Аңдатпа
Aim. To characterize the longitudinal relationships of baseline ACE2 and within-patient ACE2 change with complement components C3 and C4 and the hypoxia-related biomarker HIF-1α between 3 and 6 months after SARS-CoV-2 infection. Materials and methods. This prospective longitudinal observational study included 69 patients with paired plasma samples collected at 3 and 6 months after COVID-19. Plasma concentrations of ACE2, C3, C4, and HIF-1α were quantified using enzyme-linked immunosorbent assays. Longitudinal changes were evaluated using paired analyses and random-intercept mixed-effects models. Secondary analyses included baseline-adjusted regression, change-score associations, Pearson and Spearman correlations, and ACE2 trajectory evaluation. Robustness was assessed using raw and log-transformed variables, influential-observation analyses, alternative trajectory thresholds, and exploratory nonlinear ACE2 models. False-discovery-rate correction was applied to prespecified hypothesis families. Results and discussion. ACE2, C3, C4, and HIF-1α showed distinct longitudinal patterns between 3 and 6 months. Time-by-baseline-ACE2 interactions differed across the three biomarkers, with C4 showing the strongest ACE2-related signal, followed by C3 and HIF-1α. Although the linear interaction terms did not remain statistically significant after false-discovery-rate correction, complementary baseline-adjusted and change-score analyses indicated differentiated complement and hypoxia-related responses. Sensitivity analyses identified C4 as the leading candidate for a complex ACE2-associated longitudinal pattern, including evidence of nonlinearity, whereas HIF-1α showed pronounced temporal change with comparatively limited ACE2-centered association. Conclusions. Complement and hypoxia-related biomarkers displayed distinct longitudinal behavior in relation to ACE2 during long COVID recovery. C4 showed the most prominent evidence of a potentially complex ACE2-associated pattern, while C3 and HIF-1α demonstrated different longitudinal profiles. These findings indicate that post-COVID biomarker evolution is pathway specific and cannot be fully represented by a single uniform linear ACE2-centered mechanism.