key: cord-0681879-3phwhis5
authors: Rodriguez Serrano, D. A.; Roy-Vallejo, E.; Zurita Cruz, N. D.; Martin Ramirez, A.; Rodriguez-Garcia, S. C.; Arevalillo-Fernandez, N.; Galvan-Roman, J. M.; Fontan Garcia-Rodrigo, L.; Vega Piris, L.; Chicot Llano, M.; Arribas Mendez, D.; Gonzalez de Marcos, B.; Hernando Santos, J.; Sanchez Azofra, A.; Avalos Perez-Urria, E.; Rodriguez-Cortes, P.; Esparcia, L.; Marcos-Jimenez, A.; Sanchez-Alonso, S.; Llorente, I.; Soriano, J. B.; Suarez Fernandez, C.; Garcia-Vicuna, R.; Ancochea, J.; Sanz, J.; Munoz-Calleja, C.; de la Camara, R.; Canabal Berlanga, A.; Gonzalez-Alvaro, I.; Cardenoso, L.
title: Detection of SARS-Cov-2 RNA in serum is associated with increased mortality risk in hospitalized COVID-19 patients.
date: 2021-01-15
journal: nan
DOI: 10.1101/2021.01.14.21249372
sha: 3f27127d0040458d3439844f10d2f20b40bf6e18
doc_id: 681879
cord_uid: 3phwhis5

