Intensive Care Unit Mortality Research Articles

Economic evaluation of critically ill adult CAR-T cell recipients—analysis from a healthcare payer perspective

Abstract Background CAR-T cell (chimeric antigen receptor T) therapy is now part of standard of care treatment of B‑cell lineage malignancies. Although it is an effective treatment, it comes along with adverse side effects and toxicities that may require intensive care therapy. The costs related to critical care therapy in critically ill patients after CAR‑T administration have not been evaluated. Patients and methods Retrospective analysis of all patients who had received CAR‑T therapy and were admitted to the intensive care unit (ICU) of a tertiary care university medical centre in Germany between 1 January 2019 and 31 December 2022. Cause of admission and ICU therapy as well as treatment and total hospitals costs were evaluated. Results Thirty patients with a history of CAR-T cell therapy for underlying haematological malignancy were included. The median age of all patients was 60 years (interquartile range [IQR] 50–70) and 37% (n = 11) were female. 93% (n = 28) of patients had non-Hodgkin lymphoma and 7% (n = 2) had multiple myeloma. The cohort was stratified whether the ICU admission was CAR‑T therapy related (i.e. within 30 days after CAR‑T therapy; 73%, n = 22) or the admission was of an other cause (> 30 days after CAR‑T therapy) (27%, n = 8). The median duration from CAR‑T therapy to ICU admission was 6 (range 5–8) days in CAR-T cell therapy associated ICU admissions compared with 52 (range 31–126) days in other admissions. The overall illness severity on admission was numerically higher in CAR-T-related ICU admission compared to other admissions (46 vs. 43 points, p = 0.18). Vasopressor therapy (50% vs. 75%; p = 0.19), invasive mechanical ventilation (27% vs. 50%; p = 0.24) and renal replacement therapy (14% vs. 50%; p < 0.05) were used in CAR-T-associated admission compared to other admissions, respectively. The ICU mortality (23% vs. 50%; p = 0.15) was higher in patients with other ICU admission. Median total costs of the entire inpatient stay in hospital were € 27,845 (range 8661–368,286 €) in CAR-T-associated ICU admissions compared to € 59,234 (range 23,182–127,044 €) in the group of other ICU admissions (costs of the CAR‑T product not included). Conclusion In relation to the total costs of CAR-T-cell therapy (production of the CAR‑T product), therapy-associated complications have a relatively low impact on the costs and utilization of ICU resources.

Maternal and Perinatal Outcomes of Induced Labor for Breech Presentations

Background: Inducing labor for breech presentation is a relatively harmless and effective approach when carefully chosen for the right cases regarding both maternal and perinatal outcomes. Objective: To determine safety of induction of labor in women with breech presentation with respect to maternal and perinatal outcomes. Material and Methods: This descriptive case series included Para 1-5 with single alive fetus, frank or complete breech with flexed head after 37 weeks. Labor induced with prostaglandin E2 vaginal pessary either alone or with balloon catheter combination. Maternal outcomes (postpartum hemorrhage, postoperative fever, perinatal trauma and hospital stay of > 24 hrs) and neonatal morbidity APGAR scores (A/S) of <7, Neonatal Intensive Care Unit admission, any trauma and mortality were recorded. Results: The study showed about 43 (86%) of women between age group of 18-35 years and 37(74%) with parity 1-3. Emergency cesarean section observed in 7 (14%) while vaginal delivery in 43(86%) cases. Maternal adverse outcomes observed were PPH in 4(8%) (P= 0.50), postoperative fever3 (6%) (P =0.68), perinatal trauma 2(4%) (P =0.68), and hospital stay 9(18%) cases (P =0.84).Observed neonatal outcomes were A/S of < 7 in 5(10%) (P =0.07), Neonatal Intensive Care Unit admission in4( 8%) (P =0.59), injury to humorous in 1(2 %)P =0.64)case. Conclusion: Induction of labor for term breech showed no statistically significant change in feto-maternal outcomes with regard to mode of delivery

