Complete Anesthesia Handover in Cardiac Surgery Leads to Worse Outcomes

When another anesthesiologist took over the case, there was a higher risk of death at both 30 days and 1 year.

Complete Anesthesia Handover in Cardiac Surgery Leads to Worse Outcomes

Completely handing a patient off from one anesthesiologist to another during cardiac surgery is associated with significantly higher risks of mortality, as well as longer stays in the ICU and hospital, a new study has shown.

Researchers suspect that even though the incoming anesthesiologist is provided with the patient’s complete history, including critical patient- and procedure-related information, some “intangible” details might be lost in the handoff, contributing to worse outcomes. 

“People always strive to do well, and anesthesiology is a bit of a detail-oriented specialty where we strive not to skip anything important, but even when you do that, it’s almost like a ‘gestalt’ or a ‘Spidey sense’ or a ‘gut feeling’ about the patient that is lost during the process,” lead investigator Louise Sun, MD, SM (University of Ottawa Heart Institute, Canada), told TCTMD. “You might forget to convey some piece of information—that happens to the best of us—but it’s like some intuitive sense [of patient care] that is lost during the handover.”

For example, some patients might respond to an odd dose of medication, something that can’t really be explained medically. “You might reach for something that works better for you because you’ve spent so many hours with the patient already,” said Sun. “Another person would have to go in and figure it all out again. If the handover happens during a critical time during the surgery, that’s not a good time to figure things out.”

Cardiothoracic surgeon Todd Rosengart, MD (Baylor College of Medicine, Houston, TX), who wasn’t involved in the study, said the transfer of care has become more common since many hospitals have tried to limit the maximum number of hours physicians can work. With these institutional changes in place, patient handover is more common, and for that reason, communication remains critical.

“Like all good things, [limiting work hours] comes with potential downsides,” Rosengart told TCTMD. “While physician fatigue can be a potential risk, if you are going to hand off a patient, it has to be done properly. It’s well known in medicine, and in other high-quality, high-reliability systems—airplane cockpits are always the classic example—that communication is essential.”

Like Sun, Rosengart said handoffs should only take place during noncritical moments. For that to happen, the surgeon and anesthesiologist also need to be in communication about when handoff is anticipated or planned so there is agreement about when it’s safest to proceed.

The new study was published online February 11, 2022, in JAMA Network Open

Balancing Physician Fatigue With Patient Care

In medicine, gone are the days of working 36 hours straight, said Sun. Handovers are now part of a hospital’s organizational structure, with policies in place to improve efficiency and to limit duty hours in order to reduce physician fatigue, prevent burnout, and provide doctors with more-predictable working hours. Previous studies have linked anesthesia handovers with worse perioperative outcomes in the setting of noncardiac surgery, but there are limited data in the area of cardiac surgery.    

Using data from ICES, an independent nonprofit research institute that collects healthcare data, the researchers identified 102,156 patients who underwent cardiac surgery between 2008 and 2019. In this study, researchers only looked at outcomes of cases that involved the complete handover of anesthesia care (the case was finished by the replacement anesthesiologist). To account for differences in characteristics between patients who were versus were not exposed to anesthesia handover, the researchers used inverse probability treatment weighting (IPTW) based on propensity scores to estimate the effect of treatment handover.

Overall, 1.9% of surgeries involved the complete transfer of anesthesia care. The rate of complete intraoperative handover increased over time, from 0.7% of cases in 2008 to 2.9% in 2019. The majority of the procedures were performed in teaching hospitals (71.3%) and nearly 20% of the anesthesiologists were women.

Among patients undergoing CABG, valve surgery, or aortic procedures, complete intraoperative handoff of anesthesia care was associated with a higher risk of death at 30 days (HR 1.89; 95% CI 1.41-2.54) and 1 year (HR 1.66; 95% CI 1.31-2.12), as well as longer stays in the ICU (RR 1.43; 95% CI 1.22-1.68) and hospital (RR 1.17; 95% CI 1.06-1.28).

The association between handover and clinical outcomes was stronger in patients who underwent complex surgery. In terms of timing, handover was associated with increased perioperative and 1-year mortality, as well as hospital length of stay, only in cases started during regular workdays. Additionally, handover of patient care during or after cardiopulmonary bypass had a stronger impact on 30-day mortality and was associated with worse patient-centered outcomes (a composite that includes severe stroke, chronic ventilator dependence, new-onset heart failure, new-onset dialysis, and long-term care admission).  

