Sepsis-associated acute kidney injury (SA-AKI) is a major complication in the intensive care unit (ICU), associated with increased mortality and progression to chronic kidney disease (CKD) and end-stage renal disease (ESRD). This review summarizes current advances in the definition, epidemiology, pathophysiology, diagnosis, experimental modeling, and therapeutic strategies of SA-AKI, with emphasis on translational challenges. The conceptual framework of SA-AKI has evolved from global renal hypoperfusion to a multifactorial, predominantly non-ischemic process characterized by inflammatory dysregulation, microcirculatory dysfunction, and adaptive metabolic reprogramming. Tubular metabolic shutdown is increasingly recognized as a protective yet maladaptive response contributing to functional decline. Regulated cell death pathways, including pyroptosis and ferroptosis, further drive tubular injury. In parallel, organ crosstalk—particularly along the lung–gut–heart–kidney axis—amplifies renal dysfunction during sepsis. Recent diagnostic advances are highlighted, including the 2023 Acute Disease Quality Initiative (ADQI) consensus and emerging biomarkers reflecting distinct pathophysiological domains: tubular injury (neutrophil gelatinase-associated lipocalin [NGAL] and kidney injury molecule-1 [KIM-1]), cell cycle arrest (Tissue Inhibitor of Metalloproteinases-2[TIMP-2]·insulin-like growth factor-binding protein 7 [IGFBP7]), and endothelial dysfunction (angiopoietin-2 [Angpt-2] and the tyrosine kinase with immunoglobulin-like and EGF-like domains 2 receptor [Tie2]), supporting early risk stratification beyond serum creatinine (sCr). Experimental models ranging from endotoxemia and cecal ligation and puncture to two-hit and comorbidity-based paradigms are critically appraised. Finally, current and investigational therapies—including hemodynamic optimization, extracorporeal cytokine removal, mitochondria-targeted interventions, and regenerative approaches—are discussed in the context of persistent barriers to clinical translation.
Background The original Sequential Organ Failure Assessment (SOFA-1) score was recently revised to SOFA-2. The kidney component of SOFA-2 is based on serum creatinine (Scr) and urine output (UO). A SOFA score based on a 6-hour urine collection was shown to be a better mortality predictor than the SOFA-1. This study aimed to compare the mortality predictive performance of a novel kidney component based on a 6-hour urine collection, vs. both the kidney component of SOFA-2, and an estimated glomerular filtration rate (eGFR)-based component. Methods This is a post hoc analysis of a single-center prospective observational study that was performed in a mixed intensive care unit (ICU) of a university-affiliated tertiary hospital. Urinary creatinine (Ucr) clearance (CrCl6hr) was calculated daily based on a 6-hour urine collection performed during the first five nights of admission. eGFR was also calculated daily. Patients were assigned to five groups (0–4) based on Kidney Disease: Improving Global Outcomes (KDIGO) chronic kidney disease definitions; modified SOFA scores were calculated accordingly. SOFA-2 kidney component was assigned as well. The predictive abilities of the different kidney components (based on CrCl6hr, eGFR, and the SOFA-2) for ICU- and 90-day mortality were compared by evaluating their area under the receiver–operator curve (AUROC) values. Results A total of 200 patients were included in the study. Ninety-day mortality was 29.5% and ICU mortality was 12.0%. The AUROC of CrCl6hr-based kidney component for 90-day mortality was significantly higher than that of SOFA-2 kidney component in all examined days (day 1: 0.70 vs. 0.55, P <0.001; day 2: 0.74 vs. 0.65; P=0.023; day 3: 0.75 vs. 0.65, P=0.013; day 4: 0.76 vs. 0.62, P=0.001; day 5: 0.73 vs. 0.51, P <0.001). It was also higher compared with an eGFR-based kidney component in day 2 (0.74 vs. 0.68, P=0.015) and day 4 (0.76 vs. 0.68, P=0.008). Regarding ICU-mortality, the AUROC of CrCl6hr-based kidney component was significantly higher than that of SOFA-2 kidney component in day 1 (0.75 vs. 0.63, P=0.017) and day 5 (0.75 vs. 0.64, P=0.022), and significantly higher than that of the eGFR kidney component in Day 4 (0.81 vs. 0,73, P=0.015). Conclusions An isolated novel SOFA kidney component based on a 6-hour urine collection is a better predictor of mortality than the isolated SOFA-2-kidney component. Mortality prediction of an eGFR-based kidney component is not significantly different from that of a CrCl-based component.Trial Registration Clinicaltrials.gov Identifier: NCT06779331
