Olfactory Performance in Post Mild-to-Moderate Acute Covid-19 across 2 Years

Background: Previous longitudinal studies (n=6) of objective olfaction performance post-acute COVID-19 have a maximum follow-up of 6-month and do not often test biomarkers. Although olfactory dysfunction appears to improve within two months of symptom onset, 4/6 longitudinal studies show persistent olfactory impairment. Method(s): PCR-confirmed COVID-19 patients in the prospective ADAPT cohort (Sydney, Australia) were assessed across 18 acute symptoms and hospitalization status: 40% mild, 50% moderate, 10% severe/hospitalised - none deceased). Blood samples were taken 2 (N=179), 4 (N=148) and 12-month (N=118) post-diagnosis. The NIH Odor Identification Test (OIT) and the Cogstate brief cognitive battery were performed. 58 also had an olfaction test at 24-month. The OIT raw data were transformed into demographically-corrected T-scores. OIT's attrition was completely random and only initial age (40+/-15 versus 47+/-15) differed between patients lost to follow-up and those in the study at 24-month. We tested peripheral neurobiomarkers (NFL, GFAP, S100B, GM-CSF) and immune markers (Interleukin-IL panel: 1-beta, 1Ralpha, 4, 5, 6, 8, 10, 12p40, 12p70, 13, and MCP-1, TNF-alpha and INF-gamma), analyzed as Log transformed and elevated/normal range using published references. Our previous analyses had shown no relationship with the kynurenine pathway, but an association of impaired olfaction and impaired cognition at 2-month only. Linear mixed effect regressions with time effect (months) tested olfaction trajectories (random subject effect) and their association with the biomarkers (main and time interaction). Result(s): At 2 months post-diagnosis 30% had impaired olfaction and those who had acute severe disease were more likely to be impaired (54% versus 26%, p=.009). 21%, 31% and 37% had impaired olfaction at 4, 12 and 24-months. Olfactory performance declined over time (p< .0001), which was dependent on the initial performance (Fig 1). Neurobiomarkers were within the normal range. IFN-gamma, IL-1Ralpha, IL-13 and TNF-alpha increased across time, p< .03-p< .0005. TNF-alpha and IFN-gamma showed a time covariance with poorer olfaction performance. Conclusion(s): Post-acute mild to moderate COVID-19 is associated with a declining olfactory performance up to 2-yr post-diagnosis, especially when initially impaired with the provisio of attrition although random. Olfactory performance decline may be mediated by upregulated immune parameters which are distinct from those driving cognitive changes. (Figure Presented).

Post-Covid Cognitive Impairment Mri Brain Scan Abnormalities: A Potential Biomarker

Background: Post-acute sequelae of SARS-COV-2 infection (PASC) is associated with cognitive impairment (CI) with unclear pathogenesis though blood brain barrier (BBB) impairment and excitotoxic injury appear significant. Post-acute sequelae of SARS-COV-2 infection (PASC) is associated with cognitive impairment (CI) with unclear pathogenesis though blood brain barrier (BBB) impairment and excitotoxic injury appear significant. We hypothesized that PASC CI patients would have brain inflammation and BBB disruption using advanced MR imaging. Method(s): In this prospective longitudinal study, 14 patients with PASC CI (mild and non-hospitalised) were enrolled (mean age of 45;10 F and 4 M) and 10 sex and age matched healthy controls. 13 had a follow up MR at 9-12 months (mean 10 months). All participants underwent DCE perfusion (an index of BBB integrity with Ktrans as the measurement), Diffusion Tensor Imaging (DTI) and single voxel MR spectroscopy (MRS) of the frontal cortex/white matter and the brainstem in addition to brain anatomical MRI. Between group analyses were used to determine which MRI outcomes were significantly different from controls in patients with PASC CI. Result(s): The PASCI CI group showed significantly increased (ie BBB impairment) Ktrans, and increased region (Frontal white matter and Brain Stem)-specific areas in the brain (p=< 0.005), reduction in NAA (ie neuronal injury) and mild reduction of Glx (ie excitotoxicity) in the frontal white matter and brain stem (p=0.004), and reduction in white matter integrity (increased diffusivity -greater radial and mean diffusivity). Increased Ktrans was correlated with increased both radial and mean diffusivity (r=0.9) in all tested brain regions. Ktrans significantly improved in the follow up MR (p= 002596 Z=-2.794872) with no difference between subjects and controls indicating BBB normalisation (p= 0.442418, z= -0.144841). White matter integrity also improved especially in the fractional anisotropy values in the executive networks (p=< 0.00045). MRS showed significant improvement in the NAA in the frontal white matter but Glx remain high as compared to the controls (p=0.0006). Conclusion(s): PASC CI was characterised by reversible diffuse BBB impairment, neuronal/axonal and excitotoxic injury. BBB impairment was associated with white matter disruption. These are suggestive biomarkers for the presence, severity and prognosis of PASC CI. Such biomarkers could underpin appropriate trial design and timing of intervention.

