OBJECTIVE: Healthcare professionals may be able to anticipate more accurately a patient\'s timing of death and assess their possibility of recovery by implementing a real-time clinical decision support system. Using such a tool, the healthcare system can better understand a patient\'s condition and make more informed judgements about distributing limited resources. This scoping review aimed to analyze various death prediction AI (Artificial Intelligence) algorithms that have been used in ICU (Intensive Care Unit) patient populations.
METHODS: The search strategy of this study involved keyword combinations of outcome and patient setting such as mortality, survival, ICU, terminal care. These terms were used to perform database searches in MEDLINE, Embase, and PubMed up to July 2022. The variables, characteristics, and performance of the identified predictive models were summarized. The accuracy of the models was compared using their Area Under the Curve (AUC) values.
RESULTS: Databases search yielded an initial pool of 8271 articles. A two-step screening process was then applied: first, titles and abstracts were reviewed for relevance, reducing the pool to 429 articles. Next, a full-text review was conducted, further narrowing down the selection to 400 key studies. Out of 400 studies on different tools or models for prediction of mortality in ICUs, 16 papers focused on AI-based models which were ultimately included in this study that have deployed different AI-based and machine learning models to make a prediction about negative patient outcome. The accuracy and performance of the different models varied depending on the patient populations and medical conditions. It was found that AI models compared with traditional tools like SAP3 or APACHE IV score were more accurate in death prediction, with some models achieving an AUC of up to 92.9%. The overall mortality rate ranged from 5% to more than 60% in different studies.
CONCLUSIONS: We found that AI-based models exhibit varying performance across different patient populations. To enhance the accuracy of mortality prediction, we recommend customizing models for specific patient groups and medical contexts. By doing so, healthcare professionals may more effectively assess mortality risk and tailor treatments accordingly. Additionally, incorporating additional variables-such as genetic information-into new models can further improve their accuracy.

BACKGROUND: Sepsis is a dysregulated host immune response stemming from a systemic inflammatory response to microbial invasion, encompassing bacteria, viruses, and other pathogens. The vascular endothelial growth factor (VEGF) was initially identified for its potent induction of endothelial permeability. Studies have proposed a therapeutic role of dopamine in mitigating VEGF-induced permeability, shedding light on its potential in acute respiratory distress syndrome (ARDS) management.
OBJECTIVE: To determine the effect of dopamine as an inhibitor of VEGF and to prevent the progression of sepsis to acute lung injury (ALI) and ARDS.
METHODS: A total of 154 critical care unit patients with a diagnosis of sepsis were randomized into two groups: Group I (control group) and Group II (Study group). Both received standard treatment, as per ICU protocol. In addition, the study group (Group II) received a dopamine infusion of 2 micrograms/kg/min. Baseline routine investigation, procalcitonin, and chest X-ray were done. Day one and day seven blood samples were stored for analysis of VEGF levels. Murray\'s score and sequential organ failure assessment (SOFA) score (organ dysfunction) were calculated from day one to day seven.
RESULTS: VEGF levels on day seven were significantly lower in the study group compared to the control group (p<0.05). The PaO2/FiO2 ratio at day seven was significantly increased in the study group than in the control group, indicating an improvement in oxygenation status in the study group. There was a mean ICU stay of 9.3 days in the study group versus 11.6 days in the control group (p<0.05). The SOFA score showed a significant improvement in the study group from day five onwards, indicating a therapeutic effect of dopamine on organ dysfunction in sepsis.
CONCLUSIONS: Dopamine reduces VEGF and lung injury mediated by increased endothelial permeability.

