Relationship Between Serum Severe Acute Respiratory Syndrome Coronavirus 2 Nucleic Acid and Organ Damage in Coronavirus 2019 Patients: A Cohort Study

Abstract

Background

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread worldwide and has the ability to damage multiple organs. However, information on serum SARS-CoV-2 nucleic acid (RNAemia) in patients affected by coronavirus disease 2019 (COVID-19) is limited.

Methods

Patients who were admitted to Zhongnan Hospital of Wuhan University with laboratory-confirmed COVID-19 were tested for SARS-COV-2 RNA in serum from 28 January 2020 to 9 February 2020. Demographic data, laboratory and radiological findings, comorbidities, and outcomes data were collected and analyzed.

Results

Eighty-five patients were included in the analysis. The viral load of throat swabs was significantly higher than of serum samples. The highest detection of SARS-CoV-2 RNA in serum samples was between 11 and 15 days after symptom onset. Analysis to compare patients with and without RNAemia provided evidence that computed tomography and some laboratory biomarkers (total protein, blood urea nitrogen, lactate dehydrogenase, hypersensitive troponin I, and D-dimer) were abnormal and that the extent of these abnormalities was generally higher in patients with RNAemia than in patients without RNAemia. Organ damage (respiratory failure, cardiac damage, renal damage, and coagulopathy) was more common in patients with RNAemia than in patients without RNAemia. Patients with vs without RNAemia had shorter durations from serum testing SARS-CoV-2 RNA. The mortality rate was higher among patients with vs without RNAemia.

Conclusions

In this study, we provide evidence to support that SARS-CoV-2 may have an important role in multiple organ damage. Our evidence suggests that RNAemia has a significant association with higher risk of in-hospital mortality.

RNAemia, SARS-CoV-2, COVID-19, organ damage

As of 30 March 2020, there were 707 416 confirmed cases of coronavirus disease 2019 (COVID-19) worldwide, including 33 272 deaths (4.7% fatality rate). Understanding clinical aspects of the pathogenesis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may make it possible to better define targets for developing new therapeutic, preventive, and disease-monitoring strategies. Based on recently published work, SARS-CoV-2 exists not only in the lower respiratory tract but also in blood [1–3]. However, the rate of detection of SARS-CoV-2 RNA in blood was found to be different by different researchers. Wang et al found that viremia was present in only 3 of 307 patients studied [4]. In contrast, Zheng et al reported that 35 of 96 patients had a detectable viral load in serum samples [5]. Meanwhile, marked biochemical abnormalities, reflecting the potential capability of the virus to produce damage in different body compartments, have been observed [6–11]. Nevertheless, studies that have explored the relationships between serum SARS-CoV-2 nucleic acid (RNAemia), biochemical abnormalities, and clinical disorders are scarce. Using a retrospective approach, we conducted an analysis to explore the potential association between RNAemia and lung damage, cardiac damage, liver damage, renal impairment, and coagulation abnormalities.

METHODS

Study Design and Patients

Consecutive patients who were admitted to Zhongnan Hospital of Wuhan University with confirmed COVID-19 were tested for SARS-COV-2 RNA in serum from 28 January 2020 to 9 February 2020 and were included in this retrospective cohort study. The COVID-19 patients enrolled in this study were diagnosed according to World Health Organization interim guidance. The cases with incomplete original reports were excluded.

Data Collection

Data on clinical characteristics, laboratory findings, radiological features, and outcomes were collected from electronic medical records. Confirmed organ damage complications were based on the discharge diagnosis in the electronic medical records. Laboratory assessments consisted of a complete blood count, C-reactive protein, procalcitonin, coagulation test, liver and renal function, and cardiac markers. All information was obtained through standardized data collection forms. Two researchers independently reviewed the data collection forms to ensure data quality.

