COVID-19 patients (n = 446) admitted to our institute between February 21, 2021, and March 17, 2022 were recruited for this study. Patients were identified and selected based on their admission records to the internal medicine department and the ICU. Inclusion criteria included a confirmed diagnosis of COVID-19 through PCR testing, and patients ranged in age from 8 days to 96 years. Informed consent was obtained from all participants or their legal guardians before enrollment. The Ethics Committee of KFSH&RC approved this study (IRB No.: 2201086), and the study was conducted in accordance with the Declaration of Helsinki, 1975. Patients were recruited from both the internal medicine department (n = 237) (53.14%) and the ICU (n = 209) (46.86%). Patients ranged in age from 8 days to 96 years (median: 54.9 years). Just over half the sample (50.45%) was male. A confirmed diagnosis of SARS-CoV-2 was performed by RNA detection of the virus using RT-PCR on both blood samples and combined nasopharyngeal and oropharyngeal swab samples. This comprehensive molecular diagnosis approach ensured accurate detection of SARS-CoV-2 RNA across different specimen types. The study excluded individuals with unconfirmed COVID-19 diagnoses.

The patient’s demographics, clinical data, clinical presentation, and routine laboratory results were obtained by accessing the patient’s electronic medical file on the day of enrollment, the third, the seventh, and the fourteenth. In accordance with hospital policies and the COVID-19 procedure, all patients were clinically categorized as belonging to four separate stages can be used to classify the severity of infection (Abudouleh et al., 2023).

These stages were defined as follows:

Stage A was the least severe, while Stage D was the most severe. The distribution of COVID-19 stages among patients was as follows: Stage C was the most common, with 244 cases (54. 83%). %), followed by Stage B with 108 cases (24.27%). Stage D had 55 cases (12.36%). and Stage A was the least frequent, with 39 cases (8.54%). Approximately 90% of the COVID-19 patients exhibited comorbidity conditions, and over 60% of them were unvaccinated ( Figure 1 ).

Blood samples were collected into tubes containing citrated blood (3.2%) and EDTA. When the samples were collected, they were all citrated, centrifuged for 15 minutes at 2000–2500 RCF, and then separated into aliquots to test for various coagulation markers (D-dimer, fibrinogen, prothrombin time, international normalized ratio (INR), and activated partial thromboplastin time (aPTT) at the time of hospital admission using STAR Max^® (Diagnostica Stago, Marseille, France). Automated SYSMEX XN-10 (Sysmex Corporation, Kobe, Japan) equipment was used to count the platelets in the EDTA samples. The remaining aliquots were promptly preserved at a temperature of -80°C until the testing process was done.

The plasma concentrations of PAI-1 (Asserachrome #00949), activated and inactivated TAFI (Asserachrome #00616), and tPA (Asserachrome #00948) were assessed using an enzyme-linked immunosorbent assay (ELISA) with colorimetric output, in accordance with the manufacturer’s protocol (Asserachrome, Diagnostica Stago, France). These assays were performed on an ELISA plate reader (SYNERGY-HTX, BioTek Instruments, USA).Furthermore, the plasma concentrations of plasminogen (Stachrom #00658) and α2AP (Stachrom #00659) levels were determined using a colorimetric (STA-Stachrom) assay, employing the synthetic chromogenic substrate method. The measurements were conducted on the STA Compact Max3 Analyzer, a fully automated clinical analyzer from Diagnostica Stago Company (France).

To detect SARS-CoV-2 in blood samples, we performed a series of steps to identify RNAemia and mutations in COVID-19 patients. RNA extraction was conducted on each patient’s blood sample, followed by screening tests on all 446 samples to detect SARS-CoV-2 RNAemia. The extraction was carried out using the MagMAX™ Viral/Pathogen II (MVP II) Nucleic Acid Isolation Kit (Catalog # A42352, Thermo Fisher Scientific, MA, USA) with the KingFisher™ Flex system (Catalog # 5400610, Thermo Fisher Scientific). The quality and quantity of the extracted RNA were assessed using a NanoDrop spectrophotometer, and nucleic acid purity was determined by calculating the absorbance ratio at 260/280 nm, with a ratio of approximately 2.0 indicating pure RNA.

SARS-CoV-2 RNAemia by RT-qPCR with viral load estimated based on CT values of the N, ORF1ab, and S genes. In-house primers, which were previously described (Alhamlan et al., 2023), were used for RNA detection. The detailed RT-qPCR protocol, including primer sequences and reaction conditions, has been outlined in our previous work (Alhamlan et al., 2023).

Whole-genome sequencing was performed for all SARS-CoV-2 RNAemia samples using the Ion AmpliSeq SARS-CoV-2 Research Panel on the Ion Torrent S5™ platform. Variants were identified using the S5 Torrent Suite™ software Variant Caller, version 5.12.

