Ventilatory Ratio is a Valuable Prognostic Indicator in a Representative Observational Cohort of Patients with Acute Respiratory Distress Syndrome

Background Estimating mortality risk is essential for prognostic enrichment. How various indices specific to respiratory compromise contribute to prognostication in patients with acute respiratory distress syndrome (ARDS) is not well-characterized in general clinical populations. The primary objective of this study was to identify variables specific to respiratory failure that add prognostic value to indicators of systemic illness severity. We tested the added benefit of respiratory variables in a representative observational cohort of patients with ARDS.Methods 50 patients with ARDS were enrolled in a single-center, prospective, observational cohort. We tested the contribution of respiratory variables (oxygenation index, ventilatory ratio [VR], and the radiographic assessment of lung edema score) to logistic regression models of 28-day mortality adjusted for indicators of systemic illness severity (the Acute Physiology and Chronic Health Evaluation [APACHE] III score or severity of shock as measured by the number of vasopressors required at baseline). We also compared a model utilizing APACHE III with one including baseline number of vasopressors using the areas under their receiver operating curves.ResultsVR significantly improved model performance by likelihood ratio testing when added to APACHE III (p = 0.04) or vasopressor number at baseline (p = 0.01). Adjusted for APACHE III, each 0.5-unit change in VR was associated with an odds ratio for 28-day mortality of 1.78 (95% CI = 0.78-3.23). The number of vasopressors required at baseline had similar prognostic discrimination to the multi-component APACHE III. A model including the number of vasopressors and VR (area under the receiver operating curve [AUROC] 0.77, 95% CI 0.64-0.90) was comparable to a model including APACHE III and VR (AUROC 0.81 (95% CI 0.68 – 0.93), p value for comparison = 0.58.) ConclusionsIn this observational cohort of patients with ARDS, the ventilatory ratio significantly improved discrimination for mortality when combined with indicators of severe systemic illness. Additionally, the number of vasopressors required at baseline and APACHE III had similar discrimination for mortality when combined with VR. The ventilatory ratio is easily obtained at the bedside and offers promise for both clinical prognostication and enriching clinical trial populations.

Prediction of Short-Term Mortality With Renal Replacement Therapy in Patients With Cardiac Surgery-Associated Acute Kidney Injury

Objective: We aimed to: (1) explore the risk factors that affect the prognosis of cardiac surgery-associated acute kidney injury (CS-AKI) in patients undergoing renal replacement therapy (RRT) and (2) investigate the predictive value of the Acute Physiology and Chronic Health Evaluation (APACHE) III score, Sequential Organ Failure Assessment (SOFA) score, and Vasoactive-Inotropic Score (VIS) for mortality risk in patients undergoing RRT.Methods: Data from patients who underwent cardiac surgery from January 2015 through February 2021 were retrospectively reviewed to calculate the APACHE III score, SOFA score, and VIS on the first postoperative day and at the start of RRT. Various risk factors influencing the prognosis of the patients during treatment were evaluated; the area under the receiver operating characteristics curve (AUCROC) was used to measure the predictive ability of the three scores. Independent risk factors influencing mortality were analyzed using multivariable binary logistic regression.Results: A total of 90 patients were included in the study, using 90-day survival as the end point. Of those patients, 36 patients survived, and 54 patients died; the mortality rate reached 60%. At the start of RRT, the AUCROC of the APACHE III score was 0.866 (95% CI: 0.795–0.937), the VIS was 0.796 (95% CI: 0.700–0.892), and the SOFA score was 0.732 (95% CI: 0.623–0.842). The AUCROC-value of the APACHE III score on the first postoperative day was 0.790 (95% CI: 0.694–0.885). After analyzing multiple factors, we obtained the final logistic regression model with five independent risk factors at the start of RRT: a high APACHE III score (OR: 1.228, 95% CI: 1.079–1.397), high VIS (OR: 1.147, 95% CI: 1.021–1.290), low mean arterial pressure (MAP) (OR: 1.170, 95% CI: 1.050–1.303), high lactate value (OR: 1.552, 95% CI: 1.032–2.333), and long time from AKI to initiation of RRT (OR: 1.014, 95% CI: 1.002–1.027).Conclusion: In this study, we showed that at the start of RRT, the APACHE III score and the VIS can accurately predict the risk of death in patients undergoing continuous RRT for CS-AKI. The APACHE III score on the first postoperative day allows early prediction of patient mortality risk. Predictors influencing patient mortality at the initiation of RRT were high APACHE III score, high VIS, low MAP, high lactate value, and long time from AKI to the start of RRT.