RELIABILITY OF APACHE-II SCORING IN PREDICTING MORTALITY AT THE TIME OF ICU ADMISSION- A CLINICAL AUDIT

Different scoring systems are used to predict ICU patients' health outcomes and mortality rates. APACHE-II score has been found to have a discriminative value in predicting the mortality rate. There are some limitations to this score, such as the patients with multiple comorbid conditions and the physiological variables are all dynamic, which may alter the predicted mortality rate. Objective: The rationale behind doing this study is to see if there is any difference between the observed mortality and predicted mortality per the APACHE-II scoring system and how the APACHE-II scoring system will help predict the length of hospital stay and mortality in ICU patients. Methods: This was a cross-sectional study. The study was conducted in the Surgical Intensive Care Unit (SICU) of Doctors Hospital and Medical Centre, Lahore, from 1st January 2022 to 31st December 2022.: Data were analyzed using IBM SPSS Software version 29.0. Descriptive statistics, including mean and standard deviation, were computed. The results from the SPSS software output file were included along with the text. A one-sample t-test was applied to assess the significance of the data. A p-value of less than 0.05 was considered statistically significant, with a 95% confidence interval. Results: A total of 305 patients were included in our study. The mean age of patients who were admitted was 53.6 years. The mean length of ICU stay over the study period was calculated to be 4.3 days. Of 305 patients, 66 expired, and 239 were stepped down to in-patient units. All groups of APACHE-II score have less observed mortality than actual mortality; only group 3(APACHE 20-30) has higher observed mortality than predicted mortality. Conclusion: The APACHE-II score effectively stratifies patients at the time of admission to ICU as having a higher or lower risk of mortality depending on the disease severity so that the treatment triage, end-of-life ICU-care, and other important decisions can be made. This scoring doesn’t have a very high sensitivity and specificity, but with an increasing score, there is a higher mortality risk. We need to improve our ICU care for patients with APACHE-II scores between 20-30, as we have more observed mortality in that group. Hopefully, this study will help the practices in the ICU care region.

Predictive Potential of ECMO Blood Flow for Hemolysis and Outcome of Patients with Severe ARDS.

Background: Treatment with veno-venous extracorporeal membrane oxygenation (VV ECMO) has become a frequently considered rescue therapy in patients with severe acute respiratory distress syndrome (ARDS). Hemolysis is a common complication in patients treated with ECMO. Currently, it is unclear whether increased ECMO blood flow (Q̇EC) contributes to mortality and might be associated with increased hemolysis. Methods: A total of 441 patients with ARDS and VV ECMO, treated in a tertiary ARDS center, were included. The Q̇EC value for a significant increase in ICU mortality was determined by binary recursive partitioning. Linear regression analysis was performed to analyze a correlation between mean Q̇EC and mean plasma concentrations of cell-free hemoglobin (CFH). Results: A Q̇EC of 4 L/min divided the cohort into two groups with significantly different ICU mortality (Q̇EC ≤ 4 L/min: 39.3% (n = 300) versus Q̇EC > 4 L/min: 71.6% (n = 141), p < 0.001). Patients with Q̇EC > 4 L/min had a higher 28-day mortality. Furthermore, a higher mean Q̇EC was associated with increased CFH and decreased haptoglobin plasma concentrations. Conclusion: In patients with ARDS and VV ECMO, a mean Q̇EC > 4 L/min is associated with increased mortality, increased CFH and decreased haptoglobin plasma concentrations. Whether increased hemolysis determines the poorer outcome associated with higher Q̇EC should be the subject of future research.

Association between stress hyperglycemia ratio and mortality in patients with heart failure complicated by sepsis

Individuals afflicted with heart failure complicated by sepsis often experience a surge in blood glucose levels, a phenomenon known as stress hyperglycemia. However, the correlation between this condition and overall mortality remains unclear. 869 individuals with heart failure complicated by sepsis were identified from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database and categorized into five cohorts based on their stress hyperglycemia ratio (SHR). The primary endpoints evaluated were mortality within the intensive care unit (ICU), all-cause mortality within 28 days, and all-cause mortality during hospitalization. Cox proportional hazards regression and restricted cubic spline analyses were employed to unravel the association between SHR and mortality. The ICU mortality, in-hospital mortality, and 28-day all-cause mortality were 10.01%, 13.69%, and 16.46%, respectively. Multivariable Cox proportional hazards regression analysis revealed a significant association between elevated SHR and all-cause mortality. After adjusting for confounding variables, elevated SHR was significantly associated with increased risk of ICU mortality (hazard ratio [HR] = 1.67; 95% confidence interval [CI], 1.03–2.70)), in-hospital mortality (HR = 1.53; 95% CI, 1.00-2.33)), and 28-day all-cause mortality (HR = 1.49; 95% CI, 1.02–2.17)). Restricted cubic spline analysis demonstrated a significant U-shaped relationship between SHR and the risk of all-cause mortality. This study revealed that stress hyperglycemia ratio is an independent prognostic factor in patients with heart failure complicated by sepsis. Notably, both very high and very low SHR values were associated with increased mortality risk.