In this investigation, we utilized GraphPad Prism Version 9 for statistical analysis. When comparing two groups, we examined the Mann-Whitney U test for skewed data and a two-sided t-test for normally distributed data. The Shapiro-Wilk test was used to determine normality. Additionally, Fisher’s exact test was performed to contingency tables. Descriptive analysis was applied using bar graphs and dot plots to assess the distribution of instances over time within each group. Significance was defined as a p-value less than 0.05, with two tails. The plasma levels of the examined biomarkers were presented as population medians, accompanied by either the 95% confidence interval (CI) or the interquartile range (IQR).

In patients admitted to the ICU, the white blood cell count, particularly neutrophils, was significantly elevated, while lymphocyte and monocyte counts were notably lower. Our study revealed that ICU patients had markedly higher levels of fibrinogen and D-dimer compared to non-ICU patients. Specifically, the median fibrinogen level was 5.04 g/L (IQR: 0.65-9.71 g/L, P < 0.001), and the median D-dimer level was 1.75 µg/ml (IQR: 0.5 to >20 µg/ml, P = 0.015). In contrast, hemoglobin levels, aPTT, platelet count, and INR did not show significant differences between ICU and non-ICU patients. Table 1 provides a detailed summary of the selected laboratory results for COVID-19 patients.

Specifically, 74 samples (16.6%) were positive in plasma, 28 samples (6.3%) in whole blood, and 21 samples (4.7%) in both whole blood and plasma Figure 2A . The cycle threshold (CT) value of the N, ORF ab1, and S genes in blood samples ranged from 24 to 36, with a median of 32 ( Figure 2B ). In contrast, the CT values in nasopharyngeal samples ranged from 19 to 30, with a median of 23.4. Despite the lower CT values in nasopharyngeal samples, blood samples were chosen for sequencing to investigate the presence of SARS-CoV-2 RNAemia and its correlation with disease severity and thrombosis. Sequencing blood samples allows for the detection of viral RNA in the bloodstream, which is associated with more severe disease outcomes and can provide insights into the systemic nature of the infection. This comprehensive analysis underscores the significance of sampling multiple sites to accurately assess the presence and extent of SARS-CoV-2 RNAemia, which may have implications for understanding the systemic nature of the infection and its impact on clinical outcomes.

There was a significant difference in the CT values between blood and nasopharyngeal samples for the N, ORF, and S genes (P < 0.001). These findings highlight the importance of considering different sample types when evaluating SARS-CoV-2 RNAemia targeted the S gene for this evaluation in COVID-19 patients ( Figure 2B ). This comprehensive analysis underscores the significance of sampling multiple sites to accurately assess the presence and extent of SARS-CoV-2 RNAemia, which may have implications for understanding the systemic nature of the infection and its impact on clinical outcomes.

Figure 3 illustrates the distribution of SARS-CoV-2 RNAemia in relation to ICU admissions, patient outcomes, disease stages, and the incidence of thrombosis among COVID-19 patients. Out of 123 patients with detectable SARS-CoV-2 RNA in their blood, a significant majority, 80 patients (65.04%), required ICU admission. This high percentage indicates that the presence of RNAemia may be associated with severe disease progression. The mortality rate among these patients was considerable, with 31 out of 123 (25.20%) not surviving. This suggests a potential link between RNAemia and poorer outcomes, underscoring the need for aggressive monitoring and management of patients with detectable viral RNA in their blood.

Thrombosis was diagnosed in 22 out of 446 of COVID-19 patients (4.99%). Of these, a subset of 123 patients were found to have RNAemia. Furthemore, thrombosis was diagnosed 8 out of 123 patients (6.5%) with RNAemia, indicating that while RNAemia is associated with severe disease and systemic infection, not all patients with RNAemia will develop thrombosis. This highlights the complexity of COVID-19 and the need for comprehensive monitoring of patients with RNAemia to identify those at higher risk of thrombotic events.

Interestingly, stage C of the disease exhibited the highest proportion of patients with detectable SARS-CoV-2 RNA in their blood. This finding suggests that RNAemia may be more prevalent in later stages of COVID-19, potentially serving as a marker for disease progression and severity.

Our study investigated differences in several coagulation markers between patients with positive SARS-CoV-2 RNAemia and those with SARS-CoV-2 RNA detected in the nasopharynx. The coagulation markers examined included plasminogen, α2AP, PAI-1, TAFI, tPA, aPTT, INR, platelets, fibrinogen, and D-dimer. Specifically, plasma levels of TAFI and tPA were significantly higher in patients with SARS-CoV-2 RNAemia compared to those with the virus detected in nasopharyngeal samples (P < 0.001 for both markers).

Additionally, plasma levels of fibrinogen showed a near-significant difference between SARS-CoV-2 RNA in blood samples and nasopharyngeal samples (P = 0.054), with higher fibrinogen levels observed in patients with RNAemia.

