Simultaneous assessment of stress hyperglycemia ratio and glycemic variability to predict mortality in patients with coronary artery disease: a retrospective cohort study from the MIMIC-IV database.

BACKGROUND: Stress hyperglycemia and glycemic variability (GV) can reflect dramatic increases and acute fluctuations in blood glucose, which are associated with adverse cardiovascular events. This study aimed to explore whether the combined assessment of the stress hyperglycemia ratio (SHR) and GV provides additional information for prognostic prediction in patients with coronary artery disease (CAD) hospitalized in the intensive care unit (ICU). METHODS: Patients diagnosed with CAD from the Medical Information Mart for Intensive Care-IV database (version 2.2) between 2008 and 2019 were retrospectively included in the analysis. The primary endpoint was 1-year mortality, and the secondary endpoint was in-hospital mortality. Levels of SHR and GV were stratified into tertiles, with the highest tertile classified as high and the lower two tertiles classified as low. The associations of SHR, GV, and their combination with mortality were determined by logistic and Cox regression analyses. RESULTS: A total of 2789 patients were included, with a mean age of 69.6 years, and 30.1% were female. Overall, 138 (4.9%) patients died in the hospital, and 404 (14.5%) patients died at 1 year. The combination of SHR and GV was superior to SHR (in-hospital mortality: 0.710 vs. 0.689, p = 0.012; 1-year mortality: 0.644 vs. 0.615, p = 0.007) and GV (in-hospital mortality: 0.710 vs. 0.632, p = 0.004; 1-year mortality: 0.644 vs. 0.603, p < 0.001) alone for predicting mortality in the receiver operating characteristic analysis. In addition, nondiabetic patients with high SHR levels and high GV were associated with the greatest risk of both in-hospital mortality (odds ratio [OR] = 10.831, 95% confidence interval [CI] 4.494-26.105) and 1-year mortality (hazard ratio [HR] = 5.830, 95% CI 3.175-10.702). However, in the diabetic population, the highest risk of in-hospital mortality (OR = 4.221, 95% CI 1.542-11.558) and 1-year mortality (HR = 2.013, 95% CI 1.224-3.311) was observed in patients with high SHR levels but low GV. CONCLUSIONS: The simultaneous evaluation of SHR and GV provides more information for risk stratification and prognostic prediction than SHR and GV alone, contributing to developing individualized strategies for glucose management in patients with CAD admitted to the ICU.

Triglyceride-glucose index: a novel evaluation tool for all-cause mortality in critically ill hemorrhagic stroke patients-a retrospective analysis of the MIMIC-IV database.

BACKGROUND: Hemorrhagic stroke (HS), including non-traumatic intracerebral hemorrhage (ICH) and subarachnoid hemorrhage (SAH), constitutes a substantial proportion of cerebrovascular incidents, accounting for around 30% of stroke cases. The triglyceride-glucose index (TyG-i) represents a precise insulin resistance (IR) indicator, a crucial metabolic disturbance. Existing literature has demonstrated an association between TyG-i and all-cause mortality (ACM) among individuals suffering from ischemic stroke (IS). Yet, the TyG-i prognostic implications for severe HS patients necessitating intensive care unit (ICU) admission are not clearly understood. Considering the notably elevated mortality and morbidity associated with HS relative to IS, investigating this association is warranted. Our primary aim was to investigate TyG-i and ACM association among critically ill HS patients within an ICU context. METHODS: Herein, patients with severe HS were identified by accessing the Medical Information Mart for Intensive Care-IV (MIMIC-IV, version 2.2) database, using the International Classification of Diseases (ICD)-9/10 as diagnostic guidelines. Subsequently, we stratified the subjects into quartiles, relying on their TyG-i scores. Moreover, we measured mortality at ICU, in-hospital, 30 days, 90 days, and 1 year as the outcomes. Cox proportional hazards regression analysis and restricted cubic splines (RCS) were deployed for elucidating the relation between the TyG-i and ACM while utilizing the Kaplan-Meier (K-M) method to estimate survival curves. The findings' robustness was assessed by conducting subgroup analysis and interaction tests employing likelihood ratio tests. RESULTS: The analysis included 1475 patients, with a male predominance of 54.4%. Observed mortality rates in the ICU, hospital, 30 days, 90 days, and 1 year were 7.3%, 10.9%, 13.8%, 19.7%, and 27.3%, respectively. Multivariate Cox regression analysis results manifested that heightened TyG-i was significantly related to ACM at 30 days (adjusted hazard ratio [aHR]: 1.32; 95% confidence interval [CI]: 1.05-1.67; P = 0.020), 90 days (aHR: 1.27; 95% CI: 1.04-1.55; P = 0.019), and 1 year (aHR: 1.22; 95% CI: 1.03-1.44; P = 0.023). The results of RCS analysis demonstrated a progressive elevation in ACM risk with rising TyG-i levels. Interaction tests found no significant effect modification in this relationship. CONCLUSION: In summary, TyG-i exhibits a significant correlation with ACM among patients enduring critical illness due to HS. This correlation underscores the probable utility of TyG-i as a prognostic tool for stratifying HS patients according to their risk of mortality. Applying TyG-i in clinical settings could enhance therapeutic decision-making and the management of disease trajectories. Additionally, this investigation augments existing research on the linkage between the TyG-i and IS, elucidating the TyG-i's role in predicting mortality across diverse stroke categories.

