Background/Objectives: Artificial intelligence (AI), machine-learning (ML), and deep-learning (DL) models are increasingly used for prognostic prediction in intensive care unit (ICU) sepsis, but their clinical readiness remains uncertain. This systematic review aimed to evaluate AI-, ML-, and DL-based models for mortality and clinically relevant outcome prediction in adult ICU patients with sepsis or septic shock. Methods: PubMed/MEDLINE, Scopus, and the Cochrane Library were searched up to April 2026. Eligible studies included adult ICU sepsis or septic shock cohorts evaluating AI/ML/DL-based prognostic models. Screening, full-text assessment, and data extraction were performed independently by two reviewers. Outcomes, model families, validation strategies, discrimination, calibration, clinical utility, explainability, comparative performance versus conventional severity scores, risk of bias, and reporting completeness were synthesized. Risk of bias was assessed using PROBAST domains supplemented by PROBAST + AI considerations, and reporting completeness was evaluated according to TRIPOD/TRIPOD + AI domains. Results: Seventy-five studies were included, comprising 50 PubMed-derived and 25 additional Scopus-derived studies. AUROC or C-statistic was extractable in 64 studies, external validation was reported in 27, prospective evaluation in three, calibration in 38, decision-curve analysis or clinical utility assessment in 37, and explainability in 64. Across 17 directly extractable within-study comparisons from nine studies, AI/ML models usually, but not uniformly, achieved higher discrimination than conventional severity scores, with a median paired ΔAUROC of +0.108 (IQR, +0.082 to +0.148; range, −0.013 to +0.203). Externally validated fixed-horizon models showed clinically relevant but heterogeneous discrimination across sepsis phenotypes, with stronger evidence in selected sepsis-induced coagulopathy cohorts and more variable transportability in respiratory and liver-injury subgroups. However, 45 studies were judged at high risk of bias, mainly because of limitations in the analysis domain. Conclusions: AI/ML models for adult ICU sepsis show a recurrent signal of prognostic discrimination and often perform comparably to or better than conventional severity scores in directly extractable within-study comparisons; however, this signal should be interpreted cautiously given clinical and methodological heterogeneity, limited prospective validation, incomplete calibration, and frequent high risk of bias. The strongest evidence comes from externally validated, phenotype-specific models, although routine clinical implementation remains limited by heterogeneous endpoints, incomplete calibration, insufficient prospective validation, and scarce workflow-level evaluation. Future studies should shift from retrospective AUROC optimization toward calibrated, externally validated, clinically actionable, and workflow-integrated decision-support tools tested in prospective ICU settings.

(1) Background: In COVID-19 patients, the occurrence of thromboembolic complications contributes to disease progression and mortality. In patients at increased risk for thrombotic complications, therapeutic enoxaparin should be considered. However, critically ill COVID-19 patients could develop resistance to enoxaparin. Bivalirudin, a thrombin inhibitor, may be an alternative. This pilot multicenter randomized controlled trial aims to ascertain if bivalirudin may reduce the time spent under invasive mechanical ventilation, as compared to enoxaparin. (2) Methods: Intubated COVID-19 patients at risk for thrombo-embolic complications were randomized to receive therapeutic doses of enoxaparin or bivalirudin. We ascertained the time spent under invasive mechanical ventilation during the first 28 days from Intensive Care Unit (ICU) admission. A standardized weaning protocol was implemented in all centers. In addition, we assessed the occurrence of thromboembolic complications, the number of patients requiring percutaneous tracheostomy, the gas exchange, the reintubation rate, the ICU length of stay, the ICU and 28-days mortalities. (3) Results: We enrolled 58 consecutive patients. Bivalirudin did not reduce the time spent under invasive mechanical ventilation as compared to enoxaparin (12 [8; 13] vs. 13 [10; 15] days, respectively; p = 0.078). Thrombotic (p = 0.056) and embolic (p = 0.423) complications, need for tracheostomy (p = 0.423) or reintubation (p = 0.999), the ICU length of stay (p = 0.076) and mortality (p = 0.777) were also similar between treatments. Patients randomized to bivalirudin showed a higher oxygenation at day 7 and 15 after randomization, when compared to enoxaparin group. (4) Conclusions: In intubated COVID-19 patients at increased risk for thromboembolic complications, bivalirudin did not reduce the time spent under invasive mechanical ventilation, nor improved any other clinical outcomes.

