Association between intra-operative fentanyl dosing and postoperative nausea/vomiting and pain: A prospective cohort study BACKGROUND Postoperative nausea and/or vomiting (PONV) is one of the anaesthesia-related effects most dreaded by patients and may delay hospital discharge. Although scores and risk factors are established, many were developed before contemporary anaesthesia regimens and without focussing on modifiable anaesthesia-related variables. OBJECTIVE To examine whether, in association with a contemporary anaesthesia regimen, there is an association between intra-operative fentanyl dose and PONV, and, second, postoperative pain within the first 24 h. DESIGN Prospective, observational cohort. As informed consent was waived, neither patients nor treating physicians were aware of the study. SETTING Single-centre university hospital. PATIENTS Inclusion criteria were opioid-naive patients without chronic pain and with a simplified Apfel score at least 2 undergoing abdominal, gynaecological or otorhinolaryngological inpatient surgery. INTERVENTION None. MAIN OUTCOME MEASURE With logistic regression, we examined three models of increasing complexity exploring the relationship between PONV and fentanyl dosing: Model 1, simplified Apfel score + intra-operative fentanyl; Model 2, Model 1 + pre-emptive antiemetic prophylaxis; Model 3, Model 2 + postoperative morphine. Model 1 was the primary analysis. Second, we explored whether or not postoperative pain scores were associated with intra-operative fentanyl dosing. RESULTS From the 363 patients, 163 (45%) experienced PONV, despite the use of total intravenous anaesthesia with propofol in more than 80% of the cohort, and some 66% of patients receiving additional antiemetic agents. After adjusting for the simplified Apfel score, higher intra-operative fentanyl dose was associated with PONV: odds ratio per μg h−1, 1.006 [95% confidence interval (CI) 1.002 to 1.010]. Including intra-operative fentanyl in the simplified Apfel score also increased the area under the receiver operator characteristics curve [0.601 (95% CI 0.555 to 0.662) vs. 0.651 (95% CI 0.594 to 0.707); P = 0.016]. Finally, a higher intra-operative fentanyl dose was associated with higher 24 h pain scores (P = 0.001) and a trend towards higher 24 h morphine requirements (P = 0.055). CONCLUSION Even when using propofol and antiemetic agents, PONV within the first 24 h remained higher than expected. Intra-operative fentanyl, a modifiable risk factor, is associated with the incidence of PONV and postoperative pain. TRIAL REGISTRATION ClinicalTrials.gov, NCT03201315. Correspondence to Eckhard Mauermann, Department for Anesthesia, Surgical Intensive Care, Prehospital Emergency Medicine and Pain Therapy, University Hospital Basel, Spitalstrasse 21, 4031 Basel, Switzerland E-mail: Eckhard.Mauermann@usb.ch Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (www.ejanaesthesiology.com). © 2019 European Society of Anaesthesiology

The use of a thromboelastometry-based algorithm reduces the need for blood product transfusion during orthotopic liver transplantation: A randomised controlled study BACKGROUND Orthotopic liver transplantation is associated with a risk of bleeding. Coagulation in cirrhotic patients is difficult to assess with standard coagulation tests because of rebalanced coagulation. This can be better assessed by thromboelastometry which can detect coagulation impairments more specifically in such patients. OBJECTIVES Our first objective was to compare the number of units of blood products transfused during liver transplantation when using an algorithm based on standard coagulation tests or a thromboelastometry-guided transfusion algorithm. DESIGN Prospective randomised controlled trial. SETTING Single-centre tertiary care hospital in France, from December 2014 to August 2016. PARTICIPANTS A total of 81 adult patients undergoing orthotopic liver transplantation were studied. Patients were excluded if they had congenital coagulopathies. INTERVENTION Transfusion management during liver transplantation was guided either by a standard coagulation test algorithm or by a thromboelastometry-guided algorithm. Transfusion, treatments and postoperative outcomes were compared between groups. MAIN OUTCOME MEASURES Total number of transfused blood product units during the operative period (1 U is one pack of red blood cells (RBCs), fresh frozen plasma (FFP) or platelets). RESULTS Median [interquartile range] intra-operative transfusion requirement was reduced in the thromboelastometry group (3 [2 to 4] vs. 7 [4 to 10] U, P = 0.005). FFP and tranexamic acid were administered less frequently in the thromboelastometry group (respectively 15 vs. 46.3%, P = 0.002 and 27.5 vs. 58.5%, P = 0.005), whereas fibrinogen was more often infused in the thromboelastometry group (72.5 vs. 29.3%, P < 0.001). Median transfusions of FFP (3 [2 to 6] vs. 4 [2 to 7] U, P = 0.448), RBCs (3 [2 to 5] vs. 4 [2 to 6] U, P = 0.330) and platelets (1 [1 to 2] vs. 1 [1 to 2] U, P = 0.910) were not different between groups. In the postoperative period, RBC or platelet transfusion, the need for revision surgery or occurrence of haemorrhage were not different between groups. CONCLUSION A transfusion algorithm based on thromboelastometry assessment of coagulation reduced the total number of blood product units transfused during liver transplantation, particularly FFP administration. TRIAL REGISTRATION ClinicalTrials.gov identifier: NCT02352181. Correspondence to Dr Aurélie Bonnet, Département d’Anesthésie et Réanimation, Hôpital de la Croix Rousse, 103 Grande Rue de la Croix Rousse, 69004 Lyon, France Tel: +33 4 26 10 93 35; e-mail: aurelie.bonnet@chu-lyon.fr © 2019 European Society of Anaesthesiology

