- This retrospective claims-based cohort of 7,898 U.S. adults received invasive mechanical ventilation in ICUs in 2018–2019, excluding those with prior psychoactive prescriptions.
- The exposure was new psychoactive medication prescribing in the 30 days after discharge, judged against diagnostic codes that could justify each prescription class.
- Of 416 patients (5.3%) who received 484 new prescriptions, 207 (42.8%) lacked a documented supporting diagnosis, including 52.4% of antipsychotic prescriptions.
- Unsupported prescribing persisted for up to 2 years, reaching 24.7% at 31–90 days and 42.7% at 181–365 days, and the authors call for prospective research.
Critical care
Research from anesthesiology, pain and critical care journals, sourced from PubMed.
Sorted by publication date, not import date. Recent filters exclude papers with only a publication year; find those under All years.
AI-generated abstract summaries can contain errors. Verify findings against the original paper before applying them to patient care.
Last successful paper check: 16 Nov 2026 17:18 UTC.
0 newly discovered papers saved; these are not necessarily published today.
2 papers awaiting AI summaries. Summary processing had an error; paper-fetch status is separate.
Background checks run while the server is active. This does not guarantee a daily run if hosting sleeps.
2 papers awaiting AI summaries. Summary processing had an error; paper-fetch status is separate.
Background checks run while the server is active. This does not guarantee a daily run if hosting sleeps.
- Retrospective multicenter observational cohort of 973 critically ill adults receiving CRRT for at least 24 hours across two European ICUs.
- The exposure was net ultrafiltration rate, grouped as none, low, moderate, or high, analyzed alongside daily fluid balance change; no intervention was assigned.
- No independent link between NUF rate and 28-day mortality was found after accounting for fluid balance, and the apparent survival benefit of higher NUF appeared mediated by a more negative balance.
- More positive fluid balance was associated with worse outcomes, including 28-day mortality, so the authors suggest using NUF to reach patient-specific fluid balance targets.
- Retrospective observational study at one tertiary academic center of adults with chronic heart failure, including stable outpatients and those with acute decompensation.
- No intervention; clinical capillary leak indices (SOFA, Capillary Leak Index, Vascular Leak Index, Hematocrit-Albumin Difference) were compared with albumin transcapillary escape rate (TERalb).
- Correlations were weak or absent across analyses, but the abstract reports no sample size, coefficients, or p values.
- The authors conclude these indices do not accurately reflect capillary permeability in heart failure and should be read as context-dependent markers.
- A quality improvement study used serial anonymous surveys at two Johns Hopkins neurocritical care units, with voluntary multidisciplinary staff surveyed from baseline through 2 years.
- A stakeholder-led bundle included a shared vision, a zero-tolerance pledge, leadership communication, recognition, team-building, social events, workflow changes, and education restructuring; no concurrent control group is described.
- Median emotional exhaustion fell from 25 to 9, depersonalization from 9 to 5, and personal accomplishment rose from 29 to 40 (all p < 0.001), with sustained gains across all Team Dynamics Survey domains.
- The mean fellow in-service examination score rose 14% from April 2024 to April 2025 (p < 0.001), which the authors describe as temporally associated with the intervention.
- Post hoc analysis of a Japanese multicenter RCT in 339 adults aged 65 or older who survived septic shock to 90 days, from 513 eligible patients.
- Frailty at 90 days was graded by Clinical Frailty Scale, and associations with baseline characteristics and treatments were assessed by ordinal regression.
- Higher presepsis CFS (aOR 2.03 per point), higher SOFA (aOR 1.16 per point), and corticosteroid use (aOR 1.72) were each independently associated with more severe frailty.
- Corticosteroid use was linked to frailty mainly in patients already frail at baseline, while SOFA mattered most with lactate of 2 mmol/L or more.
- Multicenter registry-based cohort of adult ICU patients receiving oxygen or respiratory support at nine University of Toronto-affiliated ICUs, 2014-2023 (n=21,714 recorded).
- AHRF was defined by PaO2/FiO2 ≤300 mm Hg, or SpO2/FiO2 ≤315 when PaO2 was unavailable, within 24 hours of admission; no intervention was assigned.
