Journal of Intensive Care Medicine最新文献

Trial registration: Clinicaltrials.gov NCT02270281. Registered October 16, 2014.

There are discrepancies in resources and expertise available between pediatric intensive care units (PICUs) in Brazil that likely significantly impact the clinical outcomes of patients. The goal of this study was to evaluate the impact of telemedicine rounding support in two public PICUs located in the North and Northeast regions of Brazil. Our intervention involves telehealth rounds connecting two "level II" PICUs with specialist doctors from a hospital of recognized excellence. A before-and-after study was carried out to evaluate telemedicine's impact on PICUs between December 2018 and July 2019. Nine hundred and forty patients were evaluated during this period (426 pre-telemedicine, 514 post-telemedicine). The intervention occurred through telerounds between the command center and the ICUs assisted by telemedicine. In unit A, the implementation of telemedicine reduced the mortality rate from 18.86% to 9.29%, while in unit B, it decreased from 10.76% to 9.72%. There was no change in the median length of stay in unit A, but in unit B, it increased from 6 to 8 days. Logistic regression analysis confirmed a significant reduction in mortality in unit A (odds ratio (OR) 0.50; 95% confidence interval (CI) 0.29-0.86). The study found a positive correlation between adherence to telemedicine recommendations and mortality reduction across both units. This suggests that telemedicine can effectively improve outcomes in PICUs, particularly in regions with limited health-care resources.

Objective: Extracorporeal cardiopulmonary resuscitation (ECPR) utilizes veno-arterial extracorporeal membrane oxygenation (VA-ECMO) in cardiac arrest patients to reduce the risk of mortality and multiorgan dysfunction from systemic hypoperfusion. We aimed to compare clinical outcomes of patients receiving ECPR versus conventional cardiopulmonary resuscitation (CCPR) for refractory cardiac arrest.

Data sources: This was a systematic review and meta-analysis. A librarian searched the main databases, Ovid MEDLINE (including epub ahead of print, in-process & other non-indexed citations), Ovid EMBASE and Ovid Cochrane Central Register of Controlled Trials from inception through July 2024.

Study selection: We included randomized controlled trials and observational studies that compared the outcomes of ECPR to CCPR in cardiac arrest patients. Primary outcomes were neurological sequelae and survival.

Data extraction: We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Two reviewers independently screened articles, extracted data on selected articles and performed risk of bias assessments using ROBINS-I for non-randomized controlled trials and the revised Cochrane risk of bias tool for randomized controlled trials with disagreements settled by a third independent reviewer.

Data synthesis: Out of 3458 studies identified and screened, 28 studies including 304,360 cardiac arrest patients met eligibility criteria and were included. Survival at hospital discharge was 20% for ECPR versus 3.3% for CCPR (OR 0.48 [CI 0.27, 0.84]). Favorable neurological outcome at hospital discharge was 11.8% for ECPR versus 1.9% for CCPR (OR 0.41 [CI 0.17, 1.01]). Complications from bleeding were ten times higher in the ECPR group (35.3% vs 3.7%; OR 0.08 [0.03, 0.24]).

Conclusions: ECPR appeared to be superior to CCPR for improved neurological outcome and survival in cardiac arrest patients, although bleeding was increased. There was large heterogeneity in the included studies and outcomes reported. Future prospective studies may improve the identification of subgroups of patients that will benefit most from ECPR.Systematic review and meta-analysis registration: PROSPERO - CRD42023394128.

Adequate fluid therapy is crucial to maintain organ function after burn trauma. Major burns lead to a systemic response with fluid loss and cardiac dysfunction. To guide fluid therapy, measurement of cardiac pre- and afterload is helpful. Whereas cardiac function is usually measured after admission to intensive care unit (ICU), in this study, hemodynamic monitoring was performed directly after arrival at hospital. We conducted a prospective cohort study with inclusion of 19 patients (male/female 13/6, 55 ± 18 years, mean total body surface area 36 ± 19%). Arterial waveform analysis (PulsioFlexProAqt^®, Getinge) was implemented immediately after admission to hospital to measure cardiac pre- and afterload and to guide resuscitation therapy. Cardiac parameters 3.75 (2.67-6.0) h after trauma were normal regarding cardiac index (3.45 ± 0.82) L/min/m², systemic vascular resistance index (1749 ± 533) dyn sec/cm⁵ m², and stroke volume (SV; 80 ± 20) mL. Stroke volume variation (SVV) was increased (21 ± 7) % and associated with mortality (mean SVV survivors vs nonsurvivors 18.92 (±6.37) % vs 27.6 (±5.68) %, P = .017). Stroke volume was associated with mortality at the time of ICU-admission (mean SV survivors vs nonsurvivors 90 (±20) mL vs 50 (±0) mL, P = .004). Changes after volume challenge were significant for SVV (24 ± 9 vs19 ± 8%, P = .01) and SV (68 ± 24 vs 76 ± 26 mL, P = .03). We described association of SVV and SV with survival of severely burned patients in an observational study. This indicates high valence of those parameters in the early postburn period. The use of an autocalibrated device enables a very early monitoring of parameters relevant to burn shock survival.

