Dataset Information

Development and Internal Validation of a New Prognostic Model Powered to Predict 28-Day All-Cause Mortality in ICU COVID-19 Patients-The COVID-SOFA Score.

ABSTRACT:

Background

The sequential organ failure assessment (SOFA) score has poor discriminative ability for death in severely or critically ill patients with Coronavirus disease 2019 (COVID-19) requiring intensive care unit (ICU) admission. Our aim was to create a new score powered to predict 28-day mortality.

Methods

Retrospective, observational, bicentric cohort study including 425 patients with COVID-19 pneumonia, acute respiratory failure and SOFA score ≥ 2 requiring ICU admission for ≥72 h. Factors with independent predictive value for 28-day mortality were identified after stepwise Cox proportional hazards (PH) regression. Based on the regression coefficients, an equation was computed representing the COVID-SOFA score. Discriminative ability was tested using receiver operating characteristic (ROC) analysis, concordance statistics and precision-recall curves. This score was internally validated.

Results

Median (Q1-Q3) age for the whole sample was 64 [55-72], with 290 (68.2%) of patients being male. The 28-day mortality was 54.58%. After stepwise Cox PH regression, age, neutrophil-to-lymphocyte ratio (NLR) and SOFA score remained in the final model. The following equation was computed: COVID-SOFA score = 10 × [0.037 × Age + 0.347 × ln(NLR) + 0.16 × SOFA]. Harrell's C-index for the COVID-SOFA score was higher than the SOFA score alone for 28-day mortality (0.697 [95% CI; 0.662-0.731] versus 0.639 [95% CI: 0.605-0.672]). Subsequently, the prediction error rate was improved up to 16.06%. Area under the ROC (AUROC) was significantly higher for the COVID-SOFA score compared with the SOFA score for 28-day mortality: 0.796 [95% CI: 0.755-0.833] versus 0.699 [95% CI: 0.653-0.742, p < 0.001]. Better predictive value was observed with repeated measurement at 48 h after ICU admission.

Conclusions

The COVID-SOFA score is better than the SOFA score alone for 28-day mortality prediction. Improvement in predictive value seen with measurements at 48 h after ICU admission suggests that the COVID-SOFA score can be used in a repetitive manner. External validation is required to support these results.

SUBMITTER: Moisa E

PROVIDER: S-EPMC9323813 | biostudies-literature | 2022 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and Internal Validation of a New Prognostic Model Powered to Predict 28-Day All-Cause Mortality in ICU COVID-19 Patients-The COVID-SOFA Score.

Moisa Emanuel E Corneci Dan D Negutu Mihai Ionut MI Filimon Cristina Raluca CR Serbu Andreea A Popescu Mihai M Negoita Silvius S Grintescu Ioana Marina IM

Journal of clinical medicine 20220718 14

Background: The sequential organ failure assessment (SOFA) score has poor discriminative ability for death in severely or critically ill patients with Coronavirus disease 2019 (COVID-19) requiring intensive care unit (ICU) admission. Our aim was to create a new score powered to predict 28-day mortality. Methods: Retrospective, observational, bicentric cohort study including 425 patients with COVID-19 pneumonia, acute respiratory failure and SOFA score ≥ 2 requiring ICU admission for ≥72 h. Facto ...[more]

PMID: 35887924

Similar Datasets

Project description:Abstract Background and Aim The efficacy of Sequential Organ Failure Assessment (SOFA) score as predictor of clinical outcomes among ICU‐admitted COVID‐19 patients is still controversial. We aimed to assess whether SOFA‐score in different time intervals could predict 28‐day mortality compared with other well‐acknowledged risk factors of COVID‐19 mortality. Methods This observational prospective cohort was conducted on 1057 patients from March 2020 to March 2022 at Masih Daneshvari Hospital, Iran. The univariate and multivariate Cox proportional analysis were performed to assess the hazards of SOFA‐score models. Receiver operating characteristic (ROC) curves were designed to estimate the predictive values. Results Mean SOFA‐score during first 96 h (HR: 3.82 [CI: 2.75–5.31]), highest SOFA‐score (HR: 2.70 [CI: 1.93–3.78]), and initial SOFA‐score (HR: 1.65 [CI: 1.30–2.11]) had strongest association with 28‐day mortality (p < .0001). In contrast, SOFA scores at 48 and 96 h as well as Δ‐SOFA: 48‐0 h and Δ‐SOFA: 96‐0 h did not show significant correlations. Among them, merely mean SOFA‐score (HR: 2.28 [CI: 2.21–3.51]; p < .001) remained as independent prognosticator on multivariate regression analysis; though having less odds of predicting value compared with age (HR: 3.81 [CI: 1.98–5.21]), hypertension (HR: 3.11 [CI: 1.26–3.81]), coronary artery disease [CAD] (HR: 2.82 [CI: 1.51–4.8]), and diabetes mellitus (HR: 2.45 [CI: 1.36–2.99]). The area under ROC (AUROC) for mean SOFA‐score (0.77) and highest SOFA‐score (0.71) were larger than other SOFA intervals. Calculating the first 96 h of SOFA trends, it was obtained that fatality rate was <12.3% if the score dropped, between 28.8% and 46.29% if the score remained unchanged, and >50.45% if the score increased. Conclusion To predict the 28‐day mortality among ICU‐admitted COVID‐19 patients, mean SOFA upon first 96 h of ICU stay is reliable; while having inadequate accuracy comparing with well‐acknowledged COVID‐19 mortality predictors (age, diabetes mellitus, hypertension, CAD). Notably, increased SOFA levels in the course of first 96 h of ICU‐admission, prognosticate at least 50% fatality regardless of initial SOFA score.

