Hospital Mortality In ICU Patients Research Articles

Acute kidney injury-attributable mortality in critically ill patients with sepsis.

BackgroundTo assess whether acute kidney injury (AKI) is independently associated with hospital mortality in ICU patients with sepsis, and estimate the excess AKI-related mortality attributable to AKI.MethodsWe analyzed adult patients from two distinct retrospective critically ill cohorts: (1) Medical Information Mart for Intensive Care IV (MIMIC IV; n = 15,610) cohort and (2) Wenzhou (n = 1,341) cohort. AKI was defined by Kidney Disease: Improving Global Outcomes (KDIGO) criteria. We applied multivariate logistic and linear regression models to assess the hospital and ICU mortality, hospital length-of-stay (LOS), and ICU LOS. The excess attributable mortality for AKI in ICU patients with sepsis was further evaluated.ResultsAKI occurred in 5,225 subjects in the MIMIC IV cohort (33.5%) and 494 in the Wenzhou cohort (36.8%). Each stage of AKI was an independent risk factor for hospital mortality in multivariate logistic regression after adjusting for baseline illness severity. The excess attributable mortality for AKI was 58.6% (95% CI [46.8%–70.3%]) in MIMIC IV and 44.6% (95% CI [12.7%–76.4%]) in Wenzhou. Additionally, AKI was independently associated with increased ICU mortality, hospital LOS, and ICU LOS.ConclusionAcute kidney injury is an independent risk factor for hospital and ICU mortality, as well as hospital and ICU LOS in critically ill patients with sepsis. Thus, AKI is associated with excess attributable mortality.

Comparative analysis of explainable machine learning prediction models for hospital mortality

BackgroundMachine learning (ML) holds the promise of becoming an essential tool for utilising the increasing amount of clinical data available for analysis and clinical decision support. However, the lack of trust in the models has limited the acceptance of this technology in healthcare. This mistrust is often credited to the shortage of model explainability and interpretability, where the relationship between the input and output of the models is unclear. Improving trust requires the development of more transparent ML methods.MethodsIn this paper, we use the publicly available eICU database to construct a number of ML models before examining their internal behaviour with SHapley Additive exPlanations (SHAP) values. Our four models predicted hospital mortality in ICU patients using a selection of the same features used to calculate the APACHE IV score and were based on random forest, logistic regression, naive Bayes, and adaptive boosting algorithms.ResultsThe results showed the models had similar discriminative abilities and mostly agreed on feature importance while calibration and impact of individual features differed considerably and did in multiple cases not correspond to common medical theory.ConclusionsWe already know that ML models treat data differently depending on the underlying algorithm. Our comparative analysis visualises implications of these differences and their importance in a healthcare setting. SHAP value analysis is a promising method for incorporating explainability in model development and usage and might yield better and more trustworthy ML models in the future.

Dynamic APACHE II Score to Predict the Outcome of Intensive Care Unit Patients.

ObjectiveThis study aims to evaluate the accuracy of the Acute Physiology and Chronic Health Evaluation (APACHE) II score on different days in predicting the mortality of critically ill patients to identify the best time point for the APACHE II score.MethodsThe demographic and clinical data are retrieved from the Medical Information Mart for Intensive Care (MIMIC)-IV dataset. APACHE II scores on days 1, 2, 3, 5, 7, 14, and 28 of hospitalization are calculated, and their performance is evaluated using the area under the receiver operating characteristic (AUROC) analysis. The cut-off for defining the high risk of mortality is determined using Youden's index. The APACHE II score on day 3 is the best time point to predict hospital mortality of ICU patients. The Hosmer-Lemeshow goodness-of-fit test is then applied to evaluate the calibration of the day 3 APACHE II score.ResultsWe recruited 6,374 eligible subjects from the MIMIC-IV database. Day 3 is the optimal time point for obtaining the APACHE II score to predict the hospital mortality of patients. The best cut-off for day 3 APACHE II score is 17. When APACHE II score ≥17, the sensitivity for the non-survivors and survivors is 92.8 and 82.2%, respectively, and the positive predictive value (PPV) is 23.1%. When APACHE II socre <17, the specificity for non-survivors and survivors is 90.1 and 80.2%, respectively, and the negative predictive value (NPV) is 87.8%. When day-3 APACHE II is used to predict the hospital mortality, the AUROC is 0.743 (P <0.001). In the ≥17 group, the sensitivity of non-survivors and survivors is 92.2 and 81.3%, respectively, and the PPV is 30.3%. In the <17 group, the specificity of non-survivors and survivors is 100.0 and 80.2%, respectively, and the NPV is 81.6%. The Hosmer-Lemeshow test indicated day-3 APACHE II has a high predicting the hospital mortality (X2 = 6.198, P = 0.625, consistency = 79.4%). However, the day-1 APACHE II has a poor calibration in predicting the hospital mortality rate (X 2 = 294.898, P <0.001).ConclusionDay-3 APACHE II score is an optimal biomarker to predict the outcomes of ICU patients; 17 is the best cut-off for defining patients at high risk of mortality.

