Calculation of Confidence Intervals for Differences in Medians Between Groups and Comparison of Methods Continuous data that are not normally distributed are typically presented in terms of median and interquartile range (IQR) for each group. High-quality anesthesia journals often require that confidence intervals are calculated and presented for all estimated associations of interest reported within a manuscript submission, and therefore, methods for calculating confidence intervals for differences in medians are vital. It is informative to present the difference in medians along with a confidence interval to provide insight about the magnitude of variability for the estimated difference. In a clinical research example using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) Pediatric database, we demonstrate how to estimate confidence intervals for the difference in medians using 3 different statistical methods: the Hodges-Lehmann estimator, bootstrap resampling with replacement, and quantile regression modeling on the median (median regression). We discuss specific recommendations regarding the methods according to the objectives of the study as well as the distribution of the data as it pertains to the assumptions of the respective methods. Quantile regression allows for covariate adjustment, which may be an advantage in situations where differences in medians between groups may be due to confounding. Accepted for publication October 7, 2019. Funding: None. The authors declare no conflicts of interest. 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.anesthesia-analgesia.org). Reprints will not be available from the authors. Address correspondence to Steven J. Staffa, MS, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, 300 Longwood Ave, Boston, MA 02115. Address e-mail to steven.staffa@childrens.harvard.edu. © 2019 International Anesthesia Research Society

A Cross-Sectional Survey to Determine the Prevalence of Burnout Syndrome Among Anesthesia Providers in Zambian Hospitals BACKGROUND: Burnout is a psychological syndrome that results from chronic exposure to job stress. It is defined by a triad of emotional exhaustion, depersonalization, and reduced personal accomplishment. In research, mostly from high-income countries, burnout is common in health care professionals, especially in anesthesiologists. Burnout can negatively impact patient safety, the physical and mental health of the anesthetist, and institutional efficiency. However, data on burnout for anesthesia providers in low- and middle-income countries are poorly described. This study sought to determine the prevalence of burnout syndrome among all anesthesia providers (physician and nonphysician) working in Zambian hospitals and to determine which sociodemographic and occupational factors were associated with burnout. METHODS: A questionnaire was sent to all Zambian anesthesia providers working in private and public hospitals. The questionnaire assessed burnout using the Maslach Burnout Inventory Human Services Survey, a validated 22-item survey widely used to measure burnout among health professionals. Sociodemographic and occupational factors postulated to be associated with burnout were also assessed. RESULTS: Surveys were distributed to all 184 anesthesia providers in Zambia; 160 were returned. This resulted in a response rate representing 87% of all anesthesia providers in the country. Eighty-six percentage of respondents were nonphysician anesthesia providers. Burnout was present in 51.3% (95% confidence interval [CI], 43.2–59.2) of participants. Logistic regression analysis revealed that “not having the right team to carry out work to an appropriate standard” (odds ratio, 2.91, 95% CI, 1.33–6.39; P = .008), and “being a nonphysician” (odds ratio, 3.4, 95% CI, 1.25–12.34; P = .019) were significantly associated with burnout in this population. CONCLUSIONS: In a cross-sectional survey of anesthesia providers in Zambia, >50% of the respondents met the criteria for burnout. The risk was particularly high among nonphysician providers who typically work in isolated rural practice. Efforts to decrease burnout rates through policy and educational initiatives to increase the quantity and quality of training for anesthesia providers should be considered. Accepted for publication August 2, 2019. Funding: A.J.'s time was funded by the Zambia Anaesthesia Development Project. D.M. receives salary support from The Ottawa Hospital Anesthesia Alternate Funds Association, the University of Ottawa, and the Canadian Anesthesiologists' Society Career Scientist Award. The authors declare no conflicts of interest. 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.anesthesia-analgesia.org). Reprints will not be available from the authors. Address correspondence to M. Dylan Bould, MBChB, Med, Department of Anesthesiology and Pain Medicine, Children's Hospital of Eastern Ontario, Ottawa, Canada. Address e-mail to dbould@cheo.on.ca. © 2019 International Anesthesia Research Society

