Journal of Anesthesia, Analgesia and Critical Care (Online)最新文献

Background: With the growing numbers of nonagenarians and centenarians admitted to intensive care, clinicians face complex decisions regarding care appropriateness and intensity. Prognostic tools such as APACHE III, SOFA, and the Australia and New Zealand Risk of Death (ANZROD) model are widely used, yet their validity in very old patients remains uncertain given the influence of frailty, multimorbidity, and age-related physiological changes. In this study, we evaluated and compared the performance of APACHE III, its Risk of Death score, SOFA, and ANZROD in a large binational cohort of ICU-admitted nonagenarians and centenarians, assessing discrimination, calibration, and predictive utility for short- and long-term mortality using AUROC, standardized mortality ratios, and decision curve analysis.

Methods: We conducted a multicenter retrospective cohort study of 40,910 patients aged 90 years and older who were admitted to ICUs across Australia and New Zealand from 2010 to 2024. Discrimination was assessed using time-dependent area under the receiver operating characteristic curve (AUROC). Calibration was evaluated with standardized mortality ratios (SMRs), and clinical utility was examined using decision curve analysis (DCA) across ICU, hospital, 1-month, 6-month, and 1-year mortality outcomes.

Results: ANZROD consistently showed the highest discriminative ability, with AUROCs of 0.870 for ICU mortality and 0.768 for 1-month mortality. APACHEIII ROD and APACHE III followed, while SOFA had the lowest AUROCs across all timepoints. SMR analysis revealed overestimation of short term and underestimation of long term mortality by all models. ANZROD had the best calibration for hospital mortality. DCA showed ANZROD provided greater net benefit than the other scores at clinically relevant thresholds.

Conclusions: In this large binational cohort of over 40,000 nonagenarian and centenarian ICU patients, ANZROD outperformed APACHE III and SOFA in short-term mortality prediction, but all models showed poor calibration and declining discrimination over longer horizons. Current scores remain useful for benchmarking but are limited for individual decision-making, underscoring the need for age-specific prognostic tools that integrate frailty, function, and geriatric priorities.

Trial registration: ACTRN12625000775415 (23/07/2025).

Objective: We examined the relationship between modeled opioid concentration and quantitative pupillary measures during remifentanil infusion sequences with particular attention to the "light-off" (LO) reflex.

Methods: Ten volunteer subjects were recruited to undergo two 10-min remifentanil infusion protocols. Pupillary unrest in ambient light (PUAL) and LO were measured at baseline and every 2.5 min during the first 10-min infusion-25-min recovery sequence, and after a wash-out period, the Neurological Pupillary index (NPi) and LO were measured during an identical infusion-recovery sequence. We tested proportional change in each parameter from baseline as indicators of dynamic opioid effect.

Results: On average, remifentanil decreased both LO dilation and PUAL by > 85%, decreased pupil diameter by > 48%, but did not significantly alter the NPi. Hypoxia occurred in 15/16 sequences. LO and PUAL both showed excellent discrimination between high-toxic versus zero-moderate opioid effect. In contrast to PUAL and LO, the scaled pupillary light reflex measurement (NPi) was not altered by opioids.

Conclusion: LO and PUAL were robust indicators of opioid effect and provided equivalent estimates of respiratory depression risk in our healthy awake subjects. Compared with PUAL, LO offers the advantage of being intuitive and easily derived at the bedside without need for specialized software.

Significance: Measurement of the pupillary LO reflex with a portable pupillometer provides a simple, discriminating measure of opioid effect. Remifentanil blocks the pupillary LO reflex.

Background: Spinal anesthesia-induced hypotension is a common complication during cesarean delivery, often requiring vasopressor support, and is associated with maternal discomfort. Caffeine, a central nervous system stimulant with well-documented cardiovascular effects, may provide a simple adjunct to enhance hemodynamic stability. We aimed to evaluate the efficacy of a single preoperative 200 mg oral caffeine dose in reducing the incidence and severity of hypotension following spinal anesthesia in healthy patients undergoing elective cesarean delivery.

Methods: In this randomized controlled trial, 90 patients classified as ASA II and scheduled for elective cesarean delivery under spinal anesthesia were assigned to receive either 200 mg oral caffeine or a placebo 30 min before the procedure. Hemodynamic parameters, the incidence and severity of hypotension, baseline and 60 min post-administration serum caffeine levels, ephedrine requirements, incidence of postoperative nausea and vomiting, and post-dural puncture headache were recorded and analyzed.

Results: Caffeine administration significantly reduced the incidence of hypotension (9% vs. 33%, p < 0.05). Severe hypotension was not observed in the caffeine group. Patients in the caffeine group demonstrated greater hemodynamic stability, with a delayed onset of hypotension and reduced ephedrine requirements. No significant differences were observed in the incidence of bradycardia, tachycardia, or reactive hypertension. Neonatal outcomes were comparable between the groups. Additionally, caffeine was associated with lower rates of postoperative nausea and vomiting (2% vs. 20%) and post-dural puncture headache (2% vs. 16%) at 24 h.

