TABLE 1.
Systematic Studies on Molecular Imaging in Cerebral Manifestations of COVID-19: Encephalopathy
| Parameter | Kas et al. (39) | Hosp et al. (18) | Blazhenets et al. (41) |
|---|---|---|---|
| Research question | Longitudinal metabolic pattern in COVID-19 encephalopathy | Neuronal correlates of neurologic and cognitive symptoms (subacute stage) | Recovery of cognitive impairment and regional hypometabolism in subacute COVID-19 during 5- to 6-mo follow-up |
| Population | |||
| Inclusion (main) | PCR+; new cognitive impairment with focal CNS sign or seizure; other infectious or autoimmune disorders excluded | PCR+; only inpatients; ≥1 new neurologic symptom (including cognition [MoCA]); PET, MRI, and neuropsychologic test battery if ≥2 new symptoms | Follow-up data of Hosp et al. (18); clinical register, 17 pts with PET during subacute phase; 9 pts without complaints refused follow-up PET |
| n | 7 | 15 (PET) | 8 |
| Age (y) | 50–72 | 65 ± 14 | 66 ± 14 |
| Selected clinical findings | All hospitalized (7); ventilated (3); executive deficit, frontal lobe changes (7); psychiatric manifestation (5); follow-up: improved (6) but residual attention/executive deficit at 4–8 mo; anxiety/depression (4); deterioration (1) | Initial ICU (7/29; 2 only observation; 2 noninvasive and 2 invasive ventilation); impaired gustation (29/29) and smell (25/29); impaired cognition (MoCA < 26; 18/26); prominent deficits in memory (7/14), executive functions (6/13), and attention (6/15); MRI mircoembolic infarcts (4/13); CSF: PCR− (4/29) | All treated as inpatients during acute phase, ICU (2); self-reported persistent cognitive deficits (4); MoCA: recovery from 19 ± 5 (1st) to 2 3± 4 (2nd examination), still impaired (5; especially memory) |
| 18F-FDG PET | |||
| Analysis | ROI; SPM: single-subject and group; normalization: scaling to pons; comparison: 32 NCs (identical protocol); P < 0.05 FWE | Single-case: visual inspection; PCA (scaled subprofile model); comparison: 45 control patients (identical protocol); plasma glucose–adjusted SUV; confirmatory analyses with SPM (normalization: white matter, P < 0.01 FDR) and PCA with 35 NC (similar scanner) | PCA: expression of previously established COVID-19–related covariance pattern; SPM: paired (within pts) and unpaired (vs. control patients); (Hosp et al. (18)) |
| Δt | Acute, 4 wk later, or 26 wk after onset | 4 ± 2 wk | 23 ± 7 wk |
| Major findings | Acute—DEC: frontal, insula, cingulate, CN (group), and posterior cortices (6/7); INC: vermis, dentate nucleus, pons (group; P < 0.05, uncorrected); 4 wk later—DEC: frontal but improved (group); 26 wk after onset—almost normal, residual DEC orbitofrontal, insula, cingulate, CN (group); almost normal (3/7), frontal (3/7), improve/decline (1/7) | Single-case: predominant frontoparietal cortical DEC (10/15), relative INC of striatum (3/15) and vermis (1/15); group PCA: negative voxel weights (DEC) in extensive neocortical regions (especially frontoparietal) and CN; positive voxel weights (interpreted as preserved metabolism) in brain stem, CBL, MTL, and white matter; confirmatory analyses: similar results; significant negative correlation (r2 = 0.62) MoCA vs. pattern expression score | PCA: pattern expression decreased (P = 0.002) but still at trend level higher than in control patients (P = 0.06); negative correlation (R2 = 0.39) between MoCA and pattern expression; SPM (paired): significant recovery of neocortical DEC (P < 0.01 FDR); SPM (unpaired): residual neocortical DEC (trend level, P < 0.005, uncorrected) |
| Hypothesis | Widespread, frontal-dominant impairment, variably reversible in most patients until 6 mo, due to para- or postinfectious immune mechanism | Cortical dysfunction due to inflammatory process trigged by systemic immune response (e.g., cytokine release), particularly affecting white matter and being possibly reversible | Slow, but evident reversibility of lasting cortical dysfunction due to subcortical perinflammatory processes (triggered by systemic inflammatory response or cytokine release) |
MoCA = Montreal Cognitive Assessment; pts = patients; ICU = intensive care unit; CSF = cerebrospinal fluid; PCR = polymerase chain reaction; ROI = region of interest; SPM = statistical parametric mapping; SPM: single-subject or group = SPM-group or single-subject analyses (usually COVID patients vs. NCs); normalization = method/reference region used for count rate normalization of PET scans; NCs = healthy controls; PCA = principal-components analysis; FDR = false-discovery rate correction; Δt = interval between symptom onset or PCR+ for SARS-CoV-2 (as available) and PET; DEC and INC = decreased and increased signal, respectively; CN = caudate nucleus; CBL = cerebellum; MTL = mesial temporal lobe.
Numbers in parentheses refer to number of subjects with specified finding (if subsample assessed is smaller than study group, sample size as indicated). 18F-FDG target parameter is glucose metabolism.