Routinely collected real time data from ICU bedside monitors is a valuable source of information which can assist the clinical team caring for a patient. This data almost always contains artifacts caused by clinical staff attending to the patient throughout the day, e.g. blood samples being taken, body washing, administration of IV medication. We present a small (10 patient) study but with accurately annotated clinical event data constructed by observing standard clinical procedures and merging this information with the physiologic data (ECG, ABP, CVP) from bedside monitors. In addition, to assess the feasibility for using accelerometers in the ICU environment, we received ethical approval to place an accelerometer [1] on the chest of all patients and the X, Y & Z accelerometer plane data was merged with the physiological and event annotation data. The non-trivial methods required to capture the data and initial characterisation of results are detailed and discussed. A baseline model cross-correlating ECG and ABP waveform signals over a 10 s window was developed and features derived from this model were used in both GLM and Random Forest (RF) models for predicting events that routinely cause movement related artifact in waveform quality time-series data. RF model performance improved when accelerometer data was added to the model for most movement related artifact types. For the GLM models, only selected models (ANY Type, Turns, Bed Movement and Blood Sample) showed an improvement in model fit with the addition of the accelerometer data (P < 0.01). Implications for observational ICU research and potential "Closed Loop" clinical modelling studies are discussed.

Early detection of clinical deterioration on surgical wards is vital to prevent adverse outcomes. Conventional ward monitoring using the Modified Early Warning Score (MEWS) relies on intermittent nurse-measured vital signs, but its role may change when continuous wearable monitoring is already available. This study evaluated the transition from mandatory to discretionary nurse-measured MEWS assessment in a continuously monitored surgical ward and compared the diagnostic performance of two wearable-derived Early Warning Scores (EWS) with conventional MEWS. In this single-center cohort study conducted between October 2024 and June 2025, postoperative patients were continuously monitored using a wearable sensor measuring heart rate, respiratory rate, and oxygen saturation. Continuous wearable monitoring was available throughout the entire study period. During Phase 1, nurse-measured MEWS assessments were mandatory according to local protocol. During Phase 2, nurse-measured MEWS assessments were performed at staff discretion. Two wearable-derived algorithms were evaluated: the Continuous Remote EWS (CREWS), based on a previously developed algorithm, and Remote 3 EWS (R3EWS), a modified version of the hospital's standard MEWS using three parameters. The primary outcome was diagnostic accuracy for clinical deterioration, defined as postoperative complications graded Clavien-Dindo ≥ II. A total of 544 admissions (516 patients) were included, of which 98 admissions (18%) experienced clinical deterioration (comprising 154 clinical deterioration events). Following the transition from mandatory to discretionary nurse-measured MEWS assessment, intermittent nurse-measured vital signs were performed less frequently, while MEWS ≥ 3 occurred more often, resulting in higher MEWS sensitivity and lower specificity. In the separate diagnostic performance analysis, R3EWS showed slightly higher AUROC values (0.61) compared with MEWS (0.58) and CREWS (0.59), although differences were modest. Wearable-derived alarm frequency increased 6-8 h before clinical deterioration. In a surgical ward where continuous wearable monitoring was already available, transition from mandatory to discretionary nurse-measured MEWS assessment was associated with fewer intermittent observations, higher MEWS sensitivity, and lower specificity. Wearable-derived EWS algorithms showed comparable but modest discriminative performance relative to conventional MEWS. The increased number of alerts highlights the need for prospective validation and clear protocols for integrating continuous monitoring data into ward escalation workflows.

