SARS-CoV-2 RNA in blood serum could predict COVID-19 mortality

By Dr. Ramya Dwivedi, Ph.D.Jan 18 2021

An interdepartmental team from the Hospital Universitario La Princesa, and the Universidad Autónoma de Madrid, Spain, analyzed the genetic material of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) – the causative pathogen of coronavirus disease 2019 (COVID-19) – in the serum of positive patients. The team’s findings suggest that the baseline SARS-CoV-2 RNA detection in the blood (CoVemia) is associated with worse outcomes in hospitalized COVID-19 patients.

The study’s findings recently published on the medRxiv* preprint server.

Study: Detection of SARS-Cov-2 RNA in serum is associated with increased mortality risk in hospitalized COVID-19 patients. Image Credit: bajars / Shutterstock

Exhaustive observations across the world indicate that COVID-19 manifests a heterogeneous clinical picture. To unburden the overworked healthcare services everywhere and to help patients survive, the researchers argue that it is critical to identify patients with poor prognosis early on. The researchers thus suggest that SARS-CoV-2 RNA detection in serum may be an effective biomarker in predicting COVID-19 mortality.

Poor prognosis may be assessed as a higher probability of clinical deterioration, higher levels of interleukin (IL)-6, IL-5 or CXCL10, intensive care unit (ICU) admission, critical disease and death. Several biomarkers – low total lymphocyte count, high LDH serum levels, increased acute phase reactants (C325 reactive protein, ferritin, fibrinogen, etc.) or increased IL-6 serum levels – are proposed in this context.

In this study, the researchers assessed CoVemia with COVID-19 severity using two different real-time reverse-transcription polymerase chain reaction (rRT-PCR) techniques and compared them with other suggested severity biomarkers. They found that the CoVemia is associated with COVID-19-related deaths and have presented how it can be used to predict mortality in patients.

The study cohort included 193 patients admitted for COVID-19 to the Hospital Universitario La Princesa (HUP) during the early weeks of the first wave of COVID-19 outbreak in Spain (4th March to 17th April of 2020). The samples were collected at 48-72 hours of admission by two techniques from Roche and Thermo Fischer Scientific (TFS).

The main outcome variables were mortality and the need for ICU admission during hospitalization for COVID-19. Only 14 patients out of the 89 admitted to ICU did not require invasive mechanical ventilation. The researchers claim to have fulfilled the STROBE standards for observational research.

Depending on the technique used in the study – Roche and TFS – the CoVemia was detected in 95 (48%; Roche) and 117 (59%; TFS) patients. They found that the correlation between Ct (Cycle threshold) in serum obtained with both techniques was very good. They also found that, conversely, the correlation of Ct between nasopharyngeal and throat swab (NPTS) and serum samples was weak either with TFS or Roche techniques.

This study supports that patients with detectable CoVemia were older, with worse arterial oxygen, lower lymphocyte count and higher LDH, IL-6, C-reactive protein and procalcitonin serum levels compared to patients without the viral RNA in their serum.

The researchers then analyzed whether a threshold of CoVemia (as Ct) could help to predict mortality during hospitalization. The best cut-off to predict mortality was 34 Ct for Roche (sensitivity 91%, specificity 38%) and 31 Ct for TFS (sensitivity 93%, specificity 32%).

Although the sensitivity of TFS technique was slightly better for SARS-CoV-2 RNA detection in serum, the researchers observed that this difference disappeared when the specific Ct cut-offs for each kit were applied to define relevant CoVemia.

To further validate CoVemia as a predictor of mortality in COVID-19, the researchers analyzed its association with clinical and laboratory parameters related to worse outcomes. They found that it correlates with several variables that have been proposed to be associated with poor evolution in COVID-19, namely old age, comorbidity, qSOFA and CURB-65, and with markers such as high IL-6 or LDH serum levels and severe lymphopenia.

The researchers found that the best predictors always included baseline “relevant CoVemia” and high LDH serum levels. They define “relevant CoVemia” as the amount of viral load that better predicts mortality obtaining 95% sensitivity and 35% specificity.

Comprehensive analysis of CoVemia as a prognostic marker of mortality in patients hospitalized for severe COVID-19. ROC curve-analysis for mortality prediction with Ct values in serum of all patients, according to Roche (A) and Thermo Fisher Scientific [TFS] (B) techniques. Proportion of deceased patients according to relevant CoVemia determined by Roche (C) and TFS (D) techniques. Survival analysis with Kaplan-Meier estimator of patients hospitalized for COVID-19 who presented (dotted lines) and patients who did not present (solid lines) relevant CoVemia according to Roche (E) and TFS (F) techniques.

This study provides the most useful biomarker in a clinical setting for predicting mortality in COVID-19 patients. The relevant CoVemia provides a hazard ratio three times higher than that of the other significant variable: high LDH serum level.

The researchers also caution that for the need for quantitative standardization of relevant CoVemia. What persistent CoVemia signifies, for example, must be addressed and evaluated in further longitudinal studies.

The results from this study reinforce previous data with a main contribution to the management of COVID-19: we have determined a semiquantitative threshold for SARS-CoV-2 RNA detection in serum early after admission that allows establishing RNA values (“relevant CoVemia”) associated with higher mortality risk.”

In summary, the researchers believe their study presents evidence that detection of CoVemia is the best biomarker to predict death in COVID-19 patients.

*Important Notice

medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.