Project description:Background and aimsInnovative treatment modalities have not yet shown a clinical benefit in patients with septic shock. To reduce severe cytokinaemia, CytoSorb as an add-on to continuous renal replacement therapy (CRRT) showed promising results in case reports. However, there are no clinical trials investigating outcomes.MethodsIn this investigator-initiated retrospective study, patients with septic shock were treated with CRRT + CytoSorb (n = 67) or CRRT alone (n = 49). The primary outcome was the 28-day all-cause mortality rate. Patients were weighted by stabilized inverse probability of treatment weights (sIPTW) to overcome differences in baseline characteristics.ResultsAt the start of therapy, CytoSorb-treated patients had higher lactate levels (p < 0.001), lower mean arterial pressure (p = 0.007) and higher levels of noradrenaline (p < 0.001) compared to the CRRT group. For CytoSorb, the mean predicted mortality rate based on a SOFA of 13.8 (n = 67) was 75% (95%CI 71-79%), while the actual 28-day mortality rate was 48% (mean difference - 27%, 95%CI - 38 to - 15%, p < 0.001). For CRRT, based on a SOFA of 12.8 (n = 49), the mean predicted versus observed mortality was 68% versus 51% (mean difference - 16.9% [95%CI - 32.6 to - 1.2%, p = 0.035]). By sIPTW analysis, patients treated with CytoSorb had a significantly lower 28-day mortality rate compared to CRRT alone (53% vs. 72%, respectively, p = 0.038). Independent predictors of 28-day mortality in the CytoSorb group were the presence of pneumosepsis (adjusted odds ratio [aOR] 5.47, p = 0.029), higher levels of lactate at the start of CytoSorb (aOR 1.15, p = 0.031) and older age (aOR per 10 years 1.67, p = 0.034).ConclusionsCytoSorb was associated with a decreased observed versus expected 28-day all-cause mortality. By IPTW analysis, intervention with CytoSorb may be associated with a decreased all-cause mortality at 28 days compared to CRRT alone.

Project description:BackgroundHigh prevalence of malnutrition was found in critically ill COVID-19 patients. The modified Nutrition Risk in the Critically ill (mNUTRIC) score is frequently used for nutritional risk assessment in intensive care unit (ICU) COVID-19 patients. The aim of this study was to investigate the role of mNUTRIC score to predict 28-day mortality in critically ill COVID-19 patients admitted to ICU.MethodsA cohort of consecutive COVID-19 critically ill patients admitted to ICU was retrospectively evaluated and the nutritional risk was assessed with the use of mNUTRIC score. A multivariable Cox regression model to predict 28-day mortality was therefore developed including the mNUTRIC as a covariate. Internal validation was performed using the bootstrap resampling technique to reduce possible bias in the estimated risks. The performance of the prediction model was assessed via calibration and discrimination.ResultsA total of 98 critically ill COVID-19 patients with a median age of 66 years (56-73 IQR), 81 (82.7%) males were included in this study. A high nutritional risk (mNUTRIC ≥5 points) was observed in 41.8% of our critically ill COVID-19 patients while a low nutritional risk (mNUTRIC <5 points) was observed in 58.2%. Forty-five patients (45.9%) died within 28 days after ICU admission. In multivariable model after internal validation, mNUTRIC ≥5 (optimism adjusted HR 2.38, 95% CI 1.08-5.25, p = 0.02) and high-sensitivity C-reactive protein values (CRP) (optimism adjusted HR 1.02, 95% CI 1.01-1.07, p = 0.005) were independent predictors of 28-day mortality.ConclusionsA high prevalence of malnutrition as revealed by mNUTRIC was found in our critically ill COVID-19 patients once admitted in ICU. After adjustment for covariables, mNUTRIC ≥5 and CRP levels were independently associated with 28-day mortality in critically ill COVID-19 patients. The final model revealed good discrimination and calibration. Nutritional risk assessment is essential for the management of critically ill COVID-19 patients as well as for outcome prediction.