Conclusion: Preoperative administration of 200 mg oral caffeine is a cost-effective strategy for reducing spinal anesthesia-induced hypotension, the incidence of postoperative nausea and vomiting, and post-dural puncture headache in healthy patients undergoing elective cesarean delivery. These findings support further investigation of the role of caffeine as an adjunct in obstetric anesthesia.

Trial registration: The study was registered by the principal investigator (M. Helmy) at ClinicalTrials.gov under the identifier NCT07076654 on July 11, 2025.

Background: Robotic liver transplantation (LT) is a recent advancement in minimally invasive surgery; however, perioperative and anaesthetic management have not yet been described in detail. This study aimed to characterise the anaesthetic course of fully robotic LT and compare the perioperative outcomes with those of propensity score-matched open LT.

Methods: We conducted an observational, retrospective, single-centre study at Modena University Hospital. Fully robotic LTs were compared with matched open LTs. Matching was performed in a 1:1 ratio according to age, sex, BMI, year of transplantation, indication, presence of hepatocellular carcinoma, and MELD score.

Results: Eighteen robotic and 457 open LTs were initially identified; after matching, 11 robotic and 11 open cases were included in the study. In the robotic group, the operative time was longer (p < 0.05) and associated (p < 0.05) with higher lactate levels, greater norepinephrine requirement, and larger blood loss, requiring a larger use of blood components compared to standard open surgery. Postoperatively, extubation timing, cardiovascular, respiratory, and liver function recovery, and renal complications were comparable between the groups. Pain control required less opioids (p < 0.05) and ICU (p > 0.05) and hospital stay (p < 0.05) were lower in robotic than in open surgery.

Conclusions: Fully robotic LT is associated with longer operative times, greater blood loss, and increased haemodynamic demands. Despite these challenges, postoperative recovery, particularly hospital stay, appears to be favourable. Larger multicentre studies are needed to validate these findings and refine the anaesthetic strategies.

Background: Ventilator-associated pneumonia (VAP) is a preventable complication of invasive mechanical ventilation (IMV) with a significant healthcare impact. Early risk prediction is crucial, but current models lack real-time adaptability. This study develops a real-time VAP prediction model using machine learning and high-resolution EHR data from the MIMIC-III database.

Methods: We analyzed 3523 ICU stays (3204 patients) from MIMIC-III, including adults who received IMV for at least 48 h. VAP was labeled based on microbiological cultures and antibiotic initiation. A real-time ensemble model of XGBoost regressors was developed to predict time to VAP onset, incorporating vital signs, ventilator data, and lab results. Two static classifiers (24 h and 48 h) were also compared.

Results: VAP occurred in 595 ICU stays (16.89%), with an incidence rate of 23.77 per 1000 IMV-days. Median VAP onset was 113.5 h post-IMV. The real-time model outperformed static models with a C-index of 0.68, AUROC of 0.71, and AUPRC of 0.36. It provided a median lead time of 53 h before VAP onset, with key predictors including temperature, respiratory rate, and minute ventilation.

Conclusion: We present a real-time VAP prediction model that outperforms static classifiers, providing actionable lead time for proactive microbiological surveillance. The model enables risk stratification for enhanced monitoring and, when clinically indicated, timely targeted antimicrobial therapy. Future work will focus on multicenter prospective validation and integration into ICU workflows to assess clinical utility and impact on patient outcomes.

Background: Idiopathic systemic capillary leak syndrome (SCLS), or Clarkson syndrome, is a rare disorder characterized by increased capillary permeability causing fluid leakage into interstitial spaces. SCLS remains underdiagnosed in both adults and children due to its resemblance to septic shock, with particularly limited pediatric data available.

Methods: We conducted a retrospective analysis of six pediatric patients with seven SCLS episodes diagnosed in two Italian PICUs over a ten-year period.

Results: Patients aged 4-10 years presented with hypovolemic shock and the characteristic triad of hypovolemia, hemoconcentration, and hypoalbuminemia without albuminuria. No monoclonal gammopathy was observed. Viral infections (75%, predominantly influenza) were identified as triggers. Common complications included rhabdomyolysis (100%), compartment syndrome (100%), acute kidney injury (71.4%), and cardiac involvement (71.4%). Cardiothoracic index measurements showed significant elevation during acute phases (mean 58% ± 1%) with reduction after treatment (mean 51% ± 1%). Mean hospitalization was 20 days (12 in PICU). One patient died from refractory ventricular fibrillation, another experienced recurrence.

Conclusions: Pediatric SCLS represents a severe condition with distinctive characteristics compared to adult cases, including absence of monoclonal gammopathy, frequent viral triggers, and universal complications of rhabdomyolysis and compartment syndrome. Early recognition using our proposed diagnostic red flags can facilitate prompt intervention and improve outcomes.

