Comparison of differences in cohort (forwards) and case control (backwards) methodological approaches for validation of the Hypotension Prediction Index.

BACKGROUND: The Hypotension Prediction Index (the index) software is a machine learning algorithm that detects physiological changes that may lead to hypotension. The original validation used a case control (backwards) analysis that has been suggested to be biased. We therefore conducted a cohort (forwards) analysis and compared this to the original validation technique. METHODS: We conducted a retrospective analysis of data from previously reported studies. All data were analysed identically with 2 different methodologies and receiver operating characteristic curves (ROC) constructed. Both backwards and forwards analyses were performed to examine differences in area under the ROC for HPI and other haemodynamic variables to predict a MAP < 65mmHg for at least 1 minute 5, 10 and 15 minutes in advance. RESULTS: Two thousand and twenty-two patients were included in the analysis, yielding 4,152,124 measurements taken at 20 second intervals. The area-under-the-curve for the index predicting hypotension analysed by backward and forward methodologies respectively was 0.957 (95% CI, 0.947-0.964) vs 0.923 (95% CI, 0.912-0.933) 5 minutes in advance, 0.933 (95% CI, 0.924-0.942) vs 0.923 (95% CI, 0.911-0.933) 10 minutes in advance , and 0.929 (95% CI, 0.918-0.938) vs. 0.926 (95% CI, 0.914-0.937) 15 minutes in advance. No other variable had an area-under-the-curve > 0.7 except for MAP. Area-under-the-curve using forward analysis for MAP predicting hypotension 5, 10, and 15 minutes in advance was 0.932 (95% CI, 0.920-0.940), 0.929 (95% CI, 0.918-0.938), and 0.932 (95% CI, 0.921-0.940). The R 2 for the variation in the index due to MAP was 0.77. CONCLUSION: Using an updated methodology, we found the utility of the HPI index to predict future hypotensive events is high, with an area under the receiver-operating-characteristics curve similar to that of the original validation method.

Association of Intravenous Neostigmine and Anticholinergics or Sugammadex with Postoperative Delirium: A Retrospective Cohort Study.

BACKGROUND: Administration of cholinesterase inhibitors in combination with anticholinergic drugs for reversal of neuromuscular blocks may precipitate delirium through impairment of central cholinergic transmission, which could be avoided by using sugammadex. Therefore, we tested the primary hypothesis that postoperative delirium is less common when neuromuscular block is reversed with sugammadex than with neostigmine combined with glycopyrrolate or atropine. METHODS: We conducted a single-center retrospective cohort study, analyzing all adult patients having general anesthesia for noncardiac surgery who received neostigmine or sugammadex from January 2016 to March 2022. Inverse propensity score weighting and propensity score calibration were used to adjust for appropriate confounders. Our primary outcome was presence of delirium within the first 4 days after surgery, defined as at least 1 positive brief Confusion Assessment Method (bCAM) screening. The secondary outcome was the presence of early delirium within 24 hours of surgery. RESULTS: Among 49,468 cases in our analysis, 6881 received sugammadex and 42,587 received neostigmine. After propensity weighting, the incidence of delirium was 1.09% in the sugammadex group and 0.82% in the neostigmine group. The odds of postoperative delirium did not differ between the sugammadex and neostigmine groups, with an estimated odds ratio (95% confidence interval) of 1.33 (0.91-1.95), P = .147. A sensitivity analysis restricted to only include cases with at least 6 bCAM measurements over postoperative day (POD) 1 to 4 had consistent results, as sugammadex compared with neostigmine was associated with an estimated odds ratio for postoperative delirium of 1.20 (0.82-1.77), P = .346. Sugammadex was significantly associated with an increased incidence of early postoperative delirium, with an estimated odds ratio of 1.71 (1.07-2.72), P = .025. Further analysis showed no treatment-by-age interaction for either postoperative delirium (P = .637) or postoperative early delirium (P = .904). CONCLUSIONS: Compared to neostigmine, use of sugammadex for reversal of neuromuscular block was not associated with an increased risk of postoperative delirium in this retrospective single-center study. Though sugammadex was associated with a statistically significant increased risk of postoperative early delirium, the difference was small and not clinically relevant, and may reflect the presence of unknown confounders.

Hypotension-Avoidance Versus Hypertension-Avoidance Strategies in Noncardiac Surgery : An International Randomized Controlled Trial.