BACKGROUND: This study aimed to evaluate the predictive accuracy of the prehospital rapid emergency medicine score (pREMS) for predicting the outcomes of hospitalized patients with traumatic brain injury (TBI) who died, were discharged, were admitted to the intensive care unit (ICU), or were admitted to the operating room (OR) within 72 h.
METHODS: A retrospective cohort analysis was performed on a sample of 513 TBI patients admitted to the emergency department (ED) of Besat Hospital in 2023. Only patients of both sexes aged 18 years or older who were not pregnant and had adequate documentation of vital signs were included in the analysis. Patients who died during transport and patients who were transferred from other hospitals were excluded. The predictive power of the pREMS for each outcome was assessed by calculating the sensitivity and specificity curves and by analyzing the area under the receiver operating characteristic curve (AUROC).
RESULTS: The mean pREMS scores for hospital discharge, death, ICU admission and OR admission were 11.97 ± 3.84, 6.32 ± 3.15, 8.24 ± 5.17 and 9.88 ± 2.02, respectively. pREMS accurately predicted hospital discharge and death (AOR = 1.62, P < 0.001) but was not a good predictor of ICU or OR admission (AOR = 1.085, P = 0.603). The AUROCs for the ability of the pREMS to predict outcomes in hospitalized TBI patients were 0.618 (optimal cutoff point = 7) for ICU admission and OR and 0.877 (optimal cutoff point = 9.5) for hospital discharge and death at 72 h.
CONCLUSIONS: The results indicate that the pREMS, a new preclinical trauma score for traumatic brain injury, is a useful tool for prehospital risk stratification (RST) in TBI patients. The pREMS showed good discriminatory power for predicting in-hospital mortality within 72 h in patients with traumatic brain injury.

Background/Objectives: The aim of this study was to investigate the feasibility and safety of neuromuscular electrical stimulation (NMES) in patients on extracorporeal membrane oxygenation (ECMO) and thoroughly assess any potential adverse events. Methods: We conducted a prospective observational study assessing safety and feasibility, including 16 ICU patients on ECMO support who were admitted to the cardiac surgery ICU from January 2022 to December 2023. The majority of patients were females (63%) on veno-arterial (VA)-ECMO (81%), while the main cause was cardiogenic shock (81%) compared to respiratory failure. Patients underwent a 45 min NMES session while on ECMO support that included a warm-up phase of 5 min, a main phase of 35 min, and a recovery phase of 5 min. NMES was implemented on vastus lateralis, vastus medialis, gastrocnemius, and peroneus longus muscles of both lower extremities. Two stimulators delivered biphasic, symmetric impulses of 75 Hz, with a 400 μsec pulse duration, 5 sec on (1.6 sec ramp up and 0.8 sec ramp down) and 21 sec off. The intensity levels aimed to cause visible contractions and be well tolerated. Primary outcomes of this study were feasibility and safety, evaluated by whether NMES sessions were successfully achieved, and by any adverse events and complications. Secondary outcomes included indices of rhabdomyolysis from biochemical blood tests 24 h after the application of NMES. Results: All patients successfully completed their NMES session, with no adverse events or complications. The majority of patients achieved type 4 and 5 qualities of muscle contraction. Conclusions: NMES is a safe and feasible exercise methodology for patients supported with ECMO.