To demonstrate a method for the identification of safe cerebral perfusion pressure (CPP) in patients with severe traumatic brain injury using brain tissue oxygen (PbtO₂) readings as a surrogate marker of cerebral blood flow. Retrospective analysis of prospectively collected data from patients with severe TBI (sTBI). A total of 9 patients with sTBI enrolled in the BONANZA-GT trial with continuous monitoring of mean arterial blood pressure (MAP), intracranial pressure (ICP), cerebral perfusion pressure (CPP) and PbtO₂. MAP, ICP, CPP and high frequency PbtO₂ signals were continuously recorded. stoCPP was calculated as the coefficient of determination (r²) of short-term changes in PbtO₂ regressed on short-term changes in CPP within non-overlapping 144-second correlation windows, assigned to 5-mmHg CPP bins using a window-median binning approach. Algorithm parameters were optimised using Bayesian optimisation with a composite scoring function benchmarked against CPPopt. Per-subject r² thresholds were derived by block-shuffle calibration (median 0.12, range 0.06-0.26). stoCPP was successfully generated for all 9 patients from raw unfiltered data. The lower limit of the stoCPP range (LLstoCPP) was defined in approximately 22% of analysis windows, as it is identifiable only during episodes of CPP perturbation below the stable range. CPPopt point estimates fell within the stoCPP range in 76.2% of observations, with a mean stoCPP width of 22.8 mmHg. Bland-Altman analysis of lower bounds (n = 102 independent estimates) demonstrated a mean bias of 4.1 mmHg (SD 7.3 mmHg; 95% limits of agreement - 10.1 to + 18.4 mmHg). Trough depth characteristics of the stoCPP range degraded significantly with temporal misalignment of the PbtO₂ signal (Kendall's W = 0.557, p < 0.001), supporting the physiological basis of the method. stoCPP was defined in all patients from raw unfiltered data using Bayesian-optimised parameters and per-subject calibrated r² thresholds, and demonstrated close correspondence with CPPopt. We propose further validation of this method utilising alternative PbtO₂ probe types and larger patient cohorts to assess the validity of time below LLstoCPP as a marker of ischaemia and determinant of outcome in sTBI.

Automated pupillometry provides a standardised, quantitative, highly reproducible measurement of the pupillary light reactivity and other pupillary variables. The Neurologic pupillary index (NPi) has shown good prognostic value in patients with acute brain injury and cardiac arrest. However, few data on the comparison of NPi with a recently introduced index (quantitative pupillary index, QPI) are available. To compare the performance of two commercially available automated pupillometry devices, and in particular QPI and NPi, in a cohort of neuro-critically ill patients. Single-center observational study, including adult (> 18 years) patients admitted to a neuro-intensive care unit over a 6-month period. Pupillary reactivity was assessed in both eyes of each patient using the two pupillometers currently available, the Neuroptics NPi-300 (Neuroptics, Irvine, CA, USA) and the NeuroLight (ID-Med, Marseilles, France). A total of 73 patients were included in the final analysis, with a median age of 65 years (interquartile range [IQR] 54-77). The mean bias between the two pupillometers was 0.44 mm (95% Limit of Agreement [LoA] - 0.43 mm to 1.31 mm) for pupil diameter, 0.57% (95% LoA - 7.63% to 8.77%) for pupillary reactivity, and - 0.82 (95% LoA - 2.70 to 1.05) for NPi vs. QPI comparison. A very strong correlation was found between the two devices in terms of pupil diameter measurement (ρ = 0.91; 95% Confidence Interval [CI] 0.86 to 0.95; p < 0.0001) and pupil reactivity (ρ = 0.92; 95% CI 0.87 to 0.95; p < 0.0001). NPi and QPI were well correlated (ρ = 0.75; 95% CI, 0.62-0.84; p < 0.0001). In this cohort study, a significant statistical correlation between the measurements provided by two available pupilometers was found; however, NPi and QPI might not be clinically interchangeable.

The purpose was to describe patient monitor alarm burden in a cardiac surgical intensive care unit and explore whether routinely available patient characteristics are associated with alarm frequency, type, and duration. Secondary objectives were to assess indirect signs of alarm fatigue and compare alarm rates with two previous studies using identical monitors in different clinical units. We retrospectively collected patient-monitor alarm data and patient characteristics over four consecutive weeks in the cardiac surgical intensive care unit of Helsinki University Hospital. The study population comprised 81 patients. Alarm data were analysed descriptively and modelled using negative-binomial regression. Patients produced 45,044 alarms, with a median alarm rate of 6.8 [4.8 to 10.8] alarms per patient per hour. Older age was associated with fewer alarms (IRR = 0.90, 95% CI 0.83 to 0.98, P < 0.05). ASA 5 classification was associated with more low- ETCO2 alarms (IRR = 9.05, 95% CI 3.37 to 27.88, P < 0.05). Smoking was associated with more low invasive blood pressure alarms (IRR = 1.94, 95% CI 1.05 to 3.90, P < 0.05). Alarm load was high, clinician interaction with alarms was limited, and many alarms were momentary. Alarm rates were similar across three clinically different units using identical monitors. Routinely available patient characteristics were associated with selected alarm patterns, but the findings are hypothesis-generating. Similar alarm rates across different units suggest that default settings may not reflect unit- or patient-specific monitoring needs. The results suggest low positive predictive value and possible alarm fatigue, supporting unit-level alarm review that incorporates patient-level alarm patterns and signal-quality improvement, with prospective validation before patient-specific alarm-management changes.