Association of the TyG index with prognosis in surgical intensive care patients: data from the MIMIC-IV.

BACKGROUND: The triglyceride-glucose (TyG) index, a tool for assessing insulin resistance, is increasingly recognized for its ability to predict cardiovascular and metabolic risks. However, its relationship with trauma and surgical patient prognosis is understudied. This study investigated the correlation between the TyG index and mortality risk in surgical/trauma ICU patients to identify high-risk individuals and improve prognostic strategies. METHODS: This study identified patients requiring trauma/surgical ICU admission from the Medical Information Mart for Intensive Care (MIMIC-IV) database, and divided them into tertiles based on the TyG index. The outcomes included 28-day mortality and 180-day mortality for short-term and long-term prognosis. The associations between the TyG index and clinical outcomes in patients were elucidated using Cox proportional hazards regression analysis and RCS models. RESULTS: A total of 2103 patients were enrolled. The 28-day mortality and 180-day mortality rates reached 18% and 24%, respectively. Multivariate Cox proportional hazards analysis revealed that an elevated TyG index was significantly related to 28-day and 180-day mortality after covariates adjusting. An elevated TyG index was significantly associated with 28-day mortality (adjusted hazard ratio, 1.19; 95% confidence interval 1.04-1.37) and 180-day mortality (adjusted hazard ratio, 1.24; 95% confidence interval 1.11-1.39). RCS models revealed that a progressively increasing risk of mortality was related to an elevated TyG index. According to our subgroup analysis, an elevated TyG index is associated with increased risk of 28-day and 180-day mortality in critically ill patients younger than 60 years old, as well as those with concomitant stroke or cardiovascular diseases. Additionally, in nondiabetic patients, an elevated TyG index is associated with 180-day mortality. CONCLUSION: An increasing risk of mortality was related to an elevated TyG index. In critically ill patients younger than 60 years old, as well as those with concomitant stroke or cardiovascular diseases, an elevated TyG index is associated with adverse short-term and long-term outcomes. Furthermore, in non-diabetic patients, an elevated TyG index is associated with adverse long-term prognosis.

Association between lactate/albumin ratio and prognosis in patients with acute myocardial infarction.

BACKGROUND: The association between the lactate/albumin ratio (L/A) as a diagnostic indicator and unfavourable clinical outcomes has been established in patients with community-acquired pneumonia, sepsis and heart failure, but the connection between L/A and all-cause mortality in patients with acute myocardial infarction (AMI) has yet to be fully understood. METHODS: This was a retrospective cohort study using MIMIC-IV (v2.2) data, with 2816 patients enrolled and all-cause mortality during hospitalization as the primary outcome. Kaplan-Meier (KM) analysis was used to compare the all-cause mortality between high-level and low-level L/A groups. Receiver operating characteristic (ROC) curve, Restricted cubic splines (RCS) and Cox proportional hazards analysis were performed to investigate the relationship between L/A ratio and in-hospital all-cause mortality. RESULTS: L/A values were significantly higher in the non-survivor groups than the survival groups (1.14 [.20] vs. .60 [.36], p < .05), and area under the ROC curve [.734 (95% confidence interval, .694-.775)] was better than other indicators. Data of COX regression analysis showed that higher L/A value supposed to be an independent risk factor for in-hospital mortality. RCS analysis showed evidence of an increasing trend and a non-linear relationship between L/A and in-hospital mortality (p-value was non-linear <.05). KM survival curves were significantly lower in the high L/A group than the low L/A group (p < .001), and the former group had an increased risk of in-hospital mortality compared with the latter one (Log Rank p < .001). CONCLUSIONS: L/A demonstrates significant independent predictive power for elevated all-cause mortality during hospitalization in patients diagnosed with AMI.

Comparative analysis of dexmedetomidine, midazolam, and propofol impact on epilepsy-related mortality in the ICU: insights from the MIMIC-IV database.

BACKGROUND: Dexmedetomidine (Dex), midazolam, and propofol are three distinct sedatives characterized by varying pharmacological properties. Previous literature has indicated the positive impact of each of these sedatives on ICU patients. However, there is a scarcity of clinical evidence comparing the efficacy of Dex, midazolam, and propofol in reducing mortality among people with epilepsy (PWE). This study aimed to assess the impact of Dex, midazolam, and propofol on the survival of PWE. METHODS: The data were retrospectively retrieved from the Medical Information Mart for Intensive Care (MIMIC)-IV database (version 2.0). PWE were categorized into Dex, midazolam, and propofol groups based on the intravenously administered sedatives. PWE without standard drug therapy were included in the control group. Comparative analyses were performed on the data among the groups. RESULTS: The Dex group exhibited a significantly lower proportion of in-hospital deaths and a markedly higher in-hospital survival time compared to the midazolam and propofol groups (p < 0.01) after propensity score matching. Kaplan-Meier curves demonstrated a significant improvement in survival rates for the Dex group compared to the control group (p = 0.025). Analysis of Variance (ANOVA) revealed no significant differences in survival rates among the Dex, midazolam, and propofol groups (F = 1.949, p = 0.143). The nomogram indicated that compared to midazolam and propofol groups, Dex was more effective in improving the survival rate of PWE. CONCLUSION: Dex might improve the survival rate of PWE in the ICU compared to no standard drug intervention. However, Dex did not exhibit superiority in improving survival rates compared to midazolam and propofol.

