ISSN NUMBER: 1938-7172
Issue 12.14

Michael A. Fiedler, PhD, CRNA

Contributing Editors:
Mary A Golinski, PhD, CRNA
Dennis Spence, PhD, CRNA

Assistant Editor
Jessica Floyd, BS

A Publication of Lifelong Learning, LLC © Copyright 2019

New health information becomes available constantly. While we strive to provide accurate information, factual and typographical errors may occur. The authors, editors, publisher, and Lifelong Learning, LLC is/are not responsible for any errors or omissions in the information presented. We endeavor to provide accurate information helpful in your clinical practice. Remember, though, that there is a lot of information out there and we are only presenting some of it here. Also, the comments of contributors represent their personal views, colored by their knowledge, understanding, experience, and judgment which may differ from yours. Their comments are written without knowing details of the clinical situation in which you may apply the information. In the end, your clinical decisions should be based upon your best judgment for each specific patient situation. We do not accept responsibility for clinical decisions or outcomes.

Table of Contents

Agitation in adults in the post-anaesthesia care unit after general anaesthesia

Triple-low alerts do not reduce mortality: a real-time randomized trial

Neuraxial anaesthesia techniques and postoperative outcomes among joint arthroplasty patients: is spinal anaesthesia the best option?

None of the editors or contributors have any real or potential conflicts of interest to disclose.
This program has been prior approved by the American Association of Nurse Anesthetists for 20 Class A CE credits; Code Number 1035464; Expiration Date 10/31/2020.

Agitation in adults in the post-anaesthesia care unit after general anaesthesia

Br J Anaesth 2018;121:1052-1058

DOI: 10.1016/j.bja.2018.07.017

Fields A, Huang J, Schroeder D, Sprung J, Weingarten T



Purpose   The purpose of this study was to assess the incidence of emergence agitation in adult patients and search for associations between agitation and prospective risk factors. A secondary objective was to describe the postoperative course of adults who experienced agitation in the PACU.


Background   After a general anesthetic, recovery of baseline cognition is a key criterion upon which PACU discharge depends. Agitation can result in delayed PACU discharge and injury to the patient and/or the healthcare staff. When agitation progresses to delirium, morbidity and increased length of hospitalization are common. Previous studies have reported the incidence of PACU agitation at between 5% and 19%. Many things patients experience in the PACU may cause or contribute to agitation, for example, pain, having an ETT in place while awake, disorientation, or fear. Notably, hypercarbia from hypoventilation is known to cause agitation.


Methodology   This was a retrospective study in which adult patients who experienced agitation in the PACU were compared with patients who had not experienced agitation that were similar in age, gender, and surgical procedure. The institution’s medical record database was searched for adult patients who had surgery with general anesthesia over a 6+ year period. “Emergence Agitation” was defined as a Richmond Agitation–Sedation Scale (RASS) score of 3+ or 4+ or the administration of haloperidol in the PACU. Agitation was then statistically correlated with a host of other factors such as demographic data (e.g. obesity) and interventions (e.g. remaining intubated in the PACU).


Result   The study examined the records of 207,569 adult patients. Of these, 0.25% (510 patients) experienced emergence agitation; or 2.5 patients per 1,000 PACU admissions. Agitation was defined by their RASS score in 88% of patients and administration of haloperidol in 31%; some met both criteria. The 510 agitated patients infrequently caused injury:

  • 6 injured themselves
  • 6 removed their IV
  • 3 removed their NG tube
  • 15 assaulted staff by striking, kicking, grabbing, or spitting

Agitated patients were more likely to have a history of mental health problems or preexisting cognitive impairment than patients without agitation. Preoperative conditions associated with agitation in the PACU included (all statistically significant, most likely at top of list):

  • substance abuse
  • cognitive impairment
  • obesity
  • psychiatric diagnosis

Intraoperative administration of either droperidol or ketamine were associated with a decreased likelihood of PACU agitation (droperidol odds ratio 0.4, P=0.002; ketamine odds ratio 0.6, P=0.001).


