Other bulletins in this series include:

Breast Surgery

Tuesday, 1 July 2025

Critical Care Bulletin: July 2025

 

Association of hospital-level continuous kidney replacement therapy use and mortality in critically ill patients with acute kidney injury

Intensive Care Medicine, Published: 30 June 2025

Purpose

In numerous high-risk medical and surgical conditions, an increased volume of patients and procedures is associated with improved processes and survival. This study examined the association of hospital-level continuous kidney replacement therapy (CKRT) utilization rates with all-cause hospital mortality in critically ill patients with acute kidney injury (AKI).

Methods

This multicenter cohort study used data from patients admitted to the intensive care unit (ICU) within the Premier Incorporated AI (PINC-AI) database. Patients were critically ill adults with AKI receiving kidney replacement therapy (KRT) in U.S. hospitals that offered both CKRT and intermittent hemodialysis. Hospitals were characterized according to their CKRT utilization in the ICU, and risk-adjusted association with all-cause hospital mortality by day 90 was estimated.

Results

Among 49,685 patients with AKI admitted to 426 acute care U.S. hospitals and treated with KRT in the ICU, a higher hospital-level CKRT utilization rate was associated with lower patient-level risk-adjusted hospital mortality. Hospitals with higher CKRT utilization rates (CKRT use in31.5% of KRT patients per year) had a 15% lower adjusted probability of death compared with hospitals with lower CKRT utilization rates (CKRT use in<8% of KRT patients per year). When compared with the first quartile of hospital-level CKRT use, the third (adjusted hazard ratio [aHR], 0.93, 95%CI: 0.89–0.98) and fourth (aHR, 0.85, 95%CI: 0.81–0.89) quartiles were associated with lower risk-adjusted hospital mortality. Findings were consistent in several sensitivity analyses.

Conclusions

Among critically ill adults with AKI requiring KRT, treatment in hospitals with higher CKRT utilization rates was associated with reduced hospital mortality.

 

Short- and long-term outcomes and 15-year time trends of kidney replacement therapy in critically ill patients with obesity: an observational cohort

Intensive Care Medicine, Published: 30 June 2025

Purpose

Obesity is a public health challenge and a risk factor for chronic and acute kidney injury (AKI). The impact of kidney replacement therapy (KRT) on mortality in this specific population remains unclear. The objectives of this study were to evaluate the association between KRT and short- and long-term outcomes in patients with obesity, and to analyze trends over time in mortality as well as AKI and KRT incidence.

Methods

A retrospective analysis was conducted on all consecutive patients with obesity (body mass index30 kg/m2) admitted to a medico-surgical intensive care unit from 2009 to 2024. AKI was defined using KDIGO criteria. The primary outcome was 90-day mortality assessed with Kaplan–Meier curves and multivariate Cox analysis. Secondary outcomes included 1-year mortality and time trends assessed on 90-day mortality, AKI, and KRT incidence, as well as SAPS II score.

Results

Among 2,192 patients with obesity included in the study, 295 (13.5%) required KRT. The 90-day mortality was significantly higher in the KRT group compared to the non-KRT group (49.8% (95% CI [44.1–55.5] vs. 18.9% (95% CI [17.2–20.7]), p<0.0001). Multivariate analysis confirmed that KRT was independently associated with increased mortality. Over the 15-year period, AKI incidence decreased (p<0.001).

Conclusion

Among critically ill obese patients, KRT was independently associated with increased 90-day mortality. These results highlight the need for prospective studies to better define optimal management strategies in this high-risk population.

 

Self-reported symptoms in ICU patients and their impact on long-term outcomes—a prospective multicenter study

Intensive Care Medicine, Published: 30 June 2025

Purpose

The aim was to identify possible associations between patient symptom subgroups, identified while in the ICU, adjusted for clinical and demographic variables and PICS outcomes, three months after ICU admission.

Methods

A prospective multi-center cohort study. Based on reported symptom prevalence in the ICU, patients were categorized into a Low, Middle or High symptom subgroup. PICS outcomes were measured using validated questionnaires (i.e., Katz Index; Cognitive Failure questionnaire, The Hospital Anxiety and Depression Scale, Impact of Event Scale) three months after ICU admission. Possible associations between ICU symptom subgroups and dichotomized PICS outcomes were modeled using conditional backward logistic regression.

