Prognostic Value of Serum Creatinine in Severe Covid-19 Patients in Wuhan, China: A Retrospective Cohort Study


 Objectives: Although the respiratory and immune systems are the major targets of SARS-CoV-2, increasing evidence revealed that kidney injury was not rare in coronavirus disease 2019 (COVID-19). However, the incidences of kidney abnormalities were significantly different, from 0.5 to 75.4% in several reports. The association of kidney injury with prognosis remain controversial.Methods:In this retrospective single center cohort study, laboratory confirmedCOVID-19inpatients with severe type were enrolled. Demographic, clinicaland laboratory data were collected. Association ofserum creatinine (SCr)with 28-days mortality in severe COVID-19 patients was analyzed.Results:18.79% (48/304) patients died during the first 28-days of hospitalization.Non-survivors had a significantly higher SCr levels than survivors (109.27μmol/L vs. 69.99μmol/L, P <0.001). The 28-days mortality in high SCr group (>76μmol/L) was significantly higher than that in low SCr group (31.7% vs. 7.5%, P <0.001). Multivariate logistic regression revealed that the independent risk factors of 28-days outcome included age(OR: 2.95, 95%CI: 1.08-8.05), WBC (OR: 6.09, 95%CI: 2.27-6.39), lymphopenia (OR: 3.49, 95%CI: 1.55-7.92), IL-6 (OR: 4.44, 95%CI: 1.64-11.99) and SCr (OR: 2.69, 95%CI: 1.18-6.11). Kaplan-Meier analysis demonstrated the survival disadvantage in patients with high SCr levels (>76μmol/L). ROC curve showed the SCr cut-off value for predicting 28-days death was 77.5 μmol/L, with the sensitivity of 68.8% and speciality of 74.1%.Conclusion: SCr was associated with poor prognosis and might be an independent risk factor for in-hospital death. The cut-off value of SCr for prognosis prediction was 77.5 μmol/L, with the sensitivity of 68.8% and speciality of 74.1%.

critical illness as high as 28.3% 1 . Although the respiratory and immune systems are the major targets of COVID-19, increasing evidence revealed that SARS-CoV-2 infection could also be found out of lung, such as digestive system, cardiovascular system and coagulant system 1 . Disorder of coagulation and injury of cardiovascular system have been proved associated with in-hospital mortality in COVID-19 patients [1][2] .
The incidences of kidney injury in COVID-19 of different regions showed large difference. Some studies revealed acute kidney injury (AKI) occurred in 0.5-7% of cases and in 2.9-23% of ICU patients [2][3][4][5] . A recent report indicated 75.4% patients had abnormal urine dipstick tests or AKI 6 . However, other report demonstrated that COVID-19 did not result in AKI 7 .
Although many patients showed mild increase of blood urea nitrogen, creatinine and emerging of proteinuria, most of them did not meet the diagnostic criteria of AKI. Transit blood urea nitrogen elevation or proteinuria might be due to many kind of factors, such as infection, in ammation, hypoxaemia, shock or drugs. Whether the AKI of COVID-19 is caused by a coronavirus-induced cytopathic effect or cytokine storm-induced systemic in ammatory response remains unclear.
The purpose of this study was to explore the potential association of kidney function with prognosis of severe COVID-19 patients, which would be bene t for early identi cation of patients at risk of deterioration.

Scr Associated With Prognosis In Severe Covid-19 Patients
As shown in Table 2, the 28-days mortality in high SCr group was signi cantly higher than that in low SCr group (31.7% vs. 7.5%, P < 0.001). Multivariate logistic regression analysis (  a Includes congestive heart disease and coronary atherosclerotic heart disease.