Sudden cardiac death (SCD) remains a major challenge in forensic medicine, representing a leading cause of natural mortality and frequently occurring in individuals without antecedent symptoms. Although conventional autopsy and histology remain the cornerstones of investigation, up to 10–15% of cases are classified as “autopsy-negative sudden unexplained death,” underscoring the need for complementary diagnostic tools. In recent years, post-mortem biochemistry and molecular approaches have become essential to narrowing this gap. Classical protein markers of myocardial necrosis (cardiac troponins, CK-MB, H-FABP, GPBB) continue to play a fundamental role, though their interpretation is influenced by post-mortem interval and sampling site. Peptide biomarkers reflecting hemodynamic stress (BNP, NT-proBNP, copeptin, sST2) offer additional insight into cardiac dysfunction and ischemic burden, while inflammatory and immunohistochemical markers (CRP, IL-6, fibronectin, desmin, C5b-9, S100A1) assist in detecting early ischemia and myocarditis when routine histology is inconclusive. Beyond these traditional markers, molecular signatures—including cardiac-specific microRNAs, exosomal RNA, proteomic alterations, and metabolomic fingerprints—provide innovative perspectives on metabolic collapse and arrhythmic mechanisms. Molecular autopsy through next-generation sequencing has further expanded diagnostic capability by identifying pathogenic variants associated with channelopathies and cardiomyopathies, enabling both cause-of-death clarification and cascade screening in families. Emerging multi-omics and artificial intelligence frameworks promise to integrate these heterogeneous data into standardized and robust interpretive models. Pre- and post-analytical considerations, together with medico-legal implications ranging from malpractice evaluation to the management of genetic information, remain essential components of this evolving field. Overall, the incorporation of validated biomarkers into harmonized international protocols, increasingly supported by AI, represents the next frontier in forensic cardiology.

Introduction: High-frequency percussive ventilation (HFPV) is a ventilation mode characterised by high-frequency breaths. This study investigated the impact of HFPV on gas exchange and clinical outcomes in acute respiratory failure (ARF) patients during spontaneous breathing, noninvasive ventilation (NIV) and invasive mechanical ventilation (iMV).Methods: This systematic review included randomised and nonrandomised studies up to August 2023. Inclusion criteria focused on adult ARF patients, HFPV application, comparisons with other ventilation modes, and outcomes related to oxygenation and clinical parameters. A pooled data analysis was performed comparing HFPV with iMV concerning gas exchange, pulmonary infection and mortality.Results: Of the 51 identified records, 29 met the inclusion criteria. HFPV was safely and effectively applied to ARF patients during spontaneous breathing or NIV, improving oxygenation. For patients who underwent iMV, HFPV significantly enhanced oxygenation and the arterial partial pressure of carbon dioxide, reduced pulmonary infection occurrence and improved survival. Barotrauma rates were not elevated with HFPV, and haemodynamic stability remained unaffected. HFPV was also utilised in patients undergoing extracorporeal membrane oxygenation, resulting in improved lung recruitment and oxygenation.Conclusion: HFPV had favourable effects on physiological and certain clinical outcomes in ARF patients. However, the overall evidence quality remains weak, necessitating large-scale randomised controlled trials for definitive conclusions.

Background: Acute respiratory distress syndrome (ARDS) presents a significant challenge in critical care settings, characterized by compromised gas exchange, necessitating in the most severe cases interventions such as veno-venous extracorporeal membrane oxygenation (vv-ECMO) when conventional therapies fail. Critically ill ARDS patients on vv-ECMO may experience several complications. Limited data exist comparing complication rates between COVID-19 and non-COVID-19 ARDS patients undergoing vv-ECMO. This retrospective observational study aimed to assess and compare complications in these patient cohorts. Methods: We retrospectively analyzed the medical records of all patients receiving vv-ECMO for ARDS between March 2020 and March 2022. We recorded the baseline characteristics, the disease course and complication (barotrauma, bleeding, thrombosis) before and after ECMO cannulation, and clinical outcomes (mechanical ventilation and ECMO duration, intensive care unit, and hospital lengths of stay and mortalities). Data were compared between COVID-19 and non-COVID-19 patients. In addition, we compared survived and deceased patients. Results: Sixty-four patients were included. COVID-19 patients (n = 25) showed higher rates of pneumothorax (28% vs. 8%, p = 0.039) with subcutaneous emphysema (24% vs. 5%, p = 0.048) and longer non-invasive ventilation duration before vv-ECMO cannulation (2 [1; 4] vs. 0 [0; 1] days, p = <0.001), compared to non-COVID-19 patients (n = 39). However, complication rates and clinical outcomes post-vv-ECMO were similar between groups. Survival analysis revealed no significant differences in pre-vv-ECMO complications, but non-surviving patients had a trend toward higher complication rates and more pleural effusions post-vv-ECMO. Conclusions: COVID-19 patients on vv-ECMO exhibit higher pneumothorax rates with subcutaneous emphysema pre-cannulation; post-cannulation complications are comparable to non-COVID-19 patients.