- 10,832 patients (50%) met AHRF criteria, 76% of them on invasive ventilation; ICU mortality was 23% overall and 47% in ventilated severe AHRF.
- Greater AHRF severity was associated with longer ICU stay and ventilation and fewer ventilator-free days, and the authors propose a standardized AHRF definition to aid recognition and comparison.
- This commentary, not a primary study, examines the widening gap between how fast medical evidence is generated and how fast new algorithms appear.
- It uses sepsis and ARDS as examples of critical care fields with many predictive models that rarely reach routine practice.
- The authors argue that evidence takes years to build, so populations, co-interventions, and standards of care may change before results arrive.
- They conclude that rapid algorithm development and personalization strain conventional proof, framing the problem as epistemic rather than purely technical.
- Multicenter, open-label, cluster-randomized trial of 381 analyzed adults with suspected septic shock managed by prehospital mobile ICU teams in France.
- The 1-hour bundle (antibiotics, up to 35 mL/kg isotonic saline, norepinephrine, and hydrocortisone as needed) was compared with usual care.
- 28-day mortality was 22% (23 of 103) with the bundle versus 27% (76 of 278) with usual care (RR 0.81, 95% CI 0.61-1.08; p = 0.16).
- The authors found no significant mortality reduction and suggest center enrollment imbalances and a small difference in care delivered may explain this.
- This observational study used a Japanese nationwide ICU registry and analyzed 18,794 propensity-matched adults admitted from 2018 to 2023.
- The exposure was RRT during the ICU stay, and a machine learning causal forest was used to estimate varying treatment effects across patients.
- Overall, RRT was associated with higher hospital mortality (risk difference 6.1 points, 95% CI 3.4 to 8.7), but the causal forest showed heterogeneity (rank-weighted average effect -8.2 points, 95% CI -9.7 to -6.7).
- Patients with elevated creatinine, reduced urine output, and higher BUN appeared to benefit from RRT, while those with preserved renal function appeared more likely to be harmed.
- This serial cross-sectional study used claims from adult Medicare fee-for-service beneficiaries with critical care bills at acute care hospitals, January 2018 to September 2024.
- The exposure was telemedicine critical care (TCC) billing, identified from provider billing codes and compared with critically ill patients not billed for TCC.
- TCC billing rose from 0.002% of critical care bills before the pandemic to 0.01% during and after it, remaining a small share overall.
- TCC billing was less likely at rural (aOR 0.70; 95% CI 0.55-0.89) and critical access hospitals (aOR 0.59; 95% CI 0.47-0.73), and more likely at safety-net hospitals (aOR 1.33; 95% CI 1.04-1.70).
- Systematic review and meta-analysis of 75 studies (2,023 critically ill patients, 642 controls) from 1988 to 2024, mostly using vastus lateralis biopsies.
- Random-effects pooling compared myofiber cross-sectional area and protein turnover in critically ill adults against non-ICU controls.
- Myofiber area was 22% lower before ICU discharge (MD -689 µm²; 95% CI -1265 to -113; p = 0.02) and after discharge (MD -775 µm²; 95% CI -1512 to -37; p = 0.04).
- Protein synthesis did not differ significantly from controls (MD 0.007%/hr; 95% CI -0.010 to 0.027; p = 0.36), while degradation markers were consistently higher.
- Nationwide cohort of 80,263 adults admitted to South Korean ICUs with cardiogenic shock from 2017 to 2022, using National Health Insurance data.
- Exposure was region-level estimated hospital arrival time, grouped as under 10, 10-30, 30-60, and over 60 minutes, rather than individually measured arrival times.
- In-hospital mortality rose with longer arrival time (29.6% to 38.8%); adjusted odds ratios vs under 10 minutes were 1.27, 1.24, and 1.51 across the longer categories.
- At 1 year, adjusted hazard ratios for all-cause mortality were 1.16, 1.16, and 1.31, and the authors call for regional shock networks and prehospital triage.
- Preclinical study in 7-week-old male C57Bl/6J mice randomized to AKI, renal ischemia-reperfusion alone, or control groups.