BACKGROUND: Among patients having noncardiac surgery, perioperative hemodynamic abnormalities are associated with vascular complications. Uncertainty remains about what intraoperative blood pressure to target and how to manage long-term antihypertensive medications perioperatively. OBJECTIVE: To compare the effects of a hypotension-avoidance and a hypertension-avoidance strategy on major vascular complications after noncardiac surgery. DESIGN: Partial factorial randomized trial of 2 perioperative blood pressure management strategies (reported here) and tranexamic acid versus placebo. (ClinicalTrials.gov: NCT03505723). SETTING: 110 hospitals in 22 countries. PATIENTS: 7490 patients having noncardiac surgery who were at risk for vascular complications and were receiving 1 or more long-term antihypertensive medications. INTERVENTION: In the hypotension-avoidance strategy group, the intraoperative mean arterial pressure target was 80 mm Hg or greater; before and for 2 days after surgery, renin-angiotensin-aldosterone system inhibitors were withheld and the other long-term antihypertensive medications were administered only for systolic blood pressures 130 mm Hg or greater, following an algorithm. In the hypertension-avoidance strategy group, the intraoperative mean arterial pressure target was 60 mm Hg or greater; all antihypertensive medications were continued before and after surgery. MEASUREMENTS: The primary outcome was a composite of vascular death and nonfatal myocardial injury after noncardiac surgery, stroke, and cardiac arrest at 30 days. Outcome adjudicators were masked to treatment assignment. RESULTS: The primary outcome occurred in 520 of 3742 patients (13.9%) in the hypotension-avoidance group and in 524 of 3748 patients (14.0%) in the hypertension-avoidance group (hazard ratio, 0.99 [95% CI, 0.88 to 1.12]; P = 0.92). Results were consistent for patients who used 1 or more than 1 antihypertensive medication in the long term. LIMITATION: Adherence to the assigned strategies was suboptimal; however, results were consistent across different adherence levels. CONCLUSION: In patients having noncardiac surgery, our hypotension-avoidance and hypertension-avoidance strategies resulted in a similar incidence of major vascular complications. PRIMARY FUNDING SOURCE: Canadian Institutes of Health Research, National Health and Medical Research Council (Australia), and Research Grant Council of Hong Kong.

Extended-age Out-of-sample Validation of Risk Stratification Index 3.0 Models Using Commercial All-payer Claims.

BACKGROUND: The authors previously reported a broad suite of individualized Risk Stratification Index 3.0 (Health Data Analytics Institute, Inc., USA) models for various meaningful outcomes in patients admitted to a hospital for medical or surgical reasons. The models used International Classification of Diseases, Tenth Revision, trajectories and were restricted to information available at hospital admission, including coding history in the previous year. The models were developed and validated in Medicare patients, mostly age 65 yr or older. The authors sought to determine how well their models predict utilization outcomes and adverse events in younger and healthier populations. METHODS: The authors' analysis was based on All Payer Claims for surgical and medical hospital admissions from Utah and Oregon. Endpoints included unplanned hospital admissions, in-hospital mortality, acute kidney injury, sepsis, pneumonia, respiratory failure, and a composite of major cardiac complications. They prospectively applied previously developed Risk Stratification Index 3.0 models to the younger and healthier 2017 Utah and Oregon state populations and compared the results to their previous out-of-sample Medicare validation analysis. RESULTS: In the Utah dataset, there were 55,109 All Payer Claims admissions across 40,710 patients. In the Oregon dataset, there were 21,213 admissions from 16,951 patients. Model performance on the two state datasets was similar or better than in Medicare patients, with an average area under the curve of 0.83 (0.71 to 0.91). Model calibration was reasonable with an R2 of 0.93 (0.84 to 0.97) for Utah and 0.85 (0.71 to 0.91) for Oregon. The mean sensitivity for the highest 5% risk population was 28% (17 to 44) for Utah and 37% (20 to 56) for Oregon. CONCLUSIONS: Predictive analytical modeling based on administrative claims history provides individualized risk profiles at hospital admission that may help guide patient management. Similar predictive performance in Medicare and in younger and healthier populations indicates that Risk Stratification Index 3.0 models are valid across a broad range of adult hospital admissions.

Goal-directed haemodynamic therapy: an imprecise umbrella term to avoid.

'Goal-directed haemodynamic therapy' describes various haemodynamic treatment strategies that have in common that interventions are titrated to achieve predefined haemodynamic targets. However, the treatment strategies differ substantially regarding the underlying haemodynamic target variables and target values, and thus presumably have different effects on outcome. It is an over-simplifying approach to lump complex and substantially differing haemodynamic treatment strategies together under the term 'goal-directed haemodynamic therapy', an imprecise umbrella term that we should thus stop using.

Endotypes of intraoperative hypotension during major abdominal surgery: a retrospective machine learning analysis of an observational cohort study.

