Background Artificial intelligence (AI) can enhance diagnostics, treatment, and workflow efficiency. However, successful integration into clinical practice depends on users' acceptance. Objective To investigate benefits,...Background Artificial intelligence (AI) can enhance diagnostics, treatment, and workflow efficiency. However, successful integration into clinical practice depends on users' acceptance. Objective To investigate benefits, barriers, and challenges of AI applications among anaesthesia and intensive care professionals. Design International online survey. Main outcome measures The survey included items on familiarity and experiences with AI applications, perceived benefits, concerns, and demographic variables. Descriptive analyses, fisher exact tests, χ²-tests, odds ratios, and Spearman rank correlations were used to explore associations between responses and demographics. Results The survey was distributed by the European Society of Anaesthesiology and Intensive Care in 2023. A total of 510 respondents completed the entire survey, primarily from Europe (78%) and Asia (14.5%), and the majority were board-certified anaesthesiologists (86.3%). 86.5% of the respondents were aware of AI applications, but only 36.8% reported regular encounters. Familiarity was higher among males and intensive care specialists. 94.5% expressed interest in AI training, particularly younger and less experienced professionals. 94.7% expressed willingness to use AI applications, citing benefits such as improved decision-making (92.7%), complication anticipation (88.6%), and workload reduction (80.1%). Younger and female respondents were more optimistic about AI's benefits. Key concerns included lack of explainability (68.4%), over-reliance on AI (80.8%), and medico-legal uncertainties (58.4%). Scepticism was attributed to insufficient validation studies and fears of inaccurate outputs, particularly among experienced clinicians. Conclusions This international survey shows cautious optimism among anaesthesia and intensive care professionals regarding AI applications. Adoption in clinical practice requires tailored training that accounts also for demographic-specific concerns, robust validation, and clear ethical and legal frameworks.
Continuous monitoring of vital signs after hospital discharge may support early recognition of deviating vital signs. However, the utility may be challenged by high alert frequencies. This exploratory study aimed to asse...Continuous monitoring of vital signs after hospital discharge may support early recognition of deviating vital signs. However, the utility may be challenged by high alert frequencies. This exploratory study aimed to assess the impact of evidence-based augmented filtering algorithms on alert frequency following discharge. Adult patients (≥ 18 years) discharged after acute medical admission were monitored continuously using wearable devices that measured heart rate, respiratory rate, blood pressure, and oxygen saturation. The primary outcome was the number of alerts per patient per day. We compared outcomes across three filtering strategies: (1) no filtering, (2) artefact removal, and (3) filtering with artefact removal and clinical criteria based upon severity and duration. Ninety-eight patients were enrolled; the total vital sign alert frequency was reduced from a median of 74 [IQR 36-125] to 5 [IQR 1-13] alerts/patient/day following application of the clinical criteria filters, corresponding to an 84% reduction (p < 0.001). Alert frequency following the three filtering approaches was 74 [IQR 36-125], 67 [IQR 33-103], and 5 [IQR 1-13] alerts/patient/day, respectively, p < 0.001. Artefact removal and the application of filters based on severity and event duration significantly reduced alert frequency in patients continuously monitored at home after hospital discharge. Further studies are needed to evaluate clinical safety and predictive value.
To compare the safety and effectiveness of distal radial artery (DRA) versus conventional radial artery (CRA) catheterization for invasive arterial blood pressure monitoring. This meta-analysis followed PRISMA guidelines...To compare the safety and effectiveness of distal radial artery (DRA) versus conventional radial artery (CRA) catheterization for invasive arterial blood pressure monitoring. This meta-analysis followed PRISMA guidelines. Randomized controlled trials published up to December 30, 2025 were systematically searched in PubMed, Embase, Web of Science, the Cochrane Library, CINAHL, CNKI, Wanfang, VIP, and SinoMed. Two reviewers independently screened studies, extracted data, and assessed risk of bias. Meta-analyses were conducted using Review Manager 5.4 and Stata 18.0, and evidence quality was evaluated with the GRADE system. 12 randomized controlled trials (RCTs) involving 1,790 participants were included. For the primary outcomes, compared with CRA, DRA was associated with lower incidences of haematoma (RR = 0.42, 95% CI: 0.24-0.73), catheter blockage (RR = 0.33, 95% CI: 0.14-0.81), and bleeding (RR = 0.25, 95% CI: 0.11-0.59), but a longer catheter insertion time (MD = 26.89, 95% CI: 5.28-48.50). For the secondary outcomes, DRA was associated with a shorter haemostasis time and reduced waveform instability, with no significant differences in first-attempt success rate or pain scores. Subgroup analysis showed a higher first-attempt success rate in patients aged ≤ 65 years (RR = 1.14, 95% CI: 1.02-1.28). DRA catheterization for invasive arterial blood pressure monitoring appears to be associated with fewer complications and more stable arterial waveform acquisition than CRA catheterization, despite a longer insertion time. DRA may represent a feasible alternative in surgical and intensive care settings; however, further high-quality studies are needed to confirm its long-term safety and generalizability.