Predictors of Mortality in Acute Kidney Injury in Elderly Patients Admitted to Geriatrics Intensive Care Unit in Egypt: A Prospective Cross-Sectional Study

Background: Acute kidney injury (AKI) is a major problem in critically ill elderly patients in intensive care units (ICUs). It increases their morbidity, mortality, and length of ICU stay. Objective: This study aims to determine the prevalence of AKI, the factors associated with increased mortality of AKI patients, and the impact of AKI on ICU outcomes. Methods: This was a multistage cross-sectional study followed by a cohort study. Included 210 patients who were admitted to the geriatric ICU. We included the elderly aged 60 and more, and we excluded patients on regular dialysis, renal transplantation, and patients who died within the first 24 hours of admission. Results: 210 elderly patients were included in the study. The prevalence of AKI was 53.8%. Overall, in-hospital mortality was 46.2%. Mortality was higher in the AKI group compared to the non-AKI group (61.9% vs. 27.8%, p=0.001). 22.1% of patients needed renal replacement therapy. Mortality for this RRT group was 84%. In multivariable analysis, chronic kidney disease (CKD), KDIGO staging III, Acute Physiology and Chronic Health Evaluation (APACHE) II, septic shock, and diabetes (DM) were independent predictors of mortality. Conclusions: AKI is common in ICU patients. Most patients were having community-acquired AKI. Chronic kidney disease (CKD), serum creatinine at AKI diagnosis, KDIGO staging III, acute physiology and chronic health evaluation (APACHE) II, septic shock, and diabetes (DM) were independent predictors of mortality.

Association between body temperature and all-cause mortality in patients with sepsis: analysis of the MIMIC-IV database

BackgroundAbnormal body temperature (fever or hypothermia) is a critical symptom in sepsis and is strongly associated with clinical prognosis and disease progression. Given the duality and variability of body temperature fluctuations throughout the disease course, further research is essential to refine clinical strategies for temperature management in sepsis patients.MethodsWe extracted clinical data of sepsis patients from the MIMIC-IV database. A restricted cubic spline (RCS) curve was employed to describe the non-linear relationship between body temperature and clinical outcomes. Based on peak temperature within the first 24 h after admission, patients were categorized into three groups: < 36 °C, 36–38 °C, and > 38 °C. We subsequently matched patients one-to-one into three cohorts using a pairwise propensity score matching (PSM) approach. Alongside clinical data, we conducted log-rank and McNemar tests, and established multiple models, including multiple Cox regression, overlap-weighted (OW) adjusted Cox regression, multiple logistic regression, and OW-adjusted multiple logistic regression, to investigate the impact of temperature on clinical outcomes.ResultsA total of 35,499 sepsis patients were included in my study: 311 with a temperature below 36 °C, 27,538 with a temperature between 36 and 38 °C, and 7650 with a temperature above 38 °C. The RCS analysis revealed a non-linear, U-shaped relationship between body temperature and 28-day, ICU, and in-hospital mortality. Patients with hypothermia had significantly higher 28-day mortality (54.34% vs. 19.28%), ICU mortality (44.37% vs. 12.89%), and in-hospital mortality (49.20% vs. 17.46%) compared to those with hyperthermia. Among patients younger than 65 years, hyperthermia was a protective factor against 28-day mortality relative to normal body temperature, while the opposite was observed in patients aged 65 and older. This trend was consistent in the analysis of ICU and in-hospital mortality.ConclusionsAmong sepsis patients admitted to the ICU, a peak temperature below 36 °C within the first 24 h of admission was associated with higher 28-day mortality. However, no significant difference in clinical prognosis was observed between normothermic and hyperthermic patients.

Comparative Analysis of Bacteremic and Non-bacteremic Sepsis: A Retrospective Study.