Sample Collection and SARS-CoV-2 RNA Extraction

All throat swabs and serum samples were collected from COVID-19 patients for SARS-CoV-2 RNA extraction. After collection, the throat swabs were placed into a collection tube that contained 200 μL of virus preservation solution; total RNA was extracted within 3 hours using a respiratory sample RNA isolation kit (Zhongzhi, Wuhan, China). Briefly, 40 μL of cell lysates were added to a collection tube and vortexed for 10 seconds. After standing at room temperature for 10 minutes, the collection tube was centrifuged at 1000 rpm for 5 minutes. The suspension was used for real-time reverse-transcriptase polymerase chain reaction (RT-PCR) assay of SARS-CoV-2 RNA.

Peripheral blood (2 mL) was collected from patients with COVID-19 into red-cap tubes. Serum was separated using centrifugation at 1500 rpm for 5 minutes. After separation, all serum samples were immediately stored at −80°C for total RNA extraction. Total RNA was extracted from 200 μL of serum sample using a commercial nucleic acid isolation kit (Daan Gene, Guangzhou, China) according to the manufacturer’s instruction. The extracted RNA was stored at −80°C for further study.

Real-time RT-PCR Assay of SARS-CoV-2 RNA

SARS-CoV-2 RNA was amplified using a real-time RT-PCR assay kits (Daan Gene, Guangzhou, China), which was carried out in an Eppendorf tube with an ABI prism 7500 (Thermo Fisher Scientific, Waltham, MA). Two target genes, including open reading frame 1ab (ORF1ab) and nucleocapsid protein (N), were simultaneously amplified and tested during the real-time RT-PCR assay. Target 1 (ORF1ab): forward primer CCCTGTGGGTTTTACACTTAA; reverse primer ACGATTGTGCATCAGCTGA; and the probe 5’-VIC-CCGTCTGCGGTATGTGGAAAGGTTATGG-BHQ1-3′. Target 2 (N): forward primer GGGGAA CTTCTCCTGCTAGAAT; reverse primer CAGACAT TTTGCTCTCAAGCTG; and the probe 5’-FAM- TTGCTG CTGCTTGACAGATT-TAMRA-3′. The final amount of real-time RT-PCR reaction mixture was 25 μL, including 17 μL of (ORF1ab/N) solution A, 3 μL of (ORF1ab/N) solution B, and 5 μL of SARS-CoV-2 RNA solution. The real-time RT-PCR assay was performed under the following conditions: incubation at 50°C for 15 minutes and 95°C for 15 minutes, 45 cycles of denaturation at 94°C for 15 seconds, and extending and collecting the fluorescence signal at 55°C for 45 seconds. Fluorescence was monitored at regular intervals during the extension phase. The cycle threshold (Ct) values obtained from the multiple real-time RT-PCR assays were adopted to determine whether SARS-CoV-2 RNA was present in the tested sample. The detection limit of the quantitative PCR reaction was 500 copies/mL. According to the manufacturer, a Ct value of <40 was defined as a positive result, and a Ct value of ≥40 was defined as a negative result. Viral load was calculated by plotting Ct values onto the standard curve constructed based on the standard product.

Chest computed tomography (CT) images were analyzed using an artificial intelligence system provided by Shanghai United Imaging Intelligence Healthcare. The software segmented and then calculated the infected regions, including the whole, left, and right lung; 5 lung lobes; and 18 sections and their corresponding infected area. The calculated parameters involved the volume–percentage of these areas and the volume–distribution at 4 Hounsfield Unit (HU) ranges ([–, −750), [−750, −300), [−300, 49), [50, +)) in infected areas. Two senior radiologists independently participated in the image processing to determine if lesion segmentation was accurate.