BACKGROUND: Intraoperative hypotension is associated with myocardial injury, acute kidney injury, and death. In routine practice, specific causes of intraoperative hypotension are often unclear. A more detailed understanding of underlying haemodynamic alterations of intraoperative hypotension may identify specific treatments. We thus aimed to use machine learning - specifically, hierarchical clustering - to identify underlying haemodynamic alterations causing intraoperative hypotension in major abdominal surgery patients. Specifically, we tested the hypothesis that there are distinct endotypes of intraoperative hypotension, which may help refine therapeutic interventions. METHODS: We conducted a secondary analysis of intraoperative haemodynamic measurements from a prospective observational study in 100 patients who had major abdominal surgery under general anaesthesia. We used stroke volume index, heart rate, cardiac index, systemic vascular resistance index, and pulse pressure variation measurements. Intraoperative hypotension was defined as any mean arterial pressure ≤65 mm Hg or a mean arterial pressure between 66 and 75 mm Hg requiring a norepinephrine infusion rate exceeding 0.1 µg kg-1 min-1. To identify endotypes of intraoperative hypotension, we used hierarchical clustering (Ward's method). RESULTS: A total of 615 episodes of intraoperative hypotension occurred in 82 patients (46 [56%] female; median age: 64 [57, 73] yr) who had surgery of a median duration of 270 (195, 335) min. Hierarchical clustering revealed six distinct intraoperative hypotension endotypes. Based on their clinical characteristics, we labelled these endotypes as (1) myocardial depression, (2) bradycardia, (3) vasodilation with cardiac index increase, (4) vasodilation without cardiac index increase, (5) hypovolaemia, and (6) mixed type. CONCLUSION: Hierarchical clustering identified six endotypes of intraoperative hypotension. If validated, considering these intraoperative hypotension endotypes may enable causal treatment of intraoperative hypotension.

Artificial Intelligence for Perioperative Medicine: Perioperative Intelligence.

The anesthesiologist's role has expanded beyond the operating room, and anesthesiologist-led care teams can deliver coordinated care that spans the entire surgical experience, from preoperative optimization to long-term recovery of surgical patients. This expanded role can help reduce postoperative morbidity and mortality, which are regrettably common, unlike rare intraoperative mortality. Postoperative mortality, if considered a disease category, will be the third leading cause of death just after heart disease and cancer. Rapid advances in technologies like artificial intelligence provide an opportunity to build safe perioperative practices. Artificial intelligence helps by analyzing complex data across disparate systems and producing actionable information. Using artificial intelligence technologies, we can critically examine every aspect of perioperative medicine and devise innovative value-based solutions that can potentially improve patient safety and care delivery, while optimizing cost of care. In this narrative review, we discuss specific applications of artificial intelligence that may help advance all aspects of perioperative medicine, including clinical care, education, quality improvement, and research. We also discuss potential limitations of technology and provide our recommendations for successful adoption.

Relative contributions of anaemia and hypotension to myocardial infarction and renal injury: Post hoc analysis of the POISE-2 trial.

BACKGROUND: Hypotension and postoperative anaemia are associated with myocardial and renal injury after noncardiac surgery, but the interaction between them remains unknown. OBJECTIVES: To test the hypothesis that a double-hit of postoperative anaemia and hypotension synergistically worsens a 30-day composite of myocardial infarction (MI) and mortality and acute kidney injury (AKI). Characterising the interaction when hypotension and anaemia occur at same time on myocardial infarction and acute kidney injury. DESIGN: Post hoc analysis of the POISE-2 trial. SETTING: Patients were enrolled between July 2010 and December 2013 at 135 hospitals in 23 countries. PATIENTS: Adults at least 45 years old with known or suspected cardiovascular disease. We excluded patients without available postoperative haemoglobin measurements or hypotension duration records. Exposures were the lowest haemoglobin concentration and the average daily duration of SBP less than 90 mmHg within the first four postoperative days. MAIN OUTCOME MEASURES: The primary outcome was a collapsed composite of nonfatal MI and all-cause mortality during the initial 30 postoperative days; our secondary outcome was AKI. RESULTS: We included 7940 patients. The mean ±â€ŠSD lowest postoperative haemoglobin was 10 ±â€Š2 g dl -1 , and 24% of the patients had SBP less than 90 mmHg with daily duration ranging from 0 to 15 h. Four hundred and nine (5.2%) patients had an infarction or died within 30 postoperative days, and 417 (6.4%) patients developed AKI. Lowest haemoglobin concentrations less than 11 g dl -1 , and duration of SBP less than 90 mmHg was associated with greater hazard of composite outcome of nonfatal MI and all-cause mortality, as well as with AKI. However, we did not find significant multiplicative interactions between haemoglobin splines and hypotension duration on the primary composite or on AKI. CONCLUSION: Postoperative anaemia and hypotension were meaningfully associated with both our primary composite and AKI. However, lack of significant interaction suggests that the effects of hypotension and anaemia are additive rather than multiplicative. TRIAL REGISTRATION: Clinicaltrials.gov: NCT01082874.