- AKI was modeled by removing the contralateral kidney before unilateral renal ischemia-reperfusion injury.
- AKI mice showed neurologic impairment, reduced spontaneous motricity, and cerebral Evans blue extravasation, none of which appeared with ischemia-reperfusion alone; 68Ga-DTPA PET/CT supported BBB disruption.
- Electron microscopy showed more extracellular vesicles in cerebral endothelial cells of AKI mice, which the authors link to early BBB permeability and transcytosis.
- Pilot multicenter randomized trial of 35 intubated adults with heatstroke across ten tertiary ICUs in China.
- Compared rapid cooling to 35.0°C held for 48 hours against conventional cooling targeting 36.5–37.5°C, with cooling only above 38.0°C.
- Survival and 28-day poor functional outcome did not differ (17.6% vs 16.7%), but grade 3-4 GI injury (52.9% vs 0%, p<0.001) and arrhythmias (41.2% vs 11.1%, p=0.045) were higher with hypothermia.
- The authors confirmed feasibility for a larger trial and concluded the hypothermia strategy should not be routinely adopted in ICU heatstroke management.
- Systematic review and meta-analysis of 15 studies (16 cohorts, about 2.6 million encounters) of children under 18 with suspected infection in hospital settings.
- Evaluated the Phoenix Sepsis Criteria for in-hospital mortality and compared them with the IPSCC criteria.
- Pooled PSC sensitivity was 0.77 (95% CI, 0.67-0.85), specificity 0.73 (95% CI, 0.51-0.87), and AUC 0.81 (95% CI, 0.75-0.84), versus an IPSCC AUC of 0.71.
- The authors conclude the PSC has good prognostic accuracy but varies by setting, with ICU specificity 0.48 versus 0.99 in the ED, and it is intended for risk stratification rather than frontline screening.
- Retrospective observational study of 568,340 adult ICU admissions across 430 Japanese hospitals from October 2020 to March 2023, using a national administrative database.
- Facility-level annual volume was counted for patients above SOFA score thresholds (0-24) and tested against in-hospital mortality with multilevel mixed-effects logistic regression.
- Total ICU volume was not significantly associated with mortality, but volume of patients with SOFA above 5 was associated with lower mortality (OR 0.997; 95% CI 0.995-0.998; p<0.001).
- Adjusted odds ratios decreased as the SOFA threshold rose, and the authors conclude the inverse volume-outcome relationship strengthened at higher severity.
- Prospective exploratory physiologic study of 30 patients with moderate-to-severe ARDS (PaO2/FiO2 150 mmHg or less), with a dose-response cohort of 20 and an electrical impedance tomography cohort of 10.
- Anterior chest wall loading at 5%, 10%, 15%, and 20% of body weight was compared with baseline, with the 5% dose also assessed by impedance tomography.
- At 5% body weight, transpulmonary driving pressure fell from 7 to 6 cm H2O (P = 0.015) and end-expiratory lung volume fell from 1,784 to 1,559 ml (P < 0.001).
- The authors conclude that the load relieves nondependent overdistension but causes dependent derecruitment and hemodynamic compromise, so it serves as a physiologic probe rather than a therapy.
- Preclinical study using human alveolar epithelial cells, SARS-CoV-2-infected mice, and a murine SARS-CoV-2-associated ARDS model.
- Compared miR-147b loss-of-function (deletion in alveolar epithelia) with nanoparticle-mediated miR-147b overexpression, and tested an ORF8 silent mutation at the binding site.
- miR-147b was the leading candidate induced during infection, and its alveolar expression was protective; the ORF8 mutation abolished this protection in vitro and in vivo. No effect sizes are reported in the abstract.
- The authors conclude that alveolar miR-147b directly targets viral RNA and protects the lung, describing an antiviral and lung-protective mechanism.
- Retrospective-style analysis of four ICU databases (MIMIC-III, MIMIC-IV, eICU, HiRID) with 203,434 admissions from more than 200 hospitals, 2001 to 2022.
- A multimodal deep-learning model combined structured data, clinical notes, and chest x-rays using the first 24 hours of ICU data to predict in-hospital mortality, with external validation in separate cohorts.