Preparation of anaesthesia workstation for trigger-free anaesthesia: Activated charcoal filters and prolonged low-flow ventilation BACKGROUND Trigger-free anaesthesia is required for patients who are susceptible to malignant hyperthermia. Therefore, all trace of volatile anaesthetics should be removed from anaesthetic machines before induction of anaesthesia. Because the washout procedure is time consuming, activated charcoal filters (ACFs) have been introduced, but never tested under minimal flow conditions. OBJECTIVE(S) The current study aims to investigate performance of ACFs during long duration (24 h) simulated ventilation. DESIGN A bench study. SETTING A Primus anaesthesia machine (Dräger) was contaminated with either 4% sevoflurane or 8% desflurane by ventilating a test lung for 90 min. The machine was briefly flushed according to manufacturer instructions, ACFs were inserted and a test lung was ventilated in a 24 h test. Trace gas concentrations were measured using a closed gas loop high-resolution ion mobility spectrometer with gas chromatographic preseparation. During the experiment reduced fresh gas flows (FGFs) were tested. At the end of each experiment the ACFs were removed and the machine was set to standby for 10 min to test for residual contamination within the circuit. and then the ACFs were reconnected into the circuit to test their ability to continue removing volatile anaesthetics (functional test) from the gas. Control experiments were conducted without ACFs. MAIN OUTCOME MEASURES Absolute concentrations of desflurane and sevoflurane. RESULTS The concentration of volatile anaesthetics dropped to less than 5 ppm (parts per million) following insertion of ACFs. In the desflurane experiments at least 1 l min−1 FGF was needed to keep the concentration below an acceptable level (<5 ppm): 0.5 l min−1 FGF was required in sevoflurane experiments. While ACFs in the sevoflurane tests passed the functional test after 24 h, ACFs in the desflurane tests failed. CONCLUSION ACFs meet the requirements for trigger-free low flow (1 l min−1) ventilation over 24 h. Minimal flow (0.5 l min−1) ventilation may be possible for sevoflurane contaminated machines. Correspondence to Dr Sebastian Heiderich, Clinic of Anaesthesiology and Intensive Care Medicine, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany E-mail: heiderich.sebastian@mh-hannover.de © 2019 European Society of Anaesthesiology

