- Parallel observational analysis of 245 HARMONY and 206 CRRTnet patients with cirrhosis or ACLF receiving CRRT in US and Canadian ICUs.
- Exposures were CRRT process factors, including prescribed dose, modality, and anticoagulation, with in-hospital mortality as the primary outcome.
- Hospital survival was similar across cohorts (34% vs 30%), and CRRTnet prescribed dose above 30 mL/kg/hr was associated with higher mortality (OR 2.69, 95% CI 1.3-5.64, p=0.01).
- In HARMONY, higher chronic liver failure-C ACLF scores were independently associated with mortality (HR 1.03, 95% CI 1.01-1.05), and 64% of CRRTnet survivors transitioned to intermittent hemodialysis.
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.
- Prospective multicenter cohort of 1,332 adults undergoing CABG or valve surgery at ten cardiac centers in the West Bank, grouped by ventilation duration of ≤6 or >6 hours.
- Hallucinations were screened over 7 postoperative days using the Questionnaire for Psychotic Experiences, with Cox and restricted mean survival time analyses comparing delayed extubation with early extubation.
- Delayed extubation was linked to lower adjusted risk of visual hallucinations (HR 0.70, 95% CI 0.62-0.80, p < .001) and auditory hallucinations (HR 0.49, 95% CI 0.31-0.77, p = .002).
- Time-varying analysis showed higher immediate postoperative hazard in the delayed group, auditory hallucination-free time was about 0.24 days shorter (p < .001), and preoperative white blood cell count was the most consistent predictor.
AI summary not available. Read the source abstract on PubMed.
- This systematic evidence map screened 6,805 PubMed and Embase records and included 1,116 studies, mostly adult cohort studies in critically ill patients.
- It mapped novel kidney biomarkers for AKI by purpose: prediction (n = 944), prognosis (n = 647), etiology (n = 109), trial enrichment (n = 6), and management (n = 12).
- Accuracy studies comparing biomarkers with AKI diagnosis were far more common than practical trials of biomarkers in clinical use.
- The authors suggest implementation guidance can be drawn mainly from surgical and nephrotoxin-exposed populations, where evidence is available.
- Prospective observational cohort of 40 adult ICU survivors who had ventilation of at least 72 hours and/or ECMO, recruited from six UK hospitals, with 33 completing assessment.
- Participants underwent a DVLA-approved comprehensive driving assessment covering physical and cognitive evaluation, driving simulator, private track, and on-road in-car testing.
- Of the 33 assessed, 26 (79%) were judged capable of driving a standard vehicle, but only 15 (45%) had resumed driving.
- Driving confidence rose from a median of 7 to 9 after assessment (p < 0.05), and 97% rated the assessment very good or excellent.
- Systematic review and dual frequentist-Bayesian meta-analysis of seven randomized trials (n=2408) in adults with septic shock.
- Compared peripheral perfusion-guided resuscitation with standard care or lactate-guided resuscitation for 28-day mortality.
- The pooled risk ratio was 0.87 (95% CI 0.76-1.01; p=0.06), with the interval crossing no effect and I² of 0%.
- The Bayesian analysis gave a 97.2% posterior probability that RR is below 1, but the authors note the magnitude of any effect remains uncertain.
- Retrospective single-center cohort of 6,451 adults in MIMIC-IV who had major cardiac surgery, PaO₂/FiO₂ ≤300 mm Hg on day one in the ICU, and survived to a 24-hour landmark.
- The exposure was a standardized composite score (ODCS-3) built from oxygenation, creatinine, and lactate, modeled with Cox regression for mortality at days 28, 90, and 365.
- Each 1 SD increase was associated with an adjusted HR of 3.10 (95% CI, 2.64-3.65) at day 28, falling to 2.09 (95% CI, 1.89-2.30) at day 365.
- Discrimination was similar to SOFA (AUCs 0.837, 0.793, and 0.734), and the authors say the score is not a replacement severity score and needs external validation.
- Retrospective hospital database study of adults meeting Sepsis-3 criteria, with 31,109 patients after excluding positive nonviral cultures and 1034 (3.3%) classified as viral sepsis.
