Study identifies renalase as a new independent predictor of COVID-19-related mortality

*Important Notices: research plaza We publish a non-peer-reviewed, preliminary scientific report and should not be taken as conclusive, to guide clinical practice/health-related actions, or to be treated as established information.

In a recent study posted on research plaza* Preprint Server, Researcher Evaluated Role kidney failure (RNLS) are independent predictors of coronavirus disease (COVID-19)-related mortality in 2019.

Background

In the past few months, a number of biomarker studies have revealed specific components of the immune response to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), raising important new questions. Individuals exhibiting fatal symptoms have been reported to exhibit a pathophysiology that distinguishes them from those exhibiting mild symptoms, due to the severely perturbed inflammatory response. Changes in some indicators, including renal impairment, appear to be temporary.

Little is known about the depletion markers in SARS-CoV-2 infection and the dynamic behavior of these biomarkers. Renalase, a peptide endogenously produced by the heart, kidney, and endothelium, has pro-survival properties, including inhibiting cytokine release in viral infections such as COVID-19 in mouse models.

About research

In the present study, researchers determined whether renalase independently predicts COVID-19-related mortality.

The team investigated COVID-19-positive adult patients admitted to an urban academic center between March 1 and June 30, 2020. Nasopharyngeal swab All patient samples revealed severe SARS-CoV-2 infection. Collection of all specimens and images was part of standard medical care.

The team utilized the School of Medicine COVID Explorer (DOM-CovX), a group of hospitalized COVID-19 patients, to collect clinical information such as socio-demographics, vital signs, comorbidities, laboratory values, treatments and procedures. Obtained from an electronic medical record system. throughout the hospital stay.

Admission date, symptoms with onset date, smoking history, immunocompromised status, cardiopulmonary resuscitation, death date, intubation date, and last completed follow-up were extracted using manual chart review. Cohort serum or plasma specimens are evaluated for inflammatory markers including interferon (IFN)-ɑ, IFN-β, IFN-λ, IFN-Ɣ, interleukin (IL)-1β, IL-6, and tumor necrosis factor It was analyzed as follows. (TNF), along with a kidney injury molecule (KIM-1).

Mortality was defined as death occurring within 180 days of initial presentation. Traditional and modern machine learning techniques were utilized. For traditional models, the team employs logistic regression, and for a more recent approach he employs XGBoost.

result

A total of 3,450 COVID-19 patients were hospitalized from March 2020 to June 2020. The study group consisted of her 473 patients who volunteered for the study, were over the age of 18, and provided sufficient blood samples. When compared with non-included hospitalized patients, the cohort population had comparable age and gender distributions. The overall cohort included 366 surviving patients and 71 deaths.

The average length of hospital stay for deceased patients was 17 days, and the patients were older and more likely to be male. Additionally, they had more comorbidities, higher beta-natriuretic peptide (BNP), creatinine, troponin, ferritin, d-dimerand procalcitonin levels, and platelets lower than in surviving individuals.

Those who died had lower mean renalase levels and tended to have higher levels of IL-1, IFN, and KIM-1 than those who survived. The team used standard logistic regression models to identify age, patient sex, and mean renalase as significant predictors of mortality. In addition to renalase, the team identified clinical factors and various conventional laboratory markers as predictors of death.

According to the XGBoost model, BNP markers with high cardiac burden were the most important predictors of mortality. Patients with the lowest RNLS and highest BNP quartile had significantly higher mortality than those with the highest RNLS and lowest BNP quartile. At the end of the stay, those with high BNP and low renalase reported higher mortality than those with low BNP and high renalase.

Renalase showed similar relevance to troponin, a biomarker of heart damage. In conventional models, troponin was observed to predict death, despite the lack of significance in his XGBoost model.

The association was less significant when renalase levels were compared with inflammatory markers such as IFN, IL-1 and IL-6. Mortality was significantly different between those with low RNLS and high IL6 and those with high RNLS and low IL6. However, comparison of similar groups for IFN and IL-1 did not reveal significant differences.

Platelet count was also identified as an important predictor of mortality in the XGBoost model. Patients with low renalase levels and low platelet counts were found to have the highest mortality compared to those with high renalase levels and high platelet counts.

Conclusion

The findings demonstrated that machine learning techniques can efficiently complement traditional statistical methods in identifying COVID-19 mortality predictors. Renalase is a reliable and independent predictor of death in her COVID-19 patients while hospitalized.

Prospective studies should investigate renalase progression, especially in combination with other markers of endothelial and cardiac dysfunction, and the potential therapeutic use of exogenous renalase.

*Important Notices: research plaza We publish a non-peer-reviewed, preliminary scientific report and should not be taken as conclusive, to guide clinical practice/health-related actions, or to be treated as established information.