The Applicability of Commonly Used Severity of Illness Scores to Tropical Infections in Australia

Many patients with leptospirosis, melioidosis, and rickettsial infection require intensive care unit (ICU) admission in tropical Australia every year. The multi-organ dysfunction associated with these infections results in significantly elevated severity of illness (SOI) scores. However, the accuracy of these SOI scores in predicting death from these tropical infections is incompletely defined. This retrospective study was performed at Cairns Hospital, a tertiary-referral hospital in tropical Australia. All patients admitted to ICU with laboratory-confirmed leptospirosis, melioidosis, and rickettsial disease between January 1, 1999 and June 30, 2020, were eligible for the study. The ability of Acute Physiology and Chronic Health Evaluation (APACHE) II, APACHE III, Simplified Acute Physiology Scores (SAPS) II, and Sequential Organ Failure Assessment (SOFA) scores to predict death before ICU discharge was evaluated. Overall, 18 (12.1%) of the 149 included patients died: 15/74 (20.3%) with melioidosis, 2/54 (3.7%) with leptospirosis and 1/21 (4.8%) with rickettsial disease. However, the APACHE II, APACHE III, SAPS II, and SOFA scores significantly overestimated the case-fatality rate of all the infections; the disparity between the predicted and observed mortality was most marked in the cases of leptospirosis and rickettsial disease. Commonly used SOI scores significantly overestimate the case-fatality rate of melioidosis, leptospirosis, and rickettsial infections in Australian ICU patients. This may be at least partly explained by the unique pathophysiology of these infections, particularly leptospirosis and rickettsial disease. However, SOI scores may still be useful in facilitating the comparison of disease severity in clinical trials that examine patients with these pathogens.

Prognostic Performance of Existing Scoring Systems among Critically Ill Patients Requiring Continuous Renal Replacement Therapy: An Observational Study

Background: Among critical patients, few studies have evaluated the discrimination of current illness scoring systems in predicting outcomes after continuous renal replacement therapy (CRRT) initiation. Methods: Patients receiving CRRT in the ICU between 2005 and 2018 from the Chang Gung Research Database were extracted. All the components of the Acute Physiology Assessment and Chronic Health Evaluation (APACHE) III, Sequential Organ Failure Assessment (SOFA), qSOFA, and MOSAIC scoring systems on days 1, 3, and 7 of CRRT were recorded. Patients older than 80 years were identified and analyzed separately. Results: We identified 3370 adult patients for analysis. The discrimination ability of the scoring systems was acceptable at day 7 after CRRT initiation, including SOFA (area under the receiver operating characteristic curve, 74.1% (95% confidence interval, 71.7–76.5%)), APACHEIII (74.7% (72.3–77.1%)), and MOSAIC (71.3% (68.8%–73.9%)). These systems were not ideal on days 1 and 3, and that of qSOFA was poor at any time point. The discrimination performance was slightly better among patients ≥80 years. Conclusions: APACHE III, MOSAIC, and SOFA can be intensivists and families’ reference to make their decision of withdrawing or withholding CRRT after a short period of treatment, especially in adults ≥80 years old.

Metaraminol and Noradrenaline in Septic Shock: A Retrospective Comparison

IntroductionVasopressor use is an important facet of septic shock management, in order to maintain hemodynamic targets and end organ perfusion. Traditionally, Noradrenaline has been the ‘gold standard’ drug of choice for septic shock. Metaraminol is an alternative vasopressor that has been used for septic shock. However, there has been minimal research in comparing the two drugs in septic patients, particularly with regards to total time spent on infusion. ObjectivesTo compare total time spent on either Metaraminol or Noradrenaline infusion by septic shock patients, whilst adjusting for baseline severity of illness. Secondary outcomes included incidence of mechanical ventilation and new requirement of renal replacement therapy, and mortality. MethodsA retrospective medical records review was undertaken, looking at all septic shock patients admitted to ICU in 2019, who received either Metaraminol or Noradrenaline. Data extracted from eRIC (the ICU database) included total time spent on infusion, APACHE III scores, incidence of mechanical ventilation, incidence of renal replacement therapy, and mortality. ResultsOur review yielded 174 patients who were eligible for further statistical analysis (63 in Metaraminol group, and 111 in the Noradrenaline group). The mean duration of infusion in the Metaraminol group was 1655 minutes, and 2663 minutes in the Noradrenaline group. The mean APACHE III Scores were 62 in the Metaraminol group and 77 in the Noradrenaline group. A one-way ANCOVA test found that there was a statistically significant [F(1, 171)=4.511, p=0.035] reduction in time spent on Metaraminol infusion, compared with Noradrenaline, after adjusting for baseline severity of illness by way of APACHE III Score. ConclusionOur study found a statistically significant reduction in time spent on a Metaraminol infusion compared with Noradrenaline by septic shock patients, after controlling for severity of illness. However, due to its retrospective study design, we were unable to account for bias and confounders, such as antibiotic and fluid administration, or clinician preference for one drug over the other. Nevertheless, our study adds to the paucity of literature comparing Metaraminol to Noradrenaline, and paves the way for future randomized trials comparing the two drugs in septic shock.