Definitions

Fever was defined as a temperature of at least 37.3°C. Septic shock was defined according to the 2016 Third International Consensus Definition for Sepsis and Septic Shock [2]. Respiratory failure was defined as arterial blood oxygen partial pressure <60 mm Hg, with or without carbon dioxide partial pressure >50 mm Hg at sea level, resting state, and breathing air conditions, excluding intracardiac anatomical shunt and primary cardiac output factors such as reduction. Cardiac injury was diagnosed if serum levels of cardiac biomarkers (eg, highly sensitive cardiac troponin I) were above the 99th percentile upper reference limit or if new abnormalities were seen on electrocardiography and echocardiography [2]. Renal injury was diagnosed according to the Kidney Disease: Improving Global Outcomes (KDIGO) clinical practice guidelines [12]. Hepatic injury was defined as alanine aminotransferase or aspartate aminotransferase ≥3 × the upper limit of normal (ULN) or total bilirubin ≥2 × the ULN, regardless of whether there was previous liver disease [13]. Coagulopathy was defined as a 3-second extension of prothrombin time or a 5-second extension of activated partial thromboplastin time. The illness severity of COVID-19 was defined according to the Chinese management guideline for COVID-19 (version 7.0) [14]. Mild cases include nonpneumonia or mild pneumonia. Severe disease refers to dyspnea, respiratory rate ≥30/min, blood oxygen saturation ≤93%, partial pressure of arterial oxygen to fraction of inspired oxygen ratio <300, or lung infiltrates >50% within 24 to 48 hours.

Statistical Analyses

Statistical analyses were performed using SPSS 23.0 (IBM Corp, Armonk, NY). Continuous and categorical variables were presented as median (interquartile range [IQR]) and n (%), respectively. The Student t test was used to compare continuous variables that conform to a normal distribution, and the χ² test was used to compare categorical variables between the 2 groups. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multiple logistic regression. Survival curves were plotted using the Kaplan-Meier method and compared between patients with vs without RNAemia using the log-rank test. Statistical significance was defined as P < .05.

The Zhongnan Hospital of Wuhan University Medical Ethics Committee reviewed and approved this study.

RESULTS

A total of 151 COVID-19 patients were tested for SARS-COV-2 RNA in serum at Zhongnan Hospital of Wuhan University from 28 January 2020 to 9 February 2020. There were 184 serum samples from 151 patients; 47 of 184 serum samples were positive and 137 were negative through real-time RT-PCR. However, the core datasets (including clinical outcomes and symptoms) of 66 patients were lacking due to incomplete original reports. Therefore, 85 patients were included in the final analysis; 32 patients were placed into the positive subgroup and 53 patients were placed into the negative subgroup. There were 118 serum samples and 356 throat swabs from 85 patients. Among them, 14 of 32 positive patients and 13 of 53 negative patients provided more than 2 throat swabs or serum tests of SARS-CoV-2 RNA (Figure 1). The total positive rate in the serum samples and throat swabs was 28.81% and 54.49%, respectively. The viral load of throat swabs was significantly higher than that of the serum samples (P < .0001). The median duration of virus in serum samples (11 days; IQR, 8–14 days) was similar to that in throat swabs (12 days; IQR, 7–17 days; P = .489). The highest detection of SARS-CoV-2 RNA in serum samples was between 11 and 15 days after symptom onset (Figure 2 and Supplementary Table 1).

Figure 1.

A, Flow chart of patient recruitment. B, Results of SARS-CoV-2 RNA detection in serum samples. C, Date distribution of positive and negative samples in 118 serum samples of 85 patients from 28 January 2020 to 9 February 2020. D, Date distribution of positive and negative samples in 356 throat swabs of 85 patients during hospitalization. Abbreviations: RNAemia, serum SARS-CoV-2 nucleic acid; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Figure 2.

A, Comparison of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA viral load by sample types. Grey bars represent means, and black bars represent standard deviation. B, Duration of detection of SARS-CoV-2 RNA by sample type. Grey bars represent medians, and black bars represent interquartile ranges. C, Comparison of the detection of SARS-CoV-2 RNA in throat and serum samples. The rate of detection of SARS-CoV-2 RNA was highest in the throat and serum samples detected between 6 and 15 days and 11 and 15 days after the symptom onset, respectively.