Continuous monitoring with wireless wearable devices enables early detection of clinical deterioration and supports post-discharge monitoring. This study evaluated data availability and patient usability of a novel upper...Continuous monitoring with wireless wearable devices enables early detection of clinical deterioration and supports post-discharge monitoring. This study evaluated data availability and patient usability of a novel upper-arm wearable sensor for continuous monitoring, on the general ward and at home, and compared its intramural performance with the current local standard. This prospective observational study was conducted at an academic hospital involving surgical and gastroenterological patients. The viQtor was used alongside ViSi Mobile during hospital admission and subsequently stand-alone for seven days post-discharge. Data availability was measured by analyzing the proportion of missing data, usability with the System Usability Scale (SUS), and data agreement by comparing in-hospital vital sign measurements between the viQtor and ViSi Mobile. Forty-five patients (median age 64 years) participated, of whom 35 (78%) continued using viQtor at home. Data gaps increased with patient activity during admission and at home. In-hospital data availability for the viQtor was similar for pulse rate (87% vs. 85%) but lower for respiratory rate (56% vs. 83%) and oxygen saturation (SpO₂) (48% vs. 84%) compared to ViSi Mobile. At home, the data availability of the viQtor for pulse rate, respiratory rate and SpO was 77%, 46%, and 39%, respectively. ViQtor achieved a mean SUS score of 71.1, indicating acceptable usability. ViQtor showed slightly higher pulse rate values, and lower respiratory rate and SpO₂ values compared to ViSi Mobile. Data availability of the viQtor was high for pulse rate but limited for respiratory rate and SpO₂, compared to the ViSi Mobile. Activity lowered data availability of all parameters, particularly at home. Patient usability was acceptable. The viQtor showed small but significant vital-sign differences compared with the ViSi Mobile. Reducing activity-related data loss may further improve applicability for continuous monitoring.
Women with endometriosis are at increased risk of severe postoperative pain due to nociceptive sensitization. While multimodal analgesia reduces opioid use, the added value of objective nociception monitoring remains unc...Women with endometriosis are at increased risk of severe postoperative pain due to nociceptive sensitization. While multimodal analgesia reduces opioid use, the added value of objective nociception monitoring remains unclear. This study evaluated whether NOL®-guided opioid titration improves perioperative outcomes within a standardized multimodal regimen. In this prospective, randomized, single-blinded trial, premenopausal women undergoing laparoscopic surgery for suspected endometriosis or adenomyosis were assigned to NOL®-guided analgesia or standard care based on clinical assessment. All patients received a standardized multimodal protocol. The primary outcome was total perioperative opioid consumption. Secondary outcomes included postoperative pain scores (NRS) and PACU length of stay. Exploratory analyses assessed the association between preoperative pain (Mankoski Pain Scale, MPS) and postoperative outcomes. A total of 111 patients were analyzed (NOL®: n = 54; control: n = 57). Total perioperative opioid consumption did not differ significantly between groups (adjusted mean difference = 14 μg for Fentanyl and 52 μg for Remifentanil; p = 0.8). Surgery duration was an independent predictor of opioid use (p < 0.001) and PACU length of stay (p = 0.01), whereas treatment group had no significant effect. Postoperative pain scores were comparable between groups at all time points. NOL®-derived metrics were not associated with opioid consumption or pain. Higher preoperative MPS scores independently predicted higher pain scores in the late PACU phase. NOL®-guided opioid titration did not reduce perioperative opioid consumption or improve early postoperative outcomes compared with standard multimodal analgesia in women undergoing laparoscopic surgery for endometriosis.