Relationship between the magnitude of haemoglobin changes and long-term mortality in patients with sepsis: a retrospective cohort study.

BACKGROUND: Sepsis is a common and severe disease with a high mortality rate in intensive care unit (ICU). The hemoglobin (HGB) level is a key parameter for oxygen supply in sepsis. Although HGB is associated with the progression of inflammation in sepsis patients, its role as a marker following sepsis treatment remains unclear. Here, we studied the correlation between early temporal changes in HGB levels and long-term mortality rates in septic patients. METHOD: In this retrospective study of data on patients with sepsis from the Medical Information Mart for Intensive Care (MIMIC) IV database, the outcome was long-term mortality. Patients were divided based on the cut-off of the HGB percentage for receiver operating characteristic (ROC) curve calculation. Kaplan-Meier (KM) survival curves and Cox proportional hazards regression models were used to analyse the associations between groups and outcomes. Propensity score matching (PSM) was used to verify the results. RESULTS: In this study, 2042 patients with sepsis and changes in HGB levels at day 4 after admission compared to day 1 were enrolled and divided into two groups: group 1 (n = 1147) for those with reduction of HGB < 7% and group 2 (n = 895) for those with dropping ≥ 7%. The long-term survival chances of sepsis with less than a 7% reduction in the proportion of HGB at day four were significantly higher than those of patients in the group with a reduction of 7% or more. After adjusting for covariates in the Cox model, the hazard ratios (HRs) with 95% confidence intervals (CIs) for long-term all-cause mortality in the group with a reduction of 7% or more were as follows: 180 days [HR = 1.41, 95% CI (1.22 to 1.63), P < 0.001]; 360 days [HR = 1.37, 95% CI (1.21 to 1.56), P < 0.001]; 540 days [HR = 1.35, 95% CI (1.20 to 1.53), P < 0.001]; 720 days [HR = 1.45, 95% CI (1.29 to 1.64), P < 0.001]. Additionally, the long-term survival rates, using Kaplan-Meier analysis, for the group with a reduction of 7% or more were lower compared to the group with less than 7% reduction at 180 days (54.3% vs. 65.3%, P < 0.001), 360 days (42.3% vs. 50.9%, P < 0.001), 540 days (40.2% vs. 48.6%, P < 0.001), and 720 days (35.5% vs. 46.1%, P < 0.001). The same trend was obtained after using PSM. CONCLUSION: A ≥ 7% decrease in HGB levels on Day 4 after admission was associated with worse long-term prognosis in sepsis patients admitted to the ICU.

Prediction of 28-Day All-Cause Mortality in Heart Failure Patients with Clostridioides difficile Infection Using Machine Learning Models: Evidence from the MIMIC-IV Database.

INTRODUCTION: Heart failure (HF) may induce bowel hypoperfusion, leading to hypoxia of the villa of the bowel wall and the occurrence of Clostridioides difficile infection (CDI). However, the risk factors for the development of CDI in HF patients have yet to be fully illustrated, especially because of a lack of evidence from real-world data. METHODS: Clinical data and survival situations of HF patients with CDI admitted to ICU were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database. For developing a model that can predict 28-day all-cause mortality in HF patients with CDI, the Recursive Feature Elimination with Cross-Validation (RFE-CV) method was used for feature selection. And nine machine learning (ML) algorithms, including logistic regression (LR), decision tree, Bayesian, adaptive boosting, random forest (RF), gradient boosting decision tree, XGBoost, light gradient boosting machine, and categorical boosting, were applied for model construction. After training and hyperparameter optimization of the models through grid search 5-fold cross-validation, the performance of models was evaluated by the area under curve (AUC), accuracy, sensitivity, specificity, precision, negative predictive value, and F1 score. Furthermore, the SHapley Additive exPlanations (SHAP) method was used to interpret the optimal model. RESULTS: A total of 526 HF patients with CDI were included in the study, of whom 99 cases (18.8%) experienced death within 28 days. Eighteen of the 57 variables were selected for the model construction algorithm for model construction. Among the ML models considered, the RF model emerged as the optimal model achieving the accuracy, F1-score, and AUC values of 0.821, 0.596, and 0.864, respectively. The net benefit of the model surpassed other models at 16%-22% threshold probabilities based on decision curve analysis. According to the importance of features in the RF model, red blood cell distribution width, blood urea nitrogen, Simplified Acute Physiology Score II, Sequential Organ Failure Assessment, and white blood cell count were highlighted as the five most influential variables. CONCLUSIONS: We developed ML models to predict 28-day all-cause mortality in HF patients associated with CDI in the ICU, which are more effective than the conventional LR model. The RF model has the best performance among all the ML models employed. It may be useful to help clinicians identify high-risk HF patients with CDI.

