Cognitive and Neuropsychiatric Features of COVID-19 Patients After Hospital Dismission: An Italian Sample

Background and Aims: Recent studies suggest cognitive, emotional, and behavioral impairments occur in patients after SARS-CoV-2 infection. However, studies are limited to case reports or case series and, to our knowledge, few of them have control groups. This study aims to assess the prevalence of neuropsychological and neuropsychiatric impairment in patients after hospitalization. Methods: We enrolled 29 COVID+ patients (M/F: 17/12; age 58.41 +/- 10.00 years; education 11.07 +/- 3.77 years, 2 left handers) who needed hospitalization but no IC, about 20 days post-dismission, and 29 COVID- healthy matched controls. Neuropsychological and neuropsychiatric assessments were conducted via teleneuropsychology using the following tests: MMSE, CPM47, RAVLT, CDT, Digit-Span Forward/Backward, Verbal fluencies; BDI-II, STAI. People with previous reported cognitive impairment and neurological or psychiatric conditions were excluded. Clinical and demographics were collected. Comparison between groups was conducted using parametric or non-parametric tests according to data distribution (T-test, Mann Withney-U test; Chi-square goodness of fit). Within COVID+ group, we also evaluated the correlation between the cognitive and behavioral assessment scores and clinical variables collected. Results: Among COVID+, 62% had at least one pathological test (vs. 13% in COVID-; p = 0.000) and significantly worst performances than COVID- in RAVLT learning (42.55 +/- 10.44 vs. 47.9 +/- 8.29, p = 0.035), RAVLT recall (8.79 +/- 3.13 vs. 10.38 +/- 2.19, p = 0.03), and recognition (13.69 +/- 1.47 vs. 14.52 +/- 0.63, p = 0.07). STAI II was higher in COVID- (32.69 +/- 7.66 vs. 39.14 +/- 7.7, p = 0.002). Chi-square on dichotomous values (normal/pathological) showed a significant difference between groups in Digit backward test (pathological 7/29 COVID+ vs. 0/29 COVID-; p = 0.005). Conclusions: Patients COVID+ assessed by teleneuropsychology showed a vulnerability in some memory and executive functions (working memory, learning, delayed recall, and recognition). Intriguingly, anxiety was higher in the control group. Our findings therefore confirm the impact of COVID-19 on cognition even in patients who did not need IC. Follow-up is needed to evaluate the evolution of COVID-19-related cognitive deficit.