Transesophageal echocardiography (TEE) has emerged as a transformative tool in the management of cardiac arrest, offering significant advantages over traditional transthoracic echocardiography (TTE) by enabling continuous, high-resolution cardiac imaging during resuscitation. Initially used in operating rooms before 2000, TEE expanded into emergency departments (EDs), intensive care units (ICUs), and even prehospital settings. Its superior imaging capability during cardiopulmonary resuscitation (CPR) supports rapid diagnosis, optimization of compression quality, and more accurate rhythm assessment, including the differentiation between asystole and fine ventricular fibrillation. TEE has been shown to influence intra-arrest clinical decision-making in up to 78% of cases, often revealing pathologies-such as aortic dissection or cardiac rupture-not detected by TTE. Importantly, TEE aids in identifying reversible causes of cardiac arrest, guiding high-quality CPR by assessing left ventricular outflow tract (LVOT) obstruction, and shortening the duration of pulse checks. It also may play a role in extracorporeal CPR (ECPR) and organ donation procedures, particularly in cannulation and monitoring during extracorporeal membrane oxygenation (ECMO) and normothermic regional perfusion (NRP). TEE use is feasible and safe during cardiac arrest, with high insertion success rates and minimal complications when performed by trained personnel. While cardiologists, anesthesiologists, and intensivists traditionally perform TEE, simplified training protocols now enable emergency physicians to safely and effectively deploy TEE in critical settings. Protocols such as "POCUS-TEE" may promote rapid acquisition of essential views, facilitating broader implementation. Despite the potential benefits, several barriers to widespread adoption remain. These include the availability of equipment, limitations in training, and concerns regarding the safety of the probe during defibrillation. Resuscitation guidelines recommend removing the transesophageal echocardiogram (TEE) probe, or at least insulating it, as a precaution during defibrillation. TEE is particularly valuable when TTE proves inadequate, such as in obese or mechanically ventilated patients. In conclusion, TEE might enhance the quality and precision of resuscitation in cardiac arrest and has the potential to improve survival and neurological outcomes in selected cases. Its broader adoption hinges on institutional support, streamlined protocols, operator training, and further research to validate its impact on patient-centered outcomes. As technology advances and clinical integration improves, TEE may become a cornerstone of advanced life support in both in-hospital and out-of-hospital settings.

The ongoing revolution in artificial intelligence (AI) is reshaping perioperative care, including obstetric anesthesia. This narrative review synthesizes major AI applications in cesarean delivery, the world's most common inpatient surgery. Integrating history, obstetric factors, physiological variables, and imaging, AI tools enhance preoperative evaluation (estimation of risks of difficult airway), prediction of adverse events, ultrasound spine evaluation for neuraxial procedure, and postpartum hemorrhage. Language models can bridge consent and education gaps, while improving detection and treatment of postoperative pain. Machine learning models improve hemodynamic management with prediction of spinal-induced hypotension, assisted fluid management, and vasopressor requirements, with reduction of hypotensive burden. Yet cesarean-specific evidence remains limited and heterogeneous, with uncertain effects on maternal-neonatal outcomes. While promising, AI cannot replace the expertise and clinical judgment of a trained obstetric anesthesiologist. It should, instead, be viewed as a valuable tool to facilitate and support our practice, and multicenter prospective trials are needed to guide implementation.

Despite the refinement in guidelines and improving outcomes, a subset of patients with septic shock fails to respond to treatment and progresses into refractory septic shock with an associated high morbidity and mortality. This population remains underrepresented in clinical trials due to their heterogeneity, in addition to time and ethical constraints. As a large specialist referral centre, we propose an updated, pragmatic, and largely inexpensive approach based on our current clinical practice, which focuses on early multimodal therapy, aiming to reduce the detrimental effects associated with high-dose vasopressors.

Background: Chronic postsurgical pain affects surgical patients with a mean incidence of approximately 20%, posing a major public health concern with substantial implications for patients and healthcare systems. Despite increasing knowledge of risk factors, the incidence of chronic postsurgical pain remains high. Hence, there is growing interest in developing individualised pain management strategies using predictive risk. A novel chronic postsurgical pain risk assessment system using machine learning is under development in Western Norway. As a first step in implementing the risk assessment system, this study explored how in-hospital healthcare professionals perceive the potential utility of access to individualised chronic postsurgical pain risk profiles for clinical practice.

Methods: This qualitative study included seven focus groups with 39 healthcare professionals from postanaesthesia care units, surgical units and orthopaedic wards across two hospitals in Norway. Data were analysed inductively using reflexive thematic analysis.

Results: Our analyses yielded two overarching themes: (1) Lack of fit of risk predictions and (2) potentials of knowing risk profiles. Participants questioned the applicability of chronic postsurgical pain predictions in the in-hospital settings, highlighting role boundaries, time constraints, and limited influence over long-term outcomes. However, they also identified the benefits of risk awareness, including improved clinical reflection, more cautious decision-making, and an enhanced potential for individualised treatment and care.

Conclusion: Healthcare professionals expressed a balanced view of chronic postsurgical pain risk profiles, recognising both scepticism about them and their potential benefits. Effective implementation will require predictive validity, clear guidance, and cross-disciplinary collaboration. Education and training will be essential to support clinicians in interpreting and acting on risk information.