Postoperative conditions associated with agitation in the PACU included (all statistically significant, most likely at top of list):

  • ETT in place
  • Foley catheter
  • NG tube
  • chest tube

Patients who experienced agitation stayed in the PACU an average of 51% longer than patients who did not (108 min. vs. 71 min., P<0.001). Agitated patients were also twice as likely to experience PONV (P<0.001). Finally patients with agitation were about 2.5 times more likely to have delirium later in their hospital stay (P<0.001).


Conclusion Patients who required an ETT, NG tube, or chest tube; and those with substance abuse or mental health problems preoperatively were most likely to experience agitation in the PACU. The lower incidence of agitation in this study compared to previous research may have been due to differing definitions of agitation from study to study. This study probably underestimated the true incidence of agitation due to a conservative definition. The association between PACU agitation and later delirium suggests that preemptive treatment of delirium might be helpful in these patients.




Preventing and quickly treating agitation is paramount for patient safety, staff safety, and preservation of the surgical outcome.


Clinicians often speak of / think about agitation and delirium as pretty much the same thing. Technically agitation is “anxiety or nervous excitement” while delirium is “a disturbed state of mind … characterized by restlessness, illusions, and incoherent thought and speech.” The investigators defined agitation by the Richmond Agitation - Sedation Scale (RASS) score. Positive RASS values describe more and more agitation. Negative RASS values describe deeper sedation. For example, +4 is a combative or violent patient who poses a danger to themselves and/or staff. We should probably all be more familiar with the RASS scoring system and make sure PACU staff use it. I’ve included a link in the notes. Agitation can progress to delirium. In fact, there is a delirium scoring tool that starts with the RASS score.


A very important tenant of research and statistical analysis is that “association does not show causation.” The fact that two things are correlated in no way means that one caused the other. This is important to remember while reading this study. After identifying patients who were agitated in the PACU, the study basically did a screening test to find anything that was associated with agitated patients. Because of this, the investigators found some correlations that were almost certainly coincidence. I can’t imagine a mechanism by which obesity or receiving a slightly larger volume of crystalloid IV fluid could cause agitation.


So, what can we learn from this study? First, we have one more reason to do our best to make sure endotracheal tubes can safely come out before patients are awake. Intubated patients were over 16 times more likely to experience agitation. Second, we can use preoperative factors to screen patients for risk of PACU agitation: substance abuse history, preexisting cognitive impairment, and a psychiatric history. When we identify at risk patients we may consider preemptive steps to prevent agitation. Anesthetists younger than some of us (ahem …) may not know that haloperidol is a pretty good antiemetic. (Haldol is related to droperidol.) One option for such preemptive treatment of both agitation and PONV would be a small dose of haloperidol. A small dose of dexmedetomidine at the end of the case has also been shown to effectively prevent PACU agitation (see issues 7.2, 7.6, & 12.15 for more on dexmedetomidine and emergence agitation). Small doses of either drug are unlikely to prolong the PACU stay by much but agitation will most certainly prolong it.


We have protocols for PONV, multimodal analgesia, difficult airway, and others. We would benefit from risk stratification during the preanesthetic evaluation and an agitation protocol. In a tightly staffed PACU, when recovery nurses must all attend to an agitated patient, the other patients are not being attended to. Lastly, when moving patients from the PACU to the ward or unit, if report doesn’t already include the fact that a patient was agitated in the PACU, it should. PACU agitation is a risk factor for later development of delirium. Detecting delirium early will help prevent patient injury.

Michael A. Fiedler, PhD, CRNA

Richmond Agitation - Sedation Scale (RASS) score calculator is available at:

© Copyright 2019 Anesthesia Abstracts · Volume 12 Number 14, January 22, 2019

Triple-low alerts do not reduce mortality: a real-time randomized trial

Anesthesiology 2019;130:72-82

DOI: 10.1097/ALN.0000000000002480

Sessler DI, Turan A, Stapelfeldt WH, Mascha EJ, Yang D, Farag E, Cywinski J, Vlah C, Kopyeva T, Keebler AL, Perilla M, Ramchandran M, Drahuschak S, Kaple K, Kurz A



Purpose   The purpose of this study was to determine the effectiveness of real-time, automated, triple-low alerts for decreasing 90-day mortality in high-risk patients. A triple-low event is a combination of mean arterial blood pressure (MAP) < 75 mm Hg, a Bispectral Index < 45, and minimum alveolar concentration (MAC) < 0.8).


