Real-time risk monitoring is critical but challenging in intensive care units (ICUs) due to the lack of real-time updates for most clinical variables. Although real-time predictions have been integrated into various risk monitoring systems, existing systems do not address uncertainties in risk assessments. We developed a novel framework based on commonly used systems like the Sequential Organ Failure Assessment (SOFA) score by incorporating uncertainties to improve the effectiveness of real-time risk monitoring.

This study included 5351 patients admitted to the Cardiothoracic ICU in the National University Hospital in Singapore. We developed machine learning models to predict long lead-time variables and computed real-time SOFA scores using predictions. We calculated intervals to capture uncertainties in risk assessments and validated the association of the estimated real-time scores and intervals with mortality and readmission.

Our model outperforms SOFA score in predicting 24-h mortality: Nagelkerke's R-squared (0.224 vs. 0.185, p < 0.001) and the area under the receiver operating characteristic curve (AUC) (0.870 vs. 0.843, p < 0.001), and significantly outperforms quick SOFA (Nagelkerke's R-squared = 0.125, AUC = 0.778). Our model also performs better in predicting 30-day readmission. We confirmed a positive net reclassification improvement (NRI) of our model over the SOFA score (0.184, p < 0.001). Similarly, we enhanced two additional scoring systems.

Incorporating uncertainties improved existing scores in real-time monitoring, which could be used to trigger on-demand laboratory tests, potentially improving early detection, reducing unnecessary testing, and thereby lowering healthcare expenditures, mortality, and readmission rates in clinical practice.

The online version contains supplementary material available at 10.1007/s13755-024-00331-5.

Sepsis, a life-threatening condition marked by organ dysfunction due to a dysregulated host response to infection, involves complex physiological and biochemical abnormalities.

To develop a multivariate model to predict 4-, 6-, and 8-week mortality risks in intensive care units (ICUs).

A retrospective cohort of 2389 sepsis patients was analysed using data captured by a clinical decision support system. Patients were randomly allocated into training (n = 1673) and validation (n = 716) sets at a 7:3 ratio. Least Absolute Shrinkage and Selection Operator (LASSO) regression identified variables incorporated into a multivariate Cox proportional hazards regression model to construct a prognostic nomogram. The area under the receiver operating characteristic curve (AUROC) assessed model accuracy, while performance was evaluated for discrimination, calibration and clinical utility.

A risk score was developed based on 11 independent predictors from 35 initial factors. Key predictors included minimum Acute Physiology and Chronic Health Evaluation II (APACHE II) score as having the greatest impact on prognosis, followed by days of mechanical ventilation, number of vasopressors, maximum and minimum Sequential Organ Failure Assessment (SOFA) scores, infection sources, Gram-positive or Gram-negative bacteria and malignancy. The nomogram demonstrated superior discriminative ability, with AUROC values of 0.882 (95% confidence interval [CI], 0.855-0.909) and 0.851 (95% CI, 0.804-0.899) at 4 weeks; 0.836 (95% CI, 0.798-0.874) and 0.820 (95% CI, 0.761-0.878) at 6 weeks; and 0.843 (95% CI, 0.800-0.887) and 0.794 (95% CI, 0.720-0.867) at 8 weeks for training and validation sets, respectively.

A validated nomogram and web-based calculator were developed to predict in-hospital mortality in ICU sepsis patients. Targeting identified risk factors may improve outcomes for critically ill patients.

The developed prediction model and nomogram offer a tool for assessing in-hospital mortality risk in ICU patients with sepsis, potentially aiding in nursing decisions and resource allocation.

Routine clinical practice with extended anaerobic antibiotic coverage (EAC) has been recently reconsidered for several infections; however, its benefits remain unclear even in patients with anaerobic bacteremia (AB). Here, we aimed to elucidate the effects of EAC on AB prognosis.

A multicenter retrospective observational study was conducted in patients with AB. Multivariate logistic regression analysis was performed to assess the effect of EAC on 30-day mortality. Inverse probability of treatment weighting analysis was performed to confirm the robustness of the findings.

In total, 483 patients were included, of whom 387 received EAC and 96 received limited anaerobic antibiotic coverage (LAC). Atypical foci of anaerobic infection, such as urinary tract infection and pneumonia, together with undetectable infection foci, comprised a larger proportion of infection foci in the LAC group than that in the EAC group (46.9% vs 30.5%). The 30-day mortality rates of the EAC and LAC groups were similar (12.5% and 14.2%, respectively; P = 0.664). Primary analysis revealed that EAC was not significantly associated with high mortality (odds ratio [OR], 1.42; 95% confidence interval [CI], 0.7-2.8), whereas source control significantly reduced this risk (OR, 0.28; 95% CI, 0.2-0.5). The sensitivity analysis results were consistent with those of the primary analyses.

This study demonstrated a less significant effect of initial EAC on AB compared with source control, particularly on AB with atypical infection foci. These findings would prompt reconsideration of the necessity of an initial EAC in several infections.

Accurate pathogen identification is critical for managing sepsis. However, traditional microbiological methods are time-consuming and exhibit limited sensitivity, particularly with blood samples. Metagenomic sequencing of plasma or whole blood was highly affected by the proportion of host nucleic acid.

We developed a Probe-Capture Metagenomic assay and established a multicentre prospective cohort to assess its clinical utility. In this study, 184 blood samples from patients suspected of sepsis were sent for blood culture and Probe-Capture Metagenomic sequencing before using antibiotics. The pathogen-positive rate and auxiliary abilities in diagnosis were compared among Probe-Capture Metagenomics, blood culture and real-time PCR (RT-PCR). Antibiotic therapy adjustments were based on the identification of pathogens, and changes in the Sequential Organ Failure Assessment (SOFA) score were monitored on days 0, 3 and 7 of admission.

A total of 184 sepsis patients were enrolled, with a mean age of 66 years (range 56-74). The Probe-Capture Metagenomics method, confirmed by RT-PCR, demonstrated a significantly higher pathogen detection rate than blood culture alone (51.6% vs. 17.4%, p < .001). When combining the results of blood culture and RT-PCR, Probe-Capture Metagenomics achieved a concordance rate of 91.8% (169/184), with a sensitivity of 100% and specificity of 87.1%. In terms of clinical impact, antibiotic therapy was adjusted for 64 patients (34.8%) based on the results from Probe-Capture Metagenomics, and 41 patients (22.3%) showed a > 2-point decrease in SOFA score following antibiotic adjustments.

Probe-Capture Metagenomics significantly enhances the ability of pathogen detection compared with traditional metagenomics. Compared to blood culture and RT-PCR in sepsis patients, it leads to improved antibiotic treatment and better patient outcomes. This study, for the first time, evaluates the clinical impact of metagenomic sequencing by integrating antibiotic adjustments and SOFA score changes, indicating that approximately one-fifth of sepsis patients benefit from this advanced diagnostic approach.

This study has been registered in clinical trials (clinicaltrials.gov) on 30 November 2018, and the registration number is NCT03760315.

Probe-Capture Metagenome had a significantly higher positive rate than blood culture (51.6% vs. 17.4%, p < .001). Combining blood culture and RT-PCR results, Probe-Capture Metagenome achieved a consistency rate of 91.8%. Antibiotics were adjusted in 34.8% of patients based on Probe-Capture Metagenome results, and 22.3% of patients experienced a more than 2-point decrease in SOFA score.

