Changes and evolution among SARS-COV-2 hospitalised patients in terms of severity, mortality and virus genome in a Spanish cohort

Abstract

A comparison between pandemic waves could help to understand the evolution of this disease. The objective of this work was to study the evolution of COVID-19 hospitalized patients on different pandemic waves in terms of severity and mortality. We performed an observational retrospective cohort study of hospitalized patients (5,220) with SARS-CoV-2 infection from February to September in Aragon, Spain. In a comparative way, we analyzed ICU admission and 30-day mortality, clinical characteristics and risk factors, of first and second waves. SARS-CoV-2 virus genome were analyzed in 236 samples. Patients in the first wave (n=2,547) were older (74 y, IQR: 60-86 vs. 70 y, IQR: 53-85; p<0.001) and showed worse clinical and analytical parameters related to severe COVID-19 than in the second wave (n=2,673). The probability of ICU admission at 30 days was 16% and 10% in the first and second wave, respectively (p<0.001). The cumulative 30-day mortality rates were 38% in the first wave and 32% in the second one (p=0.007). Survival differences were observed among patients aged 60 to 80 years. There was variability among death risk factors and virus genome between waves. Therefore, the two COVID-19 pandemic waves analyzed were different, in terms of disease severity and mortality.

A Clinical Decision Web to Predict ICU Admission or Death for Patients Hospitalised with COVID-19 Using Machine Learning Algorithms

“…Gong et al [38] (severe COVID-19 infection within minimum 15 days in inpatients with confirmed COVID-19 at admission) and Xie et al [39] (mortality in hospital in inpatients with confirmed COVID-19 at admission) performed external validation and achieved good performance, though their validation was performed for different centres, possibly due to being early prognostic model studies of the pandemic (with a population from China); the populations studied were patients admitted between January and March 2020 (first pandemic wave). However, subsequent studies [24] have shown that there were differences between the pandemic waves in terms of patient characteristics and severity. In addition to model discrimination, calibration should also be assessed, though this is not always done or not done correctly.…”

“…In addition, the sample sizes have been limited and most models lack external validation or calibration. There have also been several pandemic waves, in which both mortality rate and severity have differed [24], so it would be necessary for models to prove useful in different scenarios over time.…”

“…Therefore, during the COVID-19 pandemic, Machine Learning has been crucial in developing tools for a variety of investigations, ranging from image analysis for diagnosing SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2) pneumonia, predicting future pandemic waves, detecting multiple predictive and prognostic factors associated with increased severity of infection, to building severity models that help health systems prioritise care [8,[21][22][23][24][25][26][27][28][29][30].…”