Sepsis remains a prevalent critical illness encountered in emergency departments and intensive care units (ICU), with culture-negative sepsis constituting 30-60% of cases. The effect of culture type on treatment and outcomes remains unclear, and conflicting evidence exists regarding disparities between Gram-positive and Gram-negative infections. To further describe and compare characteristics and outcomes of culture-positive versus culture-negative sepsis. This retrospective cohort study included 1375 patients admitted to the ICU of a single tertiary care hospital between 2016 and 2019 with a diagnosis of sepsis or septic shock. Patients who did not meet the screening criteria, lacked drawn or documented cultures, or had documented non-bacterial infections, were excluded. The primary outcome was disease severity and secondary outcomes included in-hospital mortality and duration of hospital and ICU stay. The principal and secondary exposure variables were blood culture status (positive vs. negative) and Gram staining (positive vs. negative), respectively. Overall, 943 patients (68.5%) were culture-negative and 432 (31.5%) were culture-positive. Gram-positive bacteria were isolated from 178 patients, Gram-negative bacteria from 199 patients, and both from 55 patients. Culture-positive patients demonstrated an almost two-fold higher likelihood of requiring vasopressors (adjusted odds ratio (OR): 1.98), a higher incidence of stress-dose steroid administration (adjusted OR, 1.68), and higher resuscitative fluid administration at six and 72 hours than culture-negative patients. No significant between-group differences emerged in the ICU or hospital length of stay, or mortality. No significant variations were observed when comparing Gram-positive and Gram-negative bacteremia. Although significant differences in illness severity existed between blood culture-negative and blood culture-positive patients with sepsis, patient-oriented secondary outcomes did not exhibit significant between-group differences. These results indicate that clinicians should not be reassured by the lack of proven bacteremia in patients with suspected sepsis, given similar outcomes.

Suspected heparin-induced thrombocytopaenia in pulmonary thromboendarterectomy: retrospective cohort.

Heparin is given for anticoagulation during and after pulmonary thromboendarterectomy for chronic thromboembolic pulmonary hypertension. Our objective was to add to the limited data available on the incidence, management and outcomes of suspected heparin-induced thrombocytopaenia after pulmonary thromboendarterectomy for chronic thromboembolic pulmonary hypertension. This retrospective single-centre study included consecutive patients with suspected heparin-induced thrombocytopaenia after pulmonary thromboendarterectomy done in 2005-2018. Confirmed heparin-induced thrombocytopaenia was defined as positive findings from both the antibody test and the platelet aggregation test. Patients with versus without confirmed heparin-induced thrombocytopaenia were compared, as well as patients with versus without heparin replacement therapy within the group with unconfirmed heparin-induced thrombocytopaenia. The platelet counts over time were compared to those in controls without suspected heparin-induced thrombocytopaenia. Heparin-induced thrombocytopaenia was suspected in 86 (6.3%) of 1360 patients and confirmed in 16 (16/86, 19%), all of whom received heparin replacement therapy and survived to intensive care unit discharge. Of the remaining 70 patients, 28 (40%) received heparin replacement therapy and less often experienced bleeding compared to the other 42 (3.6% vs 21.4%, P = 0.043). Intensive care unit mortality was 17/70 (24.3%) and was lower in the subgroup given heparin replacement therapy (10.7% vs 33.3%; P = 0.046). Confirmed heparin-induced thrombocytopaenia was associated with a sharp platelet-count drop on Day 5. In unconfirmed suspected heparin-induced thrombocytopaenia, the early platelet-count decline was similar to that in the controls without suspected heparin-induced thrombocytopaenia, but the baseline count was lower. Clinical features suggesting heparin-induced thrombocytopaenia after pulmonary thromboendarterectomy for chronic thromboembolic pulmonary hypertension were associated with excess mortality. Relay heparin replacement therapy was associated with lower mortality and fewer bleeding events.

Association between phosphorus-to-calcium ratio at ICU admission and all-cause mortality in acute pancreatitis: Insights from the MIMIC-IV database.