Preventing hypothermia in outpatient plastic surgery by self-warming or forced-air-warming blanket: A randomised controlled trial BACKGROUND In our outpatient post anaesthesia unit patients reported that they were feeling cold with or without shivering. Anaesthetic agents cause reduced thermoregulation, initially by redistribution of blood flow from core to periphery, later by negative balance between thermogenesis and heat loss. Even mild peri-operative hypothermia increases the risk of surgical wound infections, bleeding, impaired cardiac function, shivering, and decreases comfort. OBJECTIVE(S) We aimed to evaluate which of our current active warming measures, self-warming blanket or forced-air-warming blanket, were most effective in preventing inadvertent intraoperative heat loss. Secondarily, we assessed whether they prevented inadvertent peri-operative hypothermia when defined as core body temperature below 36 °C. DESIGN Randomised controlled trial, parallel group design. SETTING Aleris Solsiden hospital for outpatient surgery, Trondheim, Norway, from March to June 2016. PATIENTS A total of 112 consecutive patients planned for outpatient plastic surgery. Reasons for noninclusion: failing to meet the criteria for outpatient surgery according to the standard of the national society of anaesthesiology. INTERVENTION(S) Patients were randomised to active warming by a self-warming blanket or a forced-air-warming blanket. All patients received routine measures to prevent hypothermia with a high temperature in the operation theatres, prewarmed fluids, cotton blankets and surgical draping outside the surgical field. MAIN OUTCOMES Temperature, measured pre-operatively, every 10 min during general anaesthesia and postoperatively with a zero-heat-flux temperature sensor. RESULTS Core temperature was significantly lower in the self-warming blanket compared with the forced-air-warming blanket group during anaesthesia, P less than 0.0001. Hypothermia (<36 °C) was recorded in 47%, n = 22, patients in the self-warming blanket group and 25%, n = 16, in the forced-air-warming blanket group during the registration period, P = 0.02. CONCLUSION An underbody forced-air-warming blanket reduced heat loss to a greater extent than a self-warming blanket. But none of the interventions were sufficient to prevent inadvertent peri-operative hypothermia. TRIAL REGISTRATION ClinicalTrials.gov identifier: NCT03163563. Correspondence to Stig S. Tyvold, Aleris Solsiden, Innherredsveien 74, N-7014 Trondheim, Norway Tel: +47 73 87 20 00; e-mail: stig.tyvold@aleris.no © 2019 European Society of Anaesthesiology

Core topics in anaesthesia and peri-operative care of the morbidly obese surgical patient No abstract available

Effects of depth of neuromuscular block on postoperative pain during laparoscopic gastrectomy: A randomised controlled trial BACKGROUND Evidence on whether the use of deep neuromuscular block (NMB) influences postoperative pain after laparoscopic surgery is limited, and existing studies have shown conflicting results. We studied the effect of the depth of NMB during laparoscopic gastrectomy on postoperative pain. OBJECTIVE The aim of this study was to evaluate the effect of depth of NMB during laparoscopic gastrectomy on postoperative pain by allocating patients randomly to either deep or moderate NMB with a standard-pressure pneumoperitoneum. DESIGN A randomised, controlled, double-blind study. SETTING A university-affiliated hospital. PARTICIPANTS One hundred patients. INTERVENTIONS Patients were allocated randomly to receive either deep (posttetanic count 1 to 2) or moderate (train-of-four count 1 to 2) levels of NMB. Following surgery, the patients were asked to rate their pain every 10 min using a visual analogue scale (VAS) (0 = no pain, 10 = most severe pain) in the postanaesthesia care unit (PACU). Patients received intravenous oxycodone, 2 mg every 10 min, until the pain intensity (VAS) had decreased to less than 3 at rest and less than 5 on wound compression, at which point the minimum effective analgesia dose (MEAD) of oxycodone was determined. MAIN OUTCOME MEASURES The primary endpoint was the MEAD of oxycodone. Secondary endpoints included area under the curve of VAS for wound pain, VAS scores for wound and shoulder pain at 6 and 24 h after the end of surgery, rescue analgesics, a five-point surgical rating scale, Rhodes index of nausea vomiting retching at 6 and 24 h after the end of surgery and duration of pneumoperitoneum. RESULTS The median value for the MEAD of oxycodone was 8 mg in both groups. Area under the curves of VAS over time were similar in both groups. Variables associated with postoperative pain including mean VAS at PACU and frequency of rescue analgesics in the ward did not differ significantly between the two groups. The duration of pneumoperitoneum was a significant variable in determining the MEAD of oxycodone (linear regression, R2 = 0.07, P = 0.008). The number of patients who reached the acceptable surgical score was not significantly different between the two groups. However, the moderate NMB group did have a significantly higher proportion of cases that required additional muscle relaxants (P < 0.001). CONCLUSION Deep, compared with moderate, NMB did not significantly reduce the MEAD of oxycodone administered in the PACU. The duration of pneumoperitoneum was positively correlated with the MEAD. TRIAL REGISTRATION ClinicalTrials.gov identifier: NCT03266419. Correspondence to Byung-Moon Choi, MD, PhD, Department of Anesthesiology and Pain Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Republic of Korea Tel: +82 2 3010 1704; fax: +82 2 3010 6790; e-mail: byungmoonchoi7@gmail.com Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (www.ejanaesthesiology.com). © 2019 European Society of Anaesthesiology