The increasing availability of advanced monitoring technologies in anaesthesia and critical care has enabled continuous assessment of neurological, haemodynamic, and oxygen delivery parameters. Despite this, integrating multiple data streams into coherent, physiology-guided clinical decisions remains challenging, and unified interpretive frameworks remain limited. To introduce the HOPFIT framework - a conceptual model linking continuous neuro-cardiovascular monitoring to structured, physiology-guided clinical reasoning across six interdependent domains. HOPFIT organises six therapeutic domains - Hypnotics, Opioids, Pressors, Fluids, Inotropes, and Transfusion - around the interdependence of brain function, cardiovascular physiology, and oxygen delivery. Rather than introducing new monitoring modalities, the framework provides a unifying structure for bedside interpretation of existing data, including processed electroencephalography, cerebral oximetry, dynamic indices of fluid responsiveness, non-invasive haemoglobin, and cardiac output monitoring. An original pharmacodynamic risk surface illustrates the interaction between hypnotic and analgesic agents, and a conceptual hypotension framework integrating EEG and haemodynamic data illustrates how multimodal information may support clinical reasoning. Tables summarise monitoring parameters, commonly used ranges, and potential clinical applications across patient groups. HOPFIT is intended as a conceptual and educational tool to support integration of multimodal physiological data at the bedside, rather than a prescriptive clinical protocol. Prospective studies are required to determine its impact on clinical decision-making and patient outcomes.

Spinal anesthesia during cesarean delivery frequently causes maternal hypotension. Early detection of hemodynamic compromise before blood pressure decline could enable timely intervention. This study evaluates the use of serial transthoracic echocardiography (TTE) as a monitoring tool for early detection of hemodynamic fluctuation. To characterize the temporal sequence of hemodynamic changes using serial TTE, with emphasis on the relationship between stroke volume (SV) reduction and subsequent hypotension, and to assess the potential of TTE‑derived parameters as physiological indicators of preload changes. Prospective observational study. Single-center trial. Thirty healthy term pregnant women (ASA I‑II) undergoing elective cesarean delivery under spinal anesthesia. Serial TTE examinations were performed at 13 predefined time points. Left ventricular diameters (LVIDd, LVIDs) were measured and volumes (EDV, ESV) were derived using the Teichholz method, from which stroke volume, cardiac output, cardiac index, and ejection fraction were calculated. Temporal relationship between SV decline and hypotension onset. After spinal anesthesia, left ventricular end-diastolic volume (EDV) decreased by 34.6%, stroke volume (SV) by 36.2%, cardiac output (CO) by 38.7% (all P < 0.001). Hypotension (SBP drop > 20%) occurred in 76.7%. Crucially, SV reduction preceded blood pressure fall: at the first post‑spinal measurement, SV had already decreased by 30.6%, while systolic BP had fallen by only 12.2%. Left ventricular ejection fraction (LVEF) remained stable throughout. After fetal delivery, all parameters partially recovered. Serial TTE monitoring detects a significant reduction in SV consistently before hypotension becomes clinically apparent, demonstrating the potential of TTE‑derived parameters as physiological indicators of preload changes. These findings establish a monitoring‑based physiological baseline for future hypothesis‑testing in obstetric anesthesia.Trial registration: Chinese Clinical Trial Registry (ChiCTR2500111576; registered 3 November 2025).

