Summary
Until 24:00 of February 7th 2020, 31774 laboratory-confirmed cases of novel coronavirus (2019-nCoV) infection have been reported, including 6101 severe cases in critical conditions and 722 deaths. The critical and urgent need at this moment is to find an effective treatment strategy with available means to prevent these thousands of severe inpatients from worsening and dying. It has been recently known that 2019-nCoV shares a common cellular mechanism with the severe acute respiratory syndrome-associated coronavirus (SARS-CoV). Thus, we surveyed a previous retrospective case study on SARS which showed that acute renal impairment was uncommon in SARS but carried a formidably high mortality (91.7%, 33 of 36 cases). Here we report an ongoing case study on kidney functions in 59 patients infected by 2019-nCoV (including 28 diagnosed as severe cases and 3 deaths). 63% (32/51) of the patients exhibited proteinuria, indicative of renal impairment. 19% (11/59) and 27% (16/59) of the patients had an elevated level of plasma creatinine and urea nitrogen respectively. The computerized tomography (CT) scan showed radiographic abnormalities of the kidneys in 100% (27/27) of the patients. Together, these multiple lines of evidence point to the idea that renal impairment is common in 2019-nCov patients, which may be one of the major causes of the illness by the virus infection and also may contribute to multi-organ failure and death eventually. Therefore, we strongly suggest exercising a high degree of caution in monitoring the kidney functions of 2019-nCoV patients and, very importantly, that applying potential interventions including continuous renal replacement therapies (CRRT) for protecting kidney functions as early as possible, particular for those with plasma creatinine rising, is key to preventing fatality.
Background
A recent seminal work by the lab of Dr. Shi1 provided strong evidence that the 2019-nCoV uses the ACE2 (angiotensin converting enzyme II) as a cell entry receptor, a cellular mechanism identical to that of the SARS-CoV2,3. Moreover, a gain-of-function experiment reported in 20154 verified that the ACE2 could be indeed used by a wildtype coronavirus of a bat origin. However, ACE2 is not exclusively expressed in the respiratory organs. Previous reports on the ACE2 expressing level in human tissues using RNA-seq method (Fig. 1) showed that the ACE2 expression in gastrointestinal organs (small intestine, duodenum) or urinary organs (kidney) was much higher (nearly 100-fold) than that in respiratory organs (lung). This result suggests that the kidney is a possible target that can be infected by either SARS-CoV or 2019-nCoV. On the other hand, according to the clinical case data, a detailed retrospective study of the 2003 SARS-CoV outbreak5 found that acute renal impairment was uncommon in SARS but carried a formidably high mortality (91.7%, 33 of 36).
Data adapted from https://www.ncbi.nlm.nih.gov/gene/?term=59272
Note a recent fact that the reported infectious cases in the ongoing 2019-nCoV outbreak (31774, until 24:00 of Feb. 7th 2020, obtained from the Chinese National Health Commission) already much greatly outnumbered the overall reported cases in the 2003 SARS-CoV outbreak (8098). Strikingly, the situation is keeping highly serious because of a large number of severe inpatients (∼19%, 6101 of 31774). Therefore, any promising strategies to prevent these patients from worsening and dying shall be urgently developed. And in turn, the entire society will benefit from relieving the stress caused by fear of death.
Participants
For this retrospective study, we randomly recruited 59 patients from Jan. 21st to Feb. 7th, 2020 at Tongji Hospital and Wuhan Pulmonary Hospital in Wuhan, Huangshi Central Hospital in Huangshi, and Southwest Hospital in Chongqing, China. According to the arrangements conducted by the Chinese Government, adult patients (aged > 18 years) were admitted to the hospitals without selectivity. All patients who were enrolled in this study were diagnosed as infected by 2019-nCoV pneumonia according to the guidance provided by the China’s National Health Commission. This study was approved by the Ethics Commissions of Tongji Hospital, Wuhan Pulmonary Hospital, Huangshi Central Hospital and Southwest Hospital.
Results
We analyzed 59 cases of 2019-nCoV inpatients from multiple hospitals in Wuhan as well as in another nearby outbreak city Huangshi, and also from Chongqing (∼900 km away from Wuhan, 426 infected cases reported until 24:00 of February 7th 2020). We focused our analysis on some key function indicators of the kidney, including urine protein, blood urea nitrogen (BUN), plasma creatinine (Cre), and renal CT scan data.