Incidence and risk factors of postoperative delirium in patients admitted to the ICU after elective intracranial surgery: A prospective cohort study BACKGROUND Postoperative delirium (POD) has been confirmed as an important complication after major surgery. However, neurosurgical patients have usually been excluded in previous studies. To date, data on POD and risk factors in patients after intracranial surgery are scarce. OBJECTIVES To determine the incidence and risk factors of POD in patients after intracranial surgery. DESIGN Prospective cohort study. SETTING A neurosurgical ICU of a university-affiliated hospital, Beijing, China. INTERVENTIONS Adult patients admitted to the ICU after elective intracranial surgery under general anaesthesia were consecutively enrolled between 1 March 2017 and 2 February 2018. Delirium was assessed using the Confusion Assessment Method for the ICU. POD was diagnosed as Confusion Assessment Method for the ICU positive on either postoperative day 1 or day 3. Patients were classified into groups with or without POD. Data were collected for univariate and multivariate analyses to determine the risk factors for POD. RESULTS A total of 800 patients were included. POD was diagnosed in 157 patients (19.6%, 95% confidence interval 16.9 to 22.4%). Independent risk factors for POD included age, nature of intracranial lesion, frontal approach craniotomy, duration of surgery, presence of an episode of low pulse oxygenation at ICU admission, presence of inadequate emergence and emergence delirium, postoperative pain and presence of immobilising events. POD was associated with adverse outcomes and high costs. CONCLUSION POD is prevalent in patients after elective intracranial surgery. The identified risk factors for and the potential association of POD with adverse outcomes suggest that a comprehensive strategy involving screening for predisposing factors and early prevention of modifiable factors should be established in this population. TRIAL REGISTRATION The study was registered at ClinicalTrials.gov (NCT03087838). Correspondence to Jian-Xin Zhou, Department of Critical Care Medicine, Beijing Tiantan Hospital, Capital Medical University, No 6, Tiantan Xili, Dongcheng District, Beijing 100050, China Tel: +86 10 67098019; fax: +86 10 67098019; e-mail: zhoujx.cn@icloud.com Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (www.ejanaesthesiology.com). © 2019 European Society of Anaesthesiology

Transcutaneous carbon dioxide measurements in fruits, vegetables and humans: A prospective observational study BACKGROUND Transcutaneous carbon dioxide measurement (TcCO2) is frequently used as a surrogate for arterial blood gas sampling in adults and children with critical illness. Data from noninvasive TcCO2 monitoring assists with clinical decisions regarding mechanical ventilation settings, estimation of metabolic consumption and determination of adequate end-organ tissue perfusion. OBJECTIVES To report TcCO2 values obtained from various fruits, vegetables and elite critical care medicine specialists. DESIGN Prospective, observational, nonblinded cohort study. SETTINGS Single-centre, tertiary paediatric referral centre and organic farmers’ market. PARTICIPANTS Vegetables and fruits included 10 samples of each of the following: red delicious apple (Malus domestica), manzano banana (Musa sapientum), key lime (Citrus aurantiifolia), miniature sweet bell pepper (Capsicum annuum), sweet potato (Ipomoea batatas) and avocado (Persea americana). Ten human controls were studied including a paediatric intensivist, a paediatric inpatient hospital physician, four paediatric resident physicians and four paediatric critical care nurses. INTERVENTIONS None. MAIN OUTCOME MEASURES TcCO2 values for each species and device response times. RESULTS TcCO2 readings were measurable in all study species except the sweet potato. Mean ± SD values of TcCO2 for human controls [4.34 ± 0.37 kPa (32.6 ± 2.8 mmHg)] were greater than apples [3.09 ± 0.19 kPa (23.2 ± 1.4 mmHg), P < 0.01], bananas [2.73 ± 0.28 kPa (20.5 ± 2.1 mmHg), P < 0.01] and limes [2.76 ± 0.52 kPa (20.7 ± 3.9 mmHg), P < 0.01] but no different to those of avocados [4.29 ± 0.44 kPa (32.2 ± 3.3 mmHg), P = 0.77] and bell peppers [4.19 ± 1.13 kPa (31.4 ± 8.5 mmHg), P = 0.68]. Transcutaneous response times did not differ between research cohorts and human controls. CONCLUSION We found nonroot, nontuberous vegetables to have TcCO2 values similar to that of healthy, human controls. Fruits yield TcCO2 readings, but substantially lower than human controls. Correspondence to Anthony A. Sochet, MD, MS, Assistant Professor of Anesthesiology and Critical Care Medicine, Division of Pediatric Critical Care Medicine, Johns Hopkins University, Johns Hopkins All Children's Hospital, 501 6th Ave S., St#: 702A, St. Petersburg, FL 33701, USA Tel: +1 727 767 2912; e-mail: anthony.sochet@jhmi.edu © 2019 European Society of Anaesthesiology

