Study explores the biology of post-COVID-19 cognitive deficits

By Pooja Toshniwal PahariaSep 4 2023Reviewed by Sophia Coveney

In a recent study published in Nature Medicine, researchers explored the biology of cognitive deficits following acute coronavirus disease 2019 (COVID-19).

Many individuals develop neuropsychiatric symptoms in the weeks and months following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, either alone or as part of a post-acute COVID-19 condition known as long COVID.

Cognitive impairments, such as 'brain fog,' are particularly concerning among post-COVID-19 symptoms; they are prevalent yet clinically complicated, with persistent subjective and objective symptoms that impair the capacity to function.

About the study

In the present study, researchers investigated the biochemical roots of cognitive impairments in the post-acute phase of COVID-19.

The study examined the patterns of associations between biomarkers measured at COVID-19-related hospital admission and post-acute cognitive deficits measured six months and a year later in 1,837 post-hospitalization COVID-19 (PHOSP-COVID) study participants (mean age, 58 years; 37% female, 58% male).

The team assessed subjective and objective cognitive impairments and occupational consequences. The findings' generalizability was assessed by attempting to replicate them in a second cohort utilizing electronic medical records (EHR) from over 90 million patients.

To examine objective and subjective cognitive impairments, the Montreal Cognitive Assessment (MoCA) scores and responses to the cognitive component of the Patient Symptom Questionnaire (C-PSQ) were assessed.

The canonical correlation analysis (CCA) was used to identify covariance dimensions between six blood biological markers measured at hospitalization (D-dimer, C-reactive protein (CRP), fibrinogen, platelet, neutrophil, and lymphocyte counts; these represent various health aspects, including immune function, inflammation, and coagulation) and 14 cognitive scores measured six months later, each of which included seven components of the MoCA and C-PSQ.

Furthermore, split-sample and leave-one-out validation studies were conducted. The hypothesis that if cognitive abilities in the pre-COVID period could predict both the acute biomarker levels and cognitive decline in the post-acute COVID-19 phase, the findings may be confounded, was investigated in three ways.

Data were obtained from a subset of PHOSP-COVID study participants who reported their cognitive function before and after COVID at six months (547 individuals) and one year (205 individuals) using C-PSQ-2 (a sub-version of C-PSQ).

The researchers investigated whether there were substantial changes in C-PSQ-2 before and after acute SARS-CoV-2 infection to see if cognitive abnormalities at six months and one year were just a reflection of pre-existing cognitive impairments.

Subsequently, the researchers investigated if pre-existing cognitive deficiencies predicted biomarker profiles, which would be required to confound the relationships. The researchers then investigated if covariation dimensions were related to changes in cognitive performance from a pre-COVID baseline.

Results

The cognitive function of individuals in the post-acute phase of COVID-19 was significantly associated with baseline characteristics such as age, comorbidities, and educational attainment. Two distinct biomarker profiles were measured during acute hospitalization, which predicted cognitive outcomes six months and one year following SARS-CoV-2 infection.

The first profile linked increased fibrinogen levels in relation to CRP with subjective and objective cognitive deficits, while the second profile linked increased D-dimer levels in relation to CRP with subjective cognitive deficits and occupational impact. Breathlessness and fatigue were mediated by these profiles, while anxiety or depression were not significantly mediated.

Cognitive function deteriorated in the post-acute COVID-19 phase six months and one year following acute SARS-CoV-2 infection. Pre-existing deficits in cognition were not related to either of the biomarker profiles for the first and second dimensions, indicating that elevated D-dimer or fibrinogen levels in relation to C-reactive protein were not more common among individuals with pre-existing deficits in cognition.

The link between high vs. low fibrinogen levels and cognitive deficits in the post-acute phase of COVID-19 was replicated in individuals without SARS-CoV-2 infection and did not show any significant moderation by SARS-CoV-2 infection status on comparing the risks to those among individuals with SARS-CoV-2 infection.

In contrast, lower D-dimer levels and cognitive deficits in the post-acute phase of COVID-19 did not attain statistical significance among non-COVID-19 individuals and were significantly moderated by SARS-CoV-2 infection status.

Younger individuals and those who were fluent in English showed significantly worse C-PSQ but better MoCA scores.

In the post-hoc analysis, individuals with SARS-CoV-2 infection and elevated D-dimer levels were at an increased venous thromboembolism risk at one month but not prone to ischemic stroke in comparison to a propensity score-matched group of people with elevated D-dimer levels but without SARS-CoV-2 infection. Secondary analyses yielded similar findings.

Conclusions

Overall, the study's findings provide insight into the complex biology of post-COVID cognitive impairments.

The findings revealed two different characteristics that connected acute hematological biomarkers with cognitive impairments in the post-acute COVID-19 phase. The first dimension associated elevated fibrinogen levels (relative to C-reactive protein) with subjective and objective cognitive impairments six months and one year after SARS-CoV-2 infection, respectively.

The second component connected elevated D-dimer levels (relative to C-reactive protein) levels to subjective-type cognitive impairments and occupational effects six months and one year after infection. Unlike the relationship with fibrinogen, the link with serological D-dimer levels was COVID-19-specific.