Cardiorenal syndrome poses significant diagnostic and therapeutic challenges. The Venous Excess Ultrasound (VExUS) grading system based on the combination of venous Doppler assessments has shown potential in predicting acute kidney injury and cardiovascular outcomes, but its relevance regarding the management of acutely decompensated heart failure (ADHF) remains to be fully understood.

In this prospective study, patients with ADHF and acute kidney injury (AKI) were enrolled from a medical intensive care unit over 20 months. The study involved echocardiography and VExUS grading at admission and 72 h later. Data collection included clinical parameters, diuretic dosages, urine output, and fluid balance. Statistical analyses focused on exploring the relationships between VExUS grades and its components, including the renal venous stasis index (RVSI), diuretic efficiency, and renal function improvement.

The cohort of 43 patients showed varied VExUS grades at admission. Higher VExUS grades were significantly associated with lower diuretic efficiency. Specifically, the mean urine output per 40 mg of furosemide was 368 ± 213 mL, with patients having VExUS grade 2 or 3 exhibiting reduced diuretic efficiency compared to those with grade 0-1 (Grade 2 vs. Grade 0-1: 333 ± 214 mL vs. 507 ± 189 mL, p = 0.02; Grade 3 vs. Grade 0-1: 270 ± 167 mL vs. 507 ± 189 mL, p = 0.004). The relationship between VExUS grade and diuretic efficiency was independent of admission creatinine and prior use of loop-diuretics (β = -106 CI: -180; -32 p = 0.006). Among the components of venous congestion assessment, the RVSI had the best ability to predict low diuretic efficiency (AUROC: 0.76 (0.60; 091) p = 0.001). Improvement in VExUS grade at 72 h was correlated with significant renal function improvement (84.6% vs. 47.1% for improved vs. non-improved VExUS grades, p = 0.03).

High VExUS and RVSI grades at admission are independently associated with reduced diuretic efficiency in ADHF patients with AKI. The findings emphasize the clinical value of venous congestion assessment in cardiorenal syndrome management including the selection of an initial diuretic dose.

The atherogenic index of plasma (AIP) can be used to reveal atherosclerosis. This study evaluated the AIP's efficacy in predicting the prognosis of acute kidney injury (AKI) and severity of acute pancreatitis (AP).

This retrospective cohort study recruited AP cases from the First College of Clinical Medical Science of China Three Gorges University between January 2019 and October 2023, including 1470 patients. AIP was computed using the formula: log10 [serum triglyceride (mmol/L)/serum high-density lipoprotein cholesterol (mmol/L)]. The AIP relationships with AKI occurrence and AP severity were validated using multivariable logistic regression models, subgroup and sensitivity analyses, and curve fitting.

Among the 1470 patients with AP, 250 (17%) developed AKI and 166 (11.3%) with severe AP. AIP was positively correlated with AKI and the severity of AP. Potential confounders were adjusted, consequently, AIP was positively linearly related to AKI (P for non-linearity: 0.731, OR 2.5, 95% CI 1.31-4.77,) and the severity of AP (P for non-linearity: 0.145, OR 3.1, 95% CI 1.53-6.27), respectively. The strength of the association between AIP and AKI, along with the severity of AP, was demonstrated through stratified analyses. Significant interactions were not observed in sex, age, hypertension, BMI, diabetes mellitus, SOFA score, BISAP score, and etiology of AP (all P for interaction > 0.05). The areas under the curves for AIP in predicting the incidence of AKI and severity of AP were 0.64 and 0.65, respectively.

This is the first study to suggest that the AIP is critical for the assessment of AKI risk, recommending early screening of severity among AP cases. Due to the observational nature of the study, the potential for residual confounding, and the need for external validation in larger, independent cohorts.

Emergent total pneumonectomy is a rare surgical intervention for patients with severe chest trauma. Patients who survive the immediate postoperative period experience prolonged, complex hospitalizations. The purpose of this case study is to review the nutrition care provided to a patient who survived total pneumonectomy and the supporting evidence. John Doe (JD) is a man aged 28 years who presented to a level I trauma center with penetrating chest trauma. He required multiple operative interventions, resulting in a partial right and total left pneumonectomy. JD's hospitalization was complicated by prolonged use of extracorporeal membrane oxygenation (ECMO) and continuous renal replacement therapy (CRRT). His surgical course and gastric feeding intolerance hampered enteral nutrition adequacy, and parenteral nutrition support was initiated on hospital day 17. Tolerance to enteral nutrition improved after jejunal access was obtained, and the patient transitioned to total enteral nutrition support. As a result of inflammatory metabolic changes and nutrition delivery challenges for the first 2 weeks of hospitalization, JD developed malnutrition. His nutrition care was further complicated by copper and carnitine deficiencies, which have been described in patients requiring ECMO and CRRT. Patients who require emergent total pneumonectomy following traumatic chest injuries will likely require complex hospital care, including extracorporeal organ support. These patients present unique nutrition challenges; however, given the relative infrequency of the intervention, there is limited research to guide clinical practice. Additional research on nutrition interventions in this population is warranted.

Patients in neurologic intensive care unit (NICU) often undergo inferior vena cava filter (IVCF) placement for venous thrombotic events. This study aims to determine the effectiveness of IVCF, filter retrieval, and mortality among patients that received IVCF in NICU.

In this single institutional, noncomparative, retrospective study, all patients who were admitted to NICU and underwent IVCF placement from April 2015 to December 2020 were reviewed. IVCF was successfully deployed in all 175 patients [100%; median age 68 years, female 84/175 (48.0%)]. The 3 most common causes for NICU admission were intracranial hemorrhage (66/175, 37.7%), ischemic stroke (62/175, 35.4%), and traumatic brain injury (16/175, 9.1%). Deep vein thrombosis and pulmonary embolism (PE) were confirmed in 155 (88.6%) and 35 (20.0%) patients at the time of filter placement, respectively. Primary outcomes of interest were postfilter placement PE, filter retrieval, and inhospital mortality. Baseline characteristics were analyzed using t-tests and chi-squared test for continuous and noncontinuous variables, respectively. Factors associated with primary outcomes were analyzed with a logistic regression model.

Post-IVCF PE occurred in 3 patients (1.7%) with a median follow-up of 3 months. Excluding 26 inhospital deaths (14.9%, none was related to PE), filters were retrieved in 31 discharged patients (20.8%) with a median filter dwelling time of 9 months. Advanced filter retrieval required a higher fluoroscopy time (median 3.3 minutes vs 8.3 minutes, p = 0.016) and contrast volume use (median 35.0 ml vs 57.5 ml, p = 0.0028) than standard technique. No procedure-related complication occurred during filter placement and retrieval. Sequential Organ Failure Assessment (SOFA, p = 0.012) and Simplified Acute Physiology Scores (SAPS, p = 0.016) were independently associated with inhospital mortality. Modified Rankin Score (mRS) at discharge was an independent predictor for filter retrieval (p < 0.001).