Machine Learning Classifier Models Can Identify Delirium in Intensive Care Units

Purpose: The aim of this study was to use machine learning to construct a model for the analysis of risk factors and prediction of delirium among ICU patients.Methods: We developed a set of real-world data to enable the comparison of the reliability and accuracy of delirium prediction models from the MIMIC-III database, the MIMIC-IV database and the eICU Collaborative Research Database. Significance tests, correlation analysis, and factor analysis were used to individually screen 80 potential risk factors. The predictive algorithms were run using the following models: Logistic regression, naive Bayesian, K-nearest neighbors, support vector machine, random forest, and eXtreme Gradient Boosting. Conventional E-PRE-DELIRIC and eighteen models, including all-factor (AF) models with all potential variables, characteristic variable (CV) models with principal component factors, and rapid predictive (RP) models without laboratory test results, were used to construct the risk prediction model for delirium. The performance of these machine learning models was measured by the area under the receiver operating characteristic curve (AUC) of tenfold cross-validation. The VIMs and SHAP algorithms, feature interpretation and sample prediction interpretation algorithms of the machine learning black box model were implemented.Results: A total of 78,365 patients were enrolled in this study, 22,159 of whom (28.28%) had positive delirium records. The E-PRE-DELIRIC model (AUC, 0.77), CV models (AUC, 0.77-0.93), CV models (AUC, 0.77-0.88) and RP models (AUC, 0.75-0.87) had discriminatory value. The random forest CV model found that the top five factors accounting for the weight of delirium were length of ICU stay, verbal response score, APACHE-III score, urine volume and hemoglobin. The SHAP values in the eXtreme Gradient Boosting CV model showed that the top three features that were negatively correlated with outcomes were verbal response score, urine volume, and hemoglobin; the top three characteristics that were positively correlated with outcomes were length of ICU stay, APACHE-III score, and alanine transaminase.Conclusion: Even with a small number of variables, machine learning has a good ability to predict delirium in critically ill patients. Characteristic variables provide direction for early intervention to reduce the risk of delirium.

Modelling hospital outcome: problems with endogeneity

Background Mortality modelling in the critical care paradigm traditionally uses logistic regression, despite the availability of estimators commonly used in alternate disciplines. Little attention has been paid to covariate endogeneity and the status of non-randomized treatment assignment. Using a large registry database, various binary outcome modelling strategies and methods to account for covariate endogeneity were explored. Methods Patient mortality data was sourced from the Australian & New Zealand Intensive Society Adult Patient Database for 2016. Hospital mortality was modelled using logistic, probit and linear probability (LPM) models with intensive care (ICU) providers as fixed (FE) and random (RE) effects. Model comparison entailed indices of discrimination and calibration, information criteria (AIC and BIC) and binned residual analysis. Suspect covariate and ventilation treatment assignment endogeneity was identified by correlation between predictor variable and hospital mortality error terms, using the Stata™ “eprobit” estimator. Marginal effects were used to demonstrate effect estimate differences between probit and “eprobit” models. Results The cohort comprised 92,693 patients from 124 intensive care units (ICU) in calendar year 2016. Patients mean age was 61.8 (SD 17.5) years, 41.6% were female and APACHE III severity of illness score 54.5(25.6); 43.7% were ventilated. Of the models considered in predicting hospital mortality, logistic regression (with or without ICU FE) and RE logistic regression dominated, more so the latter using information criteria indices. The LPM suffered from many predictions outside the unit [0,1] interval and both poor discrimination and calibration. Error terms of hospital length of stay, an independent risk of death score and ventilation status were correlated with the mortality error term. Marked differences in the ventilation mortality marginal effect was demonstrated between the probit and the "eprobit" models which were scenario dependent. Endogeneity was not demonstrated for the APACHE III score. Conclusions Logistic regression accounting for provider effects was the preferred estimator for hospital mortality modelling. Endogeneity of covariates and treatment variables may be identified using appropriate modelling, but failure to do so yields problematic effect estimates.