Background Consensus-based recommendations have recently been developed to guide the delivery of quality care for very old patients (age ≥80 years) in the intensive care unit (ICU). To examine the influence of healthcare provider gender on agreement with consensus-based recommendations for the management of very old patients in ICUs. Methods Exploratory secondary analysis of survey responses from an international multiprofessional Delphi consensus on management of very old patients in ICUs by the European Society of Intensive Care Medicine. The primary analysis compared mean agreement scores between female and male participants using independent-samples t-tests. Secondary analyses included comparisons of strong agreement proportions and adjusted analyses using two-way analysis of variance and multivariable regression models including provider category and age. Results The analysis included 128 participants with available gender and clinical activity data (53 female and 75 male respondents). Female participants reported significantly higher agreement scores for several recommendations involving multidisciplinary input, including medication review, clinical pharmacist involvement, transfer to geriatric wards, dietitian support, and speech and language therapy (mean differences of approximately 7.4–11.0 points on a 0–100 scale, P <0.050). In contrast, male participants reported higher agreement for decisions to limit life-sustaining therapy based on expected functional outcome. Analysis of response proportions showed similar patterns, with females more frequently expressing strong agreement. In adjusted analyses, gender– provider specialty interactions were largely non-significant, indicating consistent effects across intensive care, emergency medicine, and geriatric medicine, except for medication review statements where a significant interaction (P=0.021) suggested variability across specialties. Conclusions In this exploratory secondary analysis, participant gender was associated with differences in agreement with selected consensus-based recommendations for the management of very old patients in ICUs, particularly those involving multidisciplinary team input. These findings highlight differences in perspectives toward specific recommendations within this international expert panel.
Background Despite substantial advances in critical care nephrology, the real-world implementation of diagnostic tools and evidence-based preventive strategies for acute kidney injury (AKI) in critically ill patients remains poorly defined. Understanding the gap between guideline recommendations and bedside practice is essential to inform targeted quality-improvement initiatives. The AKI-PURIFY survey was designed to provide a contemporary international overview of AKI diagnosis and management in the ICU, with emphasis on implementation of the Kidney Disease: Improving Global Outcomes (KDIGO) preventive strategies and determinants with its adoption. We hypothesized that implementation of evidence-based preventive care varies across healthcare settings and is influenced by clinician training and organizational capacity. Methods AKI-PURIFY was a multicenter, international online survey endorsed by the European Society of Intensive Care Medicine (ESICM), conducted between February and April 2024. Healthcare professionals involved in intensive care unit (ICU) management of AKI were invited to participate. The survey explored diagnostic strategies, implementation of the KDIGO kidney prevention strategy, renal replacement therapy practices, and perceived organizational and educational barriers. Multivariable logistic regression analyses were conducted to identify factors independently associated with the use of diagnostic tools and the implementation of preventive strategies. Results A total of 703 respondents completed the survey. Routine implementation of the KDIGO kidney prevention strategy was reported by 36.3% of participants (47.0% among valid responses). Diagnostic practices varied across ICU and resource settings, and the use of AKI damage biomarkers remained low. In adjusted analyses, inadequate training (odds ratio [OR]=0.56, 95% confidence interval [CI]: 0.37 to 0.84, P=0.007) and higher organizational barrier scores (OR=0.78, 95% CI: 0.63 to 0.97, P=0.030) were independently associated with lower implementation. Clear income-level gradients were observed across diagnostic strategies, training adequacy, and organizational capacity. Conclusions Adoption of guideline-recommended AKI prevention strategies remains inconsistent worldwide and is influenced by modifiable educational and system-level factors. Addressing this gap will require coordinated, context-sensitive strategies integrating structured training and scalable quality-improvement frameworks.