Prognostic impact of sustained new-onset atrial fibrillation in critically ill patients.

The development of new-onset atrial fibrillation (AF) in critically ill patients may be associated with poor outcomes. However, it is unknown whether sustained new-onset AF contributes to worse outcome. The aim of this study was to assess whether sustained new-onset AF is associated with stroke and death and to look for a possible dose-response relationship between AF duration and death. In a prospective cohort study conducted in 32 intensive care units in Japan from 2017 to 2018, we enrolled adult patients with new-onset AF. We compared patients with AF duration longer than 48h with those with AF duration shorter than 48h. To assess a dose-response relationship between AF duration and hospital mortality, we conducted landmark analysis and time-dependent Cox regression analysis. Among a total of 423 new-onset AF patients, hospital mortality was 25%, and the incidence of in-hospital stroke was 4.6%. AF duration longer than 48h was not independently associated with hospital mortality (adjusted odds ratio: 1.52; 95% Confidence Interval: 0.87-2.64). The incidence of in-hospital stroke was 7.6% in patients with AF duration longer than 48h and 3.8% in those with AF duration shorter than 48h (p = 0.154). When analyzing time more continuously, we observed a time-dependent association between AF duration and hospital mortality (p = 0.005 by landmark analysis and p = 0.019 by Cox analysis). Sustained new-onset AF was time-dependently associated with hospital mortality in ICU patients, albeit with some uncertainty since AF duration longer than 48h was not independently associated with in-hospital death or stroke.

Center Effects in Hospital Mortality of Critically Ill Patients With Hematologic Malignancies.

We aimed to investigate center effects on hospital mortality of ICU patients with hematologic malignancies and to explore interactions between center and patients characteristics. Multicenter prospective cohort. Seventeen ICUs across France and Belgium. One-thousand eleven patients with hematologic malignancies hospitalized in ICUs. Reanalysis of the original data using state-of-the-art statistical methods with permutation procedures for testing multiple random effects. Average crude mortality was 39% and varied from 11% to 58% across centers. There was a significant center effect on the mean hospital mortality, after adjustment on individual prognostic factors (p < 0.001; median adjusted odds ratio for center effect 1.57 [interquartile range, 1.24-2.18]). There was also a quantitative interaction between center and the effect of the Sequential Organ Failure Assessment score: higher scores were associated with higher mortality (odds ratio for 1 point = 1.24 on average; 95% CI, 1.15-1.33) but with a magnitude that depended on center (p = 0.028). Between-center heterogeneity in hospital mortality was confirmed after adjustment for individual prognostic factors. It was partially explained by center experience in treating oncology patients. Interestingly, center effect was similar in magnitude to that of known mortality risk factors.

The Efficacy of Glasgow Coma Scale (GCS) Score and Acute Physiology and Chronic Health Evaluation (APACHE) II for Predicting Hospital Mortality of ICU Patients with Acute Traumatic Brain Injury.