Risk Stratification Index 3.0, a Broad Set of Models for Predicting Adverse Events during and after Hospital Admission.

BACKGROUND: Risk stratification helps guide appropriate clinical care. Our goal was to develop and validate a broad suite of predictive tools based on International Classification of Diseases, Tenth Revision, diagnostic and procedural codes for predicting adverse events and care utilization outcomes for hospitalized patients. METHODS: Endpoints included unplanned hospital admissions, discharge status, excess length of stay, in-hospital and 90-day mortality, acute kidney injury, sepsis, pneumonia, respiratory failure, and a composite of major cardiac complications. Patient demographic and coding history in the year before admission provided features used to predict utilization and adverse events through 90 days after admission. Models were trained and refined on 2017 to 2018 Medicare admissions data using an 80 to 20 learn to test split sample. Models were then prospectively tested on 2019 out-of-sample Medicare admissions. Predictions based on logistic regression were compared with those from five commonly used machine learning methods using a limited dataset. RESULTS: The 2017 to 2018 development set included 9,085,968 patients who had 18,899,224 inpatient admissions, and there were 5,336,265 patients who had 9,205,835 inpatient admissions in the 2019 validation dataset. Model performance on the validation set had an average area under the curve of 0.76 (range, 0.70 to 0.82). Model calibration was strong with an average R 2 for the 99% of patients at lowest risk of 1.00. Excess length of stay had a root-mean-square error of 0.19 and R 2 of 0.99. The mean sensitivity for the highest 5% risk population was 19.2% (range, 11.6 to 30.1); for positive predictive value, it was 37.2% (14.6 to 87.7); and for lift (enrichment ratio), it was 3.8 (2.3 to 6.1). Predictive accuracies from regression and machine learning techniques were generally similar. CONCLUSIONS: Predictive analytical modeling based on administrative claims history can provide individualized risk profiles at hospital admission that may help guide patient management. Similar results from six different modeling approaches suggest that we have identified both the value and ceiling for predictive information derived from medical claims history.

Pain and Opioid Consumption and Mobilization after Surgery: Post Hoc Analysis of Two Randomized Trials.

BACKGROUND: Early mobilization is incorporated into many enhanced recovery pathways. Inadequate analgesia or excessive opioids may restrict postoperative mobilization. The authors tested the hypotheses that in adults recovering from abdominal surgery, postoperative pain and opioid consumption are inversely related to postoperative mobilization, and that postoperative mobilization is associated with fewer potentially related complications. METHODS: The authors conducted a subanalysis of two trials that enrolled adults recovering from abdominal surgery. Posture and movement were continuously monitored for 48 postoperative hours using noninvasive untethered monitors. Mobilization was defined as the fraction of monitored time spent sitting or standing. RESULTS: A total of 673 patients spent a median [interquartile range] of 7% [3 to 13%] of monitored time sitting or standing. Mobilization time was 1.9 [1.0 to 3.6] h/day for patients with average pain scores 3 or lower, but only 1.2 [0.5 to 2.6] h/day in those with average scores 6 or greater. Each unit increase in average pain score was associated with a decrease in mobilization time of 0.12 (97.5% CI, 0.02 to 0.24; P = 0.009) h/day. In contrast, there was no association between postoperative opioid consumption and mobilization time. The incidence of the composite of postoperative complications was 6.0% (10 of 168) in the lower mobilization quartile, 4.2% (7 of 168) in the second quartile, and 0% among 337 patients in the highest two quartiles (P = 0.009). CONCLUSIONS: Patients recovering from abdominal surgery spent only 7% of their time mobilized, which is considerably less than recommended. Lower pain scores are associated with increased mobility, independently of opioid consumption. Complications were more common in patients who mobilized poorly.

Preoperative Vitamin D Deficiency Is Associated With Postoperative Delirium in Critically Ill Patients.