Background COVID-19 has overloaded national health services worldwide. Thus, early identification of patients at risk of poor outcomes is critical. Our objective was to analyse SARS-CoV-2 RNA detection in serum as a severity biomarker in COVID-19. Methods and Findings Retrospective observational study including 193 patients admitted for COVID-19. Detection of SARS-CoV-2 RNA in serum (CoVemia) was performed with samples collected at 48-72 hours of admission by two techniques from Roche and Thermo Fischer Scientific (TFS). Main outcome variables were mortality and need for ICU admission during hospitalization for COVID-19. CoVemia was detected in 50-60% of patients depending on technique. The correlation of Ct in serum between both techniques was good (intraclass correlation coefficient: 0.612; p < 0.001). Patients with CoVemia were older (p = 0.006), had poorer baseline oxygenation (PaO2/FiO2; p < 0.001), more severe lymphopenia (p < 0.001) and higher LDH (p < 0.001), IL-6 (p = 0.021), C-reactive protein (CRP; p = 0.022) and procalcitonin (p = 0.002) serum levels. We defined "relevant CoVemia" when detection Ct was < 34 with Roche and < 31 for TFS. These thresholds had 95% sensitivity and 35 % specificity. Relevant CoVemia predicted death during hospitalization (OR 9.2 [3.8 - 22.6] for Roche, OR 10.3 [3.6 - 29.3] for TFS; p < 0.001). Cox regression models, adjusted by age, sex and Charlson index, identified increased LDH serum levels and relevant CoVemia (HR = 9.87 [4.13-23.57] for TFS viremia and HR = 7.09 [3.3-14.82] for Roche viremia) as the best markers to predict mortality. Conclusions CoVemia assessment at admission is the most useful biomarker for predicting mortality in COVID-19 patients. CoVemia is highly reproducible with two different techniques (TFS and Roche), has a good consistency with other severity biomarkers for COVID-19 and better predictive accuracy.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 AUTHOR SUMMARY 84 COVID-19 shows a very heterogeneous clinical picture. In addition, it has overloaded 85 national health services worldwide. Therefore, early identification of patients with 86 poor prognosis is critical to improve the use of limited health resources. In this work, 87 we evaluated whether baseline SARS-CoV2 RNA detection in blood (CoVemia) is 88 associated with worse outcomes. We studied almost 200 patients admitted to our 89 hospital and about 50-60% of them showed positive CoVemia. Patients with positive 90 CoVemia were older and had more severe disease; CoVemia was also more frequent in 91 patients requiring admission to the ICU. Moreover, we defined "relevant CoVemia", as 92 the amount of viral load that better predicted mortality obtaining 95% sensitivity and 93 35% specificity. In addition, relevant CoVemia was a better predictor than other 94 biomarkers such as LDH, lymphocyte count, interleukin-6, and indexes used in ICU such 95 as qSOFA and CURB65. 96 In summary, detection of CoVemia is the best biomarker to predict death in patients. Furthermore, it is easy to be implemented and is reproducible with two 98 techniques (Roche and Thermo Fisher Scientific) that are currently used for diagnosis 99 in nasopharyngeal swabs samples. 108 The wide spectrum of COVID-19 severity ranges from asymptomatic to critical cases, 109 albeit less than 10% of patients develop a severe disease [1] . Even though only a minority 110 of patients need hospitalization for COVID-19, the higher transmission rate of SARS-Cov-111 2 compared to other viruses, the absence of previous immunity in the population, and 112 the high incidence of this disease in a short period of time are collapsing health care 113 systems worldwide [2] . One remaining challenge from COVID-19 is the difficulty of 114 predicting individual prognosis since determinants of disease severity remain unclear. 115 Previous studies have suggested that age, male sex, obesity, hypertension, and 116 underlying diseases like hematologic malignancies are associated with worse prognosis 117 [3] [4] [5] . Likewise, some blood biomarkers are able to predict the emergence of the 118 cytokine storm leading to severe acute respiratory syndrome, the most frequent cause 119 of clinical deterioration in COVID-19 patients [6] [7] [8] . 120 121 Despite the tropism of SARS-CoV-2 for the upper respiratory tissue [9] , the relevance of 122 its viral load in nasopharyngeal samples remains controversial [10] . However, several 123 authors have reported the detection of SARS-CoV-2 RNA in serum or plasma samples 124 (henceforth CoVemia) associated with a worse prognosis, assessed as higher probability 125 of clinical deterioration, higher levels of interleukin (IL)-6, IL-5 or CXCL10, intensive care 126 unit (ICU) admission, critical disease and death [8, [11] [12] [13] . Notwithstanding, the 127 detection of CoVemia with more sensitive techniques was not associated with mortality, 128 but with immune suppression status [14] . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint was more sensitive than by Roche (S1A Figure) . The agreement to detect CoVemia was 158 75.5% and the intraclass correlation coefficient was 0.612 (p < 0.001). Conversely, the 159 correlation of Ct between nasopharyngeal and throat swab (NPTS) and serum samples 160 was weak either with TFS or Roche techniques (S1B and 1C Figure, is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint S1 Figure (Table 1 and S1 Table) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. 187 In order to evaluate the relationship between COVID-19 severity and viremia we first 188 studied the relative frequency of CoVemia in patients who required ICU admission and 189 those who did not. As shown in Figure 1 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint CoVemia was proved to be the best mortality predictor in our population since it 356 provides a hazard ratio three times higher than that of the other significant variable, 357 high LDH serum level. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint the recent description of lower cytotoxic potential of NK cells and senescence of CD8+ T 364 lymphocytes associated to high IL-6 levels, which can be restored by tocilizumab 365 treatment [20] . Second, association between lymphopenia and CoVemia could be 366 related with lymphocyte viral infection and cellular lysis by cytotoxic cells. However, 367 lymphopenia is more likely to be related with increased cellular migration into the lungs, 368 as we have recently described for dendritic cells [21] . CoVemia at admission to predict mortality. This finding could be associated to the 379 variability over time of SARS-CoV-2 RNA detection in serum in COVID-19 patients [22] . [13, 14, 17] . Therefore, sequential assessment of CoVemia may be needed to improve its 385 specificity to predict mortality. Last but not least, our findings are restricted to the so-386 called "first wave", when an overloaded health system could not cover all the needs to 387 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint In summary, the study presented here has established the usefulness of SARS-CoV-2 392 RNA detection in blood in the initial assessment of patients admitted for COVID-19 due 393 to its capability to predict mortality. This assessment would be easily implemented since 394 it is reproducible, regardless of the commercially available kit used for SARS-CoV-2 RNA 395 detection. Accordingly, information from tests widely used for diagnosis could be readily 396 used for prognosis evaluation. However, the need for quantitative standardization of is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 platform for nucleic acid extraction and RT-PCR amplification and detection. Serum is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 one gene target was detected by one test, the mean Ct value of amplified genes was 468 calculated. There was a high consistency among the results of the three genes analyzed 469 with TFS and a slightly lower one for the two genes analyzed in the Roche system [19] . 470 Since none of the two systems (Roche and TFS) provided positive results with Ct > 40, 471 by consensus decision, we assigned Ct = 42 to those samples reported as negative (for 472 all the genes) by the system. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint To determine the variables associated with the need for ICU admission, we performed 505 a multivariable logistic regression analysis that was first modeled by adding all the 506 variables with a p-value lower than 0.15 to the bivariable analysis, namely low 507 lymphocyte count, D-dimer, high LDH, low PaO2/FiO2 (< 250), COPD, high qSOFA, high 508 CURB65, hypertension, C-reactive protein, and high IL-6 level. The final model was 509 reached through backward stepwise removal of variables with p-value higher than 0.15 510 and using Wald tests to demonstrate that each model was better than its previous 511 iteration. Once the final model was obtained, the dichotomic variables CoVemia (both 512 Roche and TFS) were forced into the model in order to determine whether they were 513 able to predict the need for ICU admission. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint Odds ratio (OR) for mortality according to the presence of CoVemia was estimated for 516 both systems with the cs command of Stata. Since differences between the two 517 techniques were detected in the sensitivity to detect viremia and the capability of 518 predicting mortality, receiver operating characteristic (ROC) analyses were performed 519 using the roctab command in order to estimate the best cut-off point for each system. To compare the predictive capability of baseline CoVemia with that of other baseline 529 parameters (qSOFA, CURB65, total lymphocyte count, LDH and IL-6 serum levels) 530 described in previous publications [12, 13, 25] we fitted different Cox regression models. 531 In the first model after adjustment by age, sex and Charlson comorbidity index, the 532 hazard ratio (HR) for all other variables including relevant viremia by TFS or Roche was 533 estimated. Since qSOFA and CURB65 did not reach a p-value < 0.15, they were excluded 534 from the analysis in the subsequent models. Model 2 was adjusted by the same variables 535 included in model 1 and low lymphocyte count, the hazard ratio for the remaining 536 variables was then estimated. Model 3 and 4 were developed as model 2 but 537 substituting low lymphocyte count by high IL-6 and high LDH, respectively. 538 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted January 15, 2021. ; https://doi.org/10.1101/2021.01.14.21249372 doi: medRxiv preprint

