Rapid Evaluation of Coronavirus Illness Severity (RECOILS) in intensive care: Development and validation of a prognostic tool for in‐hospital mortality

Abstract

Background The prediction of in‐hospital mortality for ICU patients with COVID‐19 is fundamental to treatment and resource allocation. The main purpose was to develop an easily implemented score for such prediction. Methods This was an observational, multicenter, development, and validation study on a national critical care dataset of COVID‐19 patients. A systematic literature review was performed to determine variables possibly important for COVID‐19 mortality prediction. Using a logistic multivariable model with a LASSO penalty, we developed the Rapid Evaluation of Coronavirus Illness Severity (RECOILS) score and compared its performance against published scores. Results Our development (validation) cohort consisted of 1480 (937) adult patients from 14 (11) Dutch ICUs admitted between March 2020 and April 2021. Median age was 65 (65) years, 31% (26%) died in hospital, 74% (72%) were males, average length of ICU stay was 7.83 (10.25) days and average length of hospital stay was 15.90 (19.92) days. Age, platelets, PaO2/FiO2 ratio, pH, blood urea nitrogen, temperature, PaCO2, Glasgow Coma Scale (GCS) score measured within +/−24 h of ICU admission were used to develop the score. The AUROC of RECOILS score was 0.75 (CI 0.71–0.78) which was higher than that of any previously reported predictive scores (0.68 [CI 0.64–0.71], 0.61 [CI 0.58–0.66], 0.67 [CI 0.63–0.70], 0.70 [CI 0.67–0.74] for ISARIC 4C Mortality Score, SOFA, SAPS‐III, and age, respectively). Conclusions Using a large dataset from multiple Dutch ICUs, we developed a predictive score for mortality of COVID‐19 patients admitted to ICU, which outperformed other predictive scores reported so far.

Assess and validate predictive performance of models for in-hospital mortality in COVID-19 patients: A retrospective cohort study in the Netherlands comparing the value of registry data with high-granular electronic health records

“…However, many studies showed high risk of bias [1] and only few temporally validate the models [27] , [28] , [29] . Our performance in the temporal validation is in line with another externally-validated model developed on EHR data [30] , as well as other studies that take into account readily availability data [31] , [32] . The importance of external validation and continuous monitoring and updated of machine-learning models to ensure their long-term safety and effectiveness has also been underlined by previous studies [33] , [34] .…”

Performance of CURB-65 and ISARIC 4C mortality scores for hospitalized patients with confirmed COVID-19 infection in Saudi Arabia

“…To compare the ISARIC 4C mortality score and the CURB-65 score, the ISARIC-4C score was standardized into three risk groups (low [0–3], medium [ [4] , [5] , [6] , [7] , [8] ], and high [ 9 , 21 ]) to avoid a different number of risk groups between both scores according to the guidelines [ 18 , 19 ]. The diagnostic accuracy of both scores was evaluated by calculating sensitivities, specificities, positive predictive value, negative predictive value, and Youden's J indices (YJI) [ 24 ].…”

“…Additionally, early identification of subgroups at increased risk of deterioration and needing ventilation or critical care is essential for more effective medical care. Appropriate and rational patient triage strategies improve the decision-making process related to resource allocation, including hospital bed occupancy, critical care resource consumption, and preservation of the most needful and expensive therapy options for critically ill patients [ [3] , [4] , [5] ].…”

Management of COVID-19-Associated Acute Respiratory Failure with Alternatives to Invasive Mechanical Ventilation: High-Flow Oxygen, Continuous Positive Airway Pressure, and Noninvasive Ventilation

“…In patients with COVID-19-associated ARF, the PaO 2 /FiO 2 ratio was able to predict severity in a cohort of 421 patients [ 26 ], as well as in a smaller cohort of 150 patients. In the latter cohort, the authors were able to calculate an optimal cut-off PaO 2 /FiO 2 ratio of 274 mmHg, which could distinguish between patients with mild disease and patients with severe disease with a sensitivity of 72% and specificity 85% [ 27 ], and it also contributed to a predictive composite score along with age, platelets, pH, blood urea nitrogen, temperature, PaCO 2 , and Glasgow Coma Scale in a cohort of 937 patients [ 28 ].…”