INTRODUCTION: Postoperative delirium is common, with a reported incidence of 11% to 80% in critically ill patients. Delirium is an independent prognostic factor for poor hospital outcomes. Low vitamin D concentrations are associated with a decline in cognitive function. We therefore tested the hypothesis that low preoperative serum 25-hydroxyvitamin D [25(OH)D] concentrations are associated with postoperative delirium in critically ill patients. METHOD: We conducted a retrospective analysis of adults in a surgical intensive care unit for at least 48 hours immediately after non-cardiac and non-neurosurgical operations at Cleveland Clinic between 2013 and 2018. Delirium was assessed by trained nurses using CAM-ICU twice daily for the initial 5 postoperative days. Any positive value was considered evidence of delirium. We assessed the association between 25(OH)D concentrations within a year before surgery and the incidence of postoperative delirium using logistic regression, adjusted for potential confounders. A linear spline term with a knot at 30 ng/ml, the threshold for normal 25(OH)D concentration, was added to accommodate a nonlinear relationship between 25(OH)D concentrations and delirium. RESULTS: We included 632 patients, who had a mean (SD) 25(OH)D concentration of 25 (15) ng/ml; 55% (346/632) experienced delirium. We observed an adjusted odds ratio of 1.4 (95% CI: [1.1, 1.8], P = 0.01) for delirium per 10 ng/ml decrease in 25(OH)D concentrations when patients' 25(OH)D concentrations were less than 30 ng/ml. In patients whose 25(OH)D concentrations were at least 30 ng/ml, the adjusted odds ratio was 0.9 (95% CI: [0.7, 1.1], P = 0.36). CONCLUSION: Preoperative 25(OH)D concentrations are associated with postoperative delirium in patients whose concentrations are below the normal threshold, but not at concentrations ≥30 ng/ml. A trial will be needed to determine whether the relationship is causal, and whether vitamin D supplementation before surgery might reduce the incidence of delirium.

Sugammadex Versus Neostigmine for Reversal of Residual Neuromuscular Blocks After Surgery: A Retrospective Cohort Analysis of Postoperative Side Effects.

BACKGROUND: Sugammadex and neostigmine given to reverse residual neuromuscular blockade can cause side effects including bradycardia, anaphylaxis, bronchospasm, and even cardiac arrest. We tested the hypothesis that sugammadex is noninferior to neostigmine on a composite of clinically meaningful side effects, or vice versa. METHODS: We analyzed medical records of patients who had general, cardiothoracic, or pediatric surgery and were given neostigmine or sugammadex from June 2016 to December 2019. Our primary outcome was a collapsed composite of bradycardia, anaphylaxis, bronchospasm, and cardiac arrest occurring between administration of the reversal agent and departure from the operation room. We a priori restricted our analysis to side effects requiring pharmacologic treatment that were therefore presumably clinically meaningful. Sugammadex would be considered noninferior to neostigmine (or vice versa) if the odds ratio for composite of side effects did not exceed 1.2. RESULTS: Among 89,753 surgeries in 70,690 patients, 16,480 (18%) were given sugammadex and 73,273 (82%) were given neostigmine. The incidence of composite outcome was 3.4% in patients given sugammadex and 3.0% in patients given neostigmine. The most common individual side effect was bradycardia (2.4% in the sugammadex group versus 2.2% neostigmine). Noninferiority was not found, with an estimated odds ratio of 1.21 (sugammadex versus neostigmine; 95% confidence interval [CI], 1.09-1.34; noninferiority P = .57), and neostigmine was superior to sugammadex with an estimated odds ratio of 0.83 (0.74-0.92), 1-side superiority P < .001. CONCLUSIONS: The composite incidence was less with neostigmine than with sugammadex, but only by 0.4% (a negligible clinical effect). Since 250 patients would need to be given neostigmine rather than sugammadex to avoid 1 episode of a minor complication such as bradycardia or bronchospasm, we conclude that sugammadex and neostigmine are comparably safe.

The Relation Between Mean Arterial Pressure and Cardiac Index in Major Abdominal Surgery Patients: A Prospective Observational Cohort Study.

BACKGROUND: Cardiac output is an important hemodynamic variable and determines oxygen delivery. In contrast to blood pressure, cardiac output is rarely measured even in high-risk surgical patients, suggesting that clinicians consider blood pressure to be a reasonable indicator of systemic blood flow. However, the relationship depends on constant vascular tone and volume, both of which routinely vary during anesthesia and surgery. We therefore tested the hypothesis that there is no clinically meaningful correlation between mean arterial pressure and cardiac index in major abdominal surgery patients. METHODS: In this prospective observational study, we assessed the relationship between mean arterial pressure and cardiac index in 100 patients having major abdominal surgery under general anesthesia. RESULTS: The pooled within-patient correlation coefficient calculated using meta-analysis methods was r = 0.34 (95% confidence interval, 0.28-0.40). Linear regression using a linear mixed effects model of cardiac index on mean arterial pressure revealed that cardiac index increases by 0.014 L·min-1·m-2 for each 1 mm Hg increase in mean arterial pressure. The 95% Wald confidence interval of this slope was 0.011 to 0.018 L·min-1·m-2·mm Hg-1 and thus within predefined equivalence margins of -0.03 and 0.03 L·min-1·m-2·mm Hg-1, thereby demonstrating lack of clinically meaningful association between mean arterial pressure and cardiac index. CONCLUSIONS: There is no clinically meaningful correlation between mean arterial pressure and cardiac index in patients having major abdominal surgery. Intraoperative blood pressure is thus a poor surrogate for cardiac index.