- The structured-data model reached AUROC 0.92 (95% CI 0.90 to 0.93); in a subset with notes and imaging, AUROC rose from 0.87 to 0.89 (DeLong P = 0.0167).
- The authors conclude that combining multiple data sources improves mortality prediction and that external validation is essential; eICU external AUROCs ranged from 0.84 to 0.92.
- Prospective randomized physiologic study of 80 patients with early septic shock in one university hospital medical ICU.
- All patients first underwent a 2-hour MAP test (65-70, then 80-85 mm Hg) measuring renal resistive index, then were randomized to low or high MAP targets for 5 days.
- Twenty-three patients (29%) were RRI responders, and among them a higher MAP target did not improve renal outcomes such as diuresis, KDIGO stage, RRT need, or creatinine.
- Nonresponders had higher urine output with the high target, an exploratory finding suggesting that no RRI drop may flag patients who gain diuresis from a higher MAP.
- Secondary analysis of the SISPCT randomized trial, including 971 patients with admission plasma H3.1 data from 33 German ICUs.
- The exposure was admission plasma H3.1 nucleosome levels, and no intervention was assigned in this analysis.
- Each one log10 increase in H3.1 was associated with higher 28-day mortality (adjusted HR 1.48, 95% CI 1.07-2.04; p = 0.02).
- H3.1 levels were also associated with later renal replacement therapy (adjusted HR 1.8, 95% CI 1.34-2.40) and were higher in septic shock than in sepsis.
AI summary not available. Read the source abstract on PubMed.
- International expert panel of 25 clinicians from four continents used a structured, iterative voting process to develop practice recommendations for postoperative and postprocedural respiratory failure.
- The panel refined 54 preliminary statements into 13 recommendations and 15 suggestions, covering screening, tobacco cessation, procedure timing, early mobilisation, oral hygiene, noninvasive support, and sedation and extubation practice.
- No comparative effect sizes or outcome data are reported, as the output is consensus-based guidance rather than a trial or cohort analysis.
- The authors conclude the recommendations offer practical prevention strategies and a framework for future study of a risk-targeted care bundle.
- This is a narrative review of mechanisms and management of right ventricular dysfunction in ARDS, not a primary study with a defined patient cohort.
- It examines how inflammatory lung injury, pulmonary vascular dysfunction, and mechanical ventilation contribute to RV-pulmonary arterial uncoupling.
- The abstract reports no pooled numbers; it describes RV dysfunction as frequent and clinically relevant, and acute cor pulmonale as associated with increased mortality.
- The authors conclude that recognising RV dysfunction can improve bedside assessment and individualised cardiopulmonary management in ARDS.
- A retrospective single-center study of ICU nurse clinical reports from 2023 to 2024 covering 2190 shifts at University Hospital Basel.
- No intervention was studied; the authors characterized aggression and violence against nurses and screened for risk factors of recurrent incidents.
- Incidents occurred in 14% of shifts, and recurrence in 27%. Recurrence was linked to longer ICU stay (adjusted OR 1.05 per hour), initial aggression (adjusted OR 6.9), and initial verbal violence (adjusted OR 4.3).
- Violence was frequent and almost always patient-caused, and staff harm was rare (1.3%), though patient complications were reported in 60% of incidents.
- Economic model comparing current mixed practice with a hypothetical full switch for eight NHS anaesthetic and critical care medicines.
- The modelled intervention is a switch from ampoules and vials to licensed ready-to-administer prefilled syringes, including wastage, preparation time, consumables and adverse drug events.
- Epinephrine, ephedrine, atropine and lidocaine 2% were associated with reduced system costs of more than £5.3 million annually, while rocuronium and midazolam became more expensive.
- The authors conclude that prefilled syringes offer safety and workflow advantages and possible economic benefit, with acquisition cost as the main barrier.
- A multicenter randomized trial at three hospitals in one Seattle health system enrolled 449 ICU patients and 505 family members, randomizing by patient.
- Nurse communication facilitators supported goal-concordant communication in the ICU and for 3 months afterward, compared with usual control care.