Normalized lactate load as an independent prognostic indicator in patients with cardiogenic shock.

BACKGROUND: Early prognosis evaluation is crucial for decision-making in cardiogenic shock (CS) patients. Dynamic lactate assessment, for example, normalized lactate load, has been a better prognosis predictor than single lactate value in septic shock. Our objective was to investigate the correlation between normalized lactate load and in-hospital mortality in patients with CS. METHODS: Data were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database. The calculation of lactate load involved the determination of the cumulative area under the lactate curve, while normalized lactate load was computed by dividing the lactate load by the corresponding period. Receiver Operating Characteristic (ROC) curves were constructed, and the evaluation of areas under the curves (AUC) for various parameters was performed using the DeLong test. RESULTS: Our study involved a cohort of 1932 CS patients, with 687 individuals (36.1%) experiencing mortality during their hospitalization. The AUC for normalized lactate load demonstrated significant superiority compared to the first lactate (0.675 vs. 0.646, P < 0.001), maximum lactate (0.675 vs. 0.651, P < 0.001), and mean lactate (0.675 vs. 0.669, P = 0.003). Notably, the AUC for normalized lactate load showed comparability to that of the Sequential Organ Failure Assessment (SOFA) score (0.675 vs. 0.695, P = 0.175). CONCLUSION: The normalized lactate load was an independently associated with the in-hospital mortality among CS patients.

Comparison of different intensive care scoring systems and Glasgow Aneurysm score for aortic aneurysm in predicting 28-day mortality: a retrospective cohort study from MIMIC-IV database.

OBJECTIVE: This study aims to assess the performance of various scoring systems in predicting the 28-day mortality of patients with aortic aneurysms (AA) admitted to the intensive care unit (ICU). METHODS: We utilized data from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) to perform a comparative analysis of various predictive systems, including the Glasgow Aneurysm Score (GAS), Simplified Acute Physiology Score (SAPS) III, SAPS II, Logical Organ Dysfunction System (LODS), Sequential Organ Failure Assessment (SOFA), Systemic Inflammatory Response Syndrome (SIRS), and The Oxford Acute Illness Severity Score (OASIS). The discrimination abilities of these systems were compared using the area under the receiver operating characteristic curve (AUROC). Additionally, a 4-knotted restricted cubic spline regression was employed to evaluate the association between the different scoring systems and the risk of 28-day mortality. Finally, we conducted a subgroup analysis focusing on patients with abdominal aortic aneurysms (AAA). RESULTS: This study enrolled 586 patients with AA (68.39% male). Among them, 26 patients (4.4%) died within 28 days. Comparative analysis revealed higher SAPS II, SAPS III, SOFA, LODS, OASIS, and SIRS scores in the deceased group, while no statistically significant difference was observed in GAS scores between the survivor and deceased groups (P = 0.148). The SAPS III system exhibited superior predictive value for the 28-day mortality rate (AUROC 0.805) compared to the LODS system (AUROC 0.771), SOFA (AUROC 0.757), SAPS II (AUROC 0.759), OASIS (AUROC 0.742), SIRS (AUROC 0.638), and GAS (AUROC 0.586) systems. The results of the univariate and multivariate logistic analyses showed that SAPS III was statistically significant for both 28-day and 1-year mortality. Subgroup analyses yielded results consistent with the overall findings. No nonlinear relationship was identified between these scoring systems and 28-day all-cause mortality (P for nonlinear > 0.05). CONCLUSION: The SAPS III system demonstrated superior discriminatory ability for both 28-day and 1-year mortality compared to the GAS, SAPS II SIRS, SOFA, and OASIS systems among patients with AA.

Construction and evaluation of neonatal respiratory failure risk prediction model for neonatal respiratory distress syndrome.

BACKGROUND: Neonatal respiratory distress syndrome (NRDS) is a common respiratory disease in preterm infants, often accompanied by respiratory failure. The aim of this study was to establish and validate a nomogram model for predicting the probability of respiratory failure in NRDS patients. METHODS: Patients diagnosed with NRDS were extracted from the MIMIC-iv database. The patients were randomly assigned to a training and a validation cohort. Univariate and stepwise Cox regression analyses were used to determine the prognostic factors of NRDS. A nomogram containing these factors was established to predict the incidence of respiratory failure in NRDS patients. The area under the receiver operating characteristic curve (AUC), receiver operating characteristic curve (ROC), calibration curves and decision curve analysis were used to determine the effectiveness of this model. RESULTS: The study included 2,705 patients with NRDS. Univariate and multivariate stepwise Cox regression analysis showed that the independent risk factors for respiratory failure in NRDS patients were gestational age, pH, partial pressure of oxygen (PO2), partial pressure of carbon dioxide (PCO2), hemoglobin, blood culture, infection, neonatal intracranial hemorrhage, Pulmonary surfactant (PS), parenteral nutrition and respiratory support. Then, the nomogram was constructed and verified. CONCLUSIONS: This study identified the independent risk factors of respiratory failure in NRDS patients and used them to construct and evaluate respiratory failure risk prediction model for NRDS. The present findings provide clinicians with the judgment of patients with respiratory failure in NRDS and help clinicians to identify and intervene in the early stage.