- Compared viral sepsis with nonviral sepsis, and examined associations with mortality including mechanical ventilation, SOFA score, lactate, and antiviral receipt.
- Viral sepsis had 15% 90-day mortality versus 20% for nonviral sepsis; after adjustment, mechanical ventilation (OR 3.142, 95% CI 1.692-5.836), SOFA score (OR 1.159), and lactate (OR 1.204) were associated with mortality.
- Antiviral medication receipt was not associated with better survival (OR 1.160, 95% CI 0.142-9.449, P = .875), per the authors.
- This narrative review draws on PubMed, Google Scholar, reference lists, and intrathecal pump device manuals to cover perioperative, obstetric, and ICU care of patients with implanted intrathecal drug delivery systems.
- It is organized into five sections: intrathecal pharmacology, implantation considerations, perioperative care, neuraxial anesthesia, and device-related safety concerns.
- Where evidence is limited, the authors supplement the literature with expert opinion, so some recommendations rest on consensus rather than data.
- The authors highlight intrathecal medications, their behavior under MRI, and precautions before any neuraxial access in these patients.
AI summary not available. Read the source abstract on PubMed.
AI summary not available. Read the source abstract on PubMed.
- Retrospective multicenter cohort of 269 critically ill children with acute cardiorespiratory failure and hyperglycemia, treated in PICUs from 2012 to 2016.
- A three-marker classifier (IL-6, IL-8, sTNFR-1) built on conventional multiplex data was applied to rapid immunoassay measurements, and a de novo rapid-assay classifier was also derived.
- The de novo classifier had an AUROC of 0.90 (95% CI, 0.86-0.95), matched latent class classifications in 241 of 269 cases (89.6%), and showed mortality of 33.3% versus 11.8% (p = 0.009).
- Rapid-assay subphenotypes differed in response to intensive insulin management (interaction p = 0.024), and the authors suggest this approach could inform future subphenotype-based precision trials.
- Double-blind randomized placebo-controlled trial in 120 mechanically ventilated adults on opioid infusion at two Melbourne ICUs, excluding cardiac surgical patients.
- Patients received low-dose ketamine infusion (0.15 mg/kg/hr) or placebo during mechanical ventilation.
- Median hourly opioid dose was 64 µg/hr with ketamine versus 77 µg/hr with placebo (median difference -13.0, 95% credible interval -26.6 to 2.4; probability of benefit 95.1%).
- Authors concluded low-dose ketamine appeared safe and reduced opioid requirements, while noting that delirium and most other secondary outcomes did not differ and larger trials are needed.
- A retrospective, nationwide registry cohort in Sweden included 92,428 COVID-19 and 21,355 influenza hospitalizations, with COVID-19 split into five pandemic-period groups.
- Cox regression, adjusted for age, sex, socioeconomic factors, comorbidities, and SARS-CoV-2 vaccination, identified predictors of ICU admission and in-hospital mortality, and influenza served as the comparator.
- Hospitalization rates were lowest in the delta period and ICU admission rates were lowest in the Omicron period, and Omicron patients most closely resembled influenza patients in demographics, stay, ICU use, and mortality.
- Advanced age was the strongest predictor of severe illness for both infections, and lack of vaccination was additionally associated with severe COVID-19.
- Systematic review of 102 cluster randomized trials in critical care, published 2004 to 2022 and indexed in MEDLINE and Embase, assessing ethical and methodological reporting.
- Compares reporting features across trial designs, including parallel-arm (47%), two-period crossover (23%), stepped-wedge (20%), and multi-period crossover (8%) designs.
- Consent was reported in 75% of trials that identified patients or caregivers as participants, with waived consent most common (49%); only 23% reported consent for healthcare providers.
- Most trials (80%) reported a sample size calculation, but only 13% of trials with fewer than 40 clusters reported a small sample correction.
- Randomized controlled trial of 172 patients after elective CABG or valve surgery, allocated to five noninvasive ventilation or high-flow interfaces versus physiotherapy alone in one tertiary ICU.