Despite safety and effectiveness, IVCF retrieval rate for NICU patients was low, particularly those with worse mRS at time of hospital discharge. Worse SOFA and SAPS scores were associated with inhospital mortality.Clinical ImpactInferior vena cava filter (IVCF) is effective preventing post-filter pulmonary embolism (PE) in neurologic intensive care unit (NICU) patients, with only 1.7% experiencing PE post-placement, underscoring its role in managing venous thrombotic events in this high-risk population; however, the low retrieval rate of IVCFs, particularly in patients with poorer functional outcomes (worse mRS at discharge), and the association of higher SOFA and SAPS scores with increased inhospital mortality, emphasize the need for improved strategies to optimize filter retrieval and patient selection in critically ill neurologic patients.

It has been shown that general intensive care nurses are able to perform an examination of the deep venous system of the lower extremities for the diagnosis of proximal deep vein thrombosis (DVT) using a compression ultrasound test with a high degree of reliability. (Skulec et al. in Eur J Intern Med 76:130-131, 2020) Another challenge for the use of vascular point-of-care ultrasound in intensive care is the diagnosis of central venous catheter-related thrombosis. It is a common problem that is often underdiagnosed. Due to the simplicity of the examination and the possible link with nursing care of inserted central venous catheters, this may be another potential diagnostic competency for critical care nurses.

Before the start of the study, each nurse participating in the study completed a two-hour training in duplex ultrasonography and examined 5 patients under supervision. Then patients in the intensive care unit (ICU) included in the study, underwent a duplex ultrasound performed by a nurse. Within 24 h, the examination was repeated by the ICU doctor. In the case of catheter insertion into the internal jugular vein (VJI) or the subclavian vein (VSC), the jugular vein, subclavian vein, and axillary vein (VA) were examined bilaterally. When the catheter was inserted into the femoral vein (VF), the patients were subjected to a duplex ultrasound of the femoral vein and the popliteal vein (VP) of both lower limbs. The examination results of each patient were blinded until both tests were performed. Calculations were used to evaluate the reliability of the test.

A total of 160 patients aged 62.9 ± 12.3 years were included. In our sample, the prevalence of CRT was found to be 41%. The overall sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of examinations performed by general intensive care nurses were 90.8%, 97.1%, 91.8%, 96.8%, and 95.5%, respectively.

The results of our study suggest that general ICU nurses are able to perform inpatient CRT duplex ultrasound with excellent specificity but only moderate sensitivity after a short, predefined training.

Evaluating the prognosis of ARDS patients using grading systems can enhance treatment decisions. This retrospective observational study evaluated the predictive accuracy of the SOFA score, APACHE II score, SpO2/FiO2 ratio, and PaO2/FiO2 ratio for mortality in ARDS patients in Vietnam. The study included 335 adult ARDS patients admitted to a central hospital from August 2015 to August 2023. Among them, 66.9% were male, the median age was 55 years, and 61.5% died in the hospital. The SOFA (AUROC: 0.651) and APACHE II scores (AUROC: 0.693) showed poor discriminatory ability for hospital mortality. The SpO2/FiO2 (AUROC: 0.595) and PaO2/FiO2 ratios (AUROC: 0.595) also displayed poor discriminatory ability. In multivariable analyses, after adjusting for the same set of confounding variables, the APACHE II score (adjusted OR: 1.152), SpO2/FiO2 ratio (adjusted OR: 0.985), and PaO2/FiO2 ratio (adjusted OR: 0.989) were independently associated with hospital mortality. Although the SOFA score (adjusted OR: 1.132) indicated a potential association with hospital mortality, it did not reach statistical significance (p = 0.081). However, a SOFA score of ≥ 10 emerged as an independent predictor (adjusted OR: 3.398) of hospital mortality. These findings emphasize the need for further studies to develop more accurate scoring systems for predicting outcomes in ARDS patients.

Severity assessment of community-acquired pneumonia (CAP) is essential for many purposes. Among these are the microbiological confirmation strategy and choice of empirical antimicrobial therapy. However, many severity assessment systems have been developed to aid clinicians to reach reliable predictions of severe outcomes.

We aimed to apply nine disease severity assessment scoring systems to a large 2016 to 2021 CAP cohort in order to achieve test sensitivity, specificity and predictive values. We used intra-hospital all-cause mortality and the need for intensive care admission as outcomes. The area under the receiver operating characteristic (ROC) curve was used to display test performance.

A total of 1.112 CAP episodes were included in the analysis, of which 91.4% were radiologically, and 43.7% were microbiologically confirmed. When intra-hospital all-cause mortality was set as outcome, tests designed for CAP severity assessment, like PSI, and CURB65 outperformed the more generic systems like NEWS2, qSOFA, SIRS and CRB65. Designated tests for CAP (PSI, IDSA/ATS and CURB65) and overall critical illness (SOFA) displayed acceptable performances as compared to non-specific tests. Comparable results were gained when intensive care admission was set as outcome. The area under the receiving operating curve was 0.948, 0.879, 0.855 and 0.726 for the SOFA, PSI, IDSA/ATS and CURB65 scoring systems, respectively.

CAP severity assessment remains important. Designated CAP severity assessment tools outperformed generic tests.

IntroductionThe cardiovascular component of the Sequential Organ Failure Assessment (SOFA) score does not correspond with contemporary clinical practice in sepsis or identify impaired cardiac function. Our aim was to develop a modified cardiovascular SOFA component that reflects cardiac dysfunction and improves the SOFA score's 30-day mortality discrimination.MethodsA cohort of sepsis patients from a previous study was divided into a training (n = 250) and test cohort (n = 253). Nine widely available measures of cardiovascular function were screened for association with 30-day mortality using natural cubic spline. High-sensitivity cardiac troponin T (hs-cTnT), N-terminal pro B-type natriuretic peptide (NT-proBNP) and heart rate (HR) were transformed into ordinal variables (0-4 points). The presence of atrial fibrillation (AF) was assigned two points. The SOFA score was extended by adding the variable points in different weights and combinations. The best-performing cardiac-extended model (CE-SOFA) was evaluated in the test cohort. Improved prognostic discrimination and calibration were assessed using logistic regression, area under receiver operating characteristic curves (AUC), Net Reclassification Improvement (NRI) index, and DeLong and Hoshmer-Lemeshow tests.ResultsIn the training cohort, all differently weighted and combined models using hs-cTnT, NT-proBNP and AF points added to the SOFA score showed improved discriminative ability (AUC 0.67-0.75) compared to the SOFA score (AUC 0.62; NRI P < .001; DeLong P ≤ .001). In the test cohort, CE-SOFA demonstrated improved 30-day mortality discrimination compared to the SOFA score (AUC 0.72 vs 0.68), exhibiting good calibration and significantly improved discrimination using the NRI index (P = .009) but not the DeLong test (P = .142).ConclusionsThe CE-SOFA model reflects cardiac dysfunction and improves 30-day mortality discrimination in sepsis. External validation is the next step to further substantiate a revised cardiovascular component in a future SOFA 2.0.

Risk factors for in-hospital mortality related to gastrointestinal complications requiring operative intervention after cardiovascular surgery have not been previously described.

Adult patients who underwent cardiovascular surgery followed by gastrointestinal surgery during the same admission between January 2010 and June 2023 were included. Multivariable logistic regression was used to identify predictors of in-hospital mortality. Kaplan-Meier survival analysis was performed to assess overall survival based on identified risk factors.