To compare the efficacy and functional outcome of Glasgow Coma Scale (GCS) score with that of Acute Physiology and Chronic Health Evaluation Score II (APACHE II) in patients with multiple trauma admitted to the ICU. This cross-sectional study included 125 patients with traumatic brain injury associated with systemic trauma admitted to the ICU of Shahid Kamyab Hospital, Mashhad, between September 2015 and December 2016. On the day of admission, data were collected from each patient to calculate GCS and APACHE II scores. Sensitivity, specificity, and correct outcome prediction was compared between GCS and APACHE II. Positive predictive value (PPV) at the cut-off points was higher in APACHE II (80.6%) compared with GCS (69.2%). However, negative predictive value (NPV) of GCS was slightly higher in comparison with APACHE II. Moreover, the area under the receiver operating characteristic (ROC) curve for sensitivity and specificity of GCS and APACHE II showed no significant difference (0.81±0.04 vs. 0.83±0.04; p=0.278 respectively). Our study suggested that there was no considerable difference between GCS and APACHE II scores for predicting mortality in head injury patients. Both scales showed acceptable PPV, while APACHE II showed better results. However, the utilization of GCS in the initial assessment is recommended over APACHE II as the former provides higher time- and cost-efficiency.

Impact of diuretics use over AKI development and hospital mortality in ICU patients

In different studies, use if diuretics have been shown not useful for AKI prevention and a possible relationship with mortality has been suggested.

Association of Body Mass Index With Hospital Mortality in ICU Patients

Department of Anesthesiology, Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, People’s Republic of China The authors have disclosed that they do not have any potential conflicts of interest.

Light and the outcome of the critically ill: an observational cohort study

IntroductionLight before and during acute illness has been associated with both benefit and harm in animal models and small human studies. Our objective was to determine the associations of light duration (photoperiod) and intensity (insolation) before and during critical illness with hospital mortality in ICU patients. Based on the 'winter immunoenhancement' theory, we tested the hypothesis that a shorter photoperiod before critical illness is associated with improved survival.MethodsWe analyzed data from 11,439 patients admitted to 8 ICUs at the University of Pittsburgh Medical Center between June 30, 1999 and July 31, 2004. Daily photoperiod and insolation prior to and after ICU admission were estimated for each patient by using data provided by the United States Naval Observatory and National Aeronautics and Space Administration and direct measurement of light gradient from outside to bedside for each ICU room. Our primary outcome was hospital mortality. The association between light and risk of death was analyzed using multivariate analyses, adjusting for potential confounders, including severity of illness, case mix, and ICU type.ResultsThe cohort had an average APACHE III of 52.9 and a hospital mortality of 10.7%. In total, 128 ICU beds were analyzed; 108 (84%) had windows. Pre-illness photoperiod ranged from 259 to 421 hours in the prior month. A shorter photoperiod was associated with a reduced risk of death: for each 1-hour decrease, the adjusted OR was 0.997 (0.994 to 0.999, p = 0.03). In the ICU, there was near complete (99.6%) degradation of natural light from outside to the ICU bed. Thus, light exposure once in the ICU approached zero; the 24-hour insolation was 0.005 ± 0.003 kWh/m2 with little diurnal variation. There was no association between ICU photoperiod or insolation and mortality.ConclusionsConsistent with the winter immunoenhancement theory, a shorter photoperiod in the month before critical illness is associated with a reduced risk of death. Once in the ICU, patients are exposed to near negligible natural light despite the presence of windows. Further studies are warranted to determine the underlying mechanisms and whether manipulating light exposure, before or during ICU admission, can enhance survival.

A comparison between the APACHE II and Charlson Index Score for predicting hospital mortality in critically ill patients

BackgroundRisk adjustment and mortality prediction in studies of critical care are usually performed using acuity of illness scores, such as Acute Physiology and Chronic Health Evaluation II (APACHE II), which emphasize physiological derangement. Common risk adjustment systems used in administrative datasets, like the Charlson index, are entirely based on the presence of co-morbid illnesses. The purpose of this study was to compare the discriminative ability of the Charlson index to the APACHE II in predicting hospital mortality in adult multisystem ICU patients.MethodsThis was a population-based cohort design. The study sample consisted of adult (>17 years of age) residents of the Calgary Health Region admitted to a multisystem ICU between April 2002 and March 2004. Clinical data were collected prospectively and linked to hospital outcome data. Multiple regression analyses were used to compare the performance of APACHE II and the Charlson index.ResultsThe Charlson index was a poor predictor of mortality (C = 0.626). There was minimal difference between a baseline model containing age, sex and acute physiology score (C = 0.74) and models containing either chronic health points (C = 0.76) or Charlson index variations (C = 0.75, 0.76, 0.77). No important improvement in prediction occurred when the Charlson index was added to the full APACHE II model (C = 0.808 to C = 0.813).ConclusionThe Charlson index does not perform as well as the APACHE II in predicting hospital mortality in ICU patients. However, when acuity of illness scores are unavailable or are not recorded in a standard way, the Charlson index might be considered as an alternative method of risk adjustment and therefore facilitate comparisons between intensive care units.