Results

We included 175 patients with a median age of 62 years (interquartile range 49–70) and 65.1% were males. Three months after ICU admission, 23.5% had physical disability, 12.1% cognitive failure, 18.3% anxiety, 16.6% depression, and 16.7% post-traumatic stress. In multivariate analyses, patients in the High symptom subgroup compared to Low/Middle symptom subgroups had higher odds for worse outcomes in all PICS domains. The odds ratios (95%CI) were 3.31 (1.25–8.78) for physical disability, 3.56 (1.20–10.60) for cognitive failure, 3.70 (1.41–9.67) for anxiety, 4.74 (1.79–12.58) for depression and 4.38 (1.49–12.84) for post-traumatic stress, respectively.

Conclusion

A subgroup of patients with high symptom burden during ICU stay had worse physical, cognitive and mental health outcomes three months after ICU admission compared to patients with medium/low symptom burden. Future studies are needed to evaluate if early interventions in the ICU can reduce long-time burden in the PICS domains in ICU survivors.

 

Prognosis of Nonagenarian ICU Patients A Bayesian analysis of prospective European studies

Annals of Intensive Care volume 15, Article number: 85 (2025)

Background

As the population ages, the number of very elderly patients (≥90 years, nonagenarians) admitted to intensive care units (ICUs) is increasing. This trend raises concerns about the appropriateness of ICU care for this age group, especially due to the uncertainty surrounding their prognosis. Some studies suggest that elderly ICU patients have outcomes similar to slightly younger patients, but skepticism remains due to clinical judgment, cultural attitudes, and resource allocation concerns.

Methods

We reassessed the 30-day mortality risk of nonagenarians admitted to ICUs using data from the VIP1, VIP2, and COVIP registries. Bayesian statistical methods, including Markov Chain Monte Carlo (MCMC) simulations, were used to estimate the relative risk (RR) of mortality for nonagenarians compared to octogenarians (80–89 years). Various prior assumptions (non-informative, pessimistic, and skeptical) were incorporated. The analysis adjusted for key variables such as SOFA score, frailty, and treatment limitations.

Results

A total of 8,408 patients were included, consisting of 807 nonagenarians and 7,601 octogenarians. The 30-day mortality rate was 45% for nonagenarians and 42% for octogenarians (p=0.12). Bayesian analysis revealed a high probability (81.197.9%) that nonagenarians face a higher 30-day mortality risk. However, the probability of a clinically significantly increase in mortality (RR>1.1) was moderate (28.934.7%), and the probability of a substantial increase (RR>1.2) was very low (0.031.9%).

Conclusion

Nonagenarians in the ICU have a slightly higher 30-day mortality risk compared to octogenarians, but the increase is unlikely to exceed clinically meaningful thresholds. Bayesian methods offer more refined mortality risk assessment, suggesting that ICU admission decisions should be based on individualized factors, not just age.

 

Impact of awake prone positioning duration on intubation or mortality in COVID-19 patients with acute respiratory failure: secondary analysis of a randomized clinical trial

Annals of Intensive Care volume 15, Article number: 84 (2025)

Background

Compared with shorter awake prone positioning (APP), prolonged APP (≥12 h daily) reduces the intubation rate in patients with COVID-19-related acute hypoxemic respiratory failure (AHRF). However, the optimal APP duration is uncertain. In this secondary analysis, we aimed to explore whether a longer APP duration is associated with improved outcomes and to identify the optimal duration of APP.

Methods

Data from a multicenter randomized controlled trial involving nonintubated COVID-19 patients with AHRF were analyzed. Daily APP duration over 7 days after randomization was recorded as the primary exposure in present study. The primary outcome was the time from randomization to APP failure, which was defined as a composite of tracheal intubation or mortality within 28 days. A Cox proportional hazards regression model was employed to elucidate the associations, and the daily duration of APP was treated as time dependent.

Results

A total of 409 patients were randomized in the original trial, and 408 were enrolled in this analysis. Among these patients, 105 (25.7%) experienced APP failure. A longer daily APP duration was associated with a lower risk of APP failure, with a hazard ratio (HR) of 0.93 (95% confidence interval (CI): 0.88–0.98), and the association was significant only during the first three days after randomization. There was a nonlinear relationship between the daily APP duration and the risk of APP failure (P=0.015 for nonlinearity). Compared with patients whose APP duration ranged from 8 to 12 h per day, patients with less than 8 h of APP per day had a greater risk of APP failure (HR 2.44, 95% CI 1.21–4.92), whereas extending APP beyond 12 h per day did not improve the outcomes further (HR 1.03, 95% CI 0.51–2.10, P=0.932).

Interpretation

A longer daily APP duration was associated with a reduced risk of APP failure in COVID-19-related AHRF patients, and the optimal APP duration was 8–12 h per day.