Desiré Navas 590 Microbiology: Laura Cardeñoso Domingo, María del Carmen Cuevas Torresano

Mar Barrio Mayo, Encarna Rubia Garrido

Surgery Department: Iñigo García Sanz

Treatment of Coronavirus Disease 2019 (COVID-19): A 608 Review

Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease 611 2019 Pneumonia in Wuhan, China

Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2

Admitted to ICUs of the Lombardy Region

Predictors of 618 mortality in hospitalized COVID-19 patients: A systematic review and meta-analysis

Outcomes of 621 patients with hematologic malignancies and COVID-19: A systematic review and meta-622 analysis of 3377 patients. Blood

IL-6 serum levels predict severity and 625 response to Tocilizumab in COVID-19: an observational study

Clinical course and risk factors for 628 mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort 629 study

Viral RNA load in plasma is associated with critical illness and a 632 dysregulated host response in COVID-19

Transmission of SARS-CoV-2: 635 A Review of Viral, Host, and Environmental Factors

Clearance and persistence of SARS-CoV-2 RNA in patients with COVID-19

Detectable Serum Severe Disease

SARS-CoV-2 RNA in serum as predictor of severe outcome in COVID-19: a 646 retrospective cohort study

Highly sensitive 649 quantification of plasma SARS-CoV-2 RNA shelds light on its potential clinical value

Risk 652 factors for SARS-CoV-2 detection in blood of critically ill patients

The role of biomarkers in 655 diagnosis of COVID-19 -A systematic review

High 658 Frequency of SARS-CoV-2 RNAemia and Association With Severe Disease

SARS-CoV-2 RNA in plasma is associated with ICU admission and mortality in patients 662 hospitalized with COVID-19

Relationship Between serum 665

SARS-CoV-2 nucleic acid(RNAemia) and Organ Damage in COVID-19 Patients: A Cohort 666

Evaluation of two RT-PCR techniques for SARS-CoV-2 RNA detection in serum for 669 microbiological diagnosis

Impaired 674 immune cell cytotoxicity in severe COVID-19 is IL-6 dependent

COVID-19 severity associates with pulmonary redistribution of 678

CD1c+ DCs and inflammatory transitional and nonclassical monocytes

Detection and Isolation of 681

Urine, and Stool Specimens of COVID-19 Patients from the 682 Republic of Korea

The 685 SOFA (Sepsis-related Organ Failure Assessment) score to describe organ 686 dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of 687 the European Society of Intensive Care Medicine

Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive 691 care units: results of a multicenter, prospective study

Viral load dynamics and disease 695 severity in patients infected with SARS-CoV-2 in Zhejiang province, China

Treatments available for the 699 management of respiratory infection by SARS-CoV-2 [Tratamientos disponibles para el 700 manejo de la infección respiratoria por SARS-CoV-2