Reduced cortical brain perfusion following COVID-19 infection: impact of COVID-19 severity and relation to memory...

Changes to brain perfusion may be contributing to long-term cognitive dysfunction following COVID-19 infection. Lower brain perfusion beyond the acute phase of infection, as measured by perfusion MRI, has been observed in both hospitalized and non-hospitalized cases. Additionally, the hippocampus has been previously shown to be particularly sensitive to changes in blood flow. Therefore, the hippocampus may be vulnerable to perfusion changes associated with COVID-19, leading to long-term cognitive sequalae. The literature suggest that this may be particularly true among those with severe respiratory symptoms. However, the relationship between COVID-19 respiratory severity, perfusion, and cognitive dysfunction has yet to be thoroughly explored, especially after longer recovery times.

Seventy-eight participants completed a neuroimaging session and two memory tests, the Mnemonic Similarity Task (MST) and Face-Name Associative Memory Exam (FNAME), tasks known to be mediated by the hippocampus. Participants were divided into four groups based on the severity of self-reported respiratory symptoms experienced during the acute infection: controls without a reported diagnosis of COVID-19, those with COVID-19 and without respiratory symptoms, those with COVID-19 and respiratory symptoms, and hospitalized participants with respiratory symptoms from COVID-19. On average, participants were 13 months from initial infection. Perfusion was measured with pseudo-continuous arterial spin labeled (pCASL) MRI. Total cortical gray and white matter perfusion, as well as gray matter perfusion in the territories of the anterior, middle, and posterior cerebral arteries were obtained using a mask adapted from previous studies.

The hospitalized group demonstrated lower total gray matter cortical perfusion compared to all groups, t's > 1.74, p's < 0.05. White matter perfusion showed no differences between groups. A similar pattern was observed within each arterial territory. Out of all memory measures, only pattern separation from the MST was related to gray matter perfusion, r = 0.30, p < 0.01, even after controlling for age and education.

Results provide evidence for global perfusion dysfunction among hospitalized participants with severe respiratory symptoms that persists for more than a year after COVID-19 infection. Additionally, there was a relationship between perfusion and pattern separation performance among all participants, suggesting that poorer perfusion may underlie some of the long-term COVID-19 cognitive symptoms.