Perioperative anaesthetic complications are a major public health concern, as they are associated with increased mortality, prolonged hospital stay, higher healthcare costs, and compromised patient safety. Despite advances in perioperative monitoring and risk stratification, early and personalised identification of patients at risk remains limited. Artificial intelligence (AI) has emerged as a promising approach to improve prediction of perioperative anaesthetic complications, although current evidence is heterogeneous. To synthesise the available evidence on the use of artificial intelligence models to predict perioperative anaesthetic complications in adult surgical patients. A scoping review was conducted following Joanna Briggs Institute methodology and reported according to PRISMA-ScR guidelines. The protocol was registered in the Open Science Framework (DOI: https://doi.org/10.17605/OSF.IO/QU6WY). PubMed, Scopus, and Web of Science were searched for studies published between 2015 and 2025. Studies applying machine learning, deep learning, or hybrid models to predict perioperative anaesthetic complications were included. Evidence was synthesised using conceptual mapping and thematic analysis. Fifty-one studies were included. Haemodynamic and cardiovascular outcomes, particularly intraoperative hypotension, were most frequently studied, followed by renal, neurological, and respiratory complications, postoperative pain, and perioperative resource utilisation. Most models showed moderate to high predictive performance, while external validation and evaluation of clinical impact were uncommon. Artificial intelligence has substantial potential to support prediction of perioperative anaesthetic complications. However, clinical implementation is limited by methodological weaknesses, including predominantly retrospective designs, limited external validation, and scarce assessment of real-world impact. Prospective multicentre studies are required.

To determine whether the subcutaneous compartment retains an interpretable glucose signal during extreme perioperative perfusion stress and whether agreement with arterial glucose differs by anatomical site. In this prospective mechanistic observational study, 20 adults undergoing pulmonary endarterectomy with deep hypothermic circulatory arrest underwent subcutaneous microdialysis at upper thoracic and periumbilical sites. Recovery-corrected interstitial glucose was converted to calibrated subcutaneous glucose using one-point calibration. Agreement with arterial glucose was evaluated using absolute relative difference (ARD) and repeated-measures Bland-Altman analysis. Exploratory mixed-effects models examined associations of ARD with procedural phase and treatment-related variables. We analysed 743 paired measurements. Overall median ARD was 12.9% (IQR 5.9-21.9), and mean ARD was 17.2% (SD 17.2). The procedural phase was associated with ARD (global P < 0.001), with the greatest discordance during rewarming and early postoperative recovery rather than during the DHCA interval. Agreement was numerically more favourable at the upper thoracic site; however, this between-site difference did not reach statistical significance in the phase-adjusted analysis (P = 0.053). Under recovery-corrected and one-point-calibrated conditions, subcutaneous glucose retained interpretable tracking of arterial glucose during extreme perioperative perfusion stress. These findings argue against a uniform loss of subcutaneous glucose information in this setting and suggest that discordance is phase-dependent and may vary by anatomical site.

To evaluate whether jugular venous NIRS (rSjvO2) and baseline-calibrated cerebral NIRS (cal-rScO2) can serve as surrogates for mixed venous oxygen saturation (SvO2), and whether these inputs can support continuous cardiac output (CO) estimation during off-pump coronary artery bypass grafting (OPCAB). In this prospective observational study, 25 adults undergoing elective OPCAB were enrolled at a single tertiary center. NIRS sensors were placed over the left internal jugular vein (IJV) under ultrasound guidance and over the bilateral forehead. Pulmonary artery catheter measurements served as the reference for SvO2 and thermodilution CO (COTD). Agreement was assessed using Bland-Altman analysis with linear mixed-effects modeling, and trending ability by error grid analysis. CO was estimated using a previously derived modified Fick equation. rSjvO2 showed modest bias versus SvO2 (2.0%) but wide limits of agreement (LoA, - 14.9 to 18.9%). Accuracy varied with jugular anatomy, with the lowest error at an IJV depth of 1.38 cm. Cal-rScO2 showed better agreement (bias 0.8%; LoA, - 7.8 to 9.4%), excellent discrimination for SvO2 <65% (AUC, 0.970), and good trending ability. In contrast, NIRS-based CO estimation showed significant proportional bias, 33.0% percentage error (95% CI, 27.4 to 41.5%), and poor trending ability. Cal-rScO2 appears to be a useful surrogate for SvO2 trending during OPCAB after simple one-point calibration. Jugular NIRS may be useful in selected patients, whereas NIRS-based CO estimation remained clinically insufficient.Trial registration: Clinical Research Information Service (CRIS), Republic of Korea (KCT0007918 November 18, 2022).