Gastrointestinal complications requiring operative intervention after cardiac surgery occurred in 151 patients, with an overall in-hospital mortality of 35.76% (n = 54). The most common diagnosis was bowel ischemia (50.33%). On multivariable logistic regression, the history of cirrhosis (odds ratio: 37.96, 95% confidence interval: 3.57-543.90) and the clinical condition at the time of emergency general surgery consultation, described by elevated lactate (odds ratio: 5.76, 95% confidence interval: 1.71-22.82), platelets <50 × 109/L (odds ratio: 11.34, 95% confidence interval: 1.60-162.37), ≥3 vasoactive medications (odds ratio: 4.93, 95% CI: 1.29-20.14), and the need for renal replacement therapy (odds ratio: 5.18, 95% confidence interval: 1.46-20.79) were predictive of in-hospital mortality. In-hospital mortality was low when none of the risk factors identified on multivariable analysis were present (2.38%, n = 1 of 42), but in-hospital mortality was universal among patients with 4-5 risk factors (100%, n = 8 of 8).

Gastrointestinal complications after cardiac surgery are disastrous when patient illness becomes severe. Clinicians should maintain a high index of suspicion for gastrointestinal complications to promote early involvement of general surgery. Knowledge of these risk factors could help guide discussions among the multidisciplinary care team, patients, and their families.

 The criteria for diagnosing sepsis-induced coagulopathy (SIC) may overlap with those of Japanese Association for Acute Medicine (JAAM) disseminated intravascular coagulation (DIC). This study determined if the diagnostic criteria of SIC overlap with JAAM DIC diagnostic criteria for identifying patients with DIC according to the International Society on Thrombosis and Haemostasis (ISTH) criteria and whether the patients diagnosed with these criteria have the same prognosis.

 This multicenter retrospective study included patients with sepsis diagnosed using the JAAM and ISTH DIC and SIC criteria on days 1 and 4. The established ISTH DIC criteria was the reference standard for primary outcome that compared the characteristics of SIC and JAAM DIC. Secondary outcomes were multiple organ dysfunction syndrome (MODS), ventilator-free and intensive care unit-free days, and in-hospital mortality.

 A total of 1,438 patients were included in this study. On day 1, the JAAM DIC and SIC criteria diagnosed almost all patients with ISTH DIC (98 and 94%, respectively), predicting ISTH DIC (area under the receiver operating curve [AUC]: 0.740 versus 0.752, p = 0.523) and MODS (AUC: 0.686 versus 0.697, p = 0.546) on day 4 and progressing to ISTH DIC in the same proportion (28.6 versus 30.1%, p = 0.622). There were no differences in survival probabilities (p = 0.196) or secondary outcomes between patients diagnosed using JAAM DIC and SIC criteria on day 1.

 SIC and JAAM DIC diagnoses were equal among patients with sepsis, suggesting that SIC criteria add little to current DIC scoring systems.

To evaluate the evidence for the use of ascorbic acid, thiamine, or a combination of both agents without corticosteroids in the management of sepsis and septic shock.

A review of the literature was conducted through August 2023 on PubMed, MEDLINE, Web of Science, and CINAHL using the following terminology: "ascorbic acid" OR "vitamin C" OR "thiamine" OR "vitamin B" OR "vitamin B 1" AND "sepsis" OR "septic shock" NOT "steroid" OR "hydrocortisone" OR "corticosteroid."

Trials that described patient outcomes, medication efficacy, and medication safety data were considered for inclusion, while reports describing the use of either or both thiamine and ascorbic acid for a non-sepsis indication and reports that were not readily translatable to English were excluded. Studies that allowed corticosteroid use in both the intervention and control cohorts as part of a standard-of-care protocol were eligible for inclusion.

Heterogeneity of data exists, marked by divergent quantifications for successful pharmacotherapy interventions. Whereas some data support changes in patient outcome scores or critical illness indices, others have failed to demonstrate any meaningful benefit to ICU length of stay, ventilator status, or mortality.

Exploring the individual and synergistic effects of ascorbic acid and thiamine on key pathways implicated in sepsis pathophysiology has not yielded unequivocal evidence supporting their use without concomitant corticosteroids.

This study aimed to describe the population pharmacokinetics of caspofungin in critically ill patients receiving extracorporeal membrane oxygenation (ECMO) and to identify dosing regimens with a high likelihood of achieving effective exposures. Serial blood samples were collected over a single-dosing interval during ECMO. Total plasma concentrations were measured by a validated chromatographic assay. A population pharmacokinetic model was built and Monte Carlo dosing simulations were performed using Monolix. The probability of target attainment (PTA) and fractional target attainment (FTA) rates were simulated for various caspofungin dosing regimens against Candida albicans, Candida glabrata, and Candida parapsilosis. In all, 64 plasma concentration-time points were obtained from 8 critically ill patients receiving ECMO. Plasma concentration-time data for caspofungin were best described by a one-compartment model with first-order elimination. Lean body weight was identified as a significant covariate of volume of distribution. The typical volume of distribution and clearance of caspofungin in this cohort were 8.13 L and 0.55 L/h, respectively. The licensed caspofungin dosing regimen (a loading dose of 70 mg on day 1 followed by a maintenance dose of either 50 mg/day or 70 mg/day) demonstrated optimal PTA rates (≥90%) against C. albicans with an MIC of ≤0.064 mg/L, C. glabrata with an MIC of ≤0.125 mg/L, and C. parapsilosis with an MIC of ≤0.064 mg/L. The FTA analysis suggested that the licensed dosing regimen is only optimal (≥95%) against Candida glabrata, regardless of lean body weight. A higher-than-standard empirical dosing regimen (e.g., a loading dose of 100 mg on day 1, followed by a maintenance dose of 100 mg daily) is likely advantageous for critically ill patients receiving ECMO.

Background: Piperacillin-tazobactam (PTZ) demonstrates time-dependent bactericidal activity, potentially increasing the need for higher dosing in obese and critically ill patients. However, limited information is available on the safety of higher dosing strategies. Objective: To evaluate the safety and clinical impact of high dose 6.75 g IV PTZ for the treatment of pneumonia in critically ill, obese (≥120 kg) patients vs standard dose 4.5 g IV PTZ. Methods: Retrospective, cohort study, multicenter in health-system consisting of four acute-care teaching hospitals. Adult patients weighing at least 120 kg on PTZ for pneumonia in the intensive care unit (ICU) from January 2013 to September 2018 were included. The primary outcome of the study was acute nephrotoxicity defined as initiation of renal replacement therapy and/or serum creatinine increase within 48 hours of last PTZ dose. Secondary outcomes included thrombocytopenia, 14-day all-cause mortality, and ICU length of stay (LOS). Results: One hundred thirty-six patients were included with 52 and 84 in 4.5 g PTZ and 6.75 g PTZ respectively. The rate of acute nephrotoxicity was comparable between cohorts (50% 4.5 g vs 40.5% 6.75 g, P = 0.277). High dose PTZ was not independently associated with acute nephrotoxicity after control for selected confounders. All secondary outcomes were similar. Concomitant vancomycin and calculated supratherapeutic vancomycin area under curve were not independently associated with increased nephrotoxicity. Conclusions: High dose PTZ was not associated with increased acute nephrotoxicity, thrombocytopenia, 14-day all-cause mortality, or ICU LOS. Additionally, more robust trials are needed to fully assess the clinical impact of 6.75 g PTZ dosing for critically ill, obese patients, for pneumonia.