The concept of peri-operative medicine to prevent major adverse events and improve outcome in surgical patients: A narrative review Peri-operative Medicine is the patient-centred and value-based multidisciplinary peri-operative care of surgical patients. Peri-operative stress, that is the collective response to stimuli occurring before, during and after surgery, is, together with pre-existing comorbidities, the pathophysiological basis of major adverse events. The ultimate goal of peri-operative medicine is to promote high quality recovery after surgery. Clinical scores and/or biomarkers should be used to identify patients at high risk of developing major adverse events throughout the peri-operative period. Allocation of high-risk patients to specific care pathways with peri-operative organ protection, close surveillance and specific early interventions is likely to improve patient-relevant outcomes, such as disability, health-related quality of life and mortality. Correspondence to Bernardo Bollen Pinto, MD, PhD, Division of Anaesthesiology, Department of Anaesthesiology, Pharmacology, Intensive Care and Emergency Medicine, University Hospitals of Geneva, Gabrielle-Perret-Gentil 4, 1205 Geneva, Switzerland Tel: +41 789401810; fax: +41 223727511; e-mail: bernardo.bollenpinto@hcuge.ch © 2019 European Society of Anaesthesiology

Comparison of neostigmine vs. sugammadex for recovery of muscle function after neuromuscular block by means of diaphragm ultrasonography in microlaryngeal surgery: A randomised controlled trial BACKGROUND Postoperative residual neuromuscular blockade or curarisation (PORC) is a risk directly related to the use of neuromuscular blocking agents during surgical procedures. Acceleromyography is distressing for conscious patients when assessing PORC. Diaphragm ultrasonography could be a valid alternative. OBJECTIVES The primary objective was to achieve a 28% lower incidence of PORC in patients who, after rocuronium administration, received neostigmine or sugammadex at 30 min after surgery. To assess PORC, diaphragm ultrasonography was used, and thickening fractioning [the difference of thickness at the end of inspiration (TEI) and at the end of expiration (TEE), normalised for TEE (TEI − TEE/TEE)] was measured. PORC was defined as thickening fractioning of 0.36 or less. The secondary object was the comparison, in the two treatment groups, of the return to baseline thickening fractioning at 30 min after surgery (ΔTF30). DESIGN Prospective, double-blind, single-centre randomised study. SETTING University Hospital Careggi, Florence, Italy. PATIENTS Patients of American Society Anesthesiologists’ physical status 1 or 2, 18 to 80 years, receiving rocuronium during microlaryngeal surgery. INTERVENTIONS At the end of surgery participants were randomised to receive neostigmine (NEO group) or sugammadex (SUG group) as the reversal drug. Thickening fractioning and ΔTF30 were evaluated at baseline and at 0, 10 and 30 min after surgery. MAIN OUTCOME MEASURES TEE and TEI at each time point. RESULTS A total of 59 patients with similar demographic characteristics were enrolled. An association between lack of recovery (thickening fractioning ≤0.36) and drug treatment was only observed at 0 min (SUG vs. NEO, P < 0.05). Concerning ΔTF, at 30 min more patients in the SUG group returned to baseline than those in the NEO group (P < 0.001), after adjusting for side (P = 0.52), baseline thickening fractioning (P < 0.0001) and time of measurement (P < 0.01). CONCLUSION We found an early (0 min) but not long-lasting (30 min) association between diaphragm failure and treatment allocation; a full recovery in baseline diaphragm function was observed only in patients receiving sugammadex. We cannot exclude that further differences have not been found due to interpatients variability in assessing diaphragm contractility by ultrasonography. TRIAL REGISTRATION EudraCT Identifier: 2013-004787-62, Clinicaltrials.gov Identifier: NCT02698969. Correspondence to Iacopo Cappellini, Section of Anesthesiology and Intensive Care, Department of Health Sciences, University of Florence, Largo Brambilla 3, 50134 Florence, Italy E-mail: jacopocappellini@gmail.com © 2019 European Society of Anaesthesiology