QUINOLINIC ACID IS A BIOMARKER of COVID-19-ASSOCIATED COGNITIVE IMPAIRMENT

Background: COVID-19 infection-associated cognitive and olfaction impairments have an unclear pathogenesis, possibly related to systemic disease severity, hypoxia, or illness-associated anxiety and depression. A biomarker for these neurocognitive changes is lacking. The kynurenine pathway (KP) is an interferon stimulated myeloid cell mediated tryptophan degradation pathway important in immune tolerance, neurotoxicity and vascular injury, that is dysregulated in COVID-19. We hypothesized that neurocognitive impairments were associated with an activated KP. Methods: The current analysis includes COVID-19 patients as part of the ADAPT study, a prospective cohort (St Vincent's Hospital Sydney, Australia). Disease severity was assessed with 18 acute symptoms and hospitalization status. Blood samples were taken 2 months (N=136) and 4 months (N=121) post diagnosis along with cognitive (Cogstate Computerized Battery, CBB;NIH toolbox Odor Identification Test, OIT) and mental health screenings (DMI-10;IESR, SPHERE-34 Psychological subscale grouped into a composite score). KP metabolites (PIC, QUIN, 3HK, 3HAA, AA, KYN, TRP, log for analyses except for TRP) were measured by GC-MS and uHPLC. The CBB and OIT data were demographically-corrected. CBB follow-up data was also corrected for practice effect. Linear mixed effect regression models with time effect (days post diagnosis) tested whether cognition, and olfaction were associated the KP (main and time interaction);while correcting for disease severity, mental health and comorbidities. Results: 136 patients: mean age=46±15;40% females;90% English speaking background;disease severity: 40% mild, 50% moderate, 10% severe/hospitalised;34% treated comorbidities. At 2 months post diagnosis, 16% had cognitive impairment, and 25% had impaired olfaction. Cognitive impairment was more common in those with anosmia (p=.05). At 4 months, 23% had cognition impairment and 20% had impaired olfaction. QUIN (p=.001), 3HAA (p<.0001) increased over the study period, while TRP decreased (p=.02). QUIN level associated with poorer cognitive scores (p=.0007;QUIN (nM) between 800-1000 was most predictive). There was no time∗QUIN interaction. QUIN association to cognition persisted when severe cases were excluded (p<.005). Conclusion: COVID-19 is associated with KP activation, and the latter with cognitive impairment. QUIN was the only biomarker associated with cognitive impairment, and may be useful in monitoring and elucidating COVID-19 neuropathogenesis and treatment.