Overfeeding and underfeeding are associated with negative outcomes during critical illness. The purpose of this retrospective study was to assess the association between nutrition intake and outcomes for patients receiving venovenous (VV) extracorporeal membrane oxygenation (ECMO).

Adults who received VV ECMO August 2017 to June 2020 were screened. Patients with <3 ECMO nutrition support days were excluded. Age, sex, height, weight, ideal body weight (IBW), body mass index, sequential organ failure assessment score, respiratory ECMO survival prediction score, energy, and protein goals were collected. All nutrition intake was collected for the first 14 days of ECMO or until death, decannulation, or oral diet initiation. Outcomes analyzed included mortality and VV ECMO duration. The relationship between nutrition delivery and outcomes was tested with multivariate analysis. Univariate analyses were conducted on obese and nonobese subgroups.

A total of 2044 nutrition days in 178 patients were analyzed. The median estimated needs were 24 (interquartile range: 22.3-28.3) kcal/kg/day and 2.25 (interquartile range: 2.25-2.77) g/kg/day of protein using IBW in patients with obesity and actual weight in patients without obesity. Patients received 83% of energy and 63.3% of protein targets. Patients with obesity who received ≥2 g/kg IBW of protein had a significantly shorter ECMO duration (P = 0.037). Increased protein intake was independently associated with a reduced risk of death (odds ratio: 0.06; 95% confidence interval: 0.01-0.43).

Higher protein intake was associated with reduced mortality. Optimal energy targets for patients receiving ECMO are currently unknown and warrant further study.

In-hospital mortality of septic critically ill patients with COVID-19 is significantly higher than in those without COVID-19. The knowledge on long-term outcomes remains scarce. In this retrospective analysis, we compare clinical characteristics, long-term functional outcomes, and survival in septic critically ill patients with and without COVID-19.

Data of septic critically ill patients without COVID-19 were collected as part of the Comprehensive Sepsis Center Dresden-Kreischa registry from 2020 to 2023. The data of septic critically ill patients with COVID-19 were collected as part of the local ARDS/COVID-19 registry over the same period. Diagnosis of sepsis was based on the Sepsis-3 definition. Variables collected for analyses were obtained from electronic health records. Long-term follow-up was performed 6-12 months after sepsis diagnosis. Survival was depicted using Kaplan-Meier curves. Associations between long-term mortality and risk factors were modeled by Cox Regression.

372 septic patients without COVID-19 and 301 with COVID-19 were enrolled. Septic patients with COVID-19 were significantly younger, had a significantly lower Charlson Comorbidity Index, and had a significantly higher SOFA score at ICU admission. Long-term follow-up showed a significantly higher mortality in septic patients with COVID-19 (73.4 % vs. 30.1 %; HR 3.4 (95 % CI 2.73-4.27; p < 0.05)). COVID-19 infection was associated with significant increased mortality (adjusted HR 3.27; 95 % CI 2.48-4.33; p < 0.05) and reduced health-related quality of life, measured by the EQ-5D-3 L Index, (0.56 (0.16-0.79) vs. 0.79 (0.69-0.99); p < 0.05).

In our cohort of septic critically ill patients, health-related quality of life and long-term survival were considerably reduced in patients with concomitant COVID-19. Furthermore, COVID-19 could be identified as an independent risk factor for higher long-term mortality in these patients.

Sepsis continues to be a leading cause of illness and mortality in children around the world. Various scoring systems have been devised to predict the outcome of pediatric sepsis. Pediatric sequential organ failure assessment (p SOFA) and lactate clearance are the two commonly used methods.

The aim of this study was to compare the p SOFA score with lactate clearance as predictors of morbidity and mortality in pediatric sepsis, to compare the initial plasma lactate level and lactate clearance, and to know which is better to predict outcomes in sepsis and septic shock.

This prospective observational study was conducted in a pediatric intensive care unit of a tertiary care teaching hospital from July 2022 to June 2024. The blood lactate level and p SOFA score were assessed at admission and at 24 and 48 hours, and lactate clearance was calculated at 24 and 48 hours of admission. The receiver operating characteristic (ROC) curve was plotted to predict deaths using p SOFA, lactate level, and lactate clearance.

A total of 71 children were enrolled in the study. All children were divided into two groups, 58 (82%) survivors and 13 (18%) non-survivors. The most common diagnosis was pneumonia, observed in 31 (43.6%) children. Compared to survivors, non-survivors had a higher prevalence of multiple organ dysfunction syndrome (MODS). The most common organ system involved was the cardiovascular, in 50 (70%) cases. For predicting mortality, p SOFA scores were statistically significant at admission and at 24 and 48 hours with a high area under the curve (AUC) at 48 hours (0.985). Lactate clearance at 24 hours was a better predictor of mortality than at 48 hours with a higher AUC (0.958).

Both p SOFA score at 48 hours and lactate clearance at 24 hours were significant predictors of mortality. Among both parameters, lactate clearance at 24 hours was superior in predicting mortality early.

To compare clinical outcomes of patients with catecholamine-resistant vasodilatory shock (CRVS) receiving continuous renal replacement therapy who receive adjunctive angiotensin II (ANGII) to those who do not.

Retrospective cohort analysis.

Multicenter, single health system consisting of one academic medical center and four community hospitals.

Critically ill adult patients with CRVS (norepinephrine or equivalent dose ≥0.5 mcg/kg/min).

Adjunctive ANGII versus standard-of-care (SOC) vasopressors alone (norepinephrine, epinephrine, vasopressin, phenylephrine, dopamine).

The primary outcome was intensive care unit mortality. Secondary outcomes included 30-day mortality, Sequential Organ Failure Assessment (SOFA) score at 72 hours, time to shock resolution, and adverse effects. A multivariate logistic regression was used for the primary analysis. The study included 265 patients, of which 70 received ANGII and 195 received SOC. Intensive care unit and 30-day mortality were lower in patients that received ANGII (61.4% v 75.4%, adjusted odds ratio 0.438, 95% confidence interval: 0.239-0.805, p = 0.008; and 67.1% v 78.5%, adjusted odds ratio 0.479, 95% confidence interval: 0.256-0.898, p = 0.022). Differences in time to shock reversal and SOFA score at 72 hours were not statistically significant. The adverse effects evaluated were not statistically significant, apart from an increase in fungal infections in the ANGII group (17.1% v 7.2%, p = 0.016).

ANGII was associated with lower mortality in patients who received renal replacement therapy compared to SOC. This evaluation reaffirms a subgroup of patients that may benefit from the addition of ANGII.

Prediction of mortality in intensive care unit (ICU) patients typically relies on black box models (that are unacceptable for use in hospitals) or hand-tuned interpretable models (that might lead to the loss in performance). We aim to bridge the gap between these 2 categories by building on modern interpretable machine learning (ML) techniques to design interpretable mortality risk scores that are as accurate as black boxes.

We developed a new algorithm, GroupFasterRisk, which has several important benefits: it uses both hard and soft direct sparsity regularization, it incorporates group sparsity to allow more cohesive models, it allows for monotonicity constraint to include domain knowledge, and it produces many equally good models, which allows domain experts to choose among them. For evaluation, we leveraged the largest existing public ICU monitoring datasets (MIMIC III and eICU).