The demographic and clinical characteristics are shown in Supplementary Table 2. The median age of the 85 patients was 56.0 years, and 43 were females; 49.41% of patients had a history of exposure to SARS-CoV-2. Hypertension was the most common comorbidity, followed by cardiovascular disease and diabetes. The distributions of comorbidities were not statistically different for the positive and negative groups. The most common symptoms on admission were cough and myalgia, arthralgia, or fatigue, followed by shortness of breath and fever. During hospital admission, a high respiratory rate and cough were significantly more common in positive cases compared with negative cases (P < .050). The meaningful variables selected using the χ ² test were evaluated using multiple logistic regression analysis. The factors that were found to be significant were temperature at hospital admission <37.5°C (OR, 0.086; P < .050) and cough (OR, 0.330; P < .050; Supplementary Table 3).

Figure 3A is a flow chart for laboratory and CT scan data collection. Since laboratory data were not available for all 118 serum samples within the specified time frame, 85 patients took the first positive or negative time in serum. Laboratory data for the 85 patients showed that 71.76% of patients had lymphopenia. Most patients demonstrated a reduced albumin-to-globulin ratio and albumin and elevated levels of C-reactive protein. Positive cases had more prominent laboratory abnormalities (eg, neutrophilia, reduced total protein, elevated blood urea nitrogen, increased lactate dehydrogenase, and increased hypersensitive troponin I and D-dimer levels) compared with negative cases (P < .050; Table 1 and Supplementary Table 4).

Table 1.

Laboratory Findings and Chest Computed Tomography Findings Using an Artificial Intelligence System for Patients With Coronavirus 2019 During Hospitalization

Laboratory Finding	Total (n = 85), No. (%)	Positive (n = 32), No. (%)	Negative (n = 53), No. (%)	P Value
Neutrophil count ≥6.3 × 109/L	27/85 (31.76)	17/32 (53.13)	10/53 (18.87)	.001
C-reactive protein level ≥10 mg/L	40/60 (66.67)	12/17 (70.59)	28/43 (65.12)	.685
Procalcitonin level ≥0.05 ng/mL	32/80 (40.00)	16/28 (57.14)	16/52 (30.77)	.022
Total protein ≤65 g/L	23/85 (27.06)	21/32 (65.63)	22/53 (41.51)	.031
Blood urea nitrogen ≥7.6 μmol/L	24/85 (28.24)	14/32 (43.75)	10/53 (18.87)	.014
Creatine kinase ≥171 U/L	12/58 (20.69)	8/23 (34.78)	4/35 (11.43)	.069
Creatine kinase-MB ≥ 25 U/L	12/70 (17.14)	7/29 (24.14)	5/41 (12.20)	.192
Lactate dehydrogenase ≥243 U/L	24/60 (40.00)	14/25 (56.00)	10/35 (28.57)	.033
Hypersensitive troponin I ≥26.2 pg/mL	11/40 (27.5)	9/19 (47.37)	2/21 (9.52)	.020
D-dimer ≥500 mg/L	32/75 (42.67)	16/27 (59.26)	16/48 (33.33)	.029
Chest Computed Tomography Findings Using an Artificial Intelligence System
Lesion distribution, median (interquartile range)	Total (n = 79)	Positive (n = 29)	Negative (n = 50)	P Value
Left upper lobe proportion (%)
Apicoposterior	0.8 (0–6.9)	0.4 (0–7.85)	0.85 (0–6.2)	.349
Anterior	0.3 (0–5.3)	1.5 (0–8.15)	0.3 (0–4.125)	.398
Superior lingular	1.6 (0–8.5)	2.9 (0–42.3)	1.55 (0–4.475)	.039
Inferior lingular	0.8 (0–5.0)	1.6 (0.1–25.05)	0.6 (0–3.7)	.063
Right lower lobe proportion (%)
Dorsal	5.9 (0.1–36.8)	11.8 (0–77.65)	3.85 (0.1–28.0)	.059
Medial basal	0.4 (0–6.9)	2.8 (0–15.85)	0.05 (0–3.35)	.077
Anterior basal	1.8 (0–21.5)	2.6 (0.05–52.15)	0.95 (0–14.85)	.035
Lateral basal	8.7 (0.2–43.1)	18.7 (3.1–59.2)	7.4 (0.175–30.725)	.157
Posterior basal consolidation	7.3 (0–24.8)	10.5 (0.45–38.1)	5.0 (0–20.825)	.149
HU [−300, 49) volume (cm3)	35.4 (4.5–110.8)	76.9 (14.6–181.55)	26.3 (3.075–82.975)	.038
HU [−300, 49) proportion (%) vascular	0.9 (0.1–3.4)	1.9 (0.3–5.8)	0.6 (0.1–2.8)	.071
HU [50 +) volume (cm3)	6.3 (0.6–27.9)	18.7 (3.65–41.15)	5.1 (0.5–20.425)	.029
HU [50 +) proportion (%)	0.2 (0–0.8)	0.5 (0.1–1.5)	0.1 (0–0.625)	.065

