No serological evidence for neuronal damage or reactive gliosis in neuro-COVID-19 patients with long-term persistent headache

Abstract

Recent studies have indicated that long-term neurological sequelae after COVID-19 are not accompanied by an increase of canonical biomarkers of central nervous system injury in blood, but subgroup stratifications are lacking. This is a particular concern in chronic headache, which can be a leading symptom of Post-COVID diseases associated with neuronal damage such as vasculitis or autoimmune encephalitis. We here compared patients with mild Post-COVID-19 syndrome and persistent headache (persistent Post-COVID-19 headache) lasting longer than 12 weeks after the initial serological diagnosis, to patients with mild and severe COVID-19 and COVID-19-negative controls. Levels of neurofilament light chain and glial fibrillary astrocytic protein, i.e. markers of neuronal damage and reactive astrogliosis, were lower in blood from patients with persistent Post-COVID-19 headache compared to patients with severe COVID-19. Hence, our pilot serological study indicates that long-term Post-COVID-19 headache may not be a sign of underlying neuronal damage or neuroinflammation.

Neurological post-acute sequelae of SARS-CoV-2 infection (PASC) are common, although direct viral infection of the central nervous system (CNS) is rare [3, 6]. Instead, inflammatory mechanisms, parenchymal hypoxia or microvascular injuries may contribute to the development of CNS injury, raising the possibility that these long-term symptoms may be accompanied by systemic biomarkers of neuronal damage or neuroinflammation.

Accordingly, recent studies evaluated neurofilament light chain (NfL) as a marker of neuronal injury and glial fibrillary acidic protein (GFAP) as a marker of reactive astrogliosis and neuroinflammation in the blood of patients with acute COVID-19 and PASC. Although patients with severe acute COVID-19 had higher concentrations of NfL and GFAP than moderately/mildly affected COVID-19 patients or controls [4], the levels of these biomarkers subsequently returned to normal levels and were not correlated with persistent neurological symptoms in patients with PASC [4]. However, these initial studies did not allow a subgroup analysis according to chief neurological complaint or primary symptoms. This is particularly relevant for persistent headache, a common and debilitating PASC symptom for several reasons [5]. First, headache was a leading symptom associated with increased NfL and GFAP levels and increased mortality in acute COVID-19 patients [1]. Second, new persistent headache may also be an initial sign of chronic CNS inflammation such as cerebral vasculitis or autoimmune encephalitis [2].

Therefore, in this pilot study we investigated NfL and GFAP levels in blood from Post-COVID-19 patients with new daily persistent headache (n = 6, all female), defined as being different from previous primary headaches (if any), having started after the initial serological diagnosis of SARS-CoV-2 infection and persisting longer than 12 weeks. The quality of Post-COVID-19 headaches was described as a pounding or squeezing sensation, and the intensity was described as fluctuating between medium-intensity and high-intensity. These patients had been classified as mild during acute infections according to the WHO definition, i.e. they did not require high flow oxygen therapy or ventilation. In comparison, we also analyzed blood NfL levels in male and female patients diagnosed with mild COVID-19 (n = 17), severe COVID-19 (n = 11), and COVID-19-seronegative control subjects (n = 14). Specimen were obtained 14 ± 24 weeks after the initial diagnosis in mild and 8 ± 19 weeks in severe COVID-19 patients, and 33 ± 17 weeks in Post-COVID-19 headache patients.

GFAP levels were analyzed in all patients with Post-COVID-19 headache, but were only available in n = 8 patients with mild COVID-19, n = 4 severe COVID-19, and n = 8 COVID-19-negative controls.

All patient characteristics are described in Table 1. All measurements were performed on a SIMOA analyzer (Quanterix) using the corresponding SIMOA assay kits.

Table 1 Patient characteristics

We found that NfL levels were similar in patients with persistent Post-COVID-19 headache, mild COVID-19 and COVID-19-seronegative controls, but significantly elevated in severe COVID-19 compared to patients with persistent Post-COVID-19 headache (Fig. 1A). Similarly, GFAP levels were comparable in patients with persistent Post-COVID-19 headache, mild COVID-19 and COVID-19-seronegative controls, but significantly elevated in severe COVID-19 compared to persistent Post-COVID-19 headache patients (Fig. 1B).

Fig. 1

NfL and GFAP levels. A NfL levels in severe COVID-19 (n = 11) are significantly higher compared to persistent Post-COVID-19 headache (n = 6), mild COVID-19 (n = 17) and COVID-19-negative controls (n = 14; Kruskal–Wallis test followed by Dunn’s multiple comparisons test). B GFAP levels in persistent Post-COVID-19 headache (n = 6) were similar to compared to mild COVID-19 (n = 8) and COVID-19-negative controls (n = 8), but significantly lower compared to severe COVID-19 (n = 4; Kruskal–Wallis test followed by Dunn’s multiple comparisons test)

Thus, in contrast to severe COVID-19, we did not detect serological signs of CNS damage or reactive astrogliosis in patients presenting with persistent headache after mild COVID-19. Therefore, our data argue against persistent headache as an indicator of ongoing or progressive parenchymal damage or neuroinflammation. Moreover, our study suggests that persistent post-COVID-19 headache may be pathophysiologically and prognostically different from headache during acute COVID-19, which is often associated with elevated NFL and GFAP levels and may indicate increased mortality [1]. On the other hand, our data indicate that patients with severe COVID-19, even without neurological manifestations, should be closely monitored for ongoing CNS damage as this subgroup exhibited increased NfL and GFAP levels even after the acute phase of COVID-19. Limitations of this pilot study include the small sample sizes, missing follow-up analyses and clinical heterogeneity of groups. However, our study supports recent analyses that reported normal levels of CNS biomarkers in blood from COVID-19 patients with ongoing neurological symptoms [1, 4].