Clinical trial ClinicalTrials.gov: NCT05677984, Registered January 3, 2023. https://register.clinicaltrials.gov/prs/app/action/SelectProtocol?sid=S000CST9&selectaction=Edit&uid=U0000YKY&ts=4&cx=-x0muek

 

 

The effects of extracorporeal blood purification (oXiris®) in patients with cardiogenic shock who require VA-ECMO (CLEAN ECMO): a prospective, open-label, randomized controlled pilot study

Critical Care volume 29, Article number: 255 (2025)

Background

A systemic inflammatory response can contribute to poor outcomes in an advanced stage of cardiogenic shock (CS). We investigated the efficacy of extracorporeal endotoxin and cytokine adsorption using oXiris in patients with CS undergoing venoarterial extracorporeal membrane oxygenation (VA-ECMO).

Methods

In this prospective, single-center, randomized, open-label pilot trial, 40 patients with CS who were undergoing VA-ECMO were randomly assigned to receive either oXiris for 24 h (n=20) or usual care (n=20). The primary endpoint was endotoxin levels at 48 h. Secondary endpoints included changes in inflammatory cytokines, vasoactive-inotropic score (VIS), ECMO weaning success, and in-hospital and 30-day mortality.

Results

The median endotoxin levels at 48 h were 0.5 (IQR 0.4–1.0) in the oXiris group and 0.4 (IQR 0.2–0.5) in the control group, with no significant difference between them (P=0.097). The oXiris group showed significant temporal reductions in GDF-15 and IL-6 levels, with IL-6 revealing significant reductions from baseline to 24 h (P=0.020) and from baseline to 7 days (P=0.003). VIS decreased significantly from baseline to 48 h (-13.63, 95% CI: -20.90 -6.34, P<0.001) and 7 days (-12.19, 95% CI: -21.0 -3.31, P=0.007) in the oXiris group, but intergroup differences were insignificant. ECMO weaning success, duration of ECMO support, and mortality rates were similar between the groups.

Conclusion

In this pilot study conducted on CS patients requiring VA-ECMO, oXiris treatment did not significantly reduce endotoxin levels or improve patient centered clinical outcomes.

 

Can we predict the future of respiratory failure prediction?

Critical Care volume 29, Article number: 253 (2025)

Background

Mortality in patients with acute respiratory failure remains high. Predicting progression of acute respiratory failure may be critical to improving patient outcomes. Machine learning, a subset of artificial intelligence is a rapidly expanding area, which is being integrated into several areas of clinical medicine. This manuscript will address the knowledge gap in predicting the onset and progression of respiratory failure, provide a review of existing prognostic strategies, and provide a clinical perspective on the implementation and future integration of machine learning into clinical care.

Main body

Existing strategies for predicting respiratory failure, such as prediction scores and biomarkers, offer both strengths and limitations. While these tools provide some prognostic value, machine learning presents a promising, data-driven approach to prognostication in the intensive care unit. Machine learning has already shown success in various areas of clinical medicine, although relatively few algorithms target respiratory failure prediction specifically. As machine learning grows in the context of respiratory failure, outcomes such as the need for invasive mechanical ventilation and escalation of respiratory support (e.g. non-invasive ventilation) have been identified as key targets. However, the development and implementation of machine learning models in clinical care involves complex challenges. Future success will depend on rigorous model validation, clinician collaboration, thoughtful trial design, and the application of implementation science to ensure integration into clinical care.

Conclusion

Machine learning holds promise for optimizing treatment strategies and potentially improving outcomes in respiratory failure. However, further research and development are necessary to fully realize its potential in clinical practice.

 

What is the optimal approach to analyse ventilator-free days? A simulation study

Critical Care volume 29, Article number: 251 (2025)

Background

Ventilator-free days (VFDs) are a composite outcome in critical care research, reflecting both survival and mechanical ventilation duration. However, analysis methods for VFDs are inconsistent, with some focusing on counts and others on time-to-event outcomes, while other approaches such as the multistate model and the win ratio have emerged. We aimed to evaluate various statistical models through simulations to identify the optimal approach for analysing VFDs.

Methods

First, 16 datasets of 300 individuals were simulated, comparing a control group to an intervention with varying survival rates and ventilation durations. Various statistical models were evaluated for statistical power and Type I error rate. Four clinical trial datasets (LIVE study, NCT02149589; ARMA study, NCT00000579; ACURASYS study, NCT00299650; COVIDICUS study, NCT04344730) were then used to apply the same statistical models to analyse VFDs. Twelve statistical methods were evaluated, including count-based, time-to-event approaches, and the win-ratio. Additionally, sensitivity analyses were conducted.