Models produced by GroupFasterRisk outperformed OASIS and SAPS II scores and performed similarly to APACHE IV/IVa while using at most a third of the parameters. For patients with sepsis/septicemia, acute myocardial infarction, heart failure, and acute kidney failure, GroupFasterRisk models outperformed OASIS and SOFA. Finally, different mortality prediction ML approaches performed better based on variables selected by GroupFasterRisk as compared to OASIS variables.

Group Faster Risk's models performed better than risk scores currently used in hospitals, and on par with black box ML models, while being orders of magnitude sparser. Because GroupFasterRisk produces a variety of risk scores, it allows design flexibility-the key enabler of practical model creation.

Group Faster Risk is a fast, accessible, and flexible procedure that allows learning a diverse set of sparse risk scores for mortality prediction.

Ulinastatin is a protease-inhibiting drug with anti-inflammatory and other pharmacological properties. Little is known regarding its role following acute type A aortic dissection (ATAAD) surgery. We perform a randomized controlled trial to investigate the protective effect of ulinastatin against negative inflammatory response and organ dysfunction in ATAAD surgery (PANDA). The primary outcome of mean daily Sequential Organ Failure Assessment (SOFA) score from baseline to 7 days of surgery is 8.80 (SD, 4.11) in the ulinastatin group and 8.61 (SD, 4.47) in the control group (mean difference between groups was 0.04; 95% confidence interval [CI], -0.24 to 0.33; p = 0.765). Systemic inflammatory response syndrome (SIRS) within 7 days of surgery is lower in the ulinastatin group than in the control group (p < 0.001). Additional ulinastatin to standard treatment is likely to reduce SIRS rates instead of preventing organ dysfunction, highlighting the potential importance of the benefits of anti-inflammatory pharmacotherapeutics. The trial is registered on clinicaltrials.org (NCT04711889).

Identifying patients in the emergency department (ED) at higher risk for in-hospital mortality can inform shared decision making and goals-of-care discussions. Electronic health record systems allow for integrated multivariable logistic regression (LR) modeling, which can provide early predictions of mortality risk in time for crucial decision making during a patient's initial care. Many commonly used LR models require blood gas analysis values, which are not frequently obtained in the ED. The goal of this study was to develop an all-cause mortality prediction model, derived from commonly collected ED data, which can assess mortality risk early in ED care.

Data were obtained for all patients, age 18 and older, admitted from the ED to Atrium Health Wake Forest Baptist from April 1, 2016, through March 31, 2020. Initial vital signs including heart rate, respiratory rate, systolic blood pressure, diastolic blood pressure, mean arterial pressure, pulse oximetry, weight, body mass index, comprehensive metabolic panel, and a complete blood count were electronically retrieved for all patients. The prediction model was developed using LR. The ED early mortality (EDEM) model was compared with the rapid Emergency Medicine Score (REMS) for performance analysis.

A total of 45,004 patients met inclusion criteria, comprising a total of 77,117 admissions. In this cohort, 52.8% of patients were male and 47.2% were female. The model used 35 variables and yielded an area under the receiver operating characteristic curve (AUC) of 0.889 (95% CI 0.874-0.905) with a sensitivity of 0.828 (95% CI 0.791-0.860), a specificity of 0.788 (95% CI 0.783-0.794), a negative predictive value of 0.995 (95% CI 0.994-0.996), and a positive predictive value of 0.084 (95% CI 0.076-0.092). This outperformed REMS in this data set, which yielded an AUC of 0.500 (95% CI 0.455-0.545).

The EDEM model was predictive of in-hospital mortality and was superior to REMS.

The present study investigated the impact of continuous infusion therapy with low-dose terlipressin (TP) combined with norepinephrine on hemodynamics, inflammatory markers, and prognosis in patients with severe septic shock.

Seventy-four patients with severe septic shock were randomly assigned to either a control group (n = 37) or an observation group (n = 37). Patients in the control group received norepinephrine alone, while those in the observation group received a continuous infusion of low-dose TP in addition to norepinephrine. To assess the effect of treatment, a set of clinical parameters was evaluated in both groups before and after treatment. These parameters included hemodynamic indicators (heart rate [HR], mean arterial pressure [MAP], central venous pressure [CVP], cardiac index [CI], and systemic vascular resistance index [SVRI]), levels of serum inflammatory markers (interleukin-8 [IL-8], tumor necrosis factor-α [TNF-α], and hypersensitivity C-reactive protein [hs-CRP]), renal function indicators (blood urea nitrogen [BUN], serum creatinine [SCr], and cystatin C [Cys-C]), serum procalcitonin (PCT), and lactate, as well as lactate clearance rate (LCR). Additionally, the acute physiology and chronic health evaluation II (APACHE II) score, 28-day mortality rate, multiple organ dysfunction syndrome (MODS) incidence rate, and adverse reaction incidence were also determined.

Compared to baseline values, MAP, CVP, CI, SVRI, and LCR increased in both groups after treatment, while HR, levels of IL-8, TNF-α, hs-CRP, BUN, SCr, PCT, and lactate all decreased. Additionally, APACHE II scores also decreased. Furthermore, the observation group exhibited higher MAP, CVP, CI, SVRI, and LCR, along with lower HR, levels of IL-8, TNF-α, hs-CRP, BUN, SCr, PCT, and lactate than the control group after treatment. The observation group also had lower APACHE II score, 28-day mortality rate, MODS incidence rate, and adverse reaction incidence than the control group after treatment (P < .05).

Continuous infusion therapy with low-dose TP combined with norepinephrine was effective in treating patients with severe septic shock, improving hemodynamic parameters, reducing the levels of inflammatory markers, promoting renal function recovery, and reducing the mortality rate.

Objective: To investigate the value of Sequential Organ Failure (SOFA) score and its dynamics (ΔSOFA) in predicting mortality in hematology care unit (HCU) . Methods: A retrospective clinical study was conducted on 79 critically ill hematologic patients admitted to the Center for Critical Care Medicine, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences, between May and June 2024. SOFA scores and ΔSOFA were calculated within 2 days before and after HCU admission. The predictive value of SOFA and ΔSOFA in mortality was assessed using receiver operating characteristic (ROC) curve analysis. Results: Among the 79 patients, the HCU mortality rate was 54.4%. The SOFA scores on days 1-3 (D1, D2, and D3) and ΔSOFA on day 1 (ΔD_1) of all patients, leukemia patients and hematopoietic stem cell transplantation (HSCT) patients were significantly higher in the death group compared with the non-death group (all P<0.05). ROC curve analysis revealed that the D_1, D_2, D_3 scores, and ΔD_1 significantly predicted mortality (P<0.001), with areas under the curve (AUCs) of 0.786, 0.866, 0.901, and 0.843, respectively. The sensitivity values were 74.36%, 57.89%, 62.85%, and 86.84%, while specificity values were 70%, 100%, 100%, and 67.65%, respectively. In the HSCT group, the D_-1, D_1, D_2, D_ 3, scores and ΔD_1 were predictive of HCU mortality, with AUCs of 0.833, 0.794, 0.871, 0.846, and 0.795, respectively. Sensitivity values for these scores were 100%, 85.71%, 71.43%, 57.14%, and 57.14%, while specificity values were 73.33%, 70.59%, 91.33%, 100%, and 100%, respectively. In the leukemia group, the D_1, D_2, D_3 scores, and ΔD_1 were predictive of HCU mortality, with AUCs of 0.760, 0.829, 0.846, and 0.756, respectively. Sensitivity values were 71.43%, 78.57%, 53.85%, and 71.43%, while specificity values were 76.19%, 78.95%, 100%, and 63.16%, respectively. For all patients, the D_3 score exhibited the highest specificity, while the ΔD_1 demonstrated the highest sensitivity. For patients in both the HSCT and leukemia groups, the sensitivity and specificity values of the D_1 and D_3 scores exceeded those of the ΔD_1. Conclusion: For patients with hematologic critical illness, including leukemia and those undergoing HSCT hospitalized in the HCU, D_1, D_2, D_ 3 scores and ΔD_1 are significantly associated with HCU mortality.