Discussion
This retrospective report identi ed the serum creatitine independently associated with poor prognosis in severe COVID-19 patients, and the cut-off value for predicting the 28-days mortality was 77.5 µmol/L, with the sensitivity of 68.8% and speciality of 74.1%. This cut-off value is signi cantly lower than normal range, which suggesting that potential kidney injury on admission represented a higher risk of deterioration. Early detection and treatment of renal abnormalities might bene t to improve the vital prognosis of severe COVID-19.
Epidemiological data indicate that at least 20% of COVID-19 patients have severe disease. Although the respiratory and immune systems are the major targets of COVID-19, kidney injury is a major complication.
It is reported showed that 6.7% of patients with SARS developed acute kidney injury (AKI) 8 . However, the incidences of kidney injury were signi cantly different, from 0.5 to 75.4% in several reports [2][3][4][5] . This might be due to the different methods of kidney injury evaluation and de nition. About 68.5% patients experienced remission of proteinuria, 45.7% experienced complete recovery of kidney function 6 . Although renal abnormalities occurred in the majority of patients with COVID-19, it is controversial how to evaluated AKI precisely in COVID-19. Classical assessment of AKI is still based on SCr and urine output, but they represent only indicators of established kidney damage. Update, it remain di cult to determined a perfect bio-marker for AKI. In this study, we used SCr as a substituted method for kidney function evaluation. The incidence of elevated serum creatitine was 24.7% (75/304), which was signi cantly higher than previous report (14.1%) 5 . It might due to the different proportion of enrolled subjects.
Previous studies have focused on the clinical features and risk factors for death. Although AKI was popular in non-survivors in many previous reports, the affect of kidney injury on COVID-19 prognosis remain controversial. Some reports indicated that kidney disease or AKI associated mortality in COVID-19 patients [5][6] . But a recent study revealed that COVID-19 did not result in acute kidney injury (AKI) 8 . In previous study of patients with H1N1 virus infection, only those cases in the AKI III category were independently associated with mortality 9 . Most recent reports in COVID-19 evaluated kidney impairment using AKI or elevated SCr 5-6, 10 . However, the cut-off value of SCr for prognosis prediction in our study is 77.5 µmol/L, which is signi cantly lower than normal range. Furthermore, the hazard ratios of elevated baseline SCr in OVID-19 patients mortality was obviously lower than hematuria (2.04 vs. 8.89) or proteinuria (2.04 vs. 6.80) 5 . In our study, the OR for a nornal SCr level (> 76 µmol/L) was 2.69 (95%CI: 1.18-6.11), which was higher than previous report. It might be due to two reasons. Firstly, the proportion of enrolled subjects was different. Our study mainly enrolled severe or critical COVID-19 patients, but previous report enrolled 57.3% mild or moderate patients 5 . Secondly, we excluded patients with chronic kidney diseases. Lastly, elevated SCr might be not a sensitive bio-marker for kidney impairment evaluation in early stage. SCr is a delayed renal functional marker when it is generally increased after sever kidney damage. A recent study demonstrated that among patients with baseline SCr ≥ 0.7 mg/dl (61.9 µmol/l) those who experienced a 0.3 mg/dl increase in SCr within 48 h had clinically meaningful differences in outcomes (including length of hospital stay and mortality) when comparing with those who experienced a 50% increase in SCr from baseline within 7 days. These ndings suggest that the KDIGO AKI de nition should be revised because the staging may equate states that are not the same. In addition, the AKI de nition should also potentially incorporate aetiology, pathophysiology and novel biomarkers 11 .
The etiology of kidney disease involvement in patients with COVID-19 is likely to be multifactorial. The direct cytopathic effects of SARS-CoV-2 on kidney tissue might be the most important reason.The receptor of SARS-CoV-2, ACE2 was found to be upregulated in patients with COVID-19, and immunostaining with SARS-CoV nucleoprotein antibody was positive in tubules [12][13] . Electron microscopic examination showed clusters of coronavirus particles with distinctive spikes in the tubular epithelium and podocytes. In addition, kidney injury may be ascribed to several causes including an in ammatory/immune reaction characterized by an enhanced release of circulating mediators able to interact and damage kidney-resident cells 8,14 . This study showed that patients with higher SCr levels (> 76 µmol/L) had lower lymphocyte counts, higher levels of IL-6, and higher hs-CRP which indicated that they were in the state of immune system dysregulation and cytokine storm. Cytokine storm associated with COVID-19 may be another important mechanism of kidney injury in these patients and it is clear that all of these mechanisms require further exploration.
Some limitations existed in this study. Firstly, it is a single-center retrospective study, which only represented a part of severe COVID-19 patients in Wuhan, China. It could not be extrapolated completely to patients with non-severe COVID-19 or patients outside of Wuhan, China, as the diversity of epidemiological feature in different areas. Secondly, the impact of potential therapies, including immunosuppressive medications, antiretrovirals, and immunologic antibody therapy, could not be assessed as these were not uniformly implemented and often unavailable.

Conclusion
In conclusion, SCr was independently associated with short-term prognosis in severe COVID-19 patients (OR: 2.69, 95%CI: 1.18-6.11). The SCr cut-off value for predicting the 28-days mortality was 77.5 µmol/L, with the sensitivity of 68.8% and speciality of 74.1%. Surveillance for SCr may help to identify patients at high risk of death, for which early intervention may be an important prevention strategy.

Declarations
Ethics approval and consent to participate This study was approved by the institutional review boards at Wuhan Tongji Hospital and The First A liated Hospital of Soochow University. Written informed consent was exempted as COVID-19 is an emerging infectious disease.

Consent for publication
Not applicable.
Availability of data and material All data will be shared after request to corresponding authors.

Funding
This work was supported byProgram of Key Talents of Medical Science in Jiangsu Province [ QNRC2016745], Suzhou science and technology development plan [ SYS202008 ].

Competing interests
The authors declare that they have no competing interests.

Authors' contributions
Zeng DX and Jiang JH conceptualized the research aims, design the study, and takeresponsibility for the integrity of the data. Zhang WY and Xu DYcollected the data, performed the statistical analysis, wrote the rst draft of the paper. Wang CG, Liu YY and Huang JA performed the statistical analysis.