A Prospective, Randomized, Double-Blinded Study of the Effect of Intravenous Ondansetron on the Effective Dose in 50% of Subjects of Prophylactic Phenylephrine Infusions for Preventing Spinal Anesthesia–Induced Hypotension During Cesarean Delivery BACKGROUND: Ondansetron has been shown to reduce the incidence of hypotension and vasopressor requirement during spinal anesthesia for obstetric and nonobstetric surgery. However, the magnitude of this effect has not been fully quantified. In this parallel-group, randomized, double-blinded study, we determined the effective dose in 50% of subjects (ED50) of a prophylactic phenylephrine infusion for preventing hypotension in patients who received a single dose of intravenous ondansetron 4 mg or saline control before combined spinal–epidural anesthesia for elective cesarean delivery. ED50 values obtained were compared to estimate the effect of ondansetron versus placebo on vasopressor requirement. METHODS: Sixty parturients were randomly assigned to receive ondansetron (group O) or saline control (group C) 10 minutes before positioning for induction of spinal anesthesia. A prophylactic phenylephrine infusion was used to prevent hypotension. The first patient in each group received a phenylephrine infusion at the rate of 0.5 µg/kg/min. The infusion rate for each subsequent patient was varied with increments or decrements of 0.05 µg/kg/min based on the response of the previous patient, and the effective dose of the phenylephrine infusion for preventing hypotension in 50% of patients (ED50) was calculated for each group and compared using up-down sequential analysis. Probit regression was applied as a backup and sensitivity analysis was used to compare ED50 values for phenylephrine between groups by comparing calculated relative mean potency. RESULTS: The ED50 (mean [95% confidence interval (CI)]) of the rate of phenylephrine infusion was lower in group O (0.24 µg/kg/min [0.10–0.38 µg/kg/min]) compared with group C (0.32 µg/kg/min [0.14–0.47 µg/kg/min]) (P < .001). The total consumption of phenylephrine (mean ± standard deviation [SD]) until delivery was lower in group O (316.5 ± 25.9 µg) than in group C (387.7 ± 14.7 µg, P = .02). The estimate of relative median potency for phenylephrine for group O versus group C was 0.74 (95% CI, 0.37–0.95). CONCLUSIONS: Under the conditions of this study, intravenous ondansetron 4 mg reduced the ED50 of a prophylactic phenylephrine infusion by approximately 26% in patients undergoing cesarean delivery under combined spinal–epidural anesthesia. Accepted for publication October 8, 2019. Funding: Supported by grants from the Jiaxing Science and Technology Bureau (No. 2018AD32108) and from the National Natural Science Foundation of China (NSFC) (No 81471126). The authors declare no conflicts of interest. Clinical Trials Registration: Chinese Clinical Trial Registration number: ChiCTR-TRC-1800018756. Reprints will not be available from the authors. Address correspondence to Xinzhong Chen, MD, Department of Anesthesia, Women’s Hospital, Zhejiang University School of Medicine, Xueshi Rd, 1#, Hangzhou, China 310006. Address e-mail to chenxinz@zju.edu.cn. © 2019 International Anesthesia Research Society

Trends in Central Venous Catheter Insertions by Anesthesia Providers: An Analysis of the Medicare Physician Supplier Procedure Summary From 2007 to 2016 BACKGROUND: Central line insertion is a core skill for anesthesiologists. Although recent technical advances have increased the safety of central line insertion and reduced the risk of central line–associated infection, noninvasive hemodynamic monitoring and improved intravenous access techniques have also reduced the need for perioperative central venous access. We hypothesized that the number of central lines inserted by anesthesiologists has decreased over the past decade. To test our hypothesis, we reviewed the Medicare Physician Supplier Procedure Summary (PSPS) database from 2007 to 2016. METHODS: Claims for central venous catheter placement were identified in the Medicare PSPS database for nontunneled and tunneled central lines. Pulmonary artery catheter insertion was included as a nontunneled line claim. We stratified line insertion claims by specialty for Anesthesiology (including Certified Registered Nurse Anesthetists and Anesthesiology Assistants), Surgery, Radiology, Pulmonary/Critical Care, Emergency Physicians, Internal Medicine, and practitioners who were not anesthesia providers such as Advanced Practice Nurses (APNs) and Physician Assistants (PAs). Utilization rates per 10,000 Medicare beneficiaries were then calculated by specialty and year. Time-based trends were analyzed using Joinpoint linear regression, and the Average Annual Percent Change (AAPC) was calculated. RESULTS: Between 2007 and 2016, total claims for central venous catheter insertions of all types decreased from 440.9 to 325.3 claims/10,000 beneficiaries (AAPC = −3.4, 95% confidence interval [CI], −3.6 to −3.2: P < .001). When analyzed by provider specialty and year, the number of nontunneled line insertion claims fell from 43.1 to 15.9 claims/10,000 (AAPC = −7.1; −7.3 to −7.0: P < .001) for surgeons, from 21.3 to 18.5 claims/10,000 (AAPC = −2.5; −2.8 to −2.1: P < .001) for radiologists, and from 117.4 to 72.7 claims/10,000 (AAPC = −5.2; 95% CI, −6.3 to −4.0: P < .001) for anesthesia providers. In contrast, line insertions increased from 18.2 to 26.0 claims/10,000 (AAPC = 3.2; 2.3–4.2: P < .001) for Emergency Physicians and from 3.2 to 9.3 claims/10,000 (AAPC = 6.0; 5.1–6.9: P < .001) for PAs and APNs who were not anesthesia providers. Among anesthesia providers, the share of line claims made by nurse anesthetists increased by 14.5% over the time period. CONCLUSIONS: We observed a 38.3% decrease in claims for nontunneled central lines placed by anesthesiologists from 2007 to 2016. These findings have implications for anesthesiology resident training and maintenance of competence among practicing clinicians. Further research is needed to clarify the effect of decreasing line insertion numbers on line insertion competence among anesthesiologists. Accepted for publication October 4, 2019. Funding: None. Conflicts of Interest: See Disclosures at the end of the article. 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.anesthesia-analgesia.org). Reprints will not be available from the authors. Address correspondence to Daniel S. Rubin, MD, MS, Department of Anesthesia and Critical Care, University of Chicago, 5841 S Maryland Ave, MC-4028, Chicago, IL 60637. Address e-mail to drubin@dacc.uchicago.edu. © 2019 International Anesthesia Research Society