Background   Intraoperative hypotension, low BIS levels, and low MAC have each been associated with increased mortality. A "triple low event,” all three values being low simultaneously, has been suggested to be a predictor of mortality in high-risk patients. Increased mortality associated with a triple-low event may be an underlying reflection of the fragility or illness of the patient, such that they cannot tolerate anesthesia and surgery. About 20% of high-risk patients will experience a triple-low event.


Decision-support systems with smart alarms for triple-low events that are built into anesthesia workstations have the potential to help decrease mortality because they may trigger earlier intervention compared to no alerts. The investigators in this study hypothesized that providing automated, triple-low event alerts would result in reduced 90-day mortality in high-risk patients. Secondarily, they examined the effects on 30-day and one year mortality, and the association between a helpful response to a triple-low event and mortality rates.


Methodology   This was a prospective, randomized, controlled trial examining the effect of real-time, automated alerts to triple-low events on mortality in high-risk patients. The study took place between June 2010 and October 2016. Patients were randomized in real-time the first-time a triple-low event was recorded in the electronic anesthesia workstation. Investigators were blinded to group assignment. Anesthesia professionals with patients assigned to the alert group received an automated alert via the anesthesia workstations decision-support system. A message via pager was also sent to the clinician and attending anesthesiologist. An alert appeared on the anesthesia workstation display stating, "a triple-low (MAP, MAC, and BIS) condition has been detected. Consider hemodynamic support". If the triple-low event remained uncorrected after 10 minutes, an additional alert was displayed and sent via pager.


Investigators educated all anesthesia professionals at the facility about the study and triple-low event decision-support system. There was no requirement to follow a specific treatment protocol. Anesthesia professionals were free to act on the alert, ignore the alert, or consider the provided information without acting. Thus the protocol reflected real-world conditions.


For the purposes of analysis, helpful responses to a triple-low event were defined as vasopressor use within 5 minutes of the alert and/or a 20% decrease in end-tidal volatile anesthetic concentration. Investigators recorded and compared the time from the first alert to any of the following conditions being met:

  • MAP > 75 mm Hg
  • BIS >45
  • MAC >0.8)
  • 20% increase in MAP within 5 and 15 minutes of the alert

The relationship between a helpful response, irrespective of group assignment, and each outcome was examined. Statistical and power analyses were appropriate.


Result   Out of 36,670 qualifying patients, there were N = 7,569 (20%) who experienced a triple-low event. Patients were randomized to an Alert (N = 3,764) or No-alert group (N = 3,805). No significant differences in clinical or demographic characteristics were identified between groups. The average age was 63 years, with 91% being ASA 3 or greater (ASA 3 = 59%; ASA 4 = 31%, ASA 5 = 1%); 11% were emergency surgeries and surgical time averaged 5.8 hours. The most common surgical procedures were digestive (38%), followed by urinary system (13%), and musculoskeletal (13%).


The rate of helpful responses to a triple-low event were statistically significantly higher in the Alert group compared to the No-alert group (51% vs. 47%, P < 0.001). No significant differences were found in 90-day, 30-day, or one year mortality between groups (Figure 1). Length of stay did not differ between the groups; averaging 7 days in both groups.


Vasopressor use within five minutes or a 20% decrease in inhaled anesthetic concentration within 15 minutes was statistically significantly more common in the Alert group (51% vs. 47%, P < 0.001). More specifically, vasopressor use within five minutes was only slightly more common in the Alert vs. No-alert group (34% vs. 30%). A resulting 20% decrease in inhaled anesthetic concentration did not differ between groups (25% vs. 25%). The median time from the detection of a triple-low event to an MAP >75 mm Hg, or BIS > 45, or MAC > 0.8 was three minutes; and did not differ between the groups [emphasis added by editor]. The number of patients with a 20% increase in MAP within five minutes (35% vs. 34%) and 15 minutes (66% vs. 65%) of a triple alert did not differ between the Alert and No-alert groups. The maximum increase in MAP after the first triple-low was the same in the Alert and No-alert groups; 12 mm Hg at five minutes and 23 mm Hg at 15 minutes.


