Abstract
Background: Increasing evidence revealed that kidney was one of the targets of SARS-CoV-2. However, the incidences of kidney abnormalities were significantly different, from 0.5 to 75.4% in coronavirus disease 2019 (COVID-19) patients. The association of kidney injury with prognosis remain controversial.Methods: In this retrospective cohort study, laboratory confirmedCOVID-19inpatients with severe type were enrolled. Demographic, clinicaland laboratory data were collected. Association of estimated glomerular fifiltration rate (eGFR)with 28-days mortality was analyzed.Results: The total 28-days mortality of hospitalizationwas 22.3% (79/354). Non-survivors had a significantly declined eGFR levels than survivors (75.95 [IQR: 47.22,92.84] ml/min/1.73m2 vs. 96.43 [IQR: 84.11,108.47] ml/min/1.73m2, P<0.001). The 28-days mortality in declined eGFR group (<90 ml/min/1.73m2) was significantly higher than that in normal eGFR group (38.5% vs. 10.7%, P <0.001). Multivariate logistic regression revealed that the independent risk factors of 28-days outcome included lower eGFR (OR: 3.97, 95%CI: 1.42-11.11), elevated WBC (OR: 7.08, 95%CI: 3.15-15.90), lymphopenia (OR: 2.58, 95%CI: 1.21-5.49)andIL-6 (OR: 7.90, 95%CI: 2.19-28.49). Kaplan-Meier analysis indicated the survival disadvantage in patients with declined eGFR. ROC curve showed the eGFR cut-off value for predicting 28-days death was 82.2 μmol/L, with the sensitivity of 76.7% and speciality of 66.3%.Conclusion: Declined eGFR was associated with poor prognosis and could be used an independent risk factor of 28-days mortality in COVID-19 patients. Early detection and surveillance for eGFR may benefit to identify patients with high-risk ofprogression.