Results

Most statistical methods effectively controlled Type I error rate, except for the zero-inflated and hurdle Poisson/negative binomial count submodels, as well as the cause-specific Cox regression model for death. The power to detect survival benefit and ventilation duration effects varied, with time-to-event approaches, the Mann–Whitney test, the proportional odds model and the win ratio generally performing best. Similar results were observed in sensitivity analyses. In the real datasets, the multistate model, the Mann–Whitney test, the proportional odds model and the win ratio generally showed a significant association between VFDs and randomisation groups.

Conclusions

The multistate model could be recommended as the optimal approach for analysing VFDs, as it outperformed the other methods and offers a more interpretable effect size than the proportional odds model and the win ratio.

 

Can we reliably use pulse pressure as a surrogate for stroke volume? Physiological background and potential clinical implications for shock resuscitation

Critical Care volume 29, Article number: 249 (2025)

Abstract

In critically ill patients, early hemodynamic assessment is essential for guiding shock resuscitation. While cardiac output (CO) is a key indicator of circulatory status, its measurement is often limited by technical and practical constraints. This perspective explores the physiological and clinical relevance of pulse pressure (PP) as a potential surrogate for stroke volume (SV), emphasizing its accessibility at the bedside. The paper discusses how factors such as arterial compliance, vascular tone, and pulse wave amplification influence the PP-SV relationship, often complicating interpretation in acute and complex hemodynamic states. It also examines the effects of vasopressors, vascular decoupling, and catheter site on PP measurements, particularly in septic shock. Despite its limitations, the review highlights how peripheral PP, when carefully interpreted, may aid in identifying low SV and guiding early resuscitation strategies.

 

Muscle weakness after critical illness: unravelling biological mechanisms and clinical hurdles

Critical Care volume 29, Article number: 248 (2025)

Abstract

Survivors of intensive care unit (ICU) are increasingly numerous because of better hospital care. However, several consequences of an ICU stay, known as post-intensive care syndrome, worsen long-term prognoses. A predominant feature in survivors is reduced muscle strength, mass, and physical function. This leads to lower exercise capacity, long-lasting physical disability, higher mortality risk, and subsequent health costs. While ICU-acquired muscle weakness has been extensively studied these past decades, underlying mechanisms of post-ICU muscle weakness remain poorly understood, and there is still no evidence-based treatment for improving long-term physical outcomes. One hypothesis, among others, could be that the pathophysiology is dynamic over time, differing between the acute ICU and post-ICU recovery periods. This narrative review aims to address the clinical, physiological and biological determinants of persistent muscle dysfunction in ICU survivors, with particular attention to the molecular, cellular and systemic mechanisms involved. Specifically, pre-ICU health factors such as obesity and sarcopenia, ICU-related complications and treatments, and post-ICU management all influence recovery. Dysfunctions in the neuroendocrine, vascular, neurological, and muscle systems contribute as physiological determinants of the muscle weakness. Complex and multifaceted biological mechanisms drive the post-ICU muscle dysfunction with mitochondrial and autophagy dysfunction, epigenetic modifications, cellular senescence, muscle inflammation with altered cell–cell communication, including dysfunction of immune cells, stem cell exhaustion and extracellular matrix remodelling. The review also sheds light on new and innovative therapeutic approaches and discusses future research directions. Emphasis is placed on the potential for multi-approach treatments that integrate nutritional, physical, and biological interventions. Addressing these aspects in a holistic and dynamic manner, from ICU to post-ICU phases, may provide avenues for mitigating the long-term burden of muscle weakness and physical disability in ICU survivors.

 

Post-extubation dysphagia in the ICU−a narrative review: epidemiology, mechanisms and clinical management (Update 2025)

Critical Care volume 29, Article number: 244 (2025) 

Abstract

Dysphagia (i.e. an impairment in swallowing function that impacts on safety or efficiency) is present in many intensive care unit (ICU) survivors, in particular following extubation (“post-extubation dysphagia”, PED). Despite the fact that pathomechanisms leading to PED are currently incompletely understood, local as well as central neurological and neuromuscular dysfunctions may be key to development of PED. Data from prospective large-scale clinical investigations with systematic screening demonstrate that PED affects about one out of five (about 20%) of mixed medical-surgical unplanned (emergency) ICU admissions. PED is associated with an increased risk for aspiration, aspiration-induced pneumonia, malnutrition, increased ICU resource use, decreased quality of life, prolonged ICU- and hospital length of stay and increased overall morbidity and mortality. Data demonstrate that PED is an independent predictor of 90-day mortality with increased risk of death up to about one year after ICU admission. PED may be a somewhat overlooked medical problem since in many ICUs, PED is currently not routinely screened for in all patients at risk (i.e. all ICU patients) following extubation. In this review, we update the available data on PED with a focus on epidemiology, risk factors, potential aetiology and treatment approaches, as well as clinical management on ICUs.