No relationship was found between helpful responses to triple-low events and mortality at 30 and 90-days, irrespective of group assignment.


Figure 1. Mortality by Group



Conclusion   Real-time decision support alerts in response to a triple-low event did not improve intraoperative management or decrease mortality at 30-days, 90-days, or at one year. Helpful responses to the alert were lower than expected. Furthermore, the time to raise MAP to >75 mm Hg or by 20% did not differ between groups. There was no relationship between helpful responses to triple-low events and mortality.




This was a very well-designed comparative-effectiveness trial on the effect of real-time triple-low alerts on mortality in high-risk patients. Comparative-effectiveness trials seek to determine which interventions work in real-world practice with a heterogenous population. In contrast, efficacy trials (i.e., drug studies) seek to answer the question, "Does the intervention work under ideal circumstances?,” in a homogenous population. This is most commonly in randomized controlled trial under a strict protocol (see


In this study an electronic alert to a triple-low event did not reduce mortality at 30-days, 90-days or at one year in patients at high-risk for morbidity and mortality postoperatively. Furthermore, the use of an automated alert only increased the response to a triple-low event by 4%, which is not clinically relevant. Nor did the automated alert decrease the time to correction of the blood pressure in either group.


So why did this occur? What I suspect is there may have been a maturation effect that effected the internal validity of the study. This study took place over six years at a single facility and I suspect the anesthesia professionals over time became more familiar with, and vigilant of, triple-low events; therefore, the use of an alert had minimal effectiveness on helpful responses and no effect on mortality. It would be interesting to see the study replicated at other institutions not as familiar with triple-low events or alerts. Another issue at play may be alarm fatigue.


Nonetheless, I think the study was novel and helps raise awareness of triple-low events (MAP < 75 mm Hg, BIS < 45, and MAC < 0.8) which may be predictive of mortality in old, frail patients undergoing major surgery. About 20% of this population will experience a triple-low event.


Dennis Spence, PhD, CRNA

The views expressed in this article are those of the author and do not reflect official policy or position of the Department of the Navy, the Department of Defense, the Uniformed Services University of the Health Sciences, or the United States Government.

© Copyright 2019 Anesthesia Abstracts · Volume 12 Number 14, January 22, 2019

Neuraxial anaesthesia techniques and postoperative outcomes among joint arthroplasty patients: is spinal anaesthesia the best option?

Br J Anaesth 2018;121:842-849

DOI: 10.1016/j.bja.2018.05.071

Weinstein SM, Baaklini LR, Liu J, Poultsides L, Cozowicz C, Poeran J, Saleh JN, Memtsoudis SG



Purpose   The purpose of this study was to compare spinal, epidural, and combined spinal-epidural (CSE) block in regards to postoperative outcomes.


Background   It is widely understood that anesthetic technique can affect the risk of adverse events and cost of care. For total joint replacements neuraxial anesthesia has been shown to reduce the likelihood of blood transfusion, wound infection, cardiovascular and pulmonary complications, and hospital length of stay compared to general anesthesia. Neuraxial anesthesia provides superior reduction of the surgical stress response and can prevent the “wind up” which increases pain intensity. Little, if any, evidence distinguishes between subarachnoid block and epidural block, however.


Methodology   This was a retrospective study of 40,852 patients who underwent total hip or total knee arthroplasty with spinal, epidural, or CSE anesthesia. Only primary joint replacements were included. Pediatric patients were excluded. Data on the use of neuraxial opioids and supplemental peripheral nerve blocks was not considered. Adverse events were identified as ICD-9 codes present at time of discharge that were not present at hospital admission. Prolonged length of hospital stay was defined as more than 4 days, which was the 25% of patients with the longest inpatient stay. When small amounts of data were missing in a patient’s record the missing values were estimated systematically