High rate of persistent symptom 4 months after community-managed COVID-19 infection

Background: The spectrum of recovery following community-managed and hospitalized SARS-CoV-2 infection remains uncertain. The aim of the ADAPT study was to determine prevalence and nature of persistent symptoms after community and hospitalised SARS-CoV-2 infection;and to evaluate lung function;health-related quality of life (HRQOL);and neurocognitive abnormalities. Methods: A prospective observational cohort study was performed at St Vincent's Hospital Sydney Australia. Adult patients with a positive SARS-CoV-2 RNA PCR test between Mar-2020 and Apr-2020 including mild, moderate, and severe acute infection were offered enrollment. The clinical outcomes included symptom prevalence at initial infection and follow-up, HRQOL measures, pulmonary function, neurocognition and COVID-19 antibody responses. Initial study assessments were performed up to 4 months after first detection of SARS-CoV-2. Results: Ninety-six patients were recruited following community-managed mild (39%) and moderate (50%), and hospitalized severe (11%) COVID-19 infection. 39.7% patients had persistent symptoms at median 72 days after diagnosis (IQR 65-87), including those in severe (77.8%), moderate (33.3%), and mild (36.7%) sub-populations. The most common persistent symptoms were fatigue (28%), shortness of breath (25%) and cough (21%). Total lung capacity (TLC) was significantly lower after severe, compared with community-managed, COVID-19, p=0.05. Abnormal diffusion capacity for carbon monoxide values were observed in 16% patients unrelated to acute illness severity. Twenty-four percent patients demonstrated anxiety/depression, as measured by SPHERE-34 item, with the highest proportion in the moderate sub-population (37%). Neurocognitive impairment was low (9%) but associated with abnormal olfaction (p=0.02). A high proportion of patients (77-85%) demonstrated positive antibody responses, on four commercial assays, at follow-up and titres were related to acute illness severity. Conclusion: A considerable proportion of patients experience persistent symptoms at 4 months after SARS-CoV-2 infection including one third of community managed patients. High rates of depression and anxiety were reported across the cohort. Outpatient follow-up to further assess those with persistent symptoms after COVID-19 is important to allow multi-disciplinary input, further investigation, and appropriate management. Data collection on the prevalence of persisting symptoms at 8 month follow-up of the ADAPT study is currently underway.

Cognitive deficits are not a byproduct of Anxio-depressive symptoms in COVID-19

Background: Cognitive deficits and anxio-depressive symptoms have been described in the recovery phase of COVID-19. Their association, or lack thereof, may assist in better understanding the long-term consequences of COVID-19. Methods: Patients underwent neurocognitive and mental health assessment at 2 months after initial SARS-CoV-2 infection as part of the St Vincent's Hospital ADAPT study, a prospective cohort study after COVID-19 disease. Cognition was assessed with the culture fair computerized Cogstate battery. A demographically-corrected composite z-score was created representing global cognitive performance, and then classified as impaired, borderline, and unimpaired. Anxio-depressive symptoms were assessed with the Depression in the Medical Ill scale-10 (DMI-10), the Somatic and Psychological HEalth Report-34 (SHPERE) Psych subscale, and the Impact of Events Scale-Revised (IESR). The scales scores were amenable to a single Principal Component explaining 80% of the variance. Female sex (p<.01) and Non-English-Speaking Background-NESB (p=.02) were associated with greater anxio-depressive symptoms but not age, education. Regression analyses tested sex and NESB unadjusted and adjusted predictions of anxio-depression to cognition. Results: 132 patients (mean age=46±15;40% women, median education=16 years, 10% NESB) were tested after predominantly community-managed COVID-19 (10% hospitalised). 17% had impaired global cognition, and 10% had borderline deficit. 25% had elevated symptoms on the DMI-10 (score>9), 13% on the IESR (score>24) and 35% on the SPHERE Psych scale (score≥2). Anxio-depression was not predictive of cognitive performance (unadjusted p=.43;adjusted p=.98) and of impaired/unimpaired status (unadjusted p=.50;adjusted p=.78). Anxio-depression tended to predict of borderline (vs. unimpaired) performance in unadjusted (p=.08) and adjusted (=.09) analyses. This was explained by the fact that women who had borderline performance tended to report higher anxio-depressive symptoms compared to their peers who were unimpaired (p<.06);further impaired women (vs. unimpaired) tended to report the least anxio-depressive symptoms (p=.09). Conclusion: Cognitive deficits are not a by-product of anxio-depressive symptoms in recovering COVID-19 patients. Women appear to have a higher degree of introspection and reaction to very mild cognitive decline. Cognitive changes appear to be a direct consequence of COVID-19.