Laboratory data were selected 1 day before and after collecting the serum samples of patients. Computed tomography images were selected 3 days before and after collecting serum samples. The Student t test was used to compare medians (interquartile range) between the 2 groups. The χ ² test was used to compare n (%) variables between the 2 groups.

Figure 3.

A, Flow chart for laboratory and CT scan data collection. Laboratory data were selected 1 day before and after collecting the serum samples of patients. CT images were selected 3 days before and after collecting the serum samples of patients. B, Distribution of disease severity and results of SARS-CoV-2 RNA detection in 85 patients. C, Kaplan-Meier survival curves for mortality during the time from serum sampling for SARS-CoV-2 RNA. D, Comparison of SARS-CoV-2 RNA viral load by disease severity in 32 serum SARS-CoV-2 nucleic acid patients. Grey bars represent means, and black bars represent standard deviation. Abbreviations: CT, computed tomography; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

CT images were taken 3 days before and after the detection of RNAemia, and 2 patients did not meet the criteria. Four patients had only X rays due to serious illness, thus, 79 patients had chest CT scans. In total, 50 parameters were obtained from each set of CT images, including the volume and percentage of lung infection and each of its substructures. Table 1 shows that the proportion of infected lung in the positive cases was more greater compared with the negative cases, which had a significant difference in the left upper superior lingular and right lower anterior basal lobes (P < .050). Furthermore, by setting different HU ranges, it was found that HU [−300,49) and HU [50+) in the positive cases had higher volumes of infection (P < .050). Results from additional laboratory tests and CT image analysis are provided in Supplementary Tables 4 and 5.

During hospital admission, organ damage (P < .001) was more common in patients with RNAemia than in those without RNAemia and included respiratory failure (P < .001), followed by cardiac damage (P < .050), renal damage (P < .050), and coagulopathy (P < .050). This result was consistent with previous laboratory tests. Patients with RNAemia vs those without RNAemia had shorter durations from serum testing SARS-CoV-2 RNA (P < .050). Severe symptoms (P < .050) and mortality rate (P < .050) were higher among patients with vs without RNAemia, as shown in Table 2 and in the Kaplan-Meier survival curves in Figure 3C. In an in-depth study of RNAemia patients, the mean viral loads in serum samples were lower in severe cases than in mild cases, but the difference was not significant (Figure 3D).

Table 2.

Complications and Clinical Outcomes of 85 Patients With Coronavirus 2019

Complication	Total (n = 85), No. (%)	Positive (n = 32), No. (%)	Negative (n = 53), No. (%)	P Value
Septic shock	5/85 (5.88)	3/32 (9.38)	2/53 (3.77)	.557
Organ injury	23/85 (27.06)	18/32 (56.25)	5/53 (9.43)	.000
Respiratory failure	18/85 (21.18)	15/32 (46.88)	3/53 (5.66)	.000
Cardiac injury	13/85 (15.29)	9/32 (28.13)	4/53 (7.55)	.025
Renal injury	9/85 (10.59)	7/32 (21.88)	2/53 (3.77)	.024
Hepatic injury	1/85 (1.18)	1/32 (3.13)	0	.798
Coagulopathy	5/85 (5.88)	5/32 (15.63)	0	.013
Multiple organ injuries (organ ≥ 2)	18/85 (21.18)	13/32 (40.63)	5/53 (9.43)	.001
Clinical outcomes
Severe	41/85 (48.24)	20/32 (62.50)	21/53 (39.62)	.041
Discharge from hospital	69/85 (81.18)	19/32 (59.38)	50/53 (94.34)	.000
Death	13/85 (15.29)	10/32 (31.25)	3/53 (5.66)	.004
Remained in hospital	4/85 (4.71)	3/32 (9.38)	1/53 (1.89)	.293