Prognostic scores in intensive care

Since the development of prognostic score systems in intensive care medicine in the 1980s score models have improved substantially and are now based on much larger databases. They have been validated in many multicenter and international studies all over the world. Prognostic scoring systems may be used for assessment of severity of illness, stratifying patients prior to randomization in clinical trials, evaluation and comparing outcome and survival (hospital mortality), quality assessment, cost-benefit analysis, and in clinical decision making. Validated time points for predicting hospital mortality of ICU patients are at admission and at 24 hours. The relationship of the observed hospital mortality rate to the estimated mortality provides the basis for clinical performance measurement. Since each ICU serves a different patient population, each score system must be calibrated in the individual hospital to ensure that the model is applicable. General scores covering more than one disease are Acute Physiology And Chronic Health Evaluation (APACHE II, APACHE III), Simplified Acute Physiology Score (SAPS) and Mortality Predicting Model (MPM). The Therapeutic Intervention Scoring System (TISS) and in part the Hannover Intensive Score (HIS) evaluate exclusively the amount of medical therapy required. The TISS-Score might serve as a possible measure of resource use for the ICU portion of the hospital stay. Disease (e.g. Trauma Score, Injury of Severity Score) and patient (e.g. PRISM = Pediatric Risk of Mortality) specific scores take into account the influence of disease and patient population in relation to outcome. They are not always of more predictive value than general score models. Score models have been criticized for a number of reasons. Outcome of ICU therapy should incorporate not only survival but should also take into account quality of life, morbidity and disability. Severity scores have no role in clinical decision making for an individual patient (e.g. patient triage for ICU admission, discharge criteria, withdrawal of life support). This is due to the current low sensitivity. Subsequent validation of variables could improve the sensitivity and the value of severity scoring in the future. Nevertheless, illness severity scores will never be indicative of absolute irreversibility of disease or impossibility of survival. Advances in computer technology should assist in achieving many of the future goals of prognostic scoring systems. Most of the physiological data are available from ICU monitors and computerized laboratory systems. By electronically interfacing with the ICU monitor an automated patient data entry is possible and will provide that prognostic scores can be made available to the clinician daily.

Why Severity Models Should be Used with Caution

There are now two validated time points for predicting hospital mortality of ICU patients--at admission and at 24 hours. The best purposes include evaluation of high clinical performance ICUs and for patients being enrolled in clinical trials. For the latter purpose, the model must be calibrated in the individual hospital to ensure that the model is applicable. This can be estimated by using goodness-of-fit testing. There are fewer uses for physiology scores and increased emphasis on converting scores to probabilities. For individual patient application, the model should be demonstrated to have high discrimination, as measured by the area under the receiver operating characteristic curve, and high calibration, as defined by goodness-of-fit testing. Although models have improved substantially and are now based on much larger databases, there is considerable uncertainty in their application for insurance purposes, triage, regulatory applications, sanctions against individual physicians, and cost containment. Current models may not adequately describe important ICU conditions such as adult respiratory distress syndrome and multi-organ dysfunction occurring after 24 hours into ICU care. For family discussions regarding prognosis of individual patients, ICU severity models must be used cautiously at admission or after 24 hours, with the understanding of the strengths and weakness of estimating probabilities of hospital mortality. The mathematical link between physiology score and estimation of hospital mortality is established only for the time point of 24 hours after ICU admission. Calibration and discrimination of the admission and 24-hour models also must be performed within each hospital in which individual probabilities are presented to families. It may be possible to customize a probability model such as MPM to achieve a high level of calibration at the individual hospital level.