UNASSIGNED: Percutaneous dilatational tracheostomy (PDT), a bedside procedure in intensive care, enhances respiratory support for critically ill patients with benefits over traditional tracheostomy, such as improved safety, ease of use, cost-effectiveness, and operational efficiency by eliminating patient transfers to the operating room. It also minimizes complications including bleeding, infection, and inflammation. Despite decades of PDT evolution and device diversification, adaptations primarily cater to larger Western patients rather than smaller-statured Korean populations. This study assesses the efficacy and appropriateness of the Ciaglia Blue Rhino (Cook Critical Care, Bloomington, IN, USA), augmented with ultrasound, flexible bronchoscopy, and microcatheter techniques, for Korean patients with short stature.
UNASSIGNED: We conducted PDT on 183 intubated adults (128 male/55 female) with severe respiratory issues at a single medical center from January 2010 to December 2022. Patients were divided into two groups for retrospective analysis: a modified group (n=133) underwent PDT with ultrasound-guided flexible bronchoscopy and microcatheter puncture, and a conventional group (n=50) received PDT using only the Ciaglia Blue Rhino device. We assessed clinical and demographic characteristics, outcomes, and complications such as pneumothorax and emphysema. The study also evaluated the suitability and effectiveness of the devices for Korean patients with short stature.
UNASSIGNED: Demographic characteristics including sex, body weight, height, body mass index, obesity status, and underlying diseases showed no significant differences between the two groups. However, the modified group was older (69.5±14.2 vs. 63.5±14.1 years; P=0.01). The sequential organ failure assessment (SOFA) and simplified acute physiology score (SAPS) II score was slightly higher in the modified groups, but no statistically significant differences were observed (7.1±2.3 vs. 6.7±2.3, P=0.31 and 46.7±9.0 vs. 44.0±9.1, P=0.08, respectively). The duration of hospital and ICU stays, as well as days post-PDT, were longer in the conventional group, yet these differences were not statistically significant (P=0.20, P=0.44, P=0.06). Total surgical time, including preparation, ultrasound, bronchoscopy, and microcatheter puncture, was significantly longer in the modified group (25.6±7.5 vs. 19.9±6.5 minutes; P<0.001), and the success rate of the first tracheal puncture was also higher (100.0% vs. 92.0%; P=0.006). Intra-operative bleeding was less frequent in the modified group (P=0.02 for tracheostomy site bleeding and P=0.002 for minor bleeding).
UNASSIGNED: PDT, performed at the bedside in intensive care settings, proves to be a swift and dependable method. Utilizing the Ciaglia Blue Rhino device, combined with ultrasound guidance, flexible bronchoscopy, and 4.0-Fr microcatheter puncture, PDT is especially effective for intubated patients who cannot be weaned from ventilation. This technique results in fewer complications than traditional tracheostomy and is particularly beneficial for patients with respiratory issues and smaller-statured Koreans, potentially reducing morbidity and mortality.

Endocrine disorders pose significant challenges in the management of critically ill patients, contributing to morbidity and mortality in intensive care settings. Timely detection of these disorders is essential to optimizing patient outcomes. Biomarkers, as measurable indicators of biological processes or disease states, play a crucial role in the early identification and monitoring of endocrine dysfunction. This comprehensive review examines the role of biomarkers in the early detection of endocrine disorders in critical illnesses. We provide an overview of common endocrine disorders encountered in the intensive care unit (ICU) and discuss the impact of endocrine dysregulation on patient outcomes. Additionally, we classify biomarkers and explore their significance in diagnosing and monitoring endocrine disorders, including thyroid dysfunction, adrenal insufficiency, and hypopituitarism. Furthermore, we discuss the clinical applications of biomarkers, including their utility in guiding therapeutic interventions, monitoring disease progression, and predicting outcomes in critical illnesses. Emerging trends and future directions in biomarker research are also highlighted, emphasizing the need for continued investigation into novel biomarkers and technological advancements. Finally, we underscore the potential of biomarkers to revolutionize the early detection and management of endocrine disorders in critical illnesses, ultimately improving patient care and outcomes in the ICU.

The rapid advancements in Artificial Intelligence of Things (AIoT) are pivotal for the healthcare sector, especially as the world approaches an aging society which will be reached by 2050. This paper presents an innovative AIoT-enabled data fusion system implemented at the CMUH Respiratory Intensive Care Unit (RICU) to address the high incidence of medical errors in ICUs, which are among the top three causes of mortality in healthcare facilities. ICU patients are particularly vulnerable to medical errors due to the complexity of their conditions and the critical nature of their care. We introduce a four-layer AIoT architecture designed to manage and deliver both real-time and non-real-time medical data within the CMUH-RICU. Our system demonstrates the capability to handle 22 TB of medical data annually with an average delay of 1.72 ms and a bandwidth of 65.66 Mbps. Additionally, we ensure the uninterrupted operation of the CMUH-RICU with a three-node streaming cluster (called Kafka), provided a failed node is repaired within 9 h, assuming a one-year node lifespan. A case study is presented where the AI application of acute respiratory distress syndrome (ARDS), leveraging our AIoT data fusion approach, significantly improved the medical diagnosis rate from 52.2% to 93.3% and reduced mortality from 56.5% to 39.5%. The results underscore the potential of AIoT in enhancing patient outcomes and operational efficiency in the ICU setting.