- Compared a conventional oronasal mask, conventional full-face mask, Wolf Mask, ELMO helmet CPAP, and high-flow nasal cannula, each given three 60-minute daily sessions alongside standard physiotherapy.
- The Wolf Mask had the highest comfort score (9.41 ± 0.68 on a 0-10 VAS) and lowest inspiratory pressure support (5.62 ± 2.08 cmH₂O); interface comfort predicted mobilization progression (p = 0.004).
- The authors conclude full-face interfaces are associated with greater comfort, and they describe the pulmonary and oxygenation findings as exploratory and hypothesis-generating, requiring multicenter confirmation.
- Single case report of a 37-year-old woman who had a repeat cesarean complicated by peripartum stress cardiomyopathy and cardiac intensive care admission.
- Three validated patient-reported instruments and a semi-structured interview assessed her recovery, care experience, and trauma symptoms early and at six weeks.
- She met screening criteria for early posttraumatic stress on postoperative day 3, despite a favorable medical outcome and a high patient-rated care experience.
- By six weeks her standardized symptom scores had improved, but her narrative still reflected grief and loss, which the authors use to support systematic trauma screening.
- Retrospective cohort of 6928 children aged 0-18 hospitalized with sepsis from 2016 to 2022, using MarketScan administrative claims data.
- No intervention was studied; the exposure was any outpatient, emergency department, or inpatient encounter in the 7 days before sepsis admission.
- 53.5% of hospitalizations had a pre-admission encounter: 33.7% outpatient, 28.2% emergency department, and 6.2% inpatient.
- Among those with an encounter, 38.9% had an infection-related diagnosis, and mechanical ventilation, ICU use, and length of stay did not differ by prior encounter status.
- This multicenter observational registry study included 379 mechanically ventilated adults with shock across 23 hospitals, and 96 (25.3%) had TELUS added to resuscitative TEE.
- Operator-reported identification of shock etiology was higher with TELUS plus TEE than with TEE alone (86.5% vs. 75.3%).
- The etiology difference was significant (OR 1.95, 95% CI 1.02-3.72, p = 0.04), but the management change difference was not (78.1% vs. 71.7%; OR 1.40, 95% CI 0.82-2.41, p = 0.22).
- TELUS added a longer median procedure time (21.5 vs. 19.0 min, p = 0.01), complications were rare and similar, and the authors call for prospective studies with patient-centered outcomes.
- Mouse study with n=4–5 per group testing whether glutamatergic medial prefrontal cortex neurons regulate immune responses in E. coli and cecal ligation and puncture sepsis.
- Optogenetic activation and chemogenetic inhibition of these neurons were paired with vagotomy, ADRB2 blockade, or splenic denervation to probe autonomic mechanisms.
- Activation reduced proinflammatory cytokines and liver, kidney, and lung injury, while inhibition worsened them; M2 macrophages rose from 14.4 ± 1.15% to 19.4 ± 0.87% (P = .023).
- Vagotomy removed this M2 difference (14.0 ± 0.89% vs 14.2 ± 1.33%, P > .999), and ADRB2 blockade or splenic denervation reversed the M1 shift.
- This perspective piece, which describes no primary data, argues that anesthesiology should expand its critical care training pipeline as U.S. ICU demand rises.
- Anesthesiologists make up less than 10% of U.S. intensivists, and recent ACCM fellowship match data show underfilled positions.
- The authors suggest low resident interest reflects perceived compensation, workload, and competing demands, which may not match current market and practice patterns.
- They propose early mentorship, structured scholarly experiences, and combined or tailored ACCM tracks to align training with career interests and aid retention.
- Systematic review and diagnostic meta-analysis of 22 studies, with 13 included in the pooled analysis (n = 1,999 patients), comparing lung ultrasound against chest radiography or CT for ARDS.
- The index test was lung ultrasonography, with chest x-ray or computed tomography as the reference standard; scanning protocols and diagnostic thresholds varied across studies.