Background The Analgesia-Nociception Index (ANI), derived from heart rate variability, reflects parasympathetic activity and is used to assess intraoperative nociception. β-blockers modulate autonomic tone and heart rate...Background The Analgesia-Nociception Index (ANI), derived from heart rate variability, reflects parasympathetic activity and is used to assess intraoperative nociception. β-blockers modulate autonomic tone and heart rate variability and may influence responses to nociceptive stimuli. This study investigated whether preoperative β-blocker use was associated with altered intraoperative ANI dynamics during laparoscopic cholecystectomy. Method The data of 49 patients with hypertension who underwent elective laparoscopic cholecystectomy under a standardized propofol-remifentanil anesthetic protocol were analyzed in this prospective observational study. Patients were classified according to chronic β-blocker use (β-blocker group, n = 24; non-β-blocker group, n = 25). Four predefined nociceptive stimuli were applied under stable anesthetic conditions. The minimum ANI value within 2 min after each stimulus (ANI_min) and the change from the pre-stimulus value (ΔANI) were analyzed using repeated-measures analysis of variance. Results Repeated-measures ANOVA demonstrated a significant main effect of time, indicating a consistent decrease in ANI in response to standardized nociceptive stimuli. No significant time × group interaction or main effect of group was observed for ANI_min or ΔANI. The temporal patterns and magnitudes of ANI changes were comparable between groups across all predefined time points. Conclusions Preoperative β-blocker therapy did not substantially alter stimulus-related ANI dynamics during laparoscopic cholecystectomy under propofol-remifentanil anesthesia. These findings suggest that ANI may remain interpretable as a dynamic within-patient autonomic response in chronically β-blocked patients under these anesthetic conditions.
This narrative review provides an overview of the evidence on transthoracic echocardiography (TTE) at rest in patients undergoing non-cardiac surgery with regard to detection of new diagnoses, TTE-driven management chang...This narrative review provides an overview of the evidence on transthoracic echocardiography (TTE) at rest in patients undergoing non-cardiac surgery with regard to detection of new diagnoses, TTE-driven management changes, and outcome impact of preoperative TTE. It summarizes the evidence on preoperative TTE and reviews the current recommendations by professional societies from both Europe and Northern America. TTE is a very important non-invasive diagnostic tool to obtain information on cardiac function before surgery. While studies on the effectiveness of preoperative TTE are limited, there is evidence that TTE before non-cardiac surgeries can detect new diagnoses in a relevant proportion of patients. Also, data on changes in management based on TTE findings is scarce. Both the American Heart Association and the American College of Cardiologists and the European Society of Cardiology have recently published updated recommendations regarding the use of TTE before non-cardiac surgery. While both societies are in favour of TTE in symptomatic patients and do not recommend TTE to be performed routinely, several differences remain. Information on outcome benefits in patients having received TTE prior to surgery is not conclusive. In this review we could detect several knowledge gaps concerning the yield and impact on management and outcome of preoperative TTE in non-cardiac surgery patients.
Spontaneous breathing activity during pressure support ventilation (PSV) is not beneficial for the assessment of static compliance (C). We designed a new scheme for determining C during PSV and assessed its accuracy usin...Spontaneous breathing activity during pressure support ventilation (PSV) is not beneficial for the assessment of static compliance (C). We designed a new scheme for determining C during PSV and assessed its accuracy using a lung model. A Hamilton C3 ventilator was connected to an ASL5000 lung simulator that simulated different lung mechanics [system's static compliance (C), 30 or 60 mL/cmHO; airway resistance (R), 5, 10, 15, or 20 cmHO/(L·s)]. PSV and volume-controlled ventilation (VCV) underwent activation with tidal volume (V) values of 5, 7, and 10 ml/kg. Respiratory mechanics were obtained and corrected with virtual extrapolation by applying RC and relevant equations. The repeated change in pressure support (PS) level was set at ± 1 cmHO. C was determined from volume fluctuations resulting from PS level changes. The classic C measurement was obtained with the end-inspiration approach during volume-controlled ventilation (C). Comparing the value of C, similar values for C were obtained at low PS level. By changing the PS level, the estimated C resulted in a relatively high error of about 10-15%.The novel scheme provided reliable estimated results for C during pressure support ventilation. Despite its limited accuracy affected by the spontaneous effort participation, this scheme may help in non-invasive, uninterrupted monitoring of C.