Association between baseline serum bicarbonate and the risk of postoperative delirium in patients undergoing cardiac surgery in the ICU: a retrospective study from the MIMIC-IV database.

BACKGROUND: Although serum bicarbonate is a reliable predictor of various disease complications, its relationship with postoperative delirium (POD) remains unclear. Our research aimed to assess the effect of baseline serum bicarbonate levels on the incidence of POD in cardiac surgery patients. METHODS: A retrospective analysis was conducted on cardiac surgery patients who met specific inclusion and exclusion criteria, using data from the Marketplace for Information in Critical Care Medicine (MIMIC-IV) database. Univariate and multivariate logistic regression models are employed to explore the correlation between serum bicarbonate levels and the risk of POD, and their predictive efficacy is assessed by means of restricted cubic spline regression models (RCS) and receiver operating characteristic curves (ROC). In addition, subgroup and sensitivity analyses are conducted to test the robustness of the results. RESULTS: In this study, 5,422 patients were included, where the incidence of POD was 13.0%. For each 1 mmol/L increase in bicarbonate, a 13% reduction in the risk of POD was observed in the fully adjusted model (OR = 0.87, 95% CI: 0.83-0.91, P < 0.001). The RCS model demonstrated a linear negative correlation between the level of bicarbonate and the risk of POD (P for nonlinearity = 0.987). The ROC curve analysis demonstrated that the bicarbonate level had moderate predictive efficacy (AUC = 0.629). Both subgroup and sensitivity analyses reaffirmed the robustness of these results. CONCLUSIONS: Lower baseline serum bicarbonate levels in cardiac surgery patients are linked to a higher risk of POD. Monitoring and adjusting serum bicarbonate levels may help identify high-risk patients and potentially improve outcomes.

Clinical nomogram prediction model to assess the risk of prolonged ICU length of stay in patients with diabetic ketoacidosis: a retrospective analysis based on the MIMIC-IV database.

BACKGROUND: The duration of hospitalization, especially in the intensive care unit (ICU), for patients with diabetic ketoacidosis (DKA) is influenced by patient prognosis and treatment costs. Reducing ICU length of stay (LOS) in patients with DKA is crucial for optimising healthcare resources utilization. This study aimed to establish a nomogram prediction model to identify the risk factors influencing prolonged LOS in ICU-managed patients with DKA, which will serve as a basis for clinical treatment, healthcare safety, and quality management research. METHODS: In this single-centre retrospective cohort study, we performed a retrospective analysis using relevant data extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Clinical data from 669 patients with DKA requiring ICU treatment were included. Variables were selected using the Least Absolute Shrinkage and Selection Operator (LASSO) binary logistic regression model. Subsequently, the selected variables were subjected to a multifactorial logistic regression analysis to determine independent risk factors for prolonged ICU LOS in patients with DKA. A nomogram prediction model was constructed based on the identified predictors. The multivariate variables included in this nomogram prediction model were the Oxford acute severity of illness score (OASIS), Glasgow coma scale (GCS), acute kidney injury (AKI) stage, vasoactive agents, and myocardial infarction. RESULTS: The prediction model had a high predictive efficacy, with an area under the curve value of 0.870 (95% confidence interval [CI], 0.831-0.908) in the training cohort and 0.858 (95% CI, 0.799-0.916) in the validation cohort. A highly accurate predictive model was depicted in both cohorts using the Hosmer-Lemeshow (H-L) test and calibration plots. CONCLUSION: The nomogram prediction model proposed in this study has a high clinical application value for predicting prolonged ICU LOS in patients with DKA. This model can help clinicians identify patients with DKA at risk of prolonged ICU LOS, thereby enhancing prompt intervention and improving prognosis.

Associations of arterial oxygen partial pressure with all­cause mortality in critically ill ischemic stroke patients: a retrospective cohort study from MIMIC IV 2.2.

