C3a and C5b-9 Differentially Predict COVID-19 Progression and Outcome

Abstract

SARS-CoV-2 infection may result in severe pneumonia leading to mechanical ventilation and intensive care (ICU) treatment. Complement activation was verified in COVID-19 and implicated as a contributor to COVID-19 pathogenesis. This study assessed the predictive potential of complement factors C3a and C5b-9 for COVID-19 progression and outcome. We grouped 80 COVID-19 patients into severe COVID-19 patients (n = 38) and critically ill (n = 42) and subdivided into non-intubated (n = 48) and intubated (n = 32), survivors (n = 57) and non-survivors (n = 23). Results: A significant increase for C3a and C5b-9 levels was observed between: severely and critically ill patients (p < 0.001 and p < 0.0001), non-intubated vs intubated (p < 0.001 and p < 0.05), survivors vs non-survivors (p < 0.001 and p < 0.01). ROC analysis for the need for ICU treatment revealed a higher AUC for C5b-9 (0.764, p < 0.001) compared to C3a (AUC = 0.739, p < 0.01). A higher AUC was observed for C3a for the need for intubation (AUC = 0.722, p < 0.001) or mortality (AUC = 0.740, p < 0.0001) compared to C5b-9 (need for intubation AUC = 0.656, p < 0.05 and mortality AUC = 0.631, p = NS). Combining the two markers revealed a powerful prediction tool for ICU admission (AUC = 0.773, p < 0.0001), intubation (AUC = 0.756, p < 0.0001) and mortality (AUC = 0.753, p < 0.001). C3a and C5b-9 may be considered as prognostic tools separately or in combination for the progression and outcome of COVID-19.

Keywords: COVID-19; biomarkers; complement; mortality.

Figure 1

Increased levels of C3a and C5b-9 in COVID-19 patients. C3a and C5b-9 levels measured on admission were higher in critically ill compared to patients with severe COVID-19 (a,b), in intubated versus non-intubated (c,d), and in non-survivors compared to survivors (e,f). Data are expressed as means ± SD. Statistical analysis was performed using the Mann–Whitney U test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 2

C3a and C5b-9 levels correlations in COVID-19 disease. C3a levels correlated positively with (a) C-reactive protein, (b) ferritin and (c) fibrinogen levels. (d) A positive correlation between C5b-9 levels with patient hospital length of stay (ward/ICU).

Figure 3

Increased levels of C3a and C5b-9 in COVID-19 patients differentially predict ICU admission, intubation and mortality. Receiver operating characteristic curves for prediction of need for intensive care treatment (a) intubation (b) and survival (c). (d) Corresponding area under the curve (AUC), p values, and optimal cut-off points with combined greatest sensitivity (%) and specificity (%) and odds ratio values are shown.

Figure 4

Prediction of ICU admission, intubation and mortality by C-reactive protein and D-dimers. Receiver operating characteristic (ROC) curves of C-reactive protein for prediction of need for intensive care (a) intubation (b) and mortality (c) and D-dimer levels (d–f). The corresponding areas under the curve (AUC), p, sensitivity and specificity, odd ratio, and cut-off values are represented in (g).