Furthermore, there was a significant difference in INR values when comparing SARS-CoV-2 RNA in blood samples to those in nasopharyngeal samples (P = 0.035),with higher INR values in patients with RNAemia. These findings suggest a potential association between plasma levels of certain coagulation factors and SARS-CoV-2 RNAemia, indicating that the presence of viral RNA in different sample types can impact coagulation profiles. The detailed analysis is depicted in Figure 4 . This underscores the importance of monitoring coagulation markers in COVID-19 patients, as differences in these markers can provide insights into the disease’s impact on the coagulation system and help guide clinical management.

In our cohort, we investigated the association between several coagulation markers and specifics mutations in the S gene in blood samples from COVID-19 patients. Notably, we focus on the N501Y (19.03%), D61G (15.04%), K417N and N440K mutations had the same prevalence (9.73% in both).

Our results revealed that the N501Y mutation is the most prevalent among the SARS-CoV-2 RNA detected in blood samples, specifically within the S gene. We found a strong association between the N501Y mutation and plasma levels of tPA, aPTT, and D-dimer in blood samples from COVID-19 patients. Specifically, plasma levels of tPA and aPTT were considerably higher in patients with the N501Y mutation (P = 0.044 and P = 0.024, respectively), whereas D-dimer levels were levels were significantly lower (P = 0.027) (see Figures 5E, F, J )). The D61G mutation, the second most frequent mutation in our blood samples, also showed significant associations. Plasma levels of INR levels were significantly higher in individuals without the D61G mutation (P = 0.018), while platelet counts were significantly higher in those with the D61G mutation (P = 0.034). Specifically, INR levels were significantly associated with the D61G mutation (P = 0.018), and platelet counts were significantly higher in patients with the D61G mutation (P = 0.034) (see Supplementary Figures S1(8G) and (H) ). Additionally, plasma levels of aPTT and D-dimer showed near-significant associations with the D61G mutation (P = 0.065 and P = 0.064, respectively) with higher levels observed in patients without the D61G mutation (see Supplementary Figures S1F, IIJ ).

The K417N and N440K mutations were found in an equal number of blood samples from COVID-19 patients. Plasma levels of tPA were significantly different in patients with the N440K mutation (P = 0.023) (see Supplementary Figure S2E ). The detection of the K417N mutation showed a near-significant association with plasma levels of tPA (P = 0.066), with higher levels observed in patients with the mutation (see Supplementary Figures S3E ). However, there were no significant differences between the K417N and N440K mutations and other coagulation markers, including plasminogen, α2AP, PAI-1, TAFI, aPTT, INR, platelets, fibrinogen, and D-dimer (P > 0.05). These findings highlight the potential impact of the N501Y, D61G, K417N, and N440K mutations on fibrinolytic and coagulopathy changes in COVID-19 patients. Monitoring these markers can aid in risk assessment and guide clinical management strategies.

Figure 6 depicts the findings of our investigation into the association between frequent mutations and the likelihood of thrombosis development in COVID-19 patients. Our results revealed that seven out of the ten mutations examined were statistically significant predictors of thrombosis. In Figure 6 , the terms “Positive” and “Negative” indicate the presence or absence of the N501Y mutation in the analyzed samples, respectively. Among COVID-19 patients who developed thrombosis (n = 22), the N501Y mutation was detected in 10 cases (45.5%). Furthermore, showing a significant association with thrombosis development (P = 0.001). Other mutations that were positively associated with thrombosis include K417N, E484K, P681R, R346T, H69del, and V70del (P = 0.002, P = 0.006, P = 0.024, P = 0.013, P = 0.017, and P = 0.015, respectively). However, statistical analysis showed no significant correlation between thrombotic events and the D614G or T547K mutations, with P-values of 0.447 and 0.257, respectively. Additionally, the N440K mutation showed a near-significant association with thrombosis (P = 0.056), suggesting a potential but not definitive link. These findings highlight the significance of specific mutations in predicting the risk of thrombosis in COVID-19 patients. Identifying these associations can enhance risk assessment and inform targeted clinical management strategies for patients with COVID-19.

The study investigated the association between frequent variants and the likelihood of thrombosis risk in COVID-19 patients. Our results showed that the BA.1.1 strain from the Omicron variant was significantly correlated with a higher risk of thrombosis (P = 0.002). In our cohort, the Omicron variant was detected in 31.80% of COVID-19 thrombosis patients. Additionally, the Delta variant was highly significant and associated with an increased risk of developing thrombosis in COVID-19 patients (P = 0.007). However, as shown in Figure 7 , the other variants did not show a significant difference between thrombosis and non-thrombosis cases in COVID-19 patients. These findings suggest that the BA.1.1 strain of the Omicron variant and the Delta variant are important predictors of thrombosis risk in COVID-19 patients, emphasizing the need for close monitoring and tailored clinical management for patients infected with these variants.