Effect of a Rapid Response Team on the Incidence of In-Hospital Mortality.

BACKGROUND: Approximately half of the life-limiting events, such as cardiopulmonary arrests or cardiac arrhythmias occurring in hospitals, are considered preventable. These critical events are usually preceded by clinical deterioration. Rapid response teams (RRTs) were introduced to intervene early in the course of clinical deterioration and possibly prevent progression to an event. An RRT was introduced at the Cleveland Clinic in 2009 and transitioned to an anesthesiologist-led system in 2012. We evaluated the association between in-hospital mortality and: (1) the introduction of the RRT in 2009 (primary analysis), and (2) introduction of the anesthesiologist-led system in 2012 and other policy changes in 2014 (secondary analyses). METHODS: We conducted a single-center, retrospective analysis using the medical records of overnight hospitalizations from March 1, 2005, to December 31, 2018, at the Cleveland Clinic. We assessed the association between the introduction of the RRT in 2009 and in-hospital mortality using segmented regression in a generalized estimating equation model to account for within-subject correlation across repeated visits. Baseline potential confounders (demographic factors and surgery type) were controlled for using inverse probability of treatment weighting on the propensity score. We assessed whether in-hospital mortality changed at the start of the intervention and whether the temporal trend (slope) differed from before to after initiation. Analogous models were used for the secondary outcomes. RESULTS: Of 628,533 hospitalizations in our data set, 177,755 occurred before and 450,778 after introduction of our RRT program. Introduction of the RRT was associated with a slight initial increase in in-hospital mortality (odds ratio [95% confidence interval {CI}], 1.17 [1.09-1.25]; P < .001). However, while the pre-RRT slope in in-hospital mortality over time was flat (odds ratio [95% CI] per year, 1.01 [0.98-1.04]; P = .60), the post-RRT slope decreased over time, with an odds ratio per additional year of 0.961 (0.955-0.968). This represented a significant improvement (P < .001) from the pre-RRT slope. CONCLUSIONS: We found a gradual decrease in mortality over a 9-year period after introduction of an RRT program. Although mechanisms underlying this decrease are unclear, possibilities include optimization of RRT implementation, anesthesiology department leadership of the RRT program, and overall improvements in health care delivery over the study period. Our findings suggest that improvements in outcome after RRT introduction may take years to manifest. Further work is needed to better understand the effects of RRT implementation on in-hospital mortality.

New-Onset Atrial Fibrillation in Patients Undergoing Liver Transplantation: Retrospective Analysis of Risk Factors and Outcomes.

OBJECTIVE: The authors aimed to identify predictors of new-onset postoperative atrial fibrillation (POAF) during the initial 90 days following liver transplantation, and to assess the association between POAF in-hospital and 1-year mortality. DESIGN: A retrospective cohort study. SETTING: At a university hospital between 2005 and 2017. PATIENTS: Adults without a history of preoperative atrial fibrillation who underwent orthotopicliver transplantation. MEASUREMENT AND MAIN RESULTS: The authors assessed the univariate association between new-onset of POAF in the postoperative period and each potential factor through a logistic regression model. Moreover, they explored predictors for POAF through stepwise selection. Finally, the authors assessed the relationship between POAF and in-hospital and 1-year mortality using logistic regression models, and whether the duration of atrial fibrillation was associated with in-hospital and 1-year mortality. Among 857 patients, 89 (10.4%) developed POAF. Using only preoperative variables, pulmonary hypertension, age, Model for End-Stage Liver Disease (MELD) score, and White race were identified as the most important predictors. Model discrimination was 0.75 (95% CI: 0.69-0.80), and incorporating intraoperative variables was 0.77 (95% CI: 0.72-0.82). In-hospital mortality was observed in 7.2% (6/83) of patients with new-onset of POAF, and in 2.8% (22/768) without, with confounder-adjusted odds ratio (OR) 1.00 (97.5% CI: 0.29,3.45; p = 0.996). One-year mortality was 22.4% (20/89) in patients who developed POAF and 8.3% (64/768) in patients who did not, confounder-adjusted OR 2.64 (97.5% CI: 1.35-5.16; p = 0.001). The duration of POAF did not affect long-term postoperative mortality. CONCLUSION: Preoperative, mostly unmodifiable comorbidities are important risk factors for new-onset POAF after liver transplantation. The POAF was not associated with in-hospital mortality, but with increased 1-year mortality. Once developed, the duration of POAF did not affect long-term mortality after a liver transplant.