- Pooled sensitivity was 0.69 (95% CI, 0.54 to 0.81; I² = 82.9%, low certainty) and pooled specificity was 0.95 (95% CI, 0.85 to 0.98; I² = 88%, moderate certainty).
- The authors conclude that lung ultrasound is better at confirming ARDS (higher specificity) than excluding it (moderate sensitivity), with substantial between-study heterogeneity.
- Systematic review and meta-analysis of 163 RCTs and observational studies including 78,053 adults receiving ECMO, with outcomes at 6 months or later.
- Pooled analyses compared long-term mortality across ECMO modalities (venovenous, venoarterial, and extracorporeal CPR) and clinical indications.
- At 1 year, pooled mortality was 37.2% (95% CI, 30.0-45.1%) for venovenous, 55.2% (95% CI, 50.2-60.1%) for venoarterial, and 74.4% (95% CI, 71.4-77.2%) for ECPR.
- Functional outcomes were reported in only 38 of 163 studies and were inconsistent, so the authors call for standardized reporting and benchmarking.
- The paper reviews anticoagulation approaches for patients with traumatic brain injury who are supported on ECMO for ARDS-related hypoxemic respiratory failure.
- The abstract draws on case reports and retrospective studies that it describes as showing efficacy for various anticoagulation strategies.
- It cites emerging literature suggesting ECMO may reduce mortality in TBI patients with concurrent ARDS, but gives no specific figures.
- The authors aim to outline these approaches and report their complications and outcomes, which the abstract does not quantify.
- Retrospective cohort of 5,684 adults undergoing open cardiac surgery, drawn from the MIMIC-IV cardiac surgical ICU database.
- Exposures were estimated pulse wave velocity (calculated from age and early mean arterial pressure) and 0-24 hour postoperative neutrophil-to-lymphocyte ratio, assessed against strict CAM-ICU delirium.
- Of patients, 17.6% developed delirium; each 1-SD higher ePWV had OR 1.46 (95% CI 1.06-2.02), and each doubling of NLR had OR 1.15 (95% CI 1.06-1.25).
- Adding both markers to a reference model raised the apparent AUC only from 0.790 to 0.791, indicating limited incremental discrimination.
- Systematic review and meta-analysis of 55 observational studies, with 21 contributing to the meta-analyses, in acutely admitted adult ICU patients.
- Exposure was pre-admission frailty or functional dependency, compared with non-frail or independent status.
- Non-frail versus frail participants had RR 0.62 (95% CI 0.58-0.67, P<0.01) for all-cause mortality; independent versus dependent had RR 0.47 (95% CI 0.40-0.54, P<0.01).
- The authors conclude that frailty and dependency were associated with higher mortality, but certainty of evidence was very low and heterogeneity was substantial.
- Systematic review and meta-analysis of RCTs in adults undergoing cardiac surgery with cardiopulmonary bypass; 105 trials identified, 39 pooled.
- Compared ventilation strategies across the preoperative, intraoperative and postoperative periods, with all-cause mortality as the primary outcome.
- Ventilation during bypass reduced postoperative pulmonary complications (RR 0.87, 95% CI 0.79 to 0.96; P=0.005; 15 studies).
- Postoperative noninvasive respiratory support showed a non-significant mortality trend (RR 0.60, 95% CI 0.34 to 1.08; five studies).
- Post hoc secondary analysis of two trials across 30 hospitals, including 1963 adults having major abdominal or thoracic surgery.
- Compared patients given antiemetic-dose intraoperative corticosteroids with those who did not, using inverse probability of treatment weighting for preoperative confounders.
- After weighting, steroids were not significantly associated with fewer postoperative pulmonary complications (OR 0.86, 95% CI 0.62-1.18, P=0.347), though the unweighted rates were lower (9.3% vs 14.0%).
- PACU PaO2/FiO2 ratio was modestly higher with steroids (adjusted mean difference 19.9 mm Hg, 95% CI 0.64-39.1, corrected P=0.043).
- This narrative review searched PubMed and OVID through August 2025 for English-language literature on the microbiome in critically ill or injured patients.