BACKGROUND: As a supportive treatment, the effectiveness of oxygen therapy in ischemic stroke (IS) patients remains unclear. This study aimed to evaluate the relationships between arterial partial pressure of oxygen (PaO2) and both consciousness at discharge and all-cause mortality risk in ICU IS patients. METHODS: Blood gas measurements for all patients diagnosed with IS were extracted from the MIMIC-IV database. Patients were classified into four groups based on their average PaO2 during the first ICU day: hypoxemia (PaO2 < 80 mmHg), normoxemia (PaO2 80-120 mmHg), mild hyperoxemia (PaO2 121-199 mmHg), and moderate/severe hyperoxemia (PaO2 ≥ 200 mmHg). The primary endpoint was 90-day all-cause mortality. Secondary outcomes included the level of consciousness at discharge, assessed by the Glasgow Coma Scale (GCS), and 30-day all-cause mortality. Multivariate Cox regression and Restricted cubic spline (RCS) analysis were used to investigate the relationship between mean PaO2 and mortality, and to assess the nonlinear association between exposure and outcomes. RESULTS: This study included a total of 946 IS patients. The cumulative incidence of 30-day and 90-day all-cause mortality increased with decreasing PaO2 levels. RCS analysis revealed a nonlinear relationship between PaO2 and the risk of 30-day all-cause mortality (nonlinear P < 0.0001, overall P < 0.0001), as well as a nonlinear association between PaO2 and 90-day all-cause mortality (nonlinear P < 0.0001, overall P < 0.0001). The results remained consistent after excluding the small subset of patients who received reperfusion therapy. Sensitivity analysis indicated that the favorable impact on survival tends to increase with the extended duration of elevated PaO2. CONCLUSIONS: For IS patients who do not receive reperfusion therapy or whose recanalization status is unknown, a lower PaO2 early during ICU admission is considered an independent risk factor for short-term and recent mortality. Adjusting respiratory parameters to maintain supraphysiological levels of PaO2 appears to be beneficial for survival, although this finding requires further validation through additional studies. TRIAL REGISTRATION: Not applicable.

Establishment and validation of a predictive model for tracheotomy in critically ill patients and analysis of the impact of different tracheotomy timing on patient prognosis.

BACKGROUND: In critically ill patients receiving invasive mechanical ventilation (IMV), it is unable to determine early which patients require tracheotomy and whether early tracheotomy is beneficial. METHODS: Clinical data of patients who were first admitted to the ICU and underwent invasive ventilation for more than 24 h in the Medical Information Marketplace in Intensive Care (MIMIC)-IV database were retrospectively collected. Patients were categorized into successful extubation and tracheotomy groups according to whether they were subsequently successfully extubated or underwent tracheotomy. The patients were randomly divided into model training set and validation set in a ratio of 7:3. Constructing predictive models and evaluating and validating the models. The tracheotomized patients were divided into the early tracheotomy group (< = 7 days) and the late tracheotomy group (> 7 days), and the prognosis of the two groups was analyzed. RESULTS: A total of 7 key variables were screened: Glasgow coma scale (GCS) score, pneumonia, traumatic intracerebral hemorrhage, hemorrhagic stroke, left and right pupil responses to light, and parenteral nutrition. The area under the receiver operator characteristic (ROC) curve of the prediction model constructed through these seven variables was 0.897 (95% CI: 0.876-0.919), and 0.896 (95% CI: 0.866-0.926) for the training and validation sets, respectively. Patients in the early tracheotomy group had a shorter length of hospital stay, IMV duration, and sedation duration compared to the late tracheotomy group (p < 0.05), but there was no statistically significant difference in survival outcomes between the two groups. CONCLUSION: The prediction model constructed and validated based on the MIMIC-IV database can accurately predict the outcome of tracheotomy in critically ill patients. Meanwhile, early tracheotomy in critically ill patients does not improve survival outcomes but has potential advantages in shortening the duration of hospitalization, IMV, and sedation.

Association between estimated pulse wave velocity and in-hospital mortality of patients with acute kidney injury: a retrospective cohort analysis of the MIMIC-IV database.

BACKGROUND: Estimated pulse wave velocity (ePWV) has been found to be an independent predictor of cardiovascular mortality and kidney injury, which can be estimated noninvasively. This study aimed to investigate the association between ePWV and in-hospital mortality in critically ill patients with acute kidney injury (AKI). METHODS: This study included 5960 patients with AKI from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The low and high ePWV groups were compared using a Kaplan-Meier survival curve to evaluate the differences in survival status. Cox proportional hazards models were used to explore the association between ePWV and in-hospital mortality in critically ill patients with AKI. To further examine the dose-response relationship, we used a restricted cubic spline (RCS) model. Stratification analyses were conducted to investigate the effect of ePWV on hospital mortality across various subgroups. RESULTS: Survival analysis indicated that patients with high ePWV had a lower survival rate than those with low ePWV. Following adjustment, high ePWV demonstrated a statistically significant association with an increased risk of in-hospital mortality among AKI patients (HR = 1.53, 95% CI = 1.36-1.71, p < 0.001). Analysis using the RCS model confirmed a linear increase in the risk of hospital mortality as the ePWV values increased (P for nonlinearity = 0.602). CONCLUSIONS: A high ePWV was significantly associated with an increased risk of in-hospital mortality among patients with AKI. Furthermore, ePWV was an independent predictor of in-hospital mortality in critically ill patients with AKI.

Predictive model of acute kidney injury in critically ill patients with acute pancreatitis: a machine learning approach using the MIMIC-IV database.