These 59 patients (34 males and 25 females, in the age of 28 to 83, median age of 52) included 28 that were diagnosed as ‘severe’ cases according to the guidance issued by the China’s National Health Commission, and 3 deceased cases (their ages were 66, 62 and 48 respectively). Demographic and symptom profiles were obtained in all patients on admission. The statistical analyses showed that the blood urea nitrogen (BUN) had a significant difference between the group of severe cases and the group of the rest (***p<0.001, Wilcoxon rank sum test), while both the urine protein and plasma creatinine had no significant difference between the groups.
1. Proteinuria: 32/51 (63%) patients exhibited proteinuria
The test of urine protein was collected from 51 patients (Fig. 2a) and proteinuria was found in 32/51 (63%) patients. The semiquantitative result showed (+) in 24 patients (47%), (++) in 5 patients (10%), and (+++) in 3 patients (6%) (Fig. 2b). Note that among these patients with proteinuria (32 cases), a large number (64%) was found to have the detected urine protein on the first day of admission, suggesting the presence of renal impairment already before or at the moment of admission, offering a potential time point for starting interventions to protect kidney functions.
a, Level values for all patients tested with respect to urine protein. Red circles correspond to proteinuria. b, Percentages of patients tested (n = 51). Note that 0 represents (-), and 1, 2, 3 represent (+), (++) and (+++) respectively.
2. Blood urea nitrogen (BUN): 16/59 (27%) patients exhibited an elevated level of BUN
In the patients with increased BUN levels (16/59 cases), the peak plasma BUN levels ranged from −0.46 to 7.40 of the normalized value, with a median value of 0.32 of the normalized value (normal value before normalization: 1.7-12.8 mmol/L; Fig. 3a, b). In 23 cases, we were able to collect the data for the change in BUN levels over days (Fig. 3c), and this result showed that 43% (10/23) of these patients underwent an increase in BUN and the duration from the onset of admission to the presence of the BUN increase was in the range of 2 to 10 days (median 4). Note that the 2/3 deceased patients had rather high levels of BUN before death (peak values: 23.61 and 32.09 mmol/L respectively; pointed out by the last two arrows in Fig. 3c).
a, Values for all patients tested with respect to BUN. Red triangles correspond to elevated levels of BUN. b, Percentages of patients exhibited normal and elevated BUN (n = 59). c, BUN values for the patients (n = 23) tested over multiple days. The color-coded triangles correspond to the BUN values of different days. The red arrows indicate the deceased cases. Note that the dotted lines indicate the normal level that has been normalized to 1.
3. Plasma creatinine (Cre): 11/59 (19%) patients exhibited an elevated level of Cre and 3 deaths had an extremely high level
In the patients with increased Cre levels (11/59 cases), we found the peak plasma Cre levels ranged from −1.23 to 5.97 of the normalized value (median 0.42; normal value before normalization: 40-115 μmol/L) (Fig. 4a, b). In 23 cases, we could monitor the change in Cre levels over days (Fig. 4c). We found that the duration from the onset of admission to the presence of the Cre increase was in the range of 1 to 10 days (median 5). Note that rather high levels of Cre before death in all 3 deceased patients (peak values: 209, 280 and 286 μmol/L respectively; pointed out by red arrows in Fig. 4c), supporting the Cre level as a risk factors predicting mortality in coronavirus-infected patients5.
a, Values for all patients tested with respect to Cre. Red squares correspond to elevated level of Cre. b, Percentages of patients exhibited normal and elevated Cre (n = 59). c, Cre values for the patients (n = 23) tested over multiple days. The color-coded squares correspond to the Cre values of different days. The red arrows indicate the death cases. Note that the dotted lines indicate the normal level that has been normalized to 1.