Serum phosphorus and serum calcium are important electrolytes in the body. The relationship between them and acute pancreatitis (AP) has been previously discussed. However, the results seem to lack credibility due to the neglect of mutual influence between them. Thus, a comprehensive indicator is needed. In this study, AP patients with intensive care unit (ICU) treatment were extracted from Medical Information Mart for Intensive Care (MIMIC) database. The outcomes included in-hospital mortality and ICU mortality. Kaplan-Meier survival analysis, Cox proportional hazard regression model and restricted cubic spline were employed to investigate the association between the phosphorus-to-calcium ratio (PCR) index and clinical outcomes. A total of 719 AP patients (57.2% male) were enrolled. The in-hospital and ICU mortality were 11.4% and 7.5%, respectively. After adjusting for confounders, Cox proportional hazard analysis indicated patients with a higher PCR index had a significant association with in-hospital mortality (adjusted hazard ratio, 2.88; 95% confidence interval, 1.34-6.19; p = .007). Restricted cubic splines revealed that a progressively increasing risk of all-cause mortality was associated with an elevated PCR index. The PCR index has a strong correlation with in-hospital and ICU all-cause mortality in AP, which provides a reference for clinical decision-making.

Impact of Haemoadsorption Therapy on Short Term Mortality and Vasopressor Dependency in Severe Septic Shock with Acute Kidney Injury: A Retrospective Cohort Study.

Background/Objectives: Sepsis, a life-threatening organ dysfunction caused by a dysregulated host response to infection, remains a major challenge in ICUs. This study evaluated whether combining haemoadsorption therapy with continuous renal replacement therapy (CRRT) reduces ICU and short-term mortality in patients with severe septic shock and acute kidney injury (AKI) requiring CRRT. Methods: A single-centre retrospective cohort study was conducted at Rambam Health Care Campus, Haifa, Israel, from January 2018 to February 2024. Data were collected from ICU patients with severe septic shock and AKI requiring CRRT. Patients were divided into two groups: those receiving haemoadsorption therapy with CRRT and those receiving CRRT alone. Primary and secondary endpoints included ICU, 30 and 60-day mortality, vasopressor dependency index (VDI), and lactate levels. Results: Out of 545 patients with septic shock, 133 developed AKI requiring CRRT, and 76 met the inclusion criteria. The haemoadsorption group (n = 47) showed significant reductions in blood lactate levels and VDI after 24 h compared to the CRRT alone group (n = 29). ICU mortality was significantly lower in the haemoadsorption group (34.0% vs. 65.5%, p = 0.008), as was 30 and 60-day mortality (34.0% vs. 62.1%, p = 0.02, and 48.9% vs. 75.9%, p = 0.002). Multivariate analysis confirmed haemoadsorption therapy as independently associated with lower ICU and 30-day but not 60-day mortality. Conclusions: Haemoadsorption therapy combined with CRRT in patients with severe septic shock and AKI requiring CRRT is associated with improved lactate clearance, reduced vasopressor requirements, and lower ICU and 30-day mortality. Further high-quality randomized controlled trials are needed to confirm these findings.

Predictive accuracy of lactate albumin ratio for mortality in intensive care units: a nationwide cohort study

BackgroundThe lactate albumin ratio (LAR), a simple prognostic marker used in intensive care units (ICUs), combines lactate and serum albumin levels to predict patient outcomes. Despite its potential, the predictive...

Early Mortality Prediction in Intensive Care Unit Patients Based on Serum Metabolomic Fingerprint.

Predicting mortality in intensive care units (ICUs) is essential for timely interventions and efficient resource use, especially during pandemics like COVID-19, where high mortality persisted even after the state of emergency ended. Current mortality prediction methods remain limited, especially for critically ill ICU patients, due to their dynamic metabolic changes and heterogeneous pathophysiological processes. This study evaluated how the serum metabolomic fingerprint, acquired through Fourier-Transform Infrared (FTIR) spectroscopy, could support mortality prediction models in COVID-19 ICU patients. A preliminary univariate analysis of serum FTIR spectra revealed significant spectral differences between 21 discharged and 23 deceased patients; however, the most significant spectral bands did not yield high-performing predictive models. By applying a Fast-Correlation-Based Filter (FCBF) for feature selection of the spectra, a set of spectral bands spanning a broader range of molecular functional groups was identified, which enabled Naïve Bayes models with AUCs of 0.79, 0.97, and 0.98 for the first 48 h of ICU admission, seven days prior, and the day of the outcome, respectively, which are, in turn, defined as either death or discharge from the ICU. These findings suggest FTIR spectroscopy as a rapid, economical, and minimally invasive diagnostic tool, but further validation is needed in larger, more diverse cohorts.

Use of vancomycin and acute kidney injury in critically ill patients with sepsis or septic shock: A retrospective observational cohort study.