The Student t test was used to compare medians (interquartile range) between the 2 groups. The χ² test was used to compare n (%) variables between the 2 groups.

DISCUSSION

In this study, we shed light on the association between the level of SARS-CoV-2 RNAemia and organ damage using data from real clinical practice. There are several important findings from our study.

First, we compared positive throat and serum samples. The viral load of throat swabs was significantly higher than that of serum samples, which was consistent with a recent study [5]. It was also found that the percentage of positive serum samples was higher than of throat swabs 0 to 15 days after symptom onset. It seems that PCR performed on serum is more sensitive than PCR performed on respiratory samples; this needs to be confirmed with a larger sample size. The peak of SARS-CoV-2 RNA in serum was at around 11 to 15 days after symptom onset; however, the peak of SARS-CoV in plasma was at around 3 days after symptom onset [15], which may be why the condition of SARS-CoV patients deteriorates faster than that of SARS-CoV-2 patients. SARS-CoV caused 8096 confirmed cases and 774 deaths (9.6% fatality rate) in 29 countries from November 2002 to July 2003 [16]. As of 30 March 2020, the fatality rate of COVID-19 was 4.7%. The fatality rate of SARS-CoV-2 is currently lower than that of SARS-CoV.

Second, in the current study, the risk of RNAemia among patients with a temperature at hospital admission that was >39.0°C was 11.63 times higher than for patients with a temperature <37.5 °C. Patients with cough were 3.03 times more likely to have RNAemia than those without cough. This was consistent with previous studies in which patients with SARS-CoV-2 were more likely to have a fever and cough [17]. Among 85 COVID-19 patients, 23 had organ damage, 18 of whom had RNAemia. One of the most serious complications was respiratory failure. By setting different HU ranges, it was found that HU [−300, 49) and HU [50+) were more infected in RNAemia patients, which represented the lung consolidation and blood vessels were more infected in RNAemia patients. This is consistent with autopsy analysis that revealed pulmonary vascular congestion and inflammatory clusters with fibrinoid material and multinucleated giant cells [18]. In our study, the next severely damaged organ was the heart. Myocardial enzyme biomarkers indicated cardiac damage due to increased lactate dehydrogenase and hypersensitive troponin I in RNAemia patients. In a cohort of 416 patients diagnosed with COVID-19, this was supported by the fact that the median values of cardiac abnormality markers were higher in patients with heart injury than in those without [6, 7]. The third serious complication was kidney, which reduced total protein and elevated blood urea nitrogen in RNAemia patients. These findings are consistent with those from a previous study [19]. In addition, RNAemia patients demonstrated high coagulopathy with elevated D-dimer levels in our study. D-dimer >1 µg/L is associated with fatal outcome from COVID-19, although its underlying mechanism is unclear [11]. It is speculated that the induction of procoagulant factors and hemodynamic changes predispose patients to ischemia and thrombosis [20].

The multiple organ damage with RNAemia mentioned above may be related to binding of the SARS-CoV-2 S protein to the membrane receptor angiotensin-converting enzyme II (ACE2) on the host cell. ACE2 receptor has been documented to be expressed in alveoli, heart, and kidneys [21] and had an approximately 10- to 20-fold higher affinity for SARS-CoV-2 than SARS-CoV because the “down” conformation of SARS-CoV-2 was angled closer to its center [22]. In addition, the SARS-CoV-2 S glycoprotein harbors a furin cleavage site at the boundary between the S1 and S2 subunits, which helps expand the spread of the virus and, coupled with the transport of blood, it is more conducive to the virus reaching the organs and causing damage [23]. Also, some scholars propose that the extensive replication of coronavirus in the alveoli results in the breakdown of the alveolar vessel; the virus then leaks into the bloodstream and spreads throughout the body [24].