4. CT data: all 27/27 patients exhibited radiographic abnormalities of the kidneys
In 27 patients, we were able to collect the plain CT scan results of the parenchyma of the kidneys in addition to the chest CT scan. 14 of these 27 patients exhibited proteinuria, 3 had an increased level of creatinine and 4 had an increased level of BUN. The CT value of the renal parenchyma was measured on the image of the largest layer of the kidney6,7. We found that the CT value of bilateral renal parenchyma was in the range of 19.5-34.97, with a median of 26.67, which was significantly lower than patients without kidney disease (35 HU) in our hospital (***p <0.001, Wilcoxon sign rank test). It was also smaller than the result of the previous study (normal kidney CT value of 38 HU)8 (see the example of two patients in Figure 5a and b). 27/27 patients had a smaller value than 35HU (Fig. 5c). The results indicate that inflammation and edema of the renal parenchyma may occur in 2019-nCoV infected patients.
a and b, CT examples of two different patients (71-year old male and 70-year old male), respectively. The left column is the lung window images of the plan chest CT scan, and the middle column is the mediastinal window images. Multiple lung segments with patchy, segmental density and slightly higher and uniform ground-glass shadows (GGO) were observed. Right column, CT images of the largest layer of kidney and CT values of bilateral kidney. c, Average CT values for all patients (n = 27). Red plusses correspond to the levels of CT values. Note that the dotted lines indicate the normal level.
Discussion and suggestion
The high degree of similarity in cellular mechanism of the 2019-nCoV and the SARS-CoV suggests that the risk factors of mortality could also be similar. In the previous SARS-CoV case study5, acute renal impairment was the top risk factor of mortality (16.91 (8.368–34.16)), even significantly higher than acute respiratory distress 10.28 (4.854–21.77). In that case study, all patients who eventually died (N = 33) had a progressive rise of plasma creatinine, and the rise of plasma creatinine was rapid in those who succumbed early in their illness. In our results, three deceased cases also showed very high levels of creatinine before death.
Compared to the 2003 SARS patients that had only a small fraction (6.7%) of cases with acute renal impairment, we found that, in the patient group that we studied, a remarkably increased number (63%, 32/51) of 2019-nCoV cases exhibited proteinuria (Fig. 2), indicative of renal impairment. In support of renal impairment, we observed an increased level of BUN (Fig. 3) and Cre (Fig. 4) in a substantial fraction of patients. In addition, the CT scan data of the kidneys suggested inflammation and edema occurring in the parenchyma in all cases tested (Fig. 5). The renal impairment could be caused by 2019-nCoV entering the cells through ACE2 that are highly expressed in the kidney (Fig. 1).
Therefore, we strongly suggest exercising a high degree of caution in monitoring the kidney functions of 2019-nCoV severe patients regardless of the past disease history. Upon day-by-day monitoring and analysis, early interventions including continuous renal replacement therapies (CRRT)9 shall be applied as soon as possible to protect kidney functions, particularly for those patients with Cre level monotonically rising. This suggestion of therapy strategy is based on the following rationales:
Any interventions (particularly including CRRT in combination with immuno-adsorption procedure) to protect kidney functions are recommended to be applied to the severe patients with 2019-nCov as early as possible after admission, due to the observation of proteinuria on the first day of admission in a large number of cases.
CRRT has been reported to be an effective treatment in a few cases on severe patients of respiratory diseases including the middle east respiratory syndrome (MERS)10.
CRRT is generally beneficial with a low risk to 2019-nCoV infected patients that already developed severe symptoms. The beneficial effects include e.g., removing inflammatory cytokines11.
CRRT devices and expert users are readily available in large numbers in many hospitals. Approved protocols are well-established and operation costs are much less than that of extracorporeal membrane oxygenation (ECMO) devices.
Data Availability
Please find all data at the following link (code: 7hjn ):
Author Declarations
This manuscript is drafted in an emergency with incomplete acquisition of clinical case data, with respect to the serious overloaded work of the clinicians at the moment. The authors declare no competing financial interests.
Acknowledgements
We greatly appreciate all the people, particularly thousands of clinicians, nurses and scientists including Dr. Li Wenliang, who are heroically battling to eradicate this disaster for days and nights. We thank all patients involved in this study. This study was supported by grants from the “1000-Talents Program for Young Scholars” of China (X. Chen) and the “100-Talents Program for Elite Engineers” of the CAS (H. Jia). X. Chen is a junior fellow of the CAS Center for Excellence in Brain Science and Intelligence Technology.
Footnotes
↵* Email: xiaowei_chen{at}tmmu.edu.cn; junan_yan{at}aliyun.com; zhouyisjtu{at}gmail.com; jiahb{at}sibet.ac.cn