The independent association of vancomycin with Acute Kidney Injury (AKI) in the critically ill patient with sepsis or septic Shock is controversial. The aim of this study was to evaluate the incidence of AKI in a cohort of patients with sepsis or septic Shock with an adequate and strict monitoring of vancomycin, guided by the area under the concentration-time curve in relation to the minimum inhibitory concentration (AUC/MIC ratio). Retrospective cohort study on 106 patients admitted to the ICU with a diagnosis of sepsis or septic shock with vancomycin treatment, consecutively from January 2017 to December 2019. AKI was defined according to Kidney Disease Improving Global Outcomes criteria. Risk factors associated with AKI were determined by multivariable logistic regression analysis. In our cohort, 28 patients out of 106 (26%) developed AKI. ICU and 30-day mortality were 18% and 22%, respectively. After multivariable logistic regression adjusted analysis, chronic liver disease was associated with AKI. In our retrospective cohort study on critical patients with sepsis and septic shock, treated with vancomycin adjusting the dose guided by a pharmacokinetic/pharmacodynamic monitoring to achieve the target AUC0-24/CMI ratio, the incidence of AKI was 26%.

Early prediction of mortality upon intensive care unit admission

BackgroundWe aimed to develop and validate models for predicting intensive care unit (ICU) mortality of critically ill adult patients as early as upon ICU admission.MethodsCombined data of 79,657 admissions from two teaching hospitals’ ICU databases were used to train and validate the machine learning models to predict ICU mortality upon ICU admission and at 24 h after ICU admission by using logistic regression, gradient boosted trees (GBT), and deep learning algorithms.ResultsIn the testing dataset for the admission models, the ICU mortality rate was 7%, and 38.4% of patients were discharged alive or dead within 1 day of ICU admission. The area under the receiver operating characteristic curve (0.856, 95% CI 0.845–0.867) and area under the precision-recall curve (0.331, 95% CI 0.323–0.339) were the highest for the admission GBT model. The ICU mortality rate was 17.4% in the 24-hour testing dataset, and the performance was the highest for the 24-hour GBT model.ConclusionThe ADM models can provide crucial information on ICU mortality as early as upon ICU admission. 24 H models can be used to improve the prediction of ICU mortality for patients discharged more than 1 day after ICU admission.

A worldwide assessment of the mechanical ventilation in patients with acute exacerbations of chronic obstructive pulmonary disease. Analysis of the VENTILAGROUP over time. A retrospective, multicenter study

BackgroundThe trend over time and across different geographical areas of outcomes and management with noninvasive ventilation or invasive mechanical ventilation in patients admitted for acute exacerbations of chronic obstructive pulmonary disease and treated with ventilatory support is unknown. The purpose of this study was to describe outcomes and identify variables associated with survival for patients admitted to an intensive care unit (ICU) with acute exacerbation of chronic obstructive pulmonary disease [aeCOPD] who received noninvasive or invasive mechanical ventilation worldwide.MethodsRetrospective, multi-national, and multicenter studies, including four observational cohort studies, were carried out in 1998, 2004, 2010, and 2016 for the VENTILAGROUP following the same methodology.ResultsA total of 1,848 patients from 1,253 ICUs in 38 countries admitted for aeCOPD and need of ventilatory support were identified in the four study cohorts and included in the study. The overall incidence of aeCOPD as a cause for ventilatory support at ICU admission significantly decreased over time and varied widely according to the gross national income. The mortality of patients admitted to ICU for aeCOPD and ventilatory support significantly decreased over time regardless of the geographical area and gross national income; however, there is a remarkable variability in ICU mortality according to geographical area and gross national income. The use of NPPV as the first attempt at ventilatory support has significantly increased over time, with a parallel reduction of invasive mechanical ventilation regardless of gross national income.ConclusionIn this worldwide observational study, including four sequential cohorts of patients over 18 years from 1998 to 2016, the mortality of patients admitted to ICU for aeCOPD and ventilatory support significantly decreased regardless of the geographical area and gross national income. Future research will need to investigate the reason for the remarkable variability in ICU mortality according to the geographical area, gross national income, and methods to select patients for the appropriate ventilatory support.

