Oxidative Damage and Post-COVID Syndrome: A Cross-Sectional Study in a Cohort of Italian Workers

Abstract

In addition to the acute symptoms after infection, patients and society are also being challenged by the long-term effects of COVID-19, known as long COVID. Oxidative stress, as a pivotal point in the pathophysiology of COVID-19, could potentially be also involved in the development of the post-COVID syndrome. The aim of the present study was to evaluate the relationship between changes in oxidative status and the persistence of long-COVID symptoms in workers with a previous mild COVID-19 infection. A cross-sectional study was conducted among 127 employees of an Italian university (80 with a previous COVID-19 infection, and 47 healthy subjects). The TBARS assay was used to detect malondialdehyde serum levels (MDA), while total hydroperoxide (TH) production was measured by a d-ROMs kit. A significant difference in mean serum MDA values was found between previously infected subjects and healthy controls and (4.9 µm vs. 2.8 µm, respectively). Receiver-operating characteristic (ROC) curves showed high specificity and good sensibility (78.7% and 67.5%, respectively) for MDA serum levels. A random forest classifier identified the hematocrit value, MDA serum levels, and IgG titer against SARS-CoV-2 as features with the highest predictive value in distinguishing 34 long-COVID from 46 asymptomatic post-COVID subjects. Oxidative damage persists in subjects with previous COVID-19 infection, suggesting a possible role of oxidative stress mediators in the pathogenesis of long COVID.

Keywords: TBARS assay; biomarker; long COVID; malondialdehyde; oxidative stress; workers.

Figure 1

Receiver–operating characteristic (ROC) analyses assessing the ability of H₂O₂ (a) and MDA (b) to discriminate COVID-19 patients from healthy controls.

Figure 2

Serum mean MDA (µm) and H₂O₂ (mg/dL), with 95% family-wise confidence level, in 34 long-COVID (LC) patients, 46 non-long-COVID (nLC), and 47 controls.

Figure 3

Importance of the random forest classification variables in distinguishing COVID-19 patients according to the presence (long-COVID patients) or not (non-long-COVID patients) of at least one symptom four months after negativization.