- It covers the microbiome across the eyes, skin, oropharynx, lung, bladder, and gut, and examines disease- and therapy-driven disruption and restoration strategies.
- The authors single out the gastrointestinal microbiome as the most active, and say its disruption is common in ICU patients and can shape recovery or failure.
- They conclude that intensivists should understand microbiome evaluation, adaptive versus maladaptive changes, and emerging interventions to restore homeostasis.
AI summary not available. Read the source abstract on PubMed.
- This narrative review critically appraises randomized trial evidence for lung-protective ventilation in patients with ARDS and in those without ARDS.
- Low tidal volumes and titrated PEEP are widely recommended in ARDS, and the review questions whether these strategies also benefit non-ARDS patients.
- The abstract reports that recent trials in non-ARDS patients showed no clear advantage and in some cases suggested potential harm.
- The authors argue that benefit is not consistent across all ARDS patients, favoring ventilator settings individualized to the patient over a uniform approach.
- A descriptive comparative study of three open-access ICU databases (MIMIC-IV, eICU-CRD, AmsterdamUMCdb) assessed their suitability for AI-based sepsis research.
- The authors used structured queries and manual semantic inspection to assess completeness, terminology, structure, and redundancy, focusing on antimicrobial administration and mortality markers.
- MIMIC-IV had 49% missing medication route and 58% missing dose, AmsterdamUMCdb had 61.4% of admissions without diagnosis, and eICU-CRD had 1.9% lacking diagnosis.
- The authors conclude these databases show substantial missingness, redundancy, and semantic variability, and none had reliable sepsis diagnosis timestamps, so rigorous preprocessing and standardized data collection are needed.
- Cross-sectional bibliometric analysis of 4,982 original research papers (4,479 from high-income and 503 from low/middle-income countries) in ten top critical care journals, 2018 to 2022.
- Compared women's authorship by country income group, with stratification by funding type; no intervention was applied.
- Women held 31.3% of all authorships, 35.4% of first and 21.4% of senior positions, and representation was higher in low/middle-income countries overall (38.2% vs 30.6%; p < 0.01).
- The authors conclude that substantial gender disparity persisted, though women were better represented in low/middle-income country publications and authorship rose over time in both groups.
- Prospective documentation by research coordinators in an international stress ulcer prophylaxis trial in invasively ventilated patients, covering 5024 patients or substitute decision-makers approached for consent.
- Consent encounter features (consent model, who requested and gave consent, method) and center characteristics were analyzed with multilevel logistic regression; the trial itself had no intervention for this analysis.
- The overall consent rate was 85.0% (range 27.3% to 100%), with deferred consent associated with higher consent rates (OR 16.56; 95% CI 11.50-23.84).
- Consent rates were also higher when patients themselves were approached (OR 2.67; 95% CI 1.63-4.37) and during or after the COVID-19 pandemic, while center characteristics were not associated.
- Systematic review and meta-analysis of 6 studies (2 RCTs, 4 cohort studies; n = 834) in adults undergoing cardiac surgery with cardiopulmonary bypass, all from East Asia.
- Compared perioperative sivelestat sodium, a neutrophil elastase inhibitor, with standard care for preventing postoperative lung injury and ARDS.
- ARDS incidence favored sivelestat but was not significant (Peto OR 0.19, 95% CI 0.03-1.16; I² 91.3%), and mortality was also nonsignificant (OR 0.69, 95% CI 0.35-1.39).
- The authors describe a consistent directional benefit limited by heterogeneity, rate the certainty of evidence as very low, and call for larger multicenter RCTs.
- Retrospective observational cohort of 5,293 ICU patient stays at four ICUs in one German university medical center, admitted 2019 to 2022.
- Compared single rooms with shared rooms of up to three beds, plus room visibility and distance to nurse and doctor charting areas, using multivariate regression adjusted for SAPS II and other factors.
- None of the room variables were associated with in-hospital mortality in the overall cohort, which had an 18% mortality rate.
- Single-room admission was associated with a 0.81-day longer mean ICU length of stay (95% CI 0.04-1.57; p = 0.039), which the authors frame as a capacity trade-off.