Esophageal pressure (PES), used as a substitute for pleural pressure in calculations of respiratory mechanics, are complex and influenced by several factors. High-resolution manometry (HRM) provides pressure readings ove...Esophageal pressure (PES), used as a substitute for pleural pressure in calculations of respiratory mechanics, are complex and influenced by several factors. High-resolution manometry (HRM) provides pressure readings over the esophageal length, which makes it possible to identify factors other than respiratory mechanics that influence esophageal pressure. The aim of this study was to describe and explore esophageal pressure in prone and supine position at different ventilatory settings. PES was measured using HRM in 20 mechanically ventilated patients before start of spinal surgery. PEEP of 5 and 12 cmHO were used in supine and prone position while data were recorded continuously. Mean end-expiratory (PES) and tidal variation of esophageal pressures (ΔPES) from sensors in different esophageal regions were compared including a selection of sensors with theoretically optimal pressure readings (PES). PES was significantly lower (MD 3.2-5.7), ΔPES was significantly higher (MD 0.8-1.6 cmHO) and cardiac oscillations were significantly smaller in prone compared to supine position. Differences were dependent on PEEP level and regions of esophagus included. Mean within patient variability of PES and ΔPES was 38-540% depending on PEEP level and body position. End-expiratory PES was lower than mean PES from larger esophageal regions and less variable. A large variation in PES is seen in both supine and prone position but depends on which sensors are included in calculations of PES. There are significant differences in absolute pressures and in influence from mediastinal organs between supine and prone position.
This paper introduces a wearable, cuffless blood pressure (BP) monitoring system for continuous, real-time measurement in the operating room. Traditional cuff-based BP monitoring is intermittent and unsuitable for surger...This paper introduces a wearable, cuffless blood pressure (BP) monitoring system for continuous, real-time measurement in the operating room. Traditional cuff-based BP monitoring is intermittent and unsuitable for surgery, while intra-arterial measurements are invasive. The system uses a finger clip with a force sensor to detect pressure changes caused by arterial pulsations, with a waveform decomposition analysis algorithm estimating systolic and diastolic BP in real time. A clinical validation study involving 46 patients in the operating room demonstrated the system's accuracy, with reference measurements from an intra-arterial catheter. The reference systolic pressure ranged from 83 mmHg to 194 mmHg, and diastolic pressure from 42 mmHg to 105 mmHg. The error between paired measurements was - 1.44 ± 6.64 mmHg for systolic and 3.35 ± 6.91 mmHg for diastolic, with the device's measurements meeting the ISO 81060-2:2018 accuracy specification of 5 ± 8 mmHg. Pearson correlation coefficients were r = 0.90 (p < 0.01) for systolic and r = 0.78 (p < 0.01) for diastolic, indicating significant correlations with intra-arterial catheter measurements. The data analysis also considered the IEEE 1708 standard, ensuring compliance with guidelines for non-invasive BP monitoring. In collaboration with the device developer, the authors designed the clinical calibration protocol and user interface to ensure compatibility with intra-operative monitoring workflows. This study represents the first clinical validation of a tactile force-sensing cuffless system against intra-arterial reference measurements, demonstrating its feasibility for real-time, non-invasive blood pressure monitoring during anesthesia.
