No abstract

BackgroundHypoxic hepatitis (HH) is a type of acute hepatic injury that is histologically characterized by centrilobular liver cell necrosis and that is caused by insufficient oxygen delivery to the hepatocytes. Typical for HH is the sudden and significant increase of aspartate aminotransferase (AST) in response to cardiac, circulatory or respiratory failure. The aim of this study is to investigate its epidemiology, causes, evolution and outcome.MethodsThe screened population consisted of all adults admitted to the intensive care unit (ICU) at the Ghent University Hospital between January 1, 2007 and September 21, 2015. HH was defined as peak AST > 5 times the upper limit of normal (ULN) after exclusion of other causes of liver injury. Thirty-five variables were retrospectively collected and used in descriptive analysis, time series plots and Kaplan–Meier survival curves with multi-group log-rank tests.ResultsHH was observed in 4.0% of the ICU admissions at our center. The study cohort comprised 1116 patients. Causes of HH were cardiac failure (49.1%), septic shock (29.8%), hypovolemic shock (9.4%), acute respiratory failure (6.4%), acute on chronic respiratory failure (3.3%), pulmonary embolism (1.4%) and hyperthermia (0.5%). The 28-day mortality associated with HH was 45.0%. Mortality rates differed significantly (P = 0.007) among the causes, ranging from 33.3% in the hyperthermia subgroup to 52.9 and 56.2% in the septic shock and pulmonary embolism subgroups, respectively. The magnitude of AST increase was also significantly correlated (P < 0.001) with mortality: 33.2, 44.4 and 55.4% for peak AST 5–10× ULN, 10–20× ULN and > 20× ULN, respectively.ConclusionThis study surpasses by far the largest cohort of critically ill patients with HH. HH is more common than previously thought with an ICU incidence of 4.0%, and it is associated with a high all-cause mortality of 45.0% at 28 days. The main causes of HH are cardiac failure and septic shock, which include more than 3/4 of all episodes. Clinicians should search actively for any underlying hemodynamic or respiratory instability even in patients with moderately increased AST levels.

BackgroundPredictive models for delayed graft function (DGF) after kidney transplantation are usually developed using logistic regression. We want to evaluate the value of machine learning methods in the prediction of DGF.Methods497 kidney transplantations from deceased donors at the Ghent University Hospital between 2005 and 2011 are included. A feature elimination procedure is applied to determine the optimal number of features, resulting in 20 selected parameters (24 parameters after conversion to indicator parameters) out of 55 retrospectively collected parameters. Subsequently, 9 distinct types of predictive models are fitted using the reduced data set: logistic regression (LR), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machines (SVMs; using linear, radial basis function and polynomial kernels), decision tree (DT), random forest (RF), and stochastic gradient boosting (SGB). Performance of the models is assessed by computing sensitivity, positive predictive values and area under the receiver operating characteristic curve (AUROC) after 10-fold stratified cross-validation. AUROCs of the models are pairwise compared using Wilcoxon signed-rank test.ResultsThe observed incidence of DGF is 12.5 %. DT is not able to discriminate between recipients with and without DGF (AUROC of 52.5 %) and is inferior to the other methods. SGB, RF and polynomial SVM are mainly able to identify recipients without DGF (AUROC of 77.2, 73.9 and 79.8 %, respectively) and only outperform DT. LDA, QDA, radial SVM and LR also have the ability to identify recipients with DGF, resulting in higher discriminative capacity (AUROC of 82.2, 79.6, 83.3 and 81.7 %, respectively), which outperforms DT and RF. Linear SVM has the highest discriminative capacity (AUROC of 84.3 %), outperforming each method, except for radial SVM, polynomial SVM and LDA. However, it is the only method superior to LR.ConclusionsThe discriminative capacities of LDA, linear SVM, radial SVM and LR are the only ones above 80 %. None of the pairwise AUROC comparisons between these models is statistically significant, except linear SVM outperforming LR. Additionally, the sensitivity of linear SVM to identify recipients with DGF is amongst the three highest of all models. Due to both reasons, the authors believe that linear SVM is most appropriate to predict DGF.

Background: While obesity confers an increased risk of death in the general population, numerous studies have reported an association between obesity and improved survival among critically ill patients. This contrary finding has been referred to as the obesity paradox. In this retrospective study, two causal inference approaches were used to address whether the survival of non-obese critically ill patients would have been improved if they had been obese. Methods: The study cohort comprised 6557 adult critically ill patients hospitalized at the Intensive Care Unit of the Ghent University Hospital between 2015 and 2017. Obesity was defined as a body mass index of ≥ 30 kg/m 2. Two causal inference approaches were used to estimate the average effect of obesity in the non-obese (AON): a traditional approach that used regression adjustment for confounding and that assumed missingness completely at random and a robust approach that used machine learning within the targeted maximum likelihood estimation framework along with multiple imputation of missing values under the assumption of missingness at random. 1754 (26.8%) patients were discarded in the traditional approach because of at least one missing value for obesity status or confounders. Results: Obesity was present in 18.9% of patients. The in-hospital mortality was 14.6% in non-obese patients and 13.5% in obese patients. The raw marginal risk difference for in-hospital mortality between obese and non-obese patients was − 1.06% (95% confidence interval (CI) − 3.23 to 1.11%, P = 0.337). The traditional approach resulted in an AON of − 2.48% (95% CI − 4.80 to − 0.15%, P = 0.037), whereas the robust approach yielded an AON of − 0.59% (95% CI − 2.77 to 1.60%, P = 0.599). Conclusions: A causal inference approach that is robust to residual confounding bias due to model misspecification and selection bias due to missing (at random) data mitigates the obesity paradox observed in critically ill patients, whereas a traditional approach results in even more paradoxical findings. The robust approach does not provide evidence that the survival of non-obese critically ill patients would have been improved if they had been obese.