In critically ill patients, compromised microcirculation causes tissue hypoxia, organ failure, and death. These pathophysiological processes occur particularly in patients with high illness severity, so reliable hypoxia biomarkers should reflect this in their occurrence. This secondary analysis of a prospective study categorized patients by their burden of organ dysfunction (BOD) using the cohort's median initial sequential organ failure assessment (SOFA) score of 8 as a cutoff. The kinetic parameters of the hypoxia markers lactate and S-adenosylhomocysteine (SAH) were analyzed for correlation with organ dysfunction severity and mortality prediction. In low BOD patients, neither marker correlated with SOFA. In high BOD patients, lactate showed a moderate correlation and SAH showed a strong correlation. Lactate correlated with organ dysfunction in survivors but not in non-survivors, while SAH correlated strongly in non-survivors but not in survivors. In univariate logistic regression, lactate predicted mortality moderately in low BOD (areas under the receiver operating characteristic curves (AUROCs) 0.7-0.8) but poorly in high BOD patients (AUROCs 0.5-0.7). SAH's prediction improved from poor to excellent (AUROCs 0.8-0.9) with higher BOD. Thus, SAH appears superior to lactate in the detection of organ dysfunction severity and mortality prediction in high BOD patients.

Rhabdomyolysis (RM) frequently gives rise to diverse complications, ultimately leading to an unfavorable prognosis for patients. Consequently, there is a pressing need for early prediction of survival rates among RM patients, yet reliable and effective predictive models are currently scarce.

All data utilized in this study were sourced from the MIMIC-IV database. A multivariable Cox regression analysis was conducted on the data, and the performance of the new model was evaluated based on the Harrell's concordance index (C-index) and the area under the receiver operating characteristic curve (AUC). Furthermore, the clinical utility of the predictive model was assessed through decision curve analysis (DCA).

A total of 725 RM patients admitted to the intensive care unit (ICU) were included in the analysis, comprising 507 patients in the training cohort and 218 patients in the testing cohort. For the development of the predictive model, 37 variables were carefully selected. Multivariable Cox regression revealed that age, phosphate max, RR mean, and SOFA score were independent predictors of survival outcomes in RM patients. In the training cohort, the AUCs of the new model for predicting 28-day, 60-day, and 90-day survival rates were 0.818 (95% CI: 0.766-0.871), 0.810 (95% CI: 0.761-0.855), and 0.819 (95% CI: 0.773-0.864), respectively. In the validation cohort, the AUCs of the new model for predicting 28-day, 60-day, and 90-day survival rates were 0.840 (95% CI: 0.772-0.900), 0.842 (95% CI: 0.780-0.899), and 0.842 (95% CI: 0.779-0.897), respectively.

This study identified crucial demographic factors, vital signs, and laboratory parameters associated with RM patient prognosis and utilized them to develop a more accurate and convenient prognostic prediction model for assessing 28-day, 60-day, and 90-day survival rates.

This study specifically targets patients with RM admitted to ICU and presents a novel clinical prediction model that surpasses the conventional SOFA score. By integrating specific prognostic indicators tailored to RM, the model significantly enhances prediction accuracy, thereby enabling a more targeted and effective approach to managing RM patients.

Management of persistent inflammation, immunosuppression, and catabolism syndrome (PICS) after sepsis remains challenging for patients in the intensive care unit, experiencing poor quality of life and death. However, immune-cell signatures in patients with PICS after sepsis remain unclear.

We determined immune-cell signatures of PICS after sepsis at single-cell resolution. Murine cecal ligation and puncture models of PICS were applied for validation.

Immune functions of two enriched monocyte subpopulations, Mono1 and Mono4, were suppressed substantially in patients with sepsis and were partially restored in patients with PICS after sepsis and exhibited immunosuppressive and pro-apoptotic effects on B and CD8T cells. Patients with PICS and sepsis had reduced naive and memory B cells and proliferated plasma cells. Besides, naive and memory B cells in patients with PICS showed an active antigen processing and presentation gene signature compared to those with sepsis. PICS patients with better prognoses exhibited more active memory B cells and IGHA1-plasma cells. CD8TEMRA displayed signs of proliferation and immune dysfunction in the PICS-death group in contrast with the PICS-alive group. Megakaryocytes proliferation was more pronounced in patients with PICS and sepsis than in healthy controls, with notable changes in the anti-inflammatory and immunomodulatory effects observed in patients with PICS and verified in mice models.

Our study evaluated PICS after sepsis at the single-cell level, identifying the heterogeneity present within immune-cell subsets, facilitating the prediction of disease progression and the development of effective intervention.

This work was supported by the National Natural Science Foundation of China, Shanghai Municipal Health Commission "Yiyuan New Star" Youth Medical Talent Cultivating Program, and Shanghai Clinical Research Center for Anesthesiology.

To investigate the trajectories and determinants of muscle strength in ICU patients and their impact on 1-, 6-, and 12-month mortality outcomes.

Prospective multicenter cohort study.

Ten ICUs across five tertiary hospitals in Fujian Province, China.

Muscle strength was assessed using handgrip strength at three time points: ICU admission, ICU discharge, and hospital discharge. Group-based trajectory modeling was employed to identify muscle strength trajectories, while multinomial logistic analysis determined predictors based on baseline characteristics. Mortality outcomes were assessed using a Cox proportional hazards model adjusted by inverse probability of treatment weighting.

Among 317 patients, with 37 deaths within 12 months, three muscle strength trajectories were identified: Low-Level Stability (38.17 %), Medium-Level Improvement (47.00 %), and High-Level Improvement (14.83 %). Older age (≥65 years), female gender, prolonged mechanical ventilation, and lower fat-free mass were associated with a higher likelihood of being in the Low-Level Stability group. After adjustment, the Medium-Level Improvement group had a 0.067 times lower 1-month mortality risk (95 % CI: 0.005-0.825) compared to the Low-Level Stability group, with no significant differences at 6 or 12 months (P > 0.05).

Three distinct muscle strength trajectories were identified: Low-Level Stability, Medium-Level Improvement, and High-Level Improvement. Older age, female sex, prolonged mechanical ventilation, and lower fat-free mass were strongly associated with the Low-Level Stability group, which had higher 1-month mortality.