This study aimed to achieve two primary objectives: (1) to evaluate the opioid-sparing effect of Surgical Pleth Index (SPI)-directed analgesia during surgery via a randomized controlled trial (RCT), and (2) to propose an...This study aimed to achieve two primary objectives: (1) to evaluate the opioid-sparing effect of Surgical Pleth Index (SPI)-directed analgesia during surgery via a randomized controlled trial (RCT), and (2) to propose and preliminarily assess a novel dynamic metric, Threshold-based Time-Weighted SPI (Tb-TW-SPI), which integrates stimulus intensity and duration, for its predictive efficacy regarding postoperative moderate-to-severe pain. Employing an RCT combined with exploratory analysis, 61 patients undergoing elective laparoscopic gynecologic surgery were randomized into an SPI-directed analgesia group or a conventional analgesia group. The primary outcome was total intraoperative remifentanil consumption. Postoperatively, an exploratory analysis of the control group data evaluated the correlation between Tb-TW-SPI and Numeric Rating Scale (NRS) pain scores in the post-anesthesia care unit (PACU), calculating its predictive value for moderate-to-severe pain (NRS ≥ 4). Results: The SPI-directed group required significantly less intraoperative remifentanil than the conventional group [median (IQR): 5.84(5.02,6.62)vs. 6.96(5.81,8.19)µg/kg/h; P = 0.016]. Postoperative pain scores did not differ significantly between groups (P > 0.05). Exploratory analysis of the conventional analgesia group revealed that Tb-TW-SPI values were significantly higher in patients with moderate-to-severe postoperative pain (NRS ≥ 4) compared to those without (P = 0.0417).The area under the ROC curve for Tb-TW-SPI predicting this pain was 0.74 (95% CI: 0.52-0.96), with 67% sensitivity and 76% specificity at an optimal cutoff of 1210. This RCT suggests that SPI-directed analgesia can safely and moderately reduce intraoperative remifentanil consumption. Furthermore, the proposed Tb-TW-SPI metric, in this exploratory analysis, suggests potential for predicting postoperative pain, though this finding requires validation in larger cohorts with higher-frequency SPI sampling, offering a new direction for SPI interpretation. Large-scale, multicenter trials are warranted to validate the predictive utility of Tb-TW-SPI. Clinical Trial Registration, China Clinical Trial Registry: ChiCTR2400088444.
Despite major technological advances and a growing evidence base, perioperative electroencephalographic (EEG) monitoring remains underused, even though general anaesthesia is fundamentally a brain-targeted intervention....Despite major technological advances and a growing evidence base, perioperative electroencephalographic (EEG) monitoring remains underused, even though general anaesthesia is fundamentally a brain-targeted intervention. Processed EEG (pEEG) captures clinically relevant brain physiology and provides actionable information to guide anaesthetic dosing, yet its adoption has been hindered by expectations of outcome certainty that exceed those applied to other standard monitors. Much of the controversy reflects misaligned endpoints-such as movement or binary awareness detection-rather than the domains in which EEG plausibly adds value, including avoidance of excessively deep anaesthesia and optimisation of drug titration. While large trials have produced heterogeneous results and have not demonstrated uniform benefit across broad surgical populations, some analyses have explored the possibility that EEG-guided anaesthesia may be more relevant in selected high-risk groups, particularly older adults vulnerable to postoperative neurocognitive disorders. Continued failure to integrate routine brain monitoring into standard anaesthetic practice may therefore pose a greater patient safety risk than adopting it in the setting of imperfect yet biologically coherent evidence.
This study aimed to compare optic nerve sheath diameter (ONSD) measurements obtained during hypercapnic and normocapnic periods in mechanically ventilated patients and to describe changes in ONSD in relation to ventilati...This study aimed to compare optic nerve sheath diameter (ONSD) measurements obtained during hypercapnic and normocapnic periods in mechanically ventilated patients and to describe changes in ONSD in relation to ventilation-associated carbon dioxide levels. The study included sixty intubated patients admitted to the intensive care unit with a diagnosis of pneumonia and who underwent mechanical ventilation. ONSD measurements were performed during periods of hypercapnia identified during routine monitoring. After adjustment of ventilator settings, arterial blood gas analysis (ABG) was obtained one hour later, and ONSD measurements were repeated when normocapnia was achieved. The mean age of the patients was 61.6 ± 17.9 years. During the hypercapnic period, the mean ONSD (average of both eyes) was 4.88 ± 0.31 mm. Following the achievement of normocapnia, the mean ONSD decreased significantly to 4.61 ± 0.24 mm (p < 0.001). Furthermore, a significant positive correlation was found between the change (∆) in PaCO and the reduction in mean ONSD (r = 0.538, p < 0.001). ONSD measurements differed significantly between hypercapnic and normocapnic periods defined by PaCO₂ levels. These findings suggest that ONSD may reflect physiological responses across different ventilatory CO₂ states.