 

Decoding portal vein pulsatility: hemodynamic determinants in a post-hoc analysis of a prospective observational trial

Annals of Intensive Care volume 15, Article number: 81 (2025) 

Background

The portal vein pulsatility index (PVPI) reflects systemic congestion and is influenced by both volume status and right ventricular (RV) function. The mean systemic filling pressure analogue (Pmsa), derived from a mathematical model, estimates the interaction between stressed blood volume and systemic vascular compliance, serving as surrogate marker of volume status. This post-hoc analysis of an observational trial investigates the combined role of Pmsa and RV function as determinants of PVPI using echocardiography. Fifty-five mechanically ventilated patients with circulatory failure were included within 6 h of ICU admission following elective open-heart surgery. Fluid-tolerant patients (PVPI<50%) underwent a passive leg raising (PLR) test; fluid-responsive patients subsequently received 7 mL/kg of Ringers lactate. PVPI and Pmsa were measured at five timepoints: baseline (T1), after PLR (T2), upon returning to baseline (T3), after fluid administration (T4), and 20 min post-infusion (T5). RV function parameters, including RV to LV end-diastolic area ratio (RVEDA/LVEDA), tricuspid lateral annular systolic velocity (RV S’), RV fractional area change (RVFAC), pulmonary acceleration time (PAT), and right myocardial performance index (RIMP)—were assessed at T1, T4, and T5. Only fluid-responsive patients were evaluated beyond T3.

Results

At T1, robust multilinear regression including all patients identified RVEDA/LVEDA (β=10.38; p<0.001), RIMP (β=6.54; p=0.002), and RV S (β=0.60; p=0.002) as significant determinants of squared PVPI. In all patients, repeated measures correlation between Pmsa and PVPI was strong across T1-to-T3 (ρ=0.785; p<0.001), increasing from a non-significant correlation at T1 (ρ=0.215; p=0.115). Generalized estimating equations conducted only in fluid-responsive patients across T1, T4, and T5 identified Pmsa (β=4.19; p<0.001), RV S (β=5.84; p<0.001), RVEDA/LVEDA (β=34.85; p=0.018), and RIMP (β=35.28; p=0.039) as significant determinants of PVPI.

Conclusion

RV function and Pmsa are key determinants of PVPI. Their combined assessment may support an individualized congestion management by guiding interventions toward volume status, RV function, or both.

 

Sodium bicarbonate administration for metabolic acidosis in the intensive care unit: a target trial emulation

Intensive Care Medicine

Published: 10 June 2025, Volume 51, pages 1–9, (2025)

Purpose

Sodium bicarbonate is commonly administered to treat metabolic acidosis in intensive care units (ICUs). There is limited evidence from randomized trials to support this practice, and observational studies show conflicting results. Our aim was to perform a target trial emulation evaluating the effect of bicarbonate therapy on mortality.

Methods

Retrospective cohort study using data from 12 Australian ICUs. Inclusion criteria were adults with pH<7.3 and PCO245 mmHg within the first three days. We excluded repeat admissions, toxicology, diabetic ketoacidosis, and pre-existing end-stage renal failure. The treatment intervention was sodium-bicarbonate administration, and the primary outcome was 30-day ICU mortality with ICU discharge as a competing event. We evaluated multiple subgroups, including patients with acute kidney injury, requirement for vasoactive therapy, and pH<7.2. The primary model utilized a parametric g-computation and rolling entry matching was performed as a sensitivity analysis.

Results

We identified 6157 eligible admissions, of which 1764 (29%) received sodium bicarbonate. Bicarbonate therapy was associated with a 1.9% absolute mortality reduction for the primary analysis [risk ratio 0.86, 95% confidence interval (CI) 0.80 to 0.91], and significant benefits were seen across all subgroups evaluated. A similar point estimate of 2.1% was observed in the sensitivity analysis, with a sustained mortality reduction seen at 30 days.

Conclusion

In this target trial emulation, bicarbonate administration was associated with a small but statistically significant reduction in mortality for patients with metabolic acidosis. Large sample sizes would be required to demonstrate this effect in a randomized trial.