Background Community-acquired adult bacterial meningitis (CABM) is a severe condition associated with significant morbidity and mortality, particularly in patients requiring intensive care unit (ICU) admission. Previous studies have identified several risk factors for poor outcomes, including advanced age, immunosuppression, and delayed treatment. However, data on ICU-specific prognostic factors remain limited. This study aimed to identify factors associated with ICU admission, predictors of 30-day mortality, and long-term outcomes in ICU-admitted patients with CABM. Methods This post hoc analysis utilized data from 27 ICUs from the COMBAT cohort, a prospective multicenter study in France from February 2013 to July 2015. The study included patients over 18 years old with confirmed CABM. Clinical, microbiological, and ICU-specific data, including Sequential Organ Failure Assessment (SOFA) scores, were collected. The primary outcome was 30-day mortality, and secondary outcomes included ICU admission predictors and 12-month functional status. Multivariable logistic regression models with pre-specified clinically relevant variables were used to identify factors associated with 30-day mortality and ICU admission; model performance was assessed using the c-statistic and Hosmer–Lemeshow goodness-of-fit test. Results Of the 210 patients admitted to the ICU, 36 (17.1%) died within 30 days. Significant predictors of mortality included elevated SOFA scores at ICU admission and on day 3 in univariate and multivariate analyses. Age and non-meningococcal CABM were associated with mortality in the univariate analysis but not in the multivariate analysis. ICU admission was associated with altered consciousness and Neisseria meningitidis CABM according to univariate and multivariate analyses. In multivariable analysis, higher SOFA scores at admission (OR=1.49, 95% CI: 1.28 to 1.74) and on day 3 (OR=1.51, 95% CI: 1.28 to 1.78) were independently associated with 30-day mortality, while altered consciousness (OR=7.51, 95% CI: 4.65 to 15.42) and N. meningitidis identification in cerebrospinal fluid (OR=3.45, 95% CI: 1.15 to 10.00) were independently associated with ICU admission. At 12 months, 51.0% of survivors had poor outcomes on the Glasgow Outcome Scale, with significant neurofunctional impairments observed. Conclusion This study identified key prognostic factors for ICU admission and 30-day mortality in CABM patients. Altered consciousness and N. meningitidis infection were associated with ICU admission, whereas elevated SOFA scores at admission and on day 3 were associated with mortality. Despite intensive care, mortality rates remain high, and long-term neurofunctional deficits are common among survivors. Trial registration ClinicalTrials.gov identifier NCT01730690.
Infections of the central nervous system can require immediate admission to an intensive care unit (ICU) because of initial severe neurologic presentation or systemic complications or may be encountered during the course of a hospital stay because of secondary complications. Brain infections include community-acquired or healthcare-associated meningitis, ventriculitis, encephalitis, and brain abscesses, which share specific clinical features but have different pathophysiology, risk factors, and outcomes. In this review, we will provide a practice-based overview of severe presentations of brain infections as encountered by the intensivist, focusing on early decision-making, sequence of diagnostic and management decisions, and encountered complications. We illustrate the review with case examples to offer real-life illustration of challenging decision-points. Key take-home points include understanding the first-hour priorities for de novo presentation of meningitis and encephalitis with focus on early initiation of antimicrobial treatment; awareness of limitations in cerebrospinal fluid (CSF)-based diagnostics for early encephalitis and in healthcare-associated meningitis and ventriculitis, and the importance of expedited, non-CSF-based diagnostic approach to brain abscesses.
Patients with spontaneous intracerebral hemorrhage (sICH) are at high risk for venous thromboembolism (VTE), a complication strongly associated with adverse clinical outcomes. While prophylactic anticoagulation has been shown to effectively reduce VTE incidence, significant uncertainty remains regarding its safety and the optimal timing of initiation. This article comprehensively reviews key aspects including anticoagulant selection, patient eligibility, therapeutic time windows, and current clinical challenges. Research gaps persist in defining appropriate anticoagulation strategies according to drug type, initiation timing, and sICH subtypes (e.g., lobar, deep, or amyloid-related hemorrhage). Future efforts should focus on conducting large-scale, multicenter, prospective clinical trials to validate the efficacy and safety of novel anticoagulants. Concurrently, integrating machine learning to develop high-precision risk prediction models, tailoring individualized treatment approaches, and establishing multidisciplinary collaborative research frameworks are essential steps toward advancing the standardization and rationalization of prophylactic anticoagulation in patients with severe sICH.