Robotic-assisted laparoscopic prostatectomy (RALP) requires steep Trendelenburg positioning and CO₂ pneumoperitoneum, which may significantly influence cerebral hemodynamic homeostasis. While postoperative neurological c...Robotic-assisted laparoscopic prostatectomy (RALP) requires steep Trendelenburg positioning and CO₂ pneumoperitoneum, which may significantly influence cerebral hemodynamic homeostasis. While postoperative neurological complications are uncommon, there is concern regarding increased intracranial pressure and potential edema during prolonged procedures. Near-infrared spectroscopy (NIRS) allows continuous monitoring of cerebral oxygenation, rSO2 reflects the balance between oxygen delivery and consumption, integrating changes of oxygenated hemoglobin (ΔHbO2) and deoxygenated hemoglobin (ΔHHb). This study aimed to evaluate phase-dependent changes in cerebral oxygenation (rSO₂, ΔO₂Hb, ΔHHb) and optic nerve sheath diameter (ONSD) during RALP, and to assess their temporal correlation. This prospective observational study enrolled forty adult patients (ASA I-III) undergoing RALP in steep Trendelenburg position (30 degrees) at Nuovo Ospedale di Prato, Italy, between November 2024 and May 2025. Cerebral oxygenation was continuously monitored using near-infrared spectroscopy (NIRS). ONSD was measured bilaterally using ultrasound at predefined time points: baseline (T0), 45 min after Trendelenburg positioning (T1), 120 min after Trendelenburg (T2), during vesicourethral anastomosis (T3), and before extubation in neutral position (T4). Pearson correlation coefficients were calculated to assess the association between NIRS-derived parameters and ONSD changes. Compared with baseline, Trendelenburg positioning was associated with significant phase-dependent changes in cerebral oxygenation and hemoglobin-derived parameters. rSO₂ and ΔO₂Hb increased during early and prolonged Trendelenburg phases, peaking during vesicourethral anastomosis (T3), and partially decreased after return to supine position (T4). ΔHHb demonstrated a progressive increase during later intraoperative phases, reaching highest values at T3 and T4. ONSD increased progressively across phases on both sides (right: 3.79 ± 0.36 mm at T0 to 4.59 ± 0.47 mm at T3; left: 3.80 ± 0.42 mm to 4.73 ± 0.47 mm). Correlation analysis demonstrated a moderate-to-strong association between ΔHHb and ONSD (right: r = 0.63, p < 10⁻¹⁶; left: r = 0.63, p < 10⁻¹⁶), supporting temporal coupling between cerebral deoxygenated hemoglobin accumulation and optic nerve sheath expansion. This study demonstrates distinct phase-dependent cerebral oxygenation patterns during RALP. The late-phase increase in ΔHHb occurring alongside ONSD enlargement may reflect cerebral venous congestion, a mechanism potentially involved in cerebral edema development during prolonged steep Trendelenburg positioning.
Purpose This study aimed to evaluate intraocular pressure (IOP) as a time-dependent physiological monitoring parameter during ventilator weaning and to explore its association with PaCO₂-related respiratory changes. Meth...Purpose This study aimed to evaluate intraocular pressure (IOP) as a time-dependent physiological monitoring parameter during ventilator weaning and to explore its association with PaCO₂-related respiratory changes. Methods In this prospective observational study, adult ICU patients receiving mechanical ventilation and undergoing weaning were enrolled. Patients with glaucoma, prior ocular surgery, or inability to tolerate weaning were excluded. IOP was measured in both eyes before weaning (T0), during spontaneous breathing (T1), and after completion of weaning (T2). Ventilator mode, positive end-expiratory pressure (PEEP), sedative agents, and spontaneous breathing trial (SBT) method (pressure support ventilation [PSV] or T-piece) were recorded. A protocol-defined IOP increase was defined as an elevation of ≥ 1.0 mmHg in at least one eye between T0 and T2. Associations between clinical variables and IOP changes were analyzed using repeated-measures analysis and multivariable logistic regression. All measurements were performed under standardized bedside ICU conditions; T1 was obtained 30 min after initiation of spontaneous breathing and T2 30 min after completion of weaning. Results Sixty-four patients were assessed, and 45 were included in the final analysis. Mean IOP values showed minimal change over time. In exploratory multivariable logistic regression, an increase in arterial carbon dioxide tension during weaning (ΔPaCO₂) was associated with the protocol-defined IOP increase (odds ratio [OR] = 1.66; 95% confidence interval [CI], 1.13-2.45; p = 0.010). No significant associations were observed with other ventilator-related or clinical variables. Conclusion The transition from mechanical ventilation to spontaneous breathing in ICU patients was not associated with a significant increase in mean IOP. However, small individual-level IOP increases were observed, and these were associated with increases in PaCO₂ during weaning. These findings should be interpreted as exploratory physiological observations. Further studies are required to determine whether these observations have clinical relevance.