Project description:The Sequential Organ Failure Assessment, also known as SOFA score, was introduced to assess organ dysfunction of critical ill patients. However, understanding the impact of missing SOFA scores in randomized controlled trials and how this affect the validity and applicability of the SOFA score as a surrogate endpoint for predicting mortality has been a matter of interest. To address this, a secondary analysis of a systematic review was conducted to quantify the relationship between SOFA scores and the prediction of mortality in critically ill adults in randomized controlled trials (RCTs). The systematic review being referred to included 87 RCTs with a total of 12,064 critically ill patients. This analysis focused on missing SOFA score data in relation to the length of stay on the intensive care unit (ICU) and the methods used to handle missing data. SOFA score measurements from the included studies were categorized into three time frames: Early (t ≤ 4 days), Intermediate (t = 5-10 days) and Late (t > 10 days) measurement. Only one study reported a complete data set for calculating the SOFA score for an Early measurement. When considering all methods used to address missing data, 32% of studies still had missing data for Early measurements, and this percentage increased to 64% for Late measurements. These findings suggested that, over time, the number of studies with incomplete data sets has been increasing. The longer a patient is treated on the ICU, the higher the number of missing data which can impact the validity of SOFA score analyses. There was no clear trend towards a specific method for compensating missing data. An exemplary calculation demonstrated that ignoring missing data may lead to an underestimated variability of the treatment effect. This, in turn, could bias the interpretation of study results by policy- and clinical decision-makers. Overall, there are several limitations that need to be considered when using SOFA score as a surrogate endpoint for mortality. When employed as an outcome, the SOFA score is frequently missing and most studies do not adequately describe the amount or nature of missing data, or the methods used to handle missing data in the analysis.

Project description:BackgroundTimely and accurate outcome prediction plays a critical role in guiding clinical decisions for hypertensive ischemic or hemorrhagic stroke patients admitted to the ICU. However, interpreting and translating the predictive models into clinical applications are as important as the prediction itself. This study aimed to develop an interpretable machine learning (IML) model that accurately predicts 28-day all-cause mortality in hypertensive ischemic or hemorrhagic stroke patients.MethodsA total of 4,274 hypertensive ischemic or hemorrhagic stroke patients admitted to the ICU in the USA from multicenter cohorts were included in this study to develop and validate the IML model. Five machine learning (ML) models were developed, including artificial neural network (ANN), gradient boosting machine (GBM), eXtreme Gradient Boosting (XGBoost), logistic regression (LR), and support vector machine (SVM), to predict mortality using the MIMIC-IV and eICU-CRD database in the USA. Feature selection was performed using the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm. Model performance was evaluated based on the area under the curve (AUC), accuracy, positive predictive value (PPV), and negative predictive value (NPV). The ML model with the best predictive performance was selected for interpretability analysis. Finally, the SHapley Additive exPlanations (SHAP) method was employed to evaluate the risk of all-cause in-hospital mortality among hypertensive ischemic or hemorrhagic stroke patients admitted to the ICU.ResultsThe XGBoost model demonstrated the best predictive performance, with the AUC values of 0.822, 0.739, and 0.700 in the training, test, and external cohorts, respectively. The analysis of feature importance revealed that age, ethnicity, white blood cell (WBC), hyperlipidemia, mean corpuscular volume (MCV), glucose, pulse oximeter oxygen saturation (SpO2), serum calcium, red blood cell distribution width (RDW), blood urea nitrogen (BUN), and bicarbonate were the 11 most important features. The SHAP plots were employed to interpret the XGBoost model.ConclusionsThe XGBoost model accurately predicted 28-day all-cause in-hospital mortality among hypertensive ischemic or hemorrhagic stroke patients admitted to the ICU. The SHAP method can provide explicit explanations of personalized risk prediction, which can aid physicians in understanding the model.

Dataset Information

Development and Internal Validation of a New Prognostic Model Powered to Predict 28-Day All-Cause Mortality in ICU COVID-19 Patients-The COVID-SOFA Score.

Background

Methods

Results

Conclusions

Publications

Development and Internal Validation of a New Prognostic Model Powered to Predict 28-Day All-Cause Mortality in ICU COVID-19 Patients-The COVID-SOFA Score.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets