Secondary Logo

Journal Logo

Patient Safety: Research Report

Attending Handoff Is Correlated with the Decision to Delay Extubation After Surgery

Anastasian, Zirka H. MD; Kim, Minjae MD, MS; Heyer, Eric J. MD, PhD; Wang, Shuang PhD; Berman, Mitchell F. MD, MPH

Author Information
doi: 10.1213/ANE.0000000000001069

Tracheal extubation at the end of a surgical case is a critical step during a patient’s emergence from general anesthesia. Postoperatively, factors that change a patient’s airway and the ability to maintain it include anatomical differences, residual pharmacologic changes, and physiologic changes. More than 20% of major complications in airway management have been shown to occur around the time of extubation, and the consequences are severe, including hypoxia and death.1,2 Although there are many guidelines focused on the intubation of an airway, literature focused on the extubation of the airway at the end of surgery is limited.3,4 Prolonged endotracheal intubation is not a benign process: It can lead to complications including bronchopulmonary infections,5,6 glottic stenosis,7 and dysphagia.8 In making the decision to extubate or to leave the patient intubated at the end of a surgical case, the anesthesiologist weighs the individual patient’s risks and benefits in light of the preoperative and intraoperative factors.

Predictors of delayed postoperative extubation have been studied in the cardiac, thoracic, neurosurgical, and spine populations. Preoperative factors including age,9 ASA physical status, increased body mass index,10,11 and baseline lung function12–14 were shown to correlate with delayed postoperative extubation. Intraoperative factors associated with delayed postoperative extubation include duration of case,11,15 case end time,6 surgical extent estimated as number of levels or extent of resection,13,16 and volume of crystalloid and blood transfusion.11,16,17 In a previous study, we found that attending handoff was a significant factor in the decision to delay extubation for prolonged spine surgery in bivariate analysis but did not reach significance in the multivariate model.6 We hypothesized that in a larger and more general population, we would be able to investigate whether there was an effect of attending handoff on the decision to delay extubation. In addition, we wanted to investigate whether the timing of the case (ending in daytime versus overnight hours) and attending handoff were associated with delayed extubation.

In this study, we retrospectively reviewed all general anesthetics over 5 years at our institution (excluding cases that routinely remain intubated) that involved the placement of an endotracheal tube. We collected variables hypothesized to identify preoperative and intraoperative factors that affect the decision of the anesthesiologist to delay extubation at the end of the surgery. We hypothesized that, among other factors, attending handoff would be an independent factor that would contribute to delayed extubation after general anesthesia.


The study was performed with the approval from the IRB at Columbia University Medical Center. We retrospectively reviewed 5 years of electronic anesthesia information management system (AIMS; CompuRecord, Philips Medical Systems, Andover, MA) records, from August 2008 to August 2013. We identified 37,824 adult patients who underwent general anesthesia for surgery, defined as having “General Anesthesia” as their anesthetic type, both alone, or in combination with an epidural or a regional block. Only cases with an endotracheal tube were included; cases were excluded if they had a laryngeal mask airway placed for their anesthetic. Surgeries in which a patient is not extubated routinely at the end of a case were excluded: tracheostomy, cardiac surgery, liver transplant surgery, and lung transplant surgery. Cases also were excluded if the patient arrived to the operating room intubated or had a preexisting tracheostomy. Patients also were excluded if they had an intraoperative death. The “intubation” time marker in the AIMS was used to identify if a patient was intubated before coming to the operating room (Fig. 1).

Figure 1:
Diagram showing the inclusion and exclusion of study participants. ETT = endotracheal tube; ICU = intensive care unit; LMA = laryngeal mask airway; MAC = monitored anesthesia care; NYPH = New York Presbyterian Hospital; OR = operating room.

Two pieces of data from the AIMS were used to identify and verify whether a patient was extubated at the end of the case or remained intubated: the “extubation” time marker and the description of the airway on arrival to the recovery room or intensive care unit. Computer-based hospital records (WebCIS, Department of Biomedical Informatics, Columbia University Medical Informatics Services, New York-Presbyterian Healthcare, NY; Durham, NC; and Eclipsys Sunrise Enterprise, Allscripts-Misys Healthcare Solutions, Inc., Nashville, TN) were used to clarify discrepancies in the AIMS data on postoperative extubation.

Preoperative risk factors for delayed extubation included in the analysis were age, gender, ASA physical status (categorized as ASA physical status I–II, ASA physical status III, and ASA physical status IV–V), and obesity (body mass index >30). Intraoperative risk factors that were included were difficult intubation (defined as planned fiberoptic intubation or as an unanticipated difficult intubation as described in the AIMS), total case duration (the time from the start of anesthetic care to the end of anesthetic care), estimated blood loss (EBL), total crystalloid administration, total colloid administration, and total blood administration (packed red blood cells [PRBCs]), attending handoff (if at least 1 handoff occurred), day of surgery (weekend versus weekday), emergency surgery, patient position during surgery (prone versus other position), and case end time of day (7:00 AM–3:00 PM, which captures normal work hours at our institution, versus 3:00 PM–7:00 AM, which captures call team work hours).

Statistical Analysis

Delayed extubations were characterized by patient demographics, preoperative variables, and intraoperative variables. Groups were compared using the t test for continuous variables and the χ2 test for categorical variables. We assessed the linearity of continuous variables by arranging the variables into deciles and then adding decile as a categorical variable in the relative risk regression. The parameter estimates (y-axis) were then plotted against the decile (x-axis), and the linearity of the plot was assessed. Age and case duration were found to be approximately linear and thus entered as continuous variables. Total volume of crystalloid, colloid, PRBC, and EBL were not found to be linear, and therefore, they are broken down into several cutpoints by their distribution to form categorical variables. Case end time was categorized as a categorical variable divided into 2 time intervals: from 7:00 AM to 3:00 PM and from 3:00 PM to 7:00 AM. The handoffs at our institution start taking place around 3:30–4:00 PM, so we divided the cases into “day cases” and “evening cases” based on the end times.

All preoperative and intraoperative variables identified as significantly associated with the outcome in bivariate analysis were entered into a stepwise regression model using a modified Poisson model with robust error variance,18 with a significance of 0.12 required to enter the model and a significance of 0.25 to remain in the model. Risk ratios and 95% confidence intervals (CIs) were estimated.

We hypothesized that the effect of attending handoff would differ during normal work hours versus the evening when cases are handed off to the call team. Therefore, we included an interaction term between attending handoff and case end time. The presence of a significant interaction indicates that the effect of one predictor variable on the response variable was different at different values of the other predictor variable. Therefore, in our model, we allowed the effect of attending handoff on extubation to vary between cases ending during daytime hours versus evening hours. Statistical analysis was performed using JMP 7 and SAS software version 9.4 (SAS Institute Inc., Cary, NC) or Excel (Microsoft Corp., Redmond, WA).


In the analysis, we included 37,824 patients. A total of 5.4% of patients (n = 2033) were not extubated in the operating room after the completion of their surgery. Table 1 shows the surgical procedure classes of the patients included in the analysis. The majority of patients were from general surgery (not otherwise specified). The remainder included neurosurgery, gynecological surgery, ear-nose-throat, noncardiac liver-thoracic transplantation, genitourinary surgery, thoracic surgery, vascular surgery, orthopedic surgery, plastic surgery, oral surgery, and surgical oncology. The percentage of patients in which extubation was delayed ranged from 1.4% in plastic surgery to 10.8% in vascular surgery (Table 1).

Table 1:
Surgical Procedure Classes of Patients Undergoing General Anesthesia

Continuous variables included age and case duration. Categorical variables included attending handoff, ASA physical status, gender, obesity, emergency surgery, weekend surgery, difficult intubation, patient position, total volume of crystalloid, total volume of colloid, total volume of PRBC, and EBL.

Bivariate analysis identified 15 factors (P < 0.05) associated with the decision to delay extubation: age, ASA physical status, gender, obesity, emergency surgery, weekend surgery, difficult intubation, patient position during surgery, case duration, case end time of day, attending handoff, EBL, crystalloid administration, colloid administration, and blood administration (Table 2).

Table 2:
Demographics of Patients Undergoing General Anesthesia

Because attending handoff was a main variable of interest, we showed the factors as related to the presence or absence of attending handoff (Table 3). Of the factors identified in the bivariate analysis, 13 factors remained significant in the relative risk regression analysis including age, case duration, EBL, crystalloid administration, colloid administration, blood administration, ASA physical status, obesity, emergency surgery, weekend surgery, difficult intubation, case end time of day, and attending handoff.

Table 3:
Demographics of Patients Undergoing Attending Handoff

Our main variable of interest was attending handoff, and in adjusted analysis, we found that the handoff increased the risk of delayed extubation (adjusted risk ratio [aRR], 1.14; 95% CI, 1.03–1.25). Clinically, we hypothesized that the effect of this variable was different during normal work hours compared with evening hours when the call team has taken over all cases started during the day. Therefore, we constructed an interaction term between attending handoff and case end time of day, and the interaction term was statistically significant (P = 0.003). In day cases, there was a greater risk of delayed extubation with attending handoff (versus no attending handoff; aRR, 1.62; 95% CI, 1.29–2.04) than in evening cases (aRR, 1.07; 95% CI, 0.97–1.19; Table 4).

Table 4:
Final Poisson Regression Analysis Displaying Relative Risk Ratio and Confidence Intervals
Table 5:
Interaction of Attending Handoff and the Other Model Variables
Table 6:
Number of Attending Anesthesiologists and Delayed Extubation

We evaluated the interaction of attending handoff and the 12 other variables in the model using a separate regression model for each variable. A Bonferroni corrected P value (0.05/12 = 0.004) was used to determine statistical significance. The interaction with attending handoff was significant for ASA, emergency class, and procedure class (Table 5). We further analyzed these interactions using adjusted analyses (Supplemental Digital Content 1, Supplemental Tables C–E, The effect of handoff was greatest for ASA physical status I–II patients (aRR, 1.84; 95% CI, 1.54–2.20). For ASA physical status IV–V patients, the risk ratio was <1 (aRR, 0.66; 95% CI, 0.54–0.82). Likewise, handoff was associated with remaining intubated in nonemergency cases (aRR, 1.38; 95% CI, 1.22–1.56) but not in emergency cases (aRR, 0.86; 95% CI, 0.73–1.01). In adjusted analyses, 5 classes of procedures were identified in which handoff was associated with remaining intubated: ear-nose-throat surgery (aRR, 1.56; 95% CI, 1.03–2.37), genitourinary surgery (aRR, 1.77; 95% CI, 1.20–2.61), gynecological surgery (aRR, 1.52; 95% CI, 1.04–2.24), neurosurgery (aRR, 1.54; 95% CI, 1.27–1.87), and orthopedic surgery (aRR, 1.65; 95% CI, 1.14–2.39). We found that the number of attendings involved per case also increased progressively the risk of being intubated at the end of the case (Table 6).


We present an observational, retrospective study that found that attending handoff is associated with the decision to delay extubation at the end of a general surgery. In particular, there was a significant association between the attending handoff and the delayed extubation in cases ending in the daytime hours (7:00 AM–3:00 PM), but there was no association in cases ending in the evening hours (4:00 PM–7:00 AM).

Handoffs, when case supervision is transferred from one physician to another, are common in various medical and surgical specialties. Surgical cases requiring long operative and anesthetic times often involve such transitions. Arbous et al.19 found that an attending handoff was associated with an increased risk of mortality after surgery. Wright et al.20 correlated adverse events that occurred intraoperatively with later timing of cases, when an attending handoff would be likely to occur. Saager et al.21 found that handoffs increased the risk for a composite score of in-hospital mortality and morbidity. Although a handoff is a good time to reflect on the overall course of the case, there may be a lack of communication between participants, and an observational study showed common predictors of prolonged intubation, including positioning, EBL, and intraoperative fluids, are not reported in handoffs from anesthesiologists to postanesthesiology care personnel.22 Part of the reluctance of an anesthesiologist to extubate an unfamiliar airway may play a factor in the decision to delay patient extubation. We found that the number of attendings involved per case, with more handoffs and presumably less direct communication about the patient’s baseline characteristics and airway, also progressively increased the risk of being intubated at the end of the case.

In our study, although there was a significant relationship between handoffs and the decision to delay extubation, it was counterintuitive to the authors that the risk of delayed extubation with attending handoff was significant only in cases ending in the daytime but not in cases ending in the evening hours. Similarly, the effect of a handoff seems to have more of an impact on patients who are healthier (as estimated by ASA classification) and nonemergency cases. This may reflect the fact that cases that are ending in the evening have a higher ASA physical status and are emergency cases that have more factors playing a role. The effect of an attending switch is diminished when the patient likely had surgery that is of a longer duration, with more blood loss, and with more fluid replacement.

It might be believed that the start time of the cases might explain these findings. For example, cases that start overnight and end on the next day may be the cases that are likely to remain intubated. However, we found that although the start time of the cases explained some of these findings, the apparent effect of the attending handoff was greater for cases that started and ended in the daytime than for cases that started overnight but finished in the daytime (Supplemental Digital Content 1, Supplemental Tables A and B, There seem to be factors in these daytime cases that are handed over that are associated with delayed extubation. Further studies are needed to investigate the effect of the handoff on extubation in these situations.

Five classes of procedures where handoffs were associated with remaining intubated included ear-nose-throat, genitourinary, gynecology, neurosurgery, and orthopedic surgery. These classes of surgeries at our institution typically involve significant fluid shifts, often are done in the prone or Trendelenburg position, or involve the airway itself. These factors may have a large effect on the airway by the time the case concludes. For example, an anesthesiologist deciding to extubate a patient with significant edema after a multilevel spine surgery may be more hesitant if they were unfamiliar with how easy the airway was before the surgery.

Another factor that we considered was that specific anesthesiologists and surgeons may be more inclined to keep their patients intubated, but a sensitivity analysis did not find evidence for this type of relationship (Supplemental Digital Content 1, Supplemental Tables F and G,

Prolonged intubation can be associated with complications.5,8 However, it is unclear whether the risks involved in leaving a patient intubated when a case end time is very late are detrimental to the medical course of a patient, particularly after a prolonged surgery with large fluid shifts in a patient who is medically complicated. Decreased resource availability in periods of time that are outside conventional work hours may affect outcome.23 The potential difference in care at night or on weekends may influence an anesthesiologist to decide to extubate a patient under more optimal conditions during daytime hours.

This study has several limitations. Our study takes place in a single institution that is a large academic medical center that has a 4-year training program for anesthesiology residents and, thus, reflects only 1 segment of anesthesia practice. The decision to extubate is a multifactorial decision on the part of the anesthesiologist and is not driven by a protocol. Further, delayed extubation may not be detrimental in our patient population. Many of the factors that correlate with extubation failure, reintubation after extubation, are the same factors that correlate with decision not to extubate.24,25 This was a retrospective study, and other unmeasured confounders may have affected the decision to extubate. In particular, the specific procedure instead of the broader procedure class would reduce bias in regression models of perioperative outcomes.26 Our database had reliable procedure codes for 28% of the sample, and in this subset, we found that using more specific procedure categories actually increased the magnitude of effect seen with the attending handoff on delayed extubation (Supplemental Digital Content 1, Supplemental Table H,, so it is not likely that the addition of procedure codes for the entire sample would nullify the effects that we observed. In addition, the attending handoff was not a randomized event so that there might be selection bias in the cases that had a handoff compared with those that did not have a handoff. However, when we used propensity score methods to account for this effect,27 we found no meaningful differences in our results (Supplemental Digital Content 1, Supplemental Table I, Finally, the database does not have any information of outcomes and if delayed extubation affected outcomes. Future studies could focus on outcomes of delayed extubation in matched cohorts.

In conclusion, attending handoff was correlated with the decision to delay extubation at the end of a general anesthetic. Guidelines for safe extubations in clinical settings3 must consider the effect of attending handoff on the delay of extubation. Outcomes of this effect need to be investigated and considered in postoperative resource planning.


Name: Zirka H. Anastasian, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Zirka H. Anastasian has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Minjae Kim, MD, MS.

Contribution: This author helped design the study, analyze the data, and write the manuscript.

Attestation: Minjae Kim has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Eric J. Heyer, MD, PhD.

Contribution: This author helped design the study and write the manuscript.

Attestation: Eric J. Heyer has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Shuang Wang, PhD.

Contribution: This author helped analyze the data.

Attestation: Shuang Wang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Mitchell F. Berman, MD, MPH.

Contribution: This author helped design the study, analyze the data, and write the manuscript.

Attestation: Mitchell F. Berman has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

This manuscript was handled by: Sorin J. Brull, MD.


1. Cook TM, Scott S, Mihai R. Litigation related to airway and respiratory complications of anaesthesia: an analysis of claims against the NHS in England 1995-2007. Anaesthesia. 2010;65:556–63
2. Cook TM, Woodall N, Frerk CFourth National Audit Project. . Major complications of airway management in the UK: results of the Fourth National Audit Project of the Royal College of Anaesthetists and the Difficult Airway Society. Part 1: anaesthesia. Br J Anaesth. 2011;106:617–31
3. Popat M, Mitchell V, Dravid R, Patel A, Swampillai C, Higgs ADifficult Airway Society Extubation Guidelines Group. . Difficult Airway Society Guidelines for the management of tracheal extubation. Anaesthesia. 2012;67:318–40
4. Apfelbaum JL, Hagberg CA, Caplan RA, Blitt CD, Connis RT, Nickinovich DG, Hagberg CA, Caplan RA, Benumof JL, Berry FA, Blitt CD, Bode RH, Cheney FW, Connis RT, Guidry OF, Nickinovich DG, Ovassapian AAmerican Society of Anesthesiologists Task Force on Management of the Difficult Airway. . Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology. 2013;118:251–70
5. Fagon JY, Chastre J, Domart Y, Trouillet JL, Pierre J, Darne C, Gibert C. Nosocomial pneumonia in patients receiving continuous mechanical ventilation. Prospective analysis of 52 episodes with use of a protected specimen brush and quantitative culture techniques. Am Rev Respir Dis. 1989;139:877–84
6. Anastasian ZH, Gaudet JG, Levitt LC, Mergeche JL, Heyer EJ, Berman MF. Factors that correlate with the decision to delay extubation after multilevel prone spine surgery. J Neurosurg Anesthesiol. 2014;26:167–71
7. Mathias DB, Wedley JR. The effects of cuffed endotracheal tubes on the tracheal wall. Br J Anaesth. 1974;46:849–52
8. Barker J, Martino R, Reichardt B, Hickey EJ, Ralph-Edwards A. Incidence and impact of dysphagia in patients receiving prolonged endotracheal intubation after cardiac surgery. Can J Surg. 2009;52:119–24
9. Cislaghi F, Condemi AM, Corona A. Predictors of prolonged mechanical ventilation in a cohort of 5123 cardiac surgical patients. Eur J Anaesthesiol. 2009;26:396–403
10. Patel N, Bagan B, Vadera S, Maltenfort MG, Deutsch H, Vaccaro AR, Harrop J, Sharan A, Ratliff JK. Obesity and spine surgery: relation to perioperative complications. J Neurosurg Spine. 2007;6:291–7
11. Epstein NE, Hollingsworth R, Nardi D, Singer J. Can airway complications following multilevel anterior cervical surgery be avoided? J Neurosurg. 2001;94:185–8
12. Udink ten Cate FE, van Royen BJ, van Heerde M, Roerdink D, Plötz FB. Incidence and risk factors of prolonged mechanical ventilation in neuromuscular scoliosis surgery. J Pediatr Orthop B. 2008;17:203–6
13. Cywinski JB, Xu M, Sessler DI, Mason D, Koch CG. Predictors of prolonged postoperative endotracheal intubation in patients undergoing thoracotomy for lung resection. J Cardiothorac Vasc Anesth. 2009;23:766–9
14. Saleh HZ, Shaw M, Al-Rawi O, Yates J, Pullan DM, Chalmers JA, Fabri BM. Outcomes and predictors of prolonged ventilation in patients undergoing elective coronary surgery. Interact Cardiovasc Thorac Surg. 2012;15:51–6
15. Cai YH, Zeng HY, Shi ZH, Shen J, Lei YN, Chen BY, Zhou JX. Factors influencing delayed extubation after infratentorial craniotomy for tumour resection: a prospective cohort study of 800 patients in a Chinese neurosurgical centre. J Int Med Res. 2013;41:208–17
16. Kwon B, Yoo JU, Furey CG, Rowbottom J, Emery SE. Risk factors for delayed extubation after single-stage, multi-level anterior cervical decompression and posterior fusion. J Spinal Disord Tech. 2006;19:389–93
17. Sagi HC, Beutler W, Carroll E, Connolly PJ. Airway complications associated with surgery on the anterior cervical spine. Spine. 2002;27:949–53
18. Spiegelman D, Hertzmark E. Easy SAS calculations for risk or prevalence ratios and differences. Am J Epidemiol. 2005;162:199–200
19. Arbous MS, Meursing AE, van Kleef JW, de Lange JJ, Spoormans HH, Touw P, Werner FM, Grobbee DE. Impact of anesthesia management characteristics on severe morbidity and mortality. Anesthesiology. 2005;102:257–68
20. Wright MC, Phillips-Bute B, Mark JB, Stafford-Smith M, Grichnik KP, Andregg BC, Taekman JM. Time of day effects on the incidence of anesthetic adverse events. Qual Saf Health Care. 2006;15:258–63
21. Saager L, Hesler BD, You J, Turan A, Mascha EJ, Sessler DI, Kurz A. Intraoperative transitions of anesthesia care and postoperative adverse outcomes. Anesthesiology. 2014;121:695–706
22. Siddiqui N, Arzola C, Iqbal M, Sritharan K, Guerina L, Chung F, Friedman Z. Deficits in information transfer between anaesthesiologist and postanaesthesia care unit staff: an analysis of patient handover. Eur J Anaesthesiol. 2012;29:438–45
23. Eschiti V, Hamilton P. Off-peak nurse staffing: critical-care nurses speak. Dimens Crit Care Nurs. 2011;30:62–9
24. Cavallone LF, Vannucci A. Review article: extubation of the difficult airway and extubation failure. Anesth Analg. 2013;116:368–83
25. Hua M, Brady JE, Li G. A scoring system to predict unplanned intubation in patients having undergone major surgical procedures. Anesth Analg. 2012;115:88–94
26. Dexter F, Dexter EU, Ledolter J. Statistical grand rounds: importance of appropriately modeling procedure and duration in logistic regression studies of perioperative morbidity and mortality. Anesth Analg. 2011;113:1197–201
27. D’Agostino RB Jr. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med. 1998;17:2265–81
© 2016 International Anesthesia Research Society