Findings from this study underscore the importance of early identification of Low-Level Stability patients, particularly those who are older, female, require prolonged mechanical ventilation, or have reduced fat-free mass. Tailored early rehabilitation in these high-risk individuals may offer substantial clinical benefit.

The COVID-19 pandemic has left an indelible mark globally, presenting numerous challenges to public health. This crisis, while disruptive and impactful, has provided a unique opportunity to gather precious clinical data extensively. In this observational, case-control study, we utilized data collected at the Azienda Sanitaria Universitaria Friuli Centrale, Italy, to comprehensively characterize the immuno-inflammatory features in COVID-19 patients. Specifically, we employed multicolor flow cytometry, cytokine assays, and inflammatory biomarkers to elucidate the interplay between the infectious agent and the host's immune status. We characterized immuno-inflammatory profiles within the first 72 hours of hospital admission, stratified by age, disease severity, and time elapsed since symptom onset. Our findings indicate that patients admitted to the hospital shortly after symptom onset exhibit a distinct pattern compared to those who arrive later, characterized by a more active immune response and heightened cytokine activity, but lower markers of tissue damage. We used univariate and multivariate logistic regression models to identify informative markers for outcome severity. Predictors incorporating the immuno-inflammatory features significantly outperformed standard baselines, identifying up to 59% of patients with positive outcomes while maintaining a false omission rate as low as 4%. Overall, our study sheds light on the immuno-inflammatory aspects observed in COVID-19 patients prior to vaccination, providing insights for guiding the clinical management of first-time infections by a novel virus.

Acute respiratory distress syndrome (ARDS) is a severe form of organ dysfunction and a common postoperative complication. This study aims to develop a predictive model for ARDS in postoperative patients with gastrointestinal perforation to facilitate early detection and effective prevention.

In this single-center retrospective study, clinical data were collected from postoperative patients with gastrointestinal perforation admitted to the ICU in Hebei Provincial People's Hospital from October 2017 to May 2024. Univariate analysis and multifactorial logistic regression analysis were used to determine the independent risk factors for developing ARDS. Nomograms were developed to show predictive models, and the discrimination, calibration, and clinical usefulness of the models were assessed using the C-index, calibration plots, and decision curve analysis (DCA).

Two hundred patients were ultimately included for analysis. In the development cohort, 38 (27.1%) of 140 patients developed ARDS, and in the internal validation cohort, 13 (21.7%) of 60 patients developed ARDS. The multivariate logistic regression analysis revealed the site of perforation (OR = 0.164, P = 0.006), the duration of surgery (OR = 0.986, P = 0.008), BMI (OR = 1.197, P = 0.015), SOFA (OR = 1.443, P = 0.001), lactate (OR = 1.500, P = 0.017), and albumin (OR = 0.889, P = 0.007) as the independent risk factors for ARDS development. The area under the curve (AUC) was 0.921 (95% CI: 0.869, 0.973) for the development cohort and 0.894 (95% CI: 0.809, 0.978) for the validation cohort. The calibration curve and decision curve analysis (DCA) demonstrate that the nomogram possesses good predictive value and clinical practicability.

Our research introduced a nomogram that integrates six independent risk factors, facilitating the precise prediction of ARDS risk in postoperative patients following gastrointestinal perforation.

We tested the hypothesis that disseminated intravascular coagulation (DIC) predicts a poor prognosis in patients with out-of-hospital cardiac arrest (OHCA) treated with veno-arterial extracorporeal membrane oxygenation (VA-ECMO). Fifty-seven patients with cardiogenic OHCA who immediately underwent VA-ECMO upon admission to the emergency department were divided into 27 non-DIC and 30 DIC patients. DIC scores were calculated on admission and 24 h later (day 1). The primary outcome measure was the all-cause in-hospital mortality. The basic characteristics did not differ between the two groups; however, patients with DIC showed higher in-hospital mortality rates. Receiver operating characteristic curve analysis showed a moderate predictive ability of DIC scores on day 1 for in-hospital mortality. A lower probability of survival was observed in patients with DIC. The adjusted odds ratio for DIC on day 1 of in-hospital death was 5.67, confirmed by the adjusted hazard ratio of 3.472. The results indicate an association between DIC diagnosis 24 h following VA-ECMO induction for OHCA and poor outcome in these patients.

Machine learning (ML) is increasingly used to predict clinical deterioration in intensive care unit (ICU) patients through scoring systems. Although promising, such algorithms often overfit their training cohort and perform worse at new hospitals. Thus, external validation is a critical - but frequently overlooked - step to establish the reliability of predicted risk scores to translate them into clinical practice. We systematically reviewed how regularly external validation of ML-based risk scores is performed and how their performance changed in external data.

We searched MEDLINE, Web of Science, and arXiv for studies using ML to predict deterioration of ICU patients from routine data. We included primary research published in English before December 2023. We summarised how many studies were externally validated, assessing differences over time, by outcome, and by data source. For validated studies, we evaluated the change in area under the receiver operating characteristic (AUROC) attributable to external validation using linear mixed-effects models.

We included 572 studies, of which 84 (14.7%) were externally validated, increasing to 23.9% by 2023. Validated studies made disproportionate use of open-source data, with two well-known US datasets (MIMIC and eICU) accounting for 83.3% of studies. On average, AUROC was reduced by -0.037 (95% CI -0.052 to -0.027) in external data, with more than 0.05 reduction in 49.5% of studies.

External validation, although increasing, remains uncommon. Performance was generally lower in external data, questioning the reliability of some recently proposed ML-based scores. Interpretation of the results was challenged by an overreliance on the same few datasets, implicit differences in case mix, and exclusive use of AUROC.

Patients with Self-Reported Penicillin Allergy (SRPA) receive alternative antibiotics, which increase the length of stay and hospital costs, but the impact of SRPA on mortality in critically ill patients is not well described.

This was a single-center, retrospective analysis of routinely gathered clinical data for all intensive care unit (ICU) admissions over nine years. The primary outcome was 28-day mortality, which was analyzed using a time-to-event approach with multivariable models to adjust for confounding factors, including age, comorbidities, sex, and admission SOFA score (as a measure of organ dysfunction). Antibiotic prescriptions during the ICU stay were also interrogated and compared.

Of 35,319 admissions, 11.7% had SRPA. Compared with non-SRPA, patients with SRPA were more likely to be female (52.2% vs. 37.4%, p < 0.001) and had more comorbidities (p < 0.001) but had similar admission SOFA scores (median: 3.5 in both groups, p = 0.839). Patients with SRPA had significantly lower 28-day mortality (9.6% vs. 10.9%, p = 0.011). After multivariable adjustment for baseline characteristics, this effect persisted for unplanned (hazard ratio [HR]: 0.76, 95% CI: 0.68-0.86, p < 0.001), but not planned admissions (HR: 1.21, 95% CI: 0.92-1.58, p = 0.172). Whilst the duration of antibiotics in ICU was similar in the SRPA and non-SRPA groups (mean: 3.4 vs. 3.4 days, p = 0.663), the antibiotics used differed, with SRPA patients being significantly more likely to receive quinolones or other anti-Gram-positive antibiotics (p < 0.001).

SRPA was associated with a survival benefit that persisted after adjustment for confounders for unplanned ICU admissions. Patients with SRPA were more likely to receive antibiotics that are not active against anaerobic bacteria.