Abstract
Long COVID comprises a broad range of persistent symptoms that emerge after COVID-19 and can last for months or years and have a direct impact on daily life. Cognitive impairment and loss of short-term memory are examples of symptoms associated with long COVID. To date, there are neither generally agreed-upon diagnostic criteria nor effective treatments available. A recent study found an association of serotonin deficiency and long COVID in the general population [1], leading to vagal dysfunction and impaired cognition. In this study, we aimed to evaluate longitudinal serotonin levels in hemodialysis (HD) patients with or without cognitive symptoms associated with long COVID.
We analyzed routinely collected serum samples (n = 102) from 16 HD patients collected between 60 days before and 60 days after a positive SARS-CoV-2 RT-PCR. Metabolites were extracted using cold methanol and subjected to reverse phase liquid chromatography/mass spectrometry (LC/MS). Serotonin was identified by matching retention time, accurate mass, and fragmentation data against an in-house LC/MS library. Log2 of serotonin signal intensity was used for quantification. The COVID-19 Yorkshire Rehabilitation Scale (C19-YRS) was administered shortly after positive RT-PCR test, as well as 1- and 6-months post-COVID. Patients were classified as having cognitive symptoms if they reported short-term memory loss or difficulties with concentration or communication problems as per C19-YRS. Unpaired t-test or Fisher's exact tests were performed to compare characteristics of patients with and without cognitive symptoms. We defined three study periods relative to the date of COVID-19 diagnosis: –60 to –15 days (baseline); –14 to +14 days (acute COVID); and +15 to +60 days (post-COVID). Differences in serotonin signal intensity (log2) between patients with and without cognitive symptoms were evaluated with unpaired t-tests. Linear mixed effects models with random intercepts were constructed to determine changes in serotonin levels over time with the baseline period as the comparator. This research was funded by the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK130067).
Demographic data are presented in Table 1a. Of the 16 patients, 7 (44%) were classified with cognitive symptoms associated with long COVID. There were no differences in age, sex, race, ethnicity, vintage, diabetes, and hypertension between patients with and without cognitive symptoms. Serotonin levels in patients with vs. without cognitive symptoms did not significantly differ at baseline and during acute COVID (Fig. 1). Of note, in the post-COVID phase, serotonin levels were significantly lower in the group with impaired cognition (p = 0.012) compared to symptoms-free patients (Fig. 1). Within each group, serotonin levels remained unchanged at acute-COVID or post-COVID periods compared to baseline (Table 1b).
We show that HD patients with impaired cognition have lower serotonin in the post-COVID phase compared to patients without cognitive symptoms. These results corroborate findings in the general population. Further investigation in a larger cohort is warranted to evaluate the potential of serotonin as a target for the prevention or as a marker of neurocognitive symptoms associated with long COVID.
Serotonin levels in hemodialysis patients with (blue) and without (red) cognitive symptoms across different SARS-CoV-2 infection phases.
Comments