Anesthesia handover is often required for very long surgeries, which can be a reflection of a case gone wrong, said Sun. The researchers did adjust for case complexity and urgency of surgery, among other variables, but the possibility of residual confounding must be considered when interpreting the results.

Still, the researchers say there are things that can be done to minimize the risks associated with intraoperative anesthesia handovers.

“We should consider finishing our own very high-risk cases if the finish times are projected to be reasonable,” said Sun. “I don’t want anybody finishing cases going on past midnight if they have to be back at 6 am the next day. That would be very dangerous. There is a physical limit to what we think people can do, but our group recently implemented a soft policy where we’re encouraged to finish our own cases expected to finish before 8 PM. That makes it a 13- or 14-hour day, maximum, and ensures more continuity of care.”

Another way to reduce risks is to avoid handing over care during a critical time of surgery, such as when the patient is highly unstable or in periods immediately before and after the patient comes off cardiopulmonary bypass, said Sun. The use of standardized handover tools based in electronic medical records would also be helpful, she said.

Communication is Critical

In an editorial accompanying the study, Laurent Glance, MD (University of Rochester School of Medicine, NY), and colleagues write that the 89% higher risk of death at 30 days with complete anesthesia handover translates into one additional death for every 52 cases. They point out that in the observational study, making a causal connection between intraoperative handoffs and mortality is fraught because of the risk of unmeasured confounding variables, but add that the IPTW analysis helps mitigate some, but not all, of those concerns.  

The increased risk observed in this study is higher than in previously published studies of patients undergoing major surgery, including cardiac surgery. For example, one study from 2020 found that anesthesia handover was associated with only a 15% higher risk of death at 30 days. In that study, the rate of complete anesthesia handover was 8.5%.

“It is difficult to speculate what could account for the large difference in risk with the possible exception that because handovers are more routine in US practice, they may be less likely to lead to adverse outcomes,” according to Glance and colleagues.

Regardless of the effect size, the data do show that complete anesthesia handover is associated with worse outcomes. Like Sun, the editorialists recommend avoiding handovers if the case will finish in a reasonable time, even if the on-call anesthesiologist is available. Anesthesiologists should also consider completing complex or uncommon cases, and no handover should occur immediately before or during stages of cases with high potential for hemodynamic instability.

Rosengart said “readback” is important when speaking with colleagues, such as the perfusionist or anesthesiologist, in the operating room so that everyone is on the same page with the information conveyed. In terms of the present study, he said he isn’t completely surprised that outcomes are worse when patients are handed off to another anesthesiologist, and this is possibly related to inadequate communication when the exchange occurs.

Michael O’Riordan is the Associate Managing Editor for TCTMD and a Senior Journalist. He completed his undergraduate degrees at Queen’s…

Read Full Bio
  • Sun, Glance, and Rosengart report no relevant conflicts of interest.



Nabil Salameh

2 years ago
It amazes me how this piece could ever be considered for publishing anywhere let alone this journal/ magazine. Not only does the author acknowledge that there are confounders and allude to the most important confounder— NOTHING is pursued to address the confounder. This article is misleading and embarrassing for not only anesthesiologists but the whole scientific community. If a case is handed off it means very likely one or all of 3 things. 1) The case is complex— leads to longer/ second/ third pump runs which in turn leads to increased morbidity/ mortality 2) The surgeon made errors or the case was more complex than the surgeon anticipated— leads to again longer/ second/ third pump runs which in turn leads to increased morbidity/mortality 3) The surgeon is fatigued, complicating the above and leading to the above outcomes. If the author had any scientific integrity, they might control for this obvious confounder by asking anesthesiologists to “completely handoff” during the first case of the day. Or “completely handoff” during cases that were stratified to be less complex. Or compare “complex” cases/ or late day longer cases that were not “completely handed off” vs “completely handed off.” Not addressing any of these issue shows a lack of understanding of cardiac surgeries and cardiac anesthesiology as a whole which should disqualify the author from any comment on the subject matter. If the previous is not the case, then this is simply a lazy take at a study with some sort of agenda. Either way this collection of words is not scientific and should not fall into the category of “study.” It is an embarrassment to your publication.