Determination of prognostic markers for COVID-19 disease severity using routine blood tests and machine learning.

The need for the identification of risk factors associated to COVID-19 disease severity remains urgent. Patients' care and resource allocation can be potentially different and are defined based on the current classification of disease severity. This classification is based on the analysis of clinical parameters and routine blood tests, which are not standardized across the globe. Some laboratory test alterations have been associated to COVID-19 severity, although these data are conflicting partly due to the different methodologies used across different studies. This study aimed to construct and validate a disease severity prediction model using machine learning (ML). Seventy-two patients admitted to a Brazilian hospital and diagnosed with COVID-19 through RT-PCR and/or ELISA, and with varying degrees of disease severity, were included in the study. Their electronic medical records and the results from daily blood tests were used to develop a ML model to predict disease severity. Using the above data set, a combination of five laboratorial biomarkers was identified as accurate predictors of COVID-19 severe disease with a ROC-AUC of 0.80 ​±â€‹ 0.13. Those biomarkers included prothrombin activity, ferritin, serum iron, ATTP and monocytes. The application of the devised ML model may help rationalize clinical decision and care.

Seroprevalence of Dengue, Chikungunya and Zika at the epicenter of the congenital microcephaly epidemic in Northeast Brazil: A population-based survey.

BACKGROUND: The four Dengue viruses (DENV) serotypes were re-introduced in Brazil's Northeast region in a couple of decades, between 1980's and 2010's, where the DENV1 was the first detected serotype and DENV4 the latest. Zika (ZIKV) and Chikungunya (CHIKV) viruses were introduced in Recife around 2014 and led to large outbreaks in 2015 and 2016, respectively. However, the true extent of the ZIKV and CHIKV outbreaks, as well as the risk factors associated with exposure to these viruses remain vague. METHODS: We conducted a stratified multistage household serosurvey among residents aged between 5 and 65 years in the city of Recife, Northeast Brazil, from August 2018 to February 2019. The city neighborhoods were stratified and divided into high, intermediate, and low socioeconomic strata (SES). Previous ZIKV, DENV and CHIKV infections were detected by IgG-based enzyme linked immunosorbent assays (ELISA). Recent ZIKV and CHIKV infections were assessed through IgG3 and IgM ELISA, respectively. Design-adjusted seroprevalence were estimated by age group, sex, and SES. The ZIKV seroprevalence was adjusted to account for the cross-reactivity with dengue. Individual and household-related risk factors were analyzed through regression models to calculate the force of infection. Odds Ratio (OR) were estimated as measure of effect. PRINCIPAL FINDINGS: A total of 2,070 residents' samples were collected and analyzed. The force of viral infection for high SES were lower as compared to low and intermediate SES. DENV seroprevalence was 88.7% (CI95%:87.0-90.4), and ranged from 81.2% (CI95%:76.9-85.6) in the high SES to 90.7% (CI95%:88.3-93.2) in the low SES. The overall adjusted ZIKV seroprevalence was 34.6% (CI95%:20.0-50.9), and ranged from 47.4% (CI95%:31.8-61.5) in the low SES to 23.4% (CI95%:12.2-33.8) in the high SES. The overall CHIKV seroprevalence was 35.7% (CI95%:32.6-38.9), and ranged from 38.6% (CI95%:33.6-43.6) in the low SES to 22.3% (CI95%:15.8-28.8) in the high SES. Surprisingly, ZIKV seroprevalence rapidly increased with age in the low and intermediate SES, while exhibited only a small increase with age in high SES. CHIKV seroprevalence according to age was stable in all SES. The prevalence of serological markers of ZIKV and CHIKV recent infections were 1.5% (CI95%:0.1-3.7) and 3.5% (CI95%:2.7-4.2), respectively. CONCLUSIONS: Our results confirmed continued DENV transmission and intense ZIKV and CHIKV transmission during the 2015/2016 epidemics followed by ongoing low-level transmission. The study also highlights that a significant proportion of the population is still susceptible to be infected by ZIKV and CHIKV. The reasons underlying a ceasing of the ZIKV epidemic in 2017/18 and the impact of antibody decay in susceptibility to future DENV and ZIKV infections may be related to the interplay between disease transmission mechanism and actual exposure in the different SES.