- Family depression on the HADS did not differ between groups over 6 months (difference -0.07, 95% CI -0.70 to 0.55).
- Relationship with the healthcare team (p=0.0162) and goal concordance (p=0.0504) moved in the expected direction, but effect sizes were small.
- Systematic review and meta-analysis of eight randomized trials (12,669 patients) comparing aggressive and conservative blood pressure lowering in acute intracerebral hemorrhage.
- Excellent functional outcome did not differ (RR 1.07; 95% CI 0.99-1.16), nor did 90-day mortality (RR 0.90; 95% CI 0.77-1.05) or hematoma growth.
- Aggressive lowering was associated with fewer adverse events (RR 0.87; 95% CI 0.76-0.99; p = 0.03), though overall heterogeneity was substantial (I2 = 59%).
- Sensitivity analyses showed a small good-recovery benefit (RR 1.05; 95% CI 1.01-1.10) that the subgroup analyses did not confirm.
- Systematic review and meta-analysis of 12 RCTs including 713 adult cardiac surgery patients undergoing cardiopulmonary bypass.
- Compared intraoperative haemoadsorption inserted into the bypass circuit with conventional bypass without haemoadsorption.
- No significant mortality benefit was found (OR 1.08, 95% CI 0.59-2.01, P=0.75), and no significant difference in hospital or ICU length of stay.
- Authors conclude there was no evidence of clinical benefit for routine use, but clinically relevant benefit or harm cannot be excluded, and only one included study had low risk of bias.
- Nationwide retrospective cohort of 588,568 admissions to all 79 Dutch ICUs from 2014 to 2022, linked to household-level socioeconomic data.
- Exposure was SES quintile, with mixed-effects logistic regression for hospital mortality, adjusted for age, sex, year, physiologic derangement, and comorbidities.
- After adjustment, the lowest SES quintile had the strongest association with mortality, OR 1.6 (95% CI 1.5-1.6) versus the middle quintile.
- The authors conclude that lower-SES ICU patients have poorer survival regardless of comorbidity, and SES should inform outcome interpretation and health policy.
- Prospective cohort of 103 mechanically ventilated adults (≥24 h) from four University of Pennsylvania ICUs, enrolled from 967 eligible patients, with 70% retained at 12 months.
- This psychometric study has no intervention; it tests the Patient Dignity Inventory's internal consistency, generalizability, and minimal clinically important difference across timepoints.
- Cronbach's alpha was 0.94-0.96 and phi was 0.93, with 75% of score variance from patient differences and 25% from timepoint.
- Factor analysis found 2-3 factors per timepoint, with physical, psychologic, and cognitive concerns early and existential and interpersonal concerns rising over time; the authors call the PDI valid and reliable.
- Randomized, assessor-blinded trial in 68 tracheostomized adults with neurologic disorders at one rehabilitation department, 34 per group.
- Both groups received airway clearance techniques with conventional treatment for 6 weeks; the experimental group also received abdominal electrical stimulation.
- Decannulation success was 93.3% with stimulation versus 73.3% in controls (OR 5.070, 95% CI 1.41-18.24, p = 0.013), and peak involuntary cough flow was higher (mean difference 39.433 L/min, 95% CI 28.83-50.04, p < 0.001).
- Infection scores and resting diaphragm mobility did not differ between groups, so the authors report benefit for cough, abdominal muscle thickness and decannulation only.
- Randomized crossover trial in 32 mechanically ventilated children aged 1 month to 18 years who completed both 24-hour monitoring phases.
- Continuous automated cuff pressure regulation targeting 15-25 cmH2O was compared with intermittent manual adjustment by bedside nurses under a predefined protocol.
- Low-pressure episodes (<15 cmH2O) occurred in 12.7% with manual monitoring versus 1.3% with automated (p<0.001), and time outside target fell from 118 to 10 hours.
- Nursing workload was lower with automation (0.08 vs 3.65 h), and the authors conclude automated control should be preferred to enhance airway safety.
- Multicenter randomized controlled trial in 172 adult ICU sepsis patients across four hospitals in central China, 86 per arm.
