesco, Thomas | Gonzalez, Sandra | Cammilleri, Serge | Ménard, Amélie | Guedj, Eric

Edité par CCSD ; Springer Verlag (Germany) [1976-....] -

International audience. PurposeA hypometabolic profile involving the limbic areas, brainstem and cerebellum has been identified in long COVID patients using [18F]fluorodeoxyglucose (FDG)-PET. This study was conducted to evaluate possible recovery of brain metabolism during the follow-up of patients with prolonged symptoms.MethodsFifty-six adults with long COVID who underwent two brain [18F]FDG-PET scans in our department between May 2020 and October 2022 were retrospectively analysed, and compared to 51 healthy subjects. On average, PET1 was performed 7 months (range 3–17) after acute COVID-19 infection, and PET2 was performed 16 months (range 8–32) after acute infection, because of persistent severe or disabling symptoms, without significant clinical recovery. Whole-brain voxel-based analysis compared PET1 and PET2 from long COVID patients to scans from healthy subjects (p-voxel < 0.001 uncorrected, p-cluster < 0.05 FWE-corrected) and PET1 to PET2 (with the same threshold, and secondarily with a less constrained threshold of p-voxel < 0.005 uncorrected, p-cluster < 0.05 uncorrected). Additionally, a region-of-interest (ROI) semiquantitative anatomical approach was performed for the same comparisons (p < 0.05, corrected).ResultsPET1 and PET2 revealed voxel-based hypometabolisms consistent with the previously reported profile in the literature. This between-group analysis comparing PET1 and PET2 showed minor improvements in the pons and cerebellum (8.4 and 5.2%, respectively, only significant under the less constrained uncorrected p-threshold); for the pons, this improvement was correlated with the PET1-PET2 interval (r = 0.21, p < 0.05). Of the 14,068 hypometabolic voxels identified on PET1, 6,503 were also hypometabolic on PET2 (46%). Of the 7,732 hypometabolic voxels identified on PET2, 6,094 were also hypometabolic on PET1 (78%). The anatomical ROI analysis confirmed the brain hypometabolism involving limbic region, the pons and cerebellum at PET1 and PET2, without significant changes between PET1 and PET2.ConclusionSubjects with persistent symptoms of long COVID exhibit durable deficits in brain metabolism, without progressive worsening.

• Description
• Sujet/Public
• Outil
• Outil d'anticipation
• Mémoire et thèse
• Gestion