Biological prognostic parameters in SARS-CoV-2 viral infection are poorly documented. The aim of the present study was to identify biological predictors of mortality in Congolese Covid-19 patients. Methods. This was a historical follow-up study of Covid-19 patients conducted at Monkole Hospital Center, between April 2020 and December 2021. The endpoint was all-cause mortality. Survival was described by the Kaplan-Meier method. Predictors of mortality were identified using multivariate Cox regression. Results. A total of 477 patients (mean age 55.5 ± 17.2 years, male sex 61%) were included. The mortality rate was 14.3%. Patients who died were significantly older and in respiratory distress. Mean values for N/L ratio, blood urea and creatinine, SGOT, LDH and CRP were significantly higher in patients who died than in those who recovered (p<0.001). Overall survival at 7 days, 21 days and 36 days was 89.6%, 74.2% and 66% respectively. Predictors of mortality included age >60 years [aHR = 2.75 (1.332 – 5.674); p = 0.006], respiratory distress [HRa = 0.138 (068 – 0.279); p <0.000], the N/L Ratio [aHR = 1.064 (1.013 – 1.117); p = 0.013], transaminases [aHR = 1.010 (1.002 – 1.018); p = 0.010], LDH [aHR =1.001 (1.000 – 1.002); p = 0.001] and urea [aHR= 1.009 (1.000 – 1.019); p = 0.039] blood concentrations. Conclusion. In the present study, Covid-19-related death was predicted by the high N/L Ratio, pathological values ​​of cell lysis markers (SGOT and LDH) and those of urea. Abnormalities in the biological parameters of patients treated for Covid-19 therefore have prognostic value in our environment, and can guide the management of these patients.