Elevation of Circulating LIGHT (TNFSF14) and Interleukin-18 Levels in Sepsis-Induced Multi-Organ Injuries

Objective The cytokines, LIGHT (TNFSF14) and Interleukin-18 (IL-18), are two important therapeutic targets due to their central roles in the function of activated T cells and inflammatory injury. LIGHT was recently shown to play a major role in COVID19 induced acute respiratory distress syndrome (ARDS), reducing mortality and hospital stay. This study aims to investigate the associations of LIGHT and IL-18 with non-COVID19 related ARDS, acute hypoxic respiratory failure (AHRF) or acute kidney injury (AKI), secondary to viral or bacterial sepsis. Research Design and Methods A cohort of 280 subjects diagnosed with sepsis, including 91 cases with sepsis triggered by viral infections, were investigated in this study and compared to healthy controls. Serum LIGHT, IL-18, and 59 other biomarkers (cytokines, chemokines and acute-phase reactants) were measured and associated with symptom severity. Results ARDS was observed in 36% of the patients, with 29% of the total patient cohort developing multi-organ failure (failure of two or more organs). We observed significantly increased LIGHT level (>2SD above mean of healthy subjects) in both bacterial sepsis patients (P=1.80E-05) and patients with sepsis from viral infections (P=1.78E-03). In bacterial sepsis, increased LIGHT level associated with ARDS, AKI and higher Apache III scores, findings also supported by correlations of LIGHT with other biomarkers of organ failures, suggesting LIGHT may be an inflammatory driver. IL-18 levels were highly variable across individuals, and consistently correlated with Apache III scores, mortality, and AKI, in both bacterial and viral sepsis. Conclusions For the first time, we demonstrate independent effects of LIGHT and IL-18 in septic organ failures. LIGHT levels are significantly elevated in non-COVID19 sepsis patients with ARDS and/or multi-organ failures suggesting that anti-LIGHT therapy may be effective therapy in a subset of patients with sepsis. Given the large variance of plasma IL-18 among septic subjects, targeting this pathway raises opportunities that require a precision application.

Detecting Potentially Ineffective Care in Critically Ill Patients in a Community Hospital: Assessing Sustained Predictive Value after Two Decades

Background: Eighteen years ago, we derived a formula to predict 100-day post-discharge mortality, utilizing the Acute Physiology and Chronic Health Evaluation III (APACHE III) data. This study was designed to reassess this formula when applied to a new cohort of patients, utilizing the updated predictive hospital mortality equations derived from APACHE IV.Methods: Compared with the 1995‒1997 cohort in our original study, this study included a cohort of intensive care unit patients from 2012‒2017, with similar demographics. Both cohorts included patients >18 years old admitted to and surviving at least five days of intensive care in the Sarasota Memorial Hospital in Sarasota, Florida, USA. Results: In the recent cohort, the formula exhibited a specificity of 99.7%, sensitivity of 17.8%, false positive rate of 0.3%, and positive predictive value of 92.6%; applied to the original cohort, the formula exhibited values of 98.7%, 33.8%, 1.3%, and 93.3%, respectively. There was no statistical difference between the two databases, except in sensitivity. Conclusions: Potentially ineffective care can be predicted with nearly the same specificity and predictive value using the formula developed in the 2002 study. If these results are reproducible at other institutions, they could assist in patient/family and palliative care discussions.

Development and validation of the predictive risk of death model for adult patients admitted to intensive care units in Japan: an approach to improve the accuracy of healthcare quality measures

Background The Acute Physiology and Chronic Health Evaluation (APACHE) III-j model is widely used to predict mortality in Japanese intensive care units (ICUs). Although the model’s discrimination is excellent, its calibration is poor. APACHE III-j overestimates the risk of death, making its evaluation of healthcare quality inaccurate. This study aimed to improve the calibration of the model and develop a Japan Risk of Death (JROD) model for benchmarking purposes. Methods A retrospective analysis was conducted using a national clinical registry of ICU patients in Japan. Adult patients admitted to an ICU between April 1, 2018, and March 31, 2019, were included. The APACHE III-j model was recalibrated with the following models: Model 1, predicting mortality with an offset variable for the linear predictor of the APACHE III-j model using a generalized linear model; model 2, predicting mortality with the linear predictor of the APACHE III-j model using a generalized linear model; and model 3, predicting mortality with the linear predictor of the APACHE III-j model using a hierarchical generalized additive model. Model performance was assessed with the area under the receiver operating characteristic curve (AUROC), the Brier score, and the modified Hosmer–Lemeshow test. To confirm model applicability to evaluating quality of care, funnel plots of the standardized mortality ratio and exponentially weighted moving average (EWMA) charts for mortality were drawn. Results In total, 33,557 patients from 44 ICUs were included in the study population. ICU mortality was 3.8%, and hospital mortality was 8.1%. The AUROC, Brier score, and modified Hosmer–Lemeshow p value of the original model and models 1, 2, and 3 were 0.915, 0.062, and < .001; 0.915, 0.047, and < .001; 0.915, 0.047, and .002; and 0.917, 0.047, and .84, respectively. Except for model 3, the funnel plots showed overdispersion. The validity of the EWMA charts for the recalibrated models was determined by visual inspection. Conclusions Model 3 showed good performance and can be adopted as the JROD model for monitoring quality of care in an ICU, although further investigation of the clinical validity of outlier detection is required. This update method may also be useful in other settings.