Background: Elevated renin has been shown to predict poor response to standard vasoactive therapies and is associated with poor outcomes in adults. Similarly, elevated renin was associated with mortality in children with septic shock. Renin concentration profiles after pediatric cardiac surgery are unknown. The purpose of this study was to characterize renin kinetics after pediatric cardiac surgery.

Methods: Single-center retrospective study of infants who underwent cardiac surgery with cardiopulmonary bypass (CPB) utilizing serum samples obtained in the perioperative period to measure plasma renin concentrations (pg/mL). Time points included pre-bypass and 1, 4, and 24 h after initiation of CPB.

Results: Fifty patients (65% male) with a median age 5 months (interquartile range (IQR) 3.5, 6.5) were included. Renin concentrations peaked 4 h after CPB. There was a significant difference in preoperative and 4 h post-CPB renin concentration (4 h post-CPB vs preoperative: mean difference 100.6, 95% confidence interval (CI) 48.9-152.4, P < .001). Median renin concentration at 24 h after CPB was lower than the preoperative baseline.

Conclusions: We describe renin kinetics in infants after CPB. Future studies based on these data can now be performed to evaluate the associations of elevated renin concentrations with adverse outcomes.

Background: Acute kidney injury (AKI) is common in sepsis and a urine output <0.5 mL/kg/h associated with increased mortality is incorporated into AKI diagnosis. We aimed to identify the urine-output threshold associated with increased AKI incidence and hypothesized that a higher urine output than a specified threshold, which differs from the predominantly used 0.5 mL/kg/h threshold, would be associated with an increased AKI incidence.

Methods: This was a post-hoc analysis of a nationwide prospective observational study. This study included adult patients newly diagnosed with sepsis and requiring intensive care. Urine output on the day of sepsis diagnosis was categorized as low, moderate, or high (<0.5, 0.5-1.0, and >1.0 mL/kg/h, respectively), and we compared AKI incidence, renal replacement therapy (RRT) requirement, and 28-day survival by category. Estimated probabilities for these outcomes were also compared after adjusting for patient background and hourly fluid administration.

Results: Among 172 eligible patients, AKI occurred in 46.3%, 48.3%, and 53.1% of those with high, moderate, and low urine output, respectively. The probability of AKI was lower in patients with high urine output than in those with low output (43.6% vs 56.5%; P = .028), whereas RRT requirement was lower in patients with high and moderate urine output (11.7% and 12.8% vs 49.1%; P < .001). Patients with low urine output demonstrated significantly lower survival (87.7% vs 82.8% and 67.8%; P = .018). Cubic spline curves for AKI, RRT, and survival prediction indicated different urine-output thresholds, including <1.2 to 1.3 mL/kg/h for AKI and <0.6 to 0.8 mL/kg/h for RRT and mortality risk.

Conclusions: Urine output >1.0 mL/kg/h on the day of sepsis diagnosis was associated with lower AKI incidence. The urine-output threshold was higher for developing AKI than for RRT requirement or mortality.

High-risk pulmonary embolism (PE) is a life-threatening disease state with current guidelines recommending reperfusion therapy with systemic thrombolytics in addition to anticoagulation. This was a prospective observational cohort study with a historical control group comparing tenecteplase to alteplase for the treatment of PE or cardiac arrest with suspected PE. The primary outcome was the incidence of institutional protocol deviations defined as incorrect thrombolytic dose administered or the incorrect product compounded. Secondary outcomes included any bleeding event, major bleeding event, all-cause mortality, and for patients with a cardiac arrest, successful return of spontaneous circulation (ROSC). Fifty-four patients were included in the study. Protocol deviations occurred in one patient receiving tenecteplase and one patient receiving alteplase (4.0% vs 3.4%; P = 1.0). There was no difference in all-cause mortality (80% vs 86.2%; P = .72), any bleed (12% vs 13.8%; P = 1.0), major bleed (8.0% vs 6.9%; P = 1.0), or ROSC achievement (22.2% vs 28.6%; P = .73) when comparing tenecteplase to alteplase. Our study demonstrates that tenecteplase may be an alternative thrombolytic to alteplase for treatment of PE or cardiac arrest with suspected PE. Further studies comparing the different systemic thrombolytic agents for PE or cardiac arrest with suspected PE are needed.