Differentiation between acute liver failure (ALF) and acute-on-chronic liver failure (ACLF) can be challenging in patients with de novo liver disease but is important to indicate the referral to a transplant center and urgency of organ allocation. Leptin, an adipocyte-derived cytokine that regulates energy storage and satiety, has multiple regulatory functions in the liver. We enrolled 160 critically ill patients with liver disease and 20 healthy individuals to measure serum leptin concentrations as a potential biomarker for diagnostic and prognostic purposes. Notably, patients with ALF had higher concentrations of serum leptin compared to patients with decompensated advanced chronic liver disease (dACLD) or ACLF (110 vs. 50 vs. 29 pg/mL, p < 0.001). Levels of serum leptin below 56 pg/mL excluded ALF in patients with acute hepatic disease, with a negative predictive value (NPV) of 98.8% in our cohort. Lastly, serum leptin did not show any dynamic changes within the first 48 h of ICU treatment, especially not in comparison with patients with ALF vs. ACLF or survivors vs. non-survivors. In conclusion, serum leptin may represent a helpful biomarker to exclude ALF in critically ill patients who present with acute liver dysfunction.

UNASSIGNED: Hypophosphatemia has been reported to impair diaphragmatic function in patients with chronic obstructive pulmonary disease (COPD). However, little is known about the role of dysphosphatemia at admission [plasmatic phosphate concentration at intensive care unit (ICU) admission (T0-Ph)] to the ICU and respiratory outcomes among patients with severe acute COPD exacerbation. We aimed to assess the value of T0-Ph as a predictive factor of invasive mechanical ventilation (MV) during ICU stay.
UNASSIGNED: We retrospectively included consecutive patients admitted to the ICU for a severe acute exacerbation of COPD between May 2015 and December 2018. Logistic multivariate regression analysis was performed to identify association between T0-Ph and the need for invasive MV during the ICU stay.
UNASSIGNED: We included 198 patients of whom 132 (67%) were male. The median age was 70 [interquartile range (IQR), 61-77] years. Nine (4.5%) patients died in the ICU. Median T0-Ph was significantly higher among patients requiring invasive MV as compared to non-intubated patients [1.23 (IQR, 1.07-1.41) and 1.09 (IQR, 0.91-1.27) mmol/L; P=0.005]. By multivariate analysis, pneumonia [odds ratio (OR) =6.42; 95% confidence interval (CI): 2.78-15.96; P<0.0001) and a history of intubation (OR =3.33; 95% CI: 0.97-11.19; P=0.05) were independently associated with the need for invasive MV, whereas T0-Ph was not (OR =1.75; 95% CI: 0.72-4.44; P=0.22).
UNASSIGNED: T0-Ph was significantly higher in patients requiring invasive MV. However, T0-Ph was not associated with the need for invasive MV in multivariate analysis.

Cefepime and piperacillin/tazobactam are antimicrobials recommended by IDSA/ATS guidelines for the empirical management of patients admitted to the intensive care unit (ICU) with community-acquired pneumonia (CAP). Concerns have been raised about which should be used in clinical practice. This study aims to compare the effect of cefepime and piperacillin/tazobactam in critically ill CAP patients through a targeted maximum likelihood estimation (TMLE). A total of 2026 ICU-admitted patients with CAP were included. Among them, (47%) presented respiratory failure, and (27%) developed septic shock. A total of (68%) received cefepime and (32%) piperacillin/tazobactam-based treatment. After running the TMLE, we found that cefepime and piperacillin/tazobactam-based treatments have comparable 28-day, hospital, and ICU mortality. Additionally, age, PTT, serum potassium and temperature were associated with preferring cefepime over piperacillin/tazobactam (OR 1.14 95% CI [1.01-1.27], p = 0.03), (OR 1.14 95% CI [1.03-1.26], p = 0.009), (OR 1.1 95% CI [1.01-1.22], p = 0.039) and (OR 1.13 95% CI [1.03-1.24], p = 0.014)]. Our study found a similar mortality rate among ICU-admitted CAP patients treated with cefepime and piperacillin/tazobactam. Clinicians may consider factors such as availability and safety profiles when making treatment decisions.