Exploring the prognostic impact of triglyceride-glucose index in critically ill patients with first-ever stroke: insights from traditional methods and machine learning-based mortality prediction

BackgroundThe incidence and mortality of first-ever strokes have risen sharply, especially in the intensive care unit (ICU). Emerging surrogate for insulin resistance, triglyceride-glucose index (TyG), has been linked to stroke prognosis. We aims to explore the relationships between TyG with ICU all-cause mortality and other prognosis, and to develop machine learning (ML) models in predicting ICU all-cause mortality in the first-ever strokes.MethodsWe included first-ever stroke patients from the eICU Collaborative Research Database in 2014–2015 as the primary analysis cohort (then divided into training and internal validation cohorts) and from local hospital’s ICUs as the external validation cohort. Multivariate Cox proportional hazards models and restricted cubic spline analyses were used to evaluate the association between TyG and ICU/hospital all-cause mortality. Linear regression and correlation analyses were performed to examine the relationships between TyG with length of ICU/hospital stay and Glasgow Coma Score.ResultsThe primary analysis cohort included 3173 first-ever strokes (median age 68.0 [55.0–68.0] years; 63.0% male), while the external validation cohort included 201 first-ever strokes (median age 71.0 [63.0–77.0] years; 62.3% male). Multivariate Cox proportional hazards models revealed that the high TyG group (TyG ≥ 9.265) was associated with higher ICU (HR 1.92, 95% CI 1.38–2.66) and hospital (HR 1.69, 95% CI 1.32–2.16) all-cause mortality, compared with low TyG group (TyG < 9.265). TyG was also correlated with ICU length of stay (r = 0.077), hospital length of stay (r = 0.042), and Glasgow Coma Score (r = -0.132). TyG and other six features were used to construct ML models. The random forest model performed best in internal validation with AUC (0.900) and G-mean (0.443), and in external validation with AUC (0.776) and G-mean (0.399).ConclusionTyG (optimal cut-off: 9.265) was identified as an independent risk factor for ICU and hospital all-cause mortality in first-ever strokes. The ML model incorporating TyG demonstrated strong predictive performance. This emphasises the importance of insulin resistance (with TyG as a surrogate measure) in the prognostic assessment and early risk stratification of first-time stroke patients.

Evaluation of Response to Weight-based Dosing Strategies of Continuous, Fixed-Rate Atracurium Infusions in Critically Ill, Obese Adults With Acute Respiratory Distress Syndrome.

Fixed-rate infusions of weight-based neuromuscular blocking agents (NMBAs) were adopted during the COVID pandemic to limit caregiver exposure during titrations. Although fixed-rate infusions are supported in studies of acute respiratory distress syndrome (ARDS), the optimal scalar for weight-based NMBAs in patients with obesity remains controversial. This study sought to compare change in oxygenation using two weight-based dosing strategies for atracurium in obese patients with ARDS. Secondary outcomes included total atracurium dose, mortality, and intensive care unit (ICU) and ventilator-free days. Following an institutional practice update to use ideal body weight (IBW) for patients with obesity, we retrospectively compared adults (≥18 years) with ARDS and a body mass index (BMI) ≥ 30 kg/m2 who received atracurium (15 µg/kg/min) based on actual body weight (ABW) with those using IBW. The primary outcome was change in PaO2/FiO2 ratio (P/F) 48 hours after atracurium initiation. Analysis-of-covariance compared change in P/F between groups after adjustment for confounders. The IBW group (n = 123), compared with the ABW group (n = 133), had lower baseline P/F (85.0 [71.0, 118.3] vs 93.3 [76.0, 128.3], P = 0.025) and sequential organ failure assessment (SOFA) score (9.7 ± 2.6 vs 10.5 ± 2.6, P = 0.015), with greater use of steroids (96% vs 89%, P = 0.032) and prone positioning (72% vs 58%, P = 0.015). No difference was detected in change in P/F at 48 hours (adjusted least squares mean [95% confidence interval, CI]: 55.8 [37.0, 74.5] vs 56.9 [39.6, 74.1], P = 0.90). Atracurium doses were higher in the ABW group (97.4 mg/h [84.4, 110.3] vs 55.4 [47.2, 65.7], P < 0.001). There was no difference in hospital mortality, ICU mortality, and ICU-free days or ventilator-free days. In patients with obesity with ARDS receiving fixed-rate atracurium infusions, the change in P/F at 48 hours did not differ based on weight. Atracurium dosed on IBW may use less total drug without compromising ability to improve oxygenation. This is the first study comparing the dosing weight used for continuous infusion atracurium in hospitalized, critically ill ARDS patients with obesity. Additional studies are warranted to optimize dosing in obese patients.