- Early Shenfu injection was compared with normal saline, given at 20 mL/hr over 5 hours, twice daily for three days from sepsis diagnosis.
- Total vasopressor use, measured as norepinephrine equivalents, did not differ significantly between groups (median 22 vs 20; p = 0.16).
- Secondary outcomes showed lower ICU mortality (3.5% vs 11.6%; p = 0.04) and 28-day mortality (15.1% vs 27.9%; p = 0.04), which the authors say needs further validation.
- Target trial emulation using registry data from 2451 adults with shockable out-of-hospital cardiac arrest who received at least three defibrillation attempts, across multiple North American EMS sites.
- Compared first antiarrhythmic drug given: lidocaine (n=987) versus amiodarone (n=1464), with inverse probability weighting and logistic regression adjustment for confounders.
- Survival to discharge was 26.2% with lidocaine versus 23.5% with amiodarone, an adjusted difference of 2.8% favoring lidocaine (95% CI -0.6 to 6.2).
- Favorable neurologic outcome (mRS 3 or less) did not differ (17.7% vs 16.2%, adjusted difference 1.5%, 95% CI -1.5 to 4.5); authors note wide CIs cannot exclude a meaningful lidocaine benefit.
- This mixed methods study interviewed 26 people who had previously been ICU patients, using questionnaires and reflexive thematic analysis.
- Participants described the ICU experience as involving confrontation with mortality, loss of control and total dependence, and as transformative.
- The median survival chance that participants considered makes ICU treatment acceptable was 30%; no confidence interval or p value is reported.
- The authors suggest that the experiential harms and risk of a negative dying experience may outweigh survival benefit when the chance of survival is low.
- A two-step Delphi consensus by 37 experts from 20 countries developed recommendations for reducing the environmental footprint of European intensive care medicine.
- Thirty-seven initial recommendations, drafted by four topic groups on energy, waste, medication, and ethics, covered renewable energy, single-use plastics, fluorinated gases, antimicrobial resistance, and a precautionary principle.
- All recommendations reached at least 75% agreement in round one, and after four amendments all exceeded 80% in round two, with 32 above 90%.
- The authors present the recommendations as a practical framework and urge institutions, procurement bodies, guideline developers, and quality programmes to adopt them.
- Retrospective multicenter cohort using the German Thorax Registry, including 2026 patients with one-lung ventilation longer than 60 minutes between 2016 and 2021.
- No intervention was tested; the exposures were intraoperative driving pressure, peak pressure, PEEP, and tidal volume per predicted body weight, analyzed against threshold values.
- Postoperative complications occurred in 34.1%, respiratory complications in 18.6%, and in-hospital mortality in 1.8%; several thresholds were linked to complications in unadjusted analyses (e.g., driving pressure >20 mbar, respiratory complications p = 0.009).
- The authors caution against routinely reducing PEEP based on these findings, noting that baseline risk and case complexity were the main contributors in adjusted models, and that mortality analyses were limited by 37 events.
- This ecological analysis used 2024 aggregate data from 200 countries, applying SpO2 targets drawn from a randomised trial to mechanically ventilated ICU adults.
- It compared a conservative SpO2 target of 88-92% with a liberal target of at least 96% for estimating oxygen costs and emissions.
- Modelled global annual oxygen expenditure would fall from 180 to 76 million int$, and emissions from 884 to 368 ktCO2eq, under the conservative target.
- The authors conclude that conservative oxygenation reduces costs and emissions, and that cleaner electricity grids could cut oxygen-related emissions by 49.4%-91.9%.
- This single-patient case report describes a child on venoarterial ECMO after cardiac arrest who was evaluated for brain death under the 2011 pediatric guideline.
- Apnea testing used controlled reduction of sweep gas flow with serial arterial blood gases, and the 2023 consensus criteria were applied retrospectively to the original data.
- Arterial CO2 rose from 43.1 to 74.5 mm Hg without respiratory effort, and pH fell to 7.19, meeting the 2023 acidemia requirement.
- The 2011-based diagnosis was concordant with the 2023 criteria, and the authors suggest ECMO-adapted apnea testing may be feasible when physiology is carefully optimized.