Background: Hand-Held Vital Microscopy (HVM) has been utilized in clinical investigations to measure microcirculation, providing valuable insights into the pathophysiology of critical illness and its treatment. However, its clinical implementation has been hampered by several shortcomings. These included the need for automatic analysis software, for instant image stabilization, and for information regarding microcirculatory oxygenation. The aim of this study is to describe an HVM which has realized these research objectives. Methods: This preclinical study introduces the OxyCam, a novel HVM device. OxyCam stabilizes images during recording and subsequently provides essential microcirculatory functional parameters through automatic analysis. Green and blue illuminated images, sequentially recorded and ratio imaged, allow calculation of the hemoglobin (Hb) oxygen saturation (µHbO2sat) of RBCs and are presented in pseudo color images. We validated the OxyCam on the hindlimb muscle of 9 male Wistar albino rats, mechanically ventilated at different levels of inspired oxygen (FiO2). Comparison was made between OxyCam-green and blue images and to its predecessor the Cytocam. OxyCam-generated µHbO2sat values were compared to HbO2sat fiber spectrophotometry (O2C). Results: Green and blue illumination from the OxyCam yielded higher total vessel density (OxyCam green [31.4±4.3] mm/mm2; OxyCam blue [32.5±4.2] mm/mm2, Cytocam [28.5±3.6] mm/mm2) and functional capillary density (OxyCam green [25.4±6.4] mm/mm2; OxyCam blue [26.2±7.9] mm/mm2; Cytocam [23.0±4.5] mm/mm2) than the Cytocam images OxyCam-blue gave higher tissue RBC perfusion values (OxyCam green [55.3±18.0] µm/min; OxyCam blue [69.5±25.5] µm/min; Cytocam [58.3±15.3] µm/min) than the OxyCam-green or Cytocam images. OxyCam-blue and green images gave better image quality than the Cytocam images. µHbO2sat imaging with the OxyCam was validated by demonstrating expected changes in µHbsat during different FiO2 levels (100%, 60%, 21%, 0%). Furthermore, a correlation between the mean µHbO2sat of images and independently measured µHbsat by tissue spectrophotometry during different levels of FiO2 was found. Finally, µHbO2sat images in pseudo-color (red to blue) are shown for the individual microvessels at different FiO2 levels. Conclusions: The OxyCam has been validated for measuring tissue perfusion and oxygenation and can be considered for point-of-care diagnosis of the microcirculation at the bedside.
Background:Effective sedation in intensive care unit (ICU) patients is critical for minimizing agitation, ensuring safety, and facilitating mechanical ventilation. Remimazolam tosylate, a novel benzodiazepine with rapid onset and short duration, metabolized by tissue esterases, represents a promising agent for ICU sedation. This study aimed to evaluate the efficacy and safety of remimazolam tosylate in short-term sedation for ICU patients. Methods:This was a multicenter, randomized, dose-finding, phase 2 study. From October 12, 2022 through April 19, 2023, ICU patients requiring mechanical ventilation and sedation for ≥6 h were randomized (1:1) into two groups: low adjustment rate (LAR) group received intravenous remimazolam tosylate with a maintenance rate adjustment of 0.1 mg/kg/h, and high adjustment rate (HAR) group with 0.2 mg/(kg·h). Both groups were initiated at an infusion rate of 0.2 mg/(kg·h) with the same loading dose (0.08 mg/kg) and maintenance range (0-2.0 mg/(kg·h)). The primary outcome was sedation success rate, defined as the percentage of time maintaining targeted sedation for ≥70% of the drug administration period without rescue sedation. Results:A total of 59 patients completed the study (LAR, n=30; HAR, n=29). Within the limited sample, all patients achieved sedation success. The median percentage of time within the target sedation maintenance relative to remimazolam administration duration was 97.9% (IQR: 93.8%-99.7%) for LAR group and 98.1% (IQR: 93.6%-99.8%) for HAR group (P=0.7617). No patients required rescue sedation with propofol, and only one patient (3.3%) in LAR group required rescue analgesia with sufentanil. Most treatment-emergent adverse events (TEAEs) were mild or moderate, with hypotension (18.6%), hypertension (18.6%), and anemia (16.9%) being the most common. One serious TEAE (3.4%, disease progression) and one death were reported in HAR group, both were assessed as unrelated to remimazolam tosylate. Conclusions:Remimazolam tosylate demonstrated high efficacy with manageable safety profile for short-term sedation in mechanically ventilated ICU patients at both 0.1 mg/(kg·h) and 0.2 mg/(kg·h) adjustment rates.Trial registration: ClinicalTrials.gov, NCT05152303.