Efficacy of haemoadsorption combined with continuous renal replacement therapy in patients with rhabdomyolysis and acute kidney injury: a retrospective study

ABSTRACT Background Clearance of circulating myoglobin is crucial to prevent further damage in patients with rhabdomyolysis (RM) and acute kidney injury (AKI). The objective of the present study was to evaluate the efficacy and safety of haemoadsorption (HA) combined with continuous renal replacement therapy (CRRT) in critically ill patients with RM and AKI. Methods Patients with RM and AKI who received CRRT + HA or CRRT with concomitant creatine kinase (CK) &amp;gt;10 000 IU/l in our intensive care unit (ICU) between May 2021 and December 2023 were retrospectively included. The primary outcome was 90-day mortality; secondary outcomes were kidney function recovery and CK decline rate. Adverse events were also evaluated, including hypotension, circuit clotting, albumin leakage and blood loss. Propensity score matching and Cox retrospective analysis were performed. Results A total of 111 RM patients with AKI were ultimately included. The ICU and in-hospital mortality were significantly lower in the CRRT + HA group compared with the CRRT group (ICU mortality: 18% versus 42%, P = .025; in-hospital mortality: 21% versus 42%, P = .048). However, the CRRT + HA group only showed a non-significant reduction in 90-day mortality compared with the CRRT group (47% versus 68%, P = .063). After treatment for 90 days, the number of patients with kidney function recovery was not significantly different between the CRRT + HA and CRRT groups (95% versus 84%, P = .639). Moreover, the incidence of hypotension and circuit clotting events did not increase during CRRT + HA treatment. In addition, the CRRT + HA group also appeared to have a higher rate of CK reduction and reduction of CK than the CRRT group at 24 and 48 hours after the initiation of CRRT. A multivariate Cox regression model demonstrated that CRRT + HA {hazard ratio [HR] 0.477 [95% confidence interval (CI) 0.234–0.972], P = .042}, mean arterial blood pressure [per 1 mmHg; HR 0.967 (95% CI 0.943–0.992), P = .009] and CRRT treatment duration [per 1 h; HR 0.995 (95% CI 0.992–0.998), P = .002] played a favourably important role in the survival prognosis of RM and AKI patients. In contrast, serum phosphate before RRT [per 1 mmol/l; HR 1.531 (95% CI 1.113–2.106), P = .009] and McMahon score [per 1 score; HR 1.15 (95% CI 1.006–1.313), P = .04] were independent risk factors for 90-day mortality. Conclusions CRRT combined with HA therapy reduced ICU and in-hospital mortality in patients with RM and AKI and also had a cleansing effect on creatine kinase without significant adverse events.

Machine Learning-Based Prediction Model for ICU Mortality After Continuous Renal Replacement Therapy Initiation in Children.

Continuous renal replacement therapy (CRRT) is the favored renal replacement therapy in critically ill patients. Predicting clinical outcomes for CRRT patients is difficult due to population heterogeneity, varying clinical practices, and limited sample sizes. We aimed to predict survival to ICUs and hospital discharge in children and young adults receiving CRRT using machine learning (ML) techniques. Patients less than 25 years of age receiving CRRT for acute kidney injury and/or volume overload from 2015 to 2021 (80%). Internal validation occurred in a testing group of patients from the dataset (20%). Retrospective international multicenter study utilizing an 80/20 training and testing cohort split, and logistic regression with L2 regularization (LR), decision tree, random forest (RF), gradient boosting machine, and support vector machine with linear kernel to predict ICU and hospital survival. Model performance was determined by the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) due to the imbalance in the dataset. Of the 933 patients included in this study, 538 (54%) were male with a median age of 8.97 years and interquartile range (1.81-15.0 yr). The ICU mortality was 35% and hospital mortality was 37%. The RF had the best performance for predicting ICU mortality (AUROC, 0.791 and AUPRC, 0.878) and LR for hospital mortality (AUROC, 0.777 and AUPRC, 0.859). The top two predictors of ICU survival were Pediatric Logistic Organ Dysfunction-2 score at CRRT initiation and admission diagnosis of respiratory failure. These are the first ML models to predict survival at ICU and hospital discharge in children and young adults receiving CRRT. RF outperformed other models for predicting ICU mortality. Future studies should expand the input variables, conduct a more sophisticated feature selection, and use deep learning algorithms to generate more precise models.