To investigate the relationship between hemoglobin concentration and peripheral perfusion index (PI) in normovolemic, hemodynamically stable patients. This prospective observational study included 90 adult emergency department patients, classified as anemic (n = 45) and non-anemic (n = 45) according to World Health Organization criteria. PI was measured using standardized pulse oximetry conditions. Group comparisons were performed using appropriate statistical tests. The association between hemoglobin and PI was evaluated using Spearman correlation. A multivariable linear regression model with log-transformed PI was used to identify independent predictors, adjusting for clinically relevant variables. Model assumptions and influential observations were assessed. PI values were significantly lower in anemic patients compared with non-anemic patients (median 2.1 vs. 3.5, p < 0.001). Hemoglobin concentration showed a moderate positive correlation with PI (r = 0.425, p < 0.001). In multivariable analysis, hemoglobin remained independently associated with PI (β = 0.306, p = 0.021). The model demonstrated modest explanatory power (adjusted R[2]=0.069). Sensitivity analysis excluding influential observations yielded consistent findings. Hemoglobin level is independently associated with PI in normovolemic patients. Although PI primarily reflects flow-related peripheral perfusion, hemoglobin contributes to its variability. These findings suggest that anemia and peripheral perfusion are interrelated but distinct physiological parameters that should be considered together in clinical interpretation.

Advanced haemodynamic monitoring of cardiac index is restricted to selected high-risk cases, leaving most patients undergoing major surgery without real-time assessment. We aimed to develop and internally validate a simplified bedside score for continuous, minute-by-minute screening of low cardiac index using routinely available parameters. Retrospective single-centre study of adults undergoing major abdominal surgery with advanced haemodynamic monitoring (January 2023 to February 2026). Low cardiac index was defined as pulse contour cardiac index < 2.2 l.min[- 1].m[- 2]. The dataset was split 80/20 at the patient level. Twelve candidate features were ranked using SHapley Additive exPlanations (SHAP) values; the top five were used to fit a generalised linear mixed model, yielding the integer-weighted Low Cardiac Index (LCI) score. A rule-out threshold (sensitivity ≥ 0.90) was selected, and test-set performance was evaluated with patient-level cluster bootstrap. The analysis cohort comprised 136 patients and 42,346 timepoints (15.5% low cardiac index prevalence). The derived formula was LCI score = 57 + (2 × DBP) + (0.5 × Age) + PPV - HR - SBP. At a threshold of ≥ 30, the score achieved an area under the receiver operating characteristic curve of 0.84 (95% confidence interval 0.75, 0.91), sensitivity 0.92 (0.80, 0.99), and negative predictive value 0.97 (0.92, 1.00). Performance was stable across 20 random splits and outperformed individual haemodynamic parameters. The LCI score is a simple bedside rule-out tool for low cardiac index, with a score below 30 yielding a negative predictive value of 0.97. External validation and prospective assessment of clinical impact are warranted.

Arterial pressure waveform analysis has been used to assess hemodynamic status, however, investigations focusing on waveform morphology remain limited. In this study, we examined changes in selected morphological features of the arterial pressure waveform following volume loading versus vasopressor administration in critically ill patients. In this retrospective observational study, we analyzed adult patients admitted to an intensive care unit between April 2022 and June 2023. Participants were included if they exhibited increased mean arterial pressure (MAP) following either blood return during hemodiafiltration discontinuation (Volume group) or intravenous phenylephrine administration (Pressor group). Arterial waveforms were extracted from one minute before to four minutes after the intervention. We compared 21 waveform-derived features and their normalized parameters at peak MAP with baseline values. Pre- and post-intervention differences for each parameter were assessed using the Wilcoxon matched-pairs signed-rank test. The Volume group comprised 26 patients, and the Pressor group included 20 patients. Systolic angle, systolic dP/dt, and systolic area under the curve (AUC) increased in both groups following intervention. After pulse-pressure normalization, however, systolic AUC increased significantly in the Volume group (median paired difference 4.6 [-0.3-10.2] ms, 95% CI 1.9 to 7.9, P = 0.0038) but not in the Pressor group. This elevation in systolic AUC occurred mainly during the systolic descending phase (4.4 [1.1-9.5] ms, 95% CI 2.7 to 9.1, P = 0.00021). The systolic descending duration exhibited a significant prolongation in the Volume group, whereas no significant change was observed in the Pressor group. Arterial waveform morphology exhibited different patterns of change following blood return versus phenylephrine administration, particularly in the systolic descending phase. Waveform morphology analysis may provide complementary physiological information for monitoring the mechanisms underlying hemodynamic changes.