Background: We assessed 2 versions of the large language model (LLM) ChatGPT-versions 3.5 and 4.0-in generating appropriate, consistent, and readable recommendations on core critical care topics. Research Question: How do successive large language models compare in terms of generating appropriate, consistent, and readable recommendations on core critical care topics? Design and Methods: A set of 50 LLM-generated responses to clinical questions were evaluated by 2 independent intensivists based on a 5-point Likert scale for appropriateness, consistency, and readability. Results: ChatGPT 4.0 showed significantly higher median appropriateness scores compared to ChatGPT 3.5 (4.0 vs 3.0, P < .001). However, there was no significant difference in consistency between the 2 versions (40% vs 28%, P = 0.291). Readability, assessed by the Flesch-Kincaid Grade Level, was also not significantly different between the 2 models (14.3 vs 14.4, P = 0.93). Interpretation: Both models produced "hallucinations"-misinformation delivered with high confidence-which highlights the risk of relying on these tools without domain expertise. Despite potential for clinical application, both models lacked consistency producing different results when asked the same question multiple times. The study underscores the need for clinicians to understand the strengths and limitations of LLMs for safe and effective implementation in critical care settings. Registration: https://osf.io/8chj7/.

Introduction: Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. Patients with cancer are at risk of developing sepsis and requiring intensive care unit (ICU) admission. We aimed to assess survival of patients with a solid tumour admitted to ICU as an emergency with sepsis, and to identify predictors of 90-day survival at admission.

Materials and methods: We conducted a retrospective cohort survival analysis. We identified adults with a solid tumour admitted to ICU with sepsis between 01/01/2011 and 31/12/2020 at a tertiary oncology centre with two hospitals (London and Surrey, UK). We defined sepsis using the Sepsis-3 definition. The primary outcome was 90-day survival. We used the parametric accelerated failure time model for multivariate analysis to generate acceleration factors (AF).

Results: 625 patients were identified and the 90-day survival rate was 59.5%(353/593).Multivariate analysis identified the presence of localized (AF 0.13, 95% CI 0.06-0.25) or regionalized disease (AF 0.21, 95% CI 0.12-0.36) compared to distant metastatic disease, unplanned surgery on the day of admission (AF 0.15, 95% CI 0.07-0.31), lactate (AF 1.25 95% CI 1.15-1.35), Sequential Organ Failure Assessment Score (AF 1.19, 95% CI 1.12-1.27), previous radiotherapy (AF 1.89, 95% CI 1.14-3.125), previous systemic anti-cancer treatment (excluding hormonal therapy) (AF 1.49, 95% CI 0.93-2.38), bacteraemia (AF 0.47, 95% CI 0.27-0.81) and serum albumin (AF 0.94, 95% CI 0.91-0.98) as independent predictors of 90-day survival.

Conclusions: This study of solid tumour patients admitted to ICU is one of the largest providing survival data to inform clinicians and patients. This data provides information on factors that should be considered when deliberating the possible outcome of ICU admission for a patient with solid malignancy and sepsis and highlights that the presence of cancer itself should not limit ICU admission for sepsis.

Background: Venous thromboembolism (VTE), whether pulmonary embolism (PE) or deep vein thrombosis (DVT), is common in patients with COVID-19. Recommendations on systematic screening in the intensive care unit (ICU) are lacking.

Research question: Is there any clinical benefit of systematic screening for DVT in critically ill patients with severe COVID-19?

Study design and methods: Single-center randomized clinical trial (RCT) of COVID-19 cases admitted to the ICU. Patients were randomized into two groups: a study group that underwent ultrasound (US) screening for DVT Mondays and Thursdays, and a control group that was treated according to the unit protocol. The primary outcome was the presence of DVT. Secondary outcomes were ICU total stay, death within 21-day follow-up and bleeding complications (minor or major). A composite outcome of poor prognosis variables was analyzed. We tested a superiority hypothesis with a confidence level of 95% and an equivalence limit of 20%.

Results: 163 patients (84 screening group, 79 control group) were enrolled between April and July 2021. There were 90 men (55.2%) with a mean ± SD age of 49.8 ± 13.58 years. In screening group 16.7% developed DVT versus 3.8% in control group (p = .007), and 3.6% versus 5.1% developed PE, respectively (p = 0.7). Poor outcome variables were male sex, age, COVID-19 vaccination status, Fibrinogen, Urea, Creatinine and Interleukin 6 (IL6) levels; Acute Physiology and Chronic Health Evaluation II (APACHE II) and Sequential Organ Failure Assessment (SOFA) scales. The superiority comparison, with a power of 95%, showed no statistically significant differences for a composite endpoint (p = .123). After adjusting by group, the OR for poor outcome is 1.966 (0.761-5.081) p = 0.163.

Interpretation: Among these patients, a strategy of systematic US screening for DVT was not associated with any significant improvements to clinical outcomes compared with usual care.

Clinical trial registration: Clinicaltrials.org registration number: NCT05028244.