Assisted Fluid Management Software Guidance for Intraoperative Fluid Administration.

BACKGROUND: Excessive or inadequate fluid administration causes complications, but despite this, fluid administration during noncardiac surgery is highly variable. Goal-directed management helps optimize the amount and timing of fluid administration; however, implementation is difficult because algorithms are complex. The authors therefore tested the performance of the Acumen Assisted Fluid Management software (Edwards Lifesciences, USA), which is designed to guide optimal intravenous fluid administration during surgery. METHODS: In this multicenter, prospective, single-arm cohort evaluation, the authors enrolled 330 adults scheduled for moderate- to high-risk noncardiac surgery that required arterial catheter insertion and mechanical ventilation. Clinicians chose a fluid strategy based on a desired 10%, 15%, or 20% increase in stroke volume (SV) in response to a fluid bolus. Dedicated fluid management software prompted "test" or "recommended" boluses, and clinicians were free to initiate a "user" bolus of 100 to 500 ml of crystalloid or colloid. Clinicians were free to accept or decline the software prompts. The authors primarily compared the fraction of software-recommended boluses that produced suitable increases in SV to a 30% reference rate. On an exploratory basis, we compared responses to software-recommended and clinician-initiated boluses. RESULTS: Four hundred twenty-four of 479 (89%) software-recommended fluid boluses and 508 of 592 (86%) clinician-initiated fluid boluses were analyzed per protocol. Of those, 66% (95% CI, 62 to 70%) of delivered fluid boluses recommended by the software resulted in desired increases in SV, compared with the 30% reference rate, whereas only 41% (95% CI, 38 to 44%) of clinician-initiated boluses did (P < 0.0001). The mean ± SD increase in SV after boluses recommended by the software was 14.2 ± 13.9% versus 8.3 ± 12.1% (P < 0.0001) for those initiated by clinicians. CONCLUSIONS: Fluid boluses recommended by the software resulted in desired SV increases more often, and with greater absolute SV increase, than clinician-initiated boluses. Automated assessment of fluid responsiveness may help clinicians optimize intraoperative fluid management during noncardiac surgery.

Postoperative delirium and neurocognitive disorders.

PURPOSE OF REVIEW: Postoperative neurocognitive disorders (NCD) are significant causes of morbidity and mortality. In this paper, we will review our understanding and potential management of postoperative NCD. RECENT FINDINGS: Postoperative delirium, delayed neurocognitive recovery and postoperative cognitive dysfunction (POCD) are recognized as a part of the continuous spectrum of postoperative NCD. Although the pathophysiology is still poorly understood, there is renewed focus on improving neurocognitive outcomes of aging surgical population. Novel methods of neurocognitive screening are developed and research in the prevention and management of NCD has gained traction. SUMMARY: A spectrum of NCD exists in surgical patients ranging from postoperative delirium, delayed neurocognitive recovery, and POCD. Identification of patients at-risk for developing NCD can help target appropriate perioperative intervention. Also, specialized care teams and the implementation of standardized protocols are crucial for the successful management perioperative NCD. Finally, large, randomized, multicenter studies are needed to confirm benefits of preventive and treatment strategies.

Association between postoperative haemoglobin concentrations and composite of non-fatal myocardial infarction and all-cause mortality in noncardiac surgical patients: post hoc analysis of the POISE-2 trial.

BACKGROUND: Myocardial infarction is the most common postoperative major vascular complication. Perioperative anaemia may contribute to cardiac supply-demand mismatch, and therefore myocardial injury. We therefore tested the hypothesis that the lowest in-hospital postoperative haemoglobin concentration is associated with a composite of non-fatal myocardial infarction and all-cause mortality within the first 30 days after noncardiac surgery. METHODS: We conducted a retrospective analysis of patients enrolled in the PeriOperative Ischemic Evaluation-2 (POISE-2) trial. We assessed the association between the lowest postoperative haemoglobin concentration during the initial hospitalisation and a composite of non-fatal myocardial infarction (Third Universal Definition) and all-cause mortality within 30 postoperative days, using a multivariable logistic regression model. RESULTS: We analysed 7227 patients from POISE-2, of whom 7.8% developed myocardial infarction; 1.5% died within 30 days. The composite primary outcome of non-fatal myocardial infarction and all-cause mortality occurred in 8.9% patients overall, ranging from 16% in patients with postoperative haemoglobin concentrations <88 g L-1 to 4.1% in patients with postoperative haemoglobin >113 g L-1. After adjusting for baseline factors, in patients with a lowest postoperative haemoglobin concentration <110 g L-1, each 10 g L-1 reduction in the lowest postoperative haemoglobin concentration was associated with a 1.46 (95% confidence interval: 1.37-1.56; P<0.001) fold increase in the odds of the composite outcome. In contrast, there was no significant relationship amongst patients with lowest postoperative haemoglobin concentration >110 g L-1. CONCLUSIONS: Postoperative anaemia may be a modifiable risk factor for non-fatal myocardial infarction and all-cause mortality.

Association between postoperative haemoglobin and myocardial injury after noncardiac surgery: a retrospective cohort analysis.

BACKGROUND: Myocardial injury after noncardiac surgery (MINS) is common, mostly silent, and a strong predictor of postoperative mortality. MINS appears to result from myocardial supply-demand mismatch. Recent data support restrictive perioperative transfusion strategies that can result in low postoperative haemoglobin concentrations. Whether low postoperative haemoglobin is associated with myocardial injury remains unknown. We therefore tested the hypothesis that anaemia is associated with an increased risk of myocardial injury in adults having noncardiac surgery. METHODS: We conducted a retrospective analysis of adults ≥45 yr old who had routine postoperative troponin T (TnT) monitoring after noncardiac surgery at the Cleverland Clinic (including those enrolled in the PeriOperative ISchemic Evaluation-2 Trial [POISE-2], the Safety of Addition of Nitrous Oxide to General Anaesthesia in At-risk Patients Having Major Non-cardiac Surgery [ENIGMA-II], Vascular Events In Noncardiac Surgery Patients Cohort Evaluation Study [VISION], and Anaesthetic Depth and Complications After Major Surgery [BALANCED] trial). Patients with baseline increase in TnT and non-ischaemic aetiologies for TnT increase were excluded. The association between postoperative haemoglobin concentration during the 3 initial postoperative days and the incidence of MINS (fourth-generation TnT ≥0.03 ng ml-1 judged as attributable to ischaemia) was assessed using a time-varying covariate Cox proportional hazards survival analysis. RESULTS: Among 6141 patients, 4480 were analysed. The incidence of MINS was 155/4480 (3.5%), ranging from 0/345 (0%) among patients whose lowest postoperative haemoglobin exceeded 13 g dl-1 to 52/611 (8.5%) in patients whose minimum postoperative haemoglobin was <8 g dl-1. The confounder-adjusted hazard ratio [95% confidence interval] for having MINS was 1.29 [1.16-1.42] for every 1 g dl-1 decrease in postoperative haemoglobin in a time-varying covariate analysis. Similar associations were identified in sensitivity analyses. CONCLUSION: Lower postoperative haemoglobin values are associated with MINS. Whether this association is modifiable by prevention or treatment of anaemia remains to be determined.

Do It Often, Do It Better: Association Between Pairs of Experienced Subspecialty Anesthesia Caregivers and Postoperative Outcomes. A Retrospective Observational Study.

BACKGROUND: Anesthesiologists typically care for patients having a broad range of procedures. Outcomes might be improved when care is provided by caregivers experienced in particular types of surgery. We tested the hypothesis that intraoperative care provided by pairs of anesthesia caregivers having significant experience with a particular type of surgery reduces a composite of in-hospital death and 6 serious complications, including bleeding, cardiac, gastrointestinal, infectious, respiratory, and urinary complications, compared to care provided by pairs of anesthesia caregivers with less experience. METHODS: We included patients having surgery lasting at least 30 minutes. Using cluster analysis, attending anesthesiologists, and Certified Registered Nurse Anesthetists (CRNAs) were identified as experienced or inexperienced caregivers for each type of surgery at the case level. We then compared surgeries for which anesthesia was provided by a pair of experienced caregivers versus a pair of inexperienced caregivers on our composite outcome. We estimated the average relative effect (ie, the exponentiated average log odds ratio) of receiving anesthesia from an experienced versus inexperienced caregiver pair across the 7 components of the composite outcome using a generalized estimating equation (GEE) model to adjust for between-component correlation and with inverse propensity score weighing to adjust for potential confounding from a host of variables. RESULTS: A total of 8968 patients who received anesthesia care by an experienced pair were compared with 25,361 patients who received care from an inexperienced pair, adjusting for potential confounding. The incidence of composite complications (ie, any component event) was 7.6% (677/8968) for experienced pairs and 12% (2976/25,361) for inexperienced pairs (P < .001). Care by experienced pairs of caregivers was associated with lower odds of the composite outcome with an estimated average relative effect odds ratio across the individual components of 0.61 (95% confidence interval [CI], 0.54-0.71), P < .001. Among the 7 components of the primary outcome, experienced pairs of providers had significantly lower estimated odds of bleeding, infection, and mortality. CONCLUSIONS: Anesthesia care by experienced pairs was associated with fewer bleeding complications, fewer infections, shorter hospitalization, and reduced in-hospital mortality.