BACKGROUND: Acute kidney injury (AKI) is a common and serious complication in severe acute pancreatitis (AP), associated with high mortality rate. Early detection of AKI is crucial for prompt intervention and better outcomes. This study aims to develop and validate predictive models using machine learning (ML) to identify the onset of AKI in patients with AP. METHODS: Patients with AP were extracted from the MIMIC-IV database. We performed feature selection using the random forest method. Model construction involved an ensemble of ML, including random forest (RF), support vector machine (SVM), k-nearest neighbors (KNN), naive Bayes (NB), neural network (NNET), generalized linear model (GLM), and gradient boosting machine (GBM). The best-performing model was fine-tuned and evaluated through split-set validation. RESULTS: We analyzed 1,235 critically ill patients with AP, of which 667 cases (54%) experienced AKI during hospitalization. We used 49 variables to construct models, including GBM, GLM, KNN, NB, NNET, RF, and SVM. The AUC for these models was 0.814 (95% CI, 0.763 to 0.865), 0.812 (95% CI, 0.769 to 0.854), 0.671 (95% CI, 0.622 to 0.719), 0.812 (95% CI, 0.780 to 0.864), 0.688 (95% CI, 0.624 to 0.752), 0.809 (95% CI, 0.766 to 0.851), and 0.810 (95% CI, 0.763 to 0.856) respectively. In the test set, the GBM's performance was consistent, with an area of 0.867 (95% CI, 0.831 to 0.903). CONCLUSIONS: The GBM model's precision is crucial, aiding clinicians in identifying high-risk patients and enabling timely interventions to reduce mortality rates in critical care.

Association between lactate/albumin ratio and prognosis in critically ill patients with acute kidney injury undergoing continuous renal replacement therapy.

BACKGROUND: The primary objective was to examine the association between the lactate/albumin ratio (LAR) and the prognosis of patients with acute kidney injury (AKI) undergoing continuous renal replacement therapy (CRRT). METHODS: Utilizing the Medical Information Mart for Intensive Care IV (MIMIC-IV, v2.0) database, we categorized 703 adult AKI patients undergoing CRRT into survival and non-survival groups based on 28-day mortality. Patients were further grouped by LAR tertiles: low (< 0.692), moderate (0.692-1.641), and high (> 1.641). Restricted cubic splines (RCS), Least Absolute Shrinkage and Selection Operator (LASSO) regression, inverse probability treatment weighting (IPTW), and Kaplan-Meier curves were employed. RESULTS: In our study, the patients had a mortality rate of 50.07% within 28 days and 62.87% within 360 days. RCS analysis revealed a non-linear correlation between LAR and the risk of mortality at both 28 and 360 days. Cox regression analysis, which was adjusted for nine variables identified by LASSO, confirmed that a high LAR (>1.641) served as an independent predictor of mortality at these specific time points (p < 0.05) in AKI patients who were receiving CRRT. These findings remained consistent even after IPTW adjustment, thereby ensuring a reliable and robust outcome. Kaplan-Meier survival curves exhibited a gradual decline in cumulative survival rates at both 28 and 360 days as the LAR values increased (log-rank test, χ2 = 48.630, p < 0.001; χ2 = 33.530, p < 0.001). CONCLUSION: A high LAR (>1.641) was found to be an autonomous predictor of mortality at both 28 and 360 days in critically ill patients with AKI undergoing CRRT.

Association between estimated pulse wave velocity and in-hospital and one-year mortality of patients with chronic kidney disease and atherosclerotic heart disease: a retrospective cohort analysis of the MIMIC-IV database.

BACKGROUND: Carotid-femoral pulse wave velocity has been identified as an autonomous predictor of cardiovascular mortality and kidney injury. This important clinical parameter can be non-invasively estimated using the calculated pulse wave velocity (ePWV). The objective of this study was to examine the correlation between ePWV and in-hospital as well as one-year mortality among critically ill patients with chronic kidney disease (CKD) and atherosclerotic heart disease (ASHD). METHODS: This study included a cohort of 1173 patients diagnosed with both CKD and ASHD, sourced from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The four groups divided into quartiles according to ePWV were compared using a Kaplan-Meier survival curve to assess variations in survival rates. Cox proportional hazards models were employed to analyze the correlation between ePWV and in-hospital as well as one-year mortality among critically ill patients with both CKD and ASHD. To further investigate the dose-response relationship, a restricted cubic splines (RCS) model was utilized. Additionally, stratification analyses were performed to examine the impact of ePWV on hospital and one-year mortality across different subgroups. RESULTS: The survival analysis results revealed a negative correlation between higher ePWV and survival rate. After adjusting for confounding factors, higher ePWV level (ePWV > 11.90 m/s) exhibited a statistically significant association with an increased risk of both in-hospital and one-year mortality among patients diagnosed with both CKD and ASHD (HR = 4.72, 95% CI = 3.01-7.39, p < 0.001; HR = 2.04, 95% CI = 1.31-3.19, p = 0.002). The analysis incorporating an RCS model confirmed a linear escalation in the risk of both in-hospital and one-year mortality with rising ePWV values (P for nonlinearity = 0.619; P for nonlinearity = 0.267). CONCLUSIONS: The ePWV may be a potential marker for the in-hospital and one-year mortality assessment of CKD with ASHD, and elevated ePWV was strongly correlated with an elevated mortality risk in patients diagnosed with both CKD and ASHD.

