Plasma Biomarkers Predict COVID Neurologic Complications

Jim Kling

August 11, 2022

SAN DIEGO – Even after recovery of an acute COVID-19 infection, some patients experience extended or even long-term symptoms that can range from mild to debilitating. Some of these symptoms are neurological: headaches, brain fog, cognitive impairment, loss of taste or smell, and even cerebrovascular complications such stroke. There are even hints that COVID-19 infection could lead to future neurodegeneration.

Those issues have prompted efforts to identify biomarkers that can help track and monitor neurological complications of COVID-19. “Throughout the course of the pandemic, it has become apparent that COVID-19 can cause various neurological symptoms. Because of this, it’s really important for us to find a way to monitor and understand neurological complications occurring in patients with COVID 19,” Jennifer Cooper said during a lecture at the Alzheimer’s Association International Conference. She presented new research suggesting that neurofilament light (NfL) and glial fibrillary acidic protein (GFAP) may prove useful.

Ms. Cooper is a master’s degree student at the University of British Columbia and Canada.

Looking for Sensitivity and Specificity in Plasma Biomarkers

The researchers turned to plasma-based markers because they can reflect underlying pathology in the central nervous system. They focused on NfL, which reflects axonal damage, and GFAP, which is a marker of astrocyte activation.

The researchers analyzed data from 209 patients with COVID-19 who were admitted to the Vancouver (B.C.) General Hospital intensive care unit. Sixty-four percent were male, and the median age was 61 years. Sixty percent were ventilated, and 17% died.

The researchers determined if an individual patient’s biomarker level at hospital admission fell within a normal biomarker reference interval. A total of 53% had NfL levels outside the normal range, and 42% had GFAP levels outside the normal range. In addition, 31% of patients had both GFAP and NfL levels outside of the normal range.

Among all patients, 12% experienced ischemia, 4% hemorrhage, 2% seizures, and 10% degeneration.

At admission, NfL predicted a neurological complication with an area under the curve (AUC) of 0.702. GFAP had an AUC of 0.722. In combination, they had an AUC of 0.743. At 1 week, NfL had an AUC of 0.802, GFAP an AUC of 0.733, and the combination an AUC of 0.812.

Using age-specific cutoff values, the researchers found increased risks for neurological complications at admission (NfL odds ratio [OR], 2.9; GFAP OR, 1.6; combined OR, 2.1) and at 1 week (NfL OR, not significant; GFAP OR, 4.8; combined OR, 6.6). “We can see that both NFL and GFAP have utility in detecting neurological complications. And combining both of our markers improves detection at both time points. NfL is a marker that provides more sensitivity, where in this cohort GFAP is a marker that provides a little bit more specificity,” said Ms. Cooper.

Will Additional Biomarkers Help?

The researchers are continuing to follow up patients at 6 months and 18 months post diagnosis, using neuropsychiatric tests and additional biomarker analysis, as well as PET and MRI scans. The patient sample is being expanded to those in the general hospital ward and some who were not hospitalized.

During the Q&A session, Ms. Cooper was asked if the group had collected reference data from patients who were admitted to the ICU with non-COVID disease. She responded that the group has some of that data, but as the pandemic went on they had difficulty finding patients who had never been infected with COVID to serve as reliable controls. To date, they have identified 33 controls who had a respiratory condition when admitted to the ICU. “What we see is the neurological biomarker levels in COVID are slightly lower than those with another respiratory condition in the ICU. But the data has a massive spread and the significance is very small between the two groups,” said Ms. Cooper.

Unanswered Questions

The study is interesting, but leaves a lot of unanswered questions, according to Wiesje van der Flier, PhD, who moderated the session where the study was presented. “There are a lot of unknowns still: Will [the biomarkers] become normal again, once the COVID is over? Also, there was an increased risk, but it was not a one-to-one correspondence, so you can also have the increased markers but not have the neurological signs or symptoms. So I thought there were lots of questions as well,” said van der Flier, professor of neurology at Amsterdam University Medical Center.

She noted that researchers at her institution in Amsterdam have observed similar relationships, and that the associations between neurological complications and plasma biomarkers over time will be an important topic of study.

The work could provide more information on neurological manifestations of long COVID, such as long-haul fatigue. “You might also think that’s some response in their brain. It would be great if we could actually capture that [using biomarkers],” said van der Flier.

Ms. Cooper and van der Flier have no relevant financial disclosures.

This story originally appeared on, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.