Background: Despite numerous observations of neuropsychological deficits immediately following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, little is known about what happens to these deficits over time and whether they are affected by changes in fatigue and any psychiatric symptoms. We aimed to assess the prevalence of neuropsychological deficits at 6-9 months and again at 12-15 months after coronavirus disease 2019 (COVID-19) and to explore whether it was associated with changes in fatigue and psychiatric symptoms.

Methods: We administered a series of neuropsychological tests and psychiatric questionnaires to 95 patients (mean age = 57.12 years, standard deviation (SD) = 10.68; 35.79% women) 222 (time point 1 (T1)) and 441 (time point 2 (T2)) days on average after infection. Patients were categorised according to the severity of their respiratory COVID-19 symptoms in the acute phase: mild (no hospitalisation), moderate (conventional hospitalisation), and severe (hospitalisation in intensive care unit (ICU) plus mechanical ventilation). We ran Monte-Carlo simulation methods at each time point to generate a simulated population and then compared the cumulative percentages of cognitive disorders displayed by the three patient subgroups with the estimated normative data. We calculated generalised estimating equations for the whole sample to assess the longitudinal associations between cumulative neuropsychological deficits, fatigue, and psychiatric data (anxiety, depressive symptoms, posttraumatic stress disorder, and apathy).

Results: Most participants (>50%) exhibited a decrease in their neuropsychological impairments, while approximately 25% showed an escalation in these cognitive deficits. At T2, patients in the mild subgroup remained free of accumulated neuropsychological impairments. Patients with moderate severity of symptoms displayed a decrease in the magnitude of cumulative deficits in perceptual and attentional functions, a persistence of executive, memory and logical reasoning deficits, and the emergence of language deficits. In patients with severe symptoms, perceptual deficits emerged and executive deficits increased, while attentional and memory deficits remained unchanged. Changes in executive functions were significantly associated with changes in depressive symptoms, but the generalised estimating equations failed to reveal any other significant effect.

Conclusion: While most cumulative neuropsychological deficits observed at T1 persisted and even worsened over time in the subgroups of patients with moderate and severe symptoms, a significant proportion of patients, mainly in the mild subgroup, exhibited improved performances. However, we identified heterogeneous neuropsychological profiles both cross-sectionally and over time, suggesting that there may be distinct patient phenotypes. Predictors of these detrimental dynamics have yet to be identified.