Independent effect of the triglyceride-glucose index on all-cause mortality in critically ill patients with chronic obstructive pulmonary disease and asthma: A retrospective cohort study.

BACKGROUND: The triglyceride-glucose (TyG) index serves as a reliable proxy for insulin resistance (IR). IR has been linked to heightened incidence, prevalence, or severity of chronic obstructive pulmonary disease (COPD) and asthma. Prior research indicates that critically ill patients are prone to developing IR. Nevertheless, few studies have delved into the correlation between IR and all-cause mortality in critically ill patients with COPD and asthma. Therefore, the aim of this study is to explore the association between the TyG index and all-cause mortality in patients with COPD and asthma, with the goal of assessing the impact of IR on the prognosis of this patient population. METHODS: This is a retrospective study, and all data are from the Medical Information Mart for Intensive Care IV (MIMIC-IV) critical care database. This study included 684 ICU patients with COPD and asthma and divided them into quartiles based on TyG index levels. The primary outcomes of this study were all-cause mortality during follow-up, encompassing mortality at 30 days, 90 days, and 180 days. The Kaplan-Meier analysis was used to compare all-cause mortality among the above four groups. Cox proportional hazards analyses were performed to examine the association between TyG index and all-cause mortality in critically ill patients with COPD and asthma. Restricted cubic spline analysis was used to assess potential nonlinear association between the TyG index and the primary outcome. RESULTS: A total of 684 patients (53.9% female) were included. The 90-days all-cause mortality rate and 180-days all-cause mortality were 11.7% and 12.3%, respectively. Kaplan-Meier analysis revealed a significant association between the TyG index and both 90-days all-cause mortality (log-rank p = .039) and 180-days all-cause mortality (log-rank p = .017). Cox proportional hazards analysis revealed a significant association between the TyG index and 90-days all-cause mortality in both the unadjusted model (HR, 1.30 [95% CI 1.08-1.57] p = .005) and the model adjusted for age, gender, and diabetes (HR, 1.38 [95% CI 1.15-1.67] p < .001). Similarly, the TyG index was associated with 180-days all-cause mortality in the unadjusted model (HR, 1.30 [95% CI 1.09-1.56] p = .004) and the model adjusted for age, sex, and diabetes (HR, 1.38 [95% CI 1.15-1.66] p < .001). The restricted cubic splines (RCS) regression model indicated a significant nonlinear association between the TyG index and both 90-days and 180-days all-cause mortality. Specifically, TyG index >4.8 was associated with an increased risk of mortality at both 90 days and 180 days. CONCLUSIONS: In summary, our results extend the utility of the TyG index to critically ill patients with COPD and asthma. Our study shows that the TyG index is a potential predictor of all-cause mortality in critically ill patients with COPD and asthma. In addition, in patients with a TyG index exceeding 4.8, there was a heightened risk of mortality. Measuring the TyG index may help with risk stratification and prognosis prediction in critically ill patients with COPD and asthma. Further prospective studies are needed to confirm our findings.

Construction and validation of prognostic models in critically Ill patients with sepsis-associated acute kidney injury: interpretable machine learning approach.

BACKGROUND: Acute kidney injury (AKI) is a common complication in critically ill patients with sepsis and is often associated with a poor prognosis. We aimed to construct and validate an interpretable prognostic prediction model for patients with sepsis-associated AKI (S-AKI) using machine learning (ML) methods. METHODS: Data on the training cohort were collected from the Medical Information Mart for Intensive Care IV database version 2.2 to build the model, and data of patients were extracted from Hangzhou First People's Hospital Affiliated to Zhejiang University School of Medicine for external validation of model. Predictors of mortality were identified using Recursive Feature Elimination (RFE). Then, random forest, extreme gradient boosting (XGBoost), multilayer perceptron classifier, support vector classifier, and logistic regression were used to establish a prognosis prediction model for 7, 14, and 28 days after intensive care unit (ICU) admission, respectively. Prediction performance was assessed using the receiver operating characteristic (ROC) curve and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) were used to interpret the ML models. RESULTS: In total, 2599 patients with S-AKI were included in the analysis. Forty variables were selected for the model development. According to the areas under the ROC curve (AUC) and DCA results for the training cohort, XGBoost model exhibited excellent performance with F1 Score of 0.847, 0.715, 0.765 and AUC (95% CI) of 0.91 (0.90, 0.92), 0.78 (0.76, 0.80), and 0.83 (0.81, 0.85) in 7 days, 14 days and 28 days group, respectively. It also demonstrated excellent discrimination in the external validation cohort. Its AUC (95% CI) was 0.81 (0.79, 0.83), 0.75 (0.73, 0.77), 0.79 (0.77, 0.81) in 7 days, 14 days and 28 days group, respectively. SHAP-based summary plot and force plot were used to interpret the XGBoost model globally and locally. CONCLUSIONS: ML is a reliable tool for predicting the prognosis of patients with S-AKI. SHAP methods were used to explain intrinsic information of the XGBoost model, which may prove clinically useful and help clinicians tailor precise management.