Trials and Tribulations of Viscoelastic-Based Determination of Fibrinogen Concentration Acquired fibrinogen deficiency is a major determinant of severe bleeding in different clinical conditions, including cardiac surgery, trauma, postpartum hemorrhage, liver surgery, and transplantation. The existing guidelines recommend to supplement fibrinogen in patients with severe bleeding when the fibrinogen concentration is <1.5 g/L. Viscoelastic tests (VETs) provide a fast determination of the fibrinogen contribution to clot firmness and allow prompt treatment of acquired fibrinogen deficiency. However, different VET devices are presently available on the market, based on different technologies and different activators and platelet inhibitors. The available tests are the functional fibrinogen (FF, thromboelastography), the fibrinogen contribution to clot firmness (fibrinogen determination [FIBTEM], thromboelastometry), and the fibrinogen contribution to clot strength (FCS, sonorheometry). All these tests have a moderate to very good correlation with the Clauss fibrinogen assays; however, when comparing VET-based fibrinogen contribution to clot firmness with Clauss fibrinogen concentration, strong differences occur within the same test under different conditions and between different tests. The most widely studied test is the thromboelastometric FIBTEM; the best predictor of a Clauss fibrinogen <1.5 g/L is placed at a maximum clot firmness around 8 mm of amplitude. Fewer data are available for thromboelastographic FF, but the correspondent value is in the range of 12 mm. Overall, due to an incomplete inhibition of platelet contribution, FF overestimates the fibrinogen contribution with respect to FIBTEM. Data on sonorheometry FCS are limited and conflicting. When addressing the correlation between different tests, it is good in general, but no fixed conversion factors can be proposed, due to a considerable dispersion of the experimental points. In conclusion, VET-based fibrinogen tests are certainly powerful tools that are presently suggested by the existing guidelines; however, when using them for clinical decision-making, users should consider the possible sources of bias, which include the different level of platelet inhibition, the role of platelet count and function, the possible different degrees of blood activation with tissue factor, the important role of factor XIII in stabilizing the fibrin clot, and others. Accepted for publication October 2, 2019. Funding: None. Conflicts of Interest: See Disclosures at the end of the article. Reprints will not be available from the authors. Address correspondence to Marco Ranucci, MD, FESC, Department of Cardiovascular Anesthesia and Intensive Care, IRCCS Policlinico San Donato, Via Morandi 30, San Donato Milanese, Milan 20097, Italy. Address e-mail to cardioanestesia@virgilio.it. © 2019 International Anesthesia Research Society