Third, viral encephalitis or viral meningitis can occur in RNAemia, but no clinical manifestations of central nervous system invasion were found in our study. However, SARS-CoV-2 was found in the cerebrospinal fluid of a patient with COVID-19 in China, and loss of patients’ taste or smell was confirmed in COVID-19 cases in Japan and Korea. Medical staff should be aware of the possibility of the central nervous system becoming infected by SARS-CoV-2 and improve its detection in cerebrospinal fluid.

Finally, the current study documents that RNAemia patients had severe symptoms and high mortality rates, suggesting that SARS-CoV-2 RNAemia levels are strongly associated with unfavorable clinical outcome. Twelve of the 32 RNAemia patients were mild cases in our study, which was inconsistent with previous studies. Two teams found that patients who were all positive by blood (6 of 57) or serum (5 of 48) were severe [3, 24]. However, SARS-CoV-2 RNA was found in serum of mild patients in our study, and the mean Ct value of severe patients was lower than that of mild patients, which means that the viral load in severe patients is relatively large, although there is no statistically significant difference. This may also be due to a “positive” PCR result, reflecting only the detection of viral RNA, and does not mean or indicate the presence of viable virus [25]. This requires further observation of viral activity by cell culture in vitro. RNAemia patients presented with increased neutrophils and C-reactive protein as well as decreased lymphocytes in our study. This indicates an inflammatory response caused by viral invasion, which might induce cytokine storms or hyperinflammation and lead to high mortality rates [26]. Chen et al suggested that an extremely high interleukin 6 level was closely correlated with the incidence of RNAemia and mortality [3], which increases the credibility of our conclusion.

Our study has some limitations. First, under the premise of controlled transmission in Wuhan, the present study lacks evidence from magnetic resonance imaging for determining the features of the kidney, heart, and other organs and gaining better insight. Second, further follow-up is needed to determine whether long-term organ damage will occur in patients who are negative for SARS-CoV-2 in blood after hospital discharge.

In conclusion, our study provides new evidence to support the hypothesis that SARS-CoV-2 may play an important role in multiple organ damage, such as respiratory failure, cardiac damage, renal damage, and coagulopathy. We did not find strong evidence that SARS-CoV-2 plays a role in damage to the liver or the central nervous system. RNAemia has a significant association with organ damage of COVID-19 patients and it is associated with a higher risk of in-hospital mortality. Although the underlying mechanism of RNAemia needs to be further explored, the findings presented here highlight the need to perform routine examination of virus in blood in the clinic, which can potentially detect RNAemia early and guide clinicians to targeted treatment in order to prevent multiple organ damage.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. Haibo Xu had full access to all of the data and takes responsibility for data integrity and the accuracy of the data analysis, including any adverse effects. Dan Xu and Wenbo Sun made substantial contributions to the study concept and design. Dan Xu was in charge of the manuscript draft. Lan Lan and Huan Li collected clinical data and confirmed data accuracy. Wenbo Sun and Feng Xiao collected computed tomography images and confirmed data accuracy. Fuling Zhou and Vijaya B. Kolachalama participated in drafting the manuscript, and revising it based on reviewers’ comments. Dan Xu and Wenbo Sun made substantial contributions to data acquisition and analysis. Ying Li made contributions to interpretation. Liangjun Chen and Yirong Li were in charge of the laboratory tasks, including sample processing and detection. Haibo Xu and Xinghuan Wang made substantial revisions to the manuscript.

Financial support. This work was supported by the National Natural Science Foundation of China (grant 81771819) and the National Key Research and Development Program of China (grants 2017YFC0108803 and 2020YFC0845500).

Potential conflicts of interest. The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

Author notes

D. X., F. Z., and W. S. contributed equally to this manuscript.