Association Between Perioperative Hypotension and Delirium in Postoperative Critically Ill Patients: A Retrospective Cohort Analysis BACKGROUND: Postoperative delirium is common in critically ill patients, with a reported incidence of 11%–43%, and is associated with significant morbidity and cost. Perioperative hypotension and consequent brain hypoperfusion may contribute. We, therefore, tested the hypotheses that intraoperative and postoperative hypotension are associated with critical care delirium. METHODS: We included 1083 postoperative patients who were admitted directly from an operating room to the surgical intensive care unit. Delirium was assessed with the Confusion Assessment Method for Intensive Care Unit patients at 12-hour intervals. We used a confounder-adjusted Cox proportional hazard survival model to assess the association between the amount of intraoperative hypotension, which was measured as the time-weighted average of mean arterial pressure <65 mm Hg, and delirium while in critical care. Thereafter, we used a Cox model with the lowest mean arterial pressure on each intensive care day as a time-varying covariate to assess the relationship between critical care hypotension and delirium, adjusted for confounders and amount of intraoperative hypotension. RESULTS: Three hundred seventy-seven (35%) patients had delirium within the first 5 postoperative days in the surgical intensive care unit. Intraoperative hypotension was moderately associated with higher odds of postoperative delirium. The adjusted hazard ratio associated with 1 mm Hg increase in time-weighted average of mean arterial pressure <65 mm Hg was 1.11 (95% confidence interval [CI], 1.03–1.20; P = .008). Postoperatively, a 10 mm Hg reduction in the lowest mean pressure on each day in the critical care unit was significantly associated with a higher hazard of delirium, with an adjusted hazard ratio 1.12 (95% CI, 1.04–1.20; P = .003). CONCLUSIONS: Both intraoperative and postoperative hypotension are associated with delirium in postoperative critical care patients. The extent to which these relationships are causal remains unknown, but to the extent that they are, hypotension prevention may help reduce delirium and should be studied in prospective clinical trials. Accepted for publication September 30, 2019. Funding: None. The authors declare no conflicts of interest. 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.anesthesia-analgesia.org). Reprints will not be available from the authors. Address correspondence to Kamal Maheshwari, MD, Department of General Anesthesiology, Anesthesiology Institute, Cleveland Clinic, 9500 Euclid Ave, E-31, Cleveland, OH 44195. Address e-mail to maheshk@ccf.org. © 2019 International Anesthesia Research Society

Obstetric Anesthesiology: An Illustrated Case-Based Approach No abstract available

Cricoid Pressure: Do No Harm, But Do It Correctly No abstract available

Transfusion of Red Blood Cells, Fresh Frozen Plasma, or Platelets Is Associated With Mortality and Infection After Cardiac Surgery in a Dose-Dependent Manner BACKGROUND: It is unclear whether transfusion of platelets or fresh frozen plasma, in addition to red blood cells, is associated with an increased risk of mortality and infection after cardiac surgery. METHODS: Patients who underwent valve surgery and/or coronary artery bypass grafting from January 1, 2011 to June 30, 2017 and September 1, 2013 to June 30, 2017 at 2 centers performing cardiac surgery were included in this retrospective study. After stratifying patients based on propensity score matching, we compared rates of mortality and infection between patients who transfused red blood cells, fresh frozen plasma, or platelets with those who did not receive such transfusions. We also compared outcomes between patients who received any of the 3 blood products and patients who received no transfusions at all. Multivariable logistic regression was used to assess associations between transfusion and outcomes. RESULTS: Of 8238 patients in this study, 109 (1.3%) died, 812 (9.9%) experienced infection, and 4937 (59.9%) received at least 1 type of blood product. Transfusion of any blood type was associated with higher rates of mortality (2.0% vs 0.18%; P < .01) and infection (13.3% vs 4.8%; P < .01). Each of the 3 blood products was independently associated with an increase in mortality per unit transfused (red blood cells, odds ratio 1.18, 95% confidence interval [CI], 1.14–1.22; fresh frozen plasma, odds ratio 1.24, 95% CI, 1.18–1.30; platelets, odds ratio 1.12, 95% CI, 1.07–1.18). Transfusing 3 units of any of the 3 blood products was associated with a dose-dependent increase in the incidence of mortality (odds ratio 1.88, 95% CI, 1.70–2.08) and infection (odds ratio 1.50, 95% CI, 1.43–1.57). CONCLUSIONS: Transfusion of red blood cells, fresh frozen plasma, or platelets is an independent risk factor of mortality and infection, and combination of the 3 blood products is associated with adverse outcomes after cardiac surgery in a dose-dependent manner. Accepted for publication October 3, 2019. Funding: This work was supported by the 1.3.5 Project for Disciplines of Excellence (ZY2016101), West China Hospital of Sichuan University, and the National and Zhejiang Health and Family Planning Commission (2018272998), the Second Affiliated Hospital of Zhejiang University. The authors declare no conflicts of interest. 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.anesthesia-analgesia.org). Reprints will not be available from the authors. Address correspondence to Min Yan, MD, PhD, Department of Anesthesiology, The Second Affiliated Hospital, Zhejiang University, No. 88 Jiefang Rd, Hangzhou, Zhejiang, 330100, China. Address e-mail to zryanmin@zju.edu.cn. © 2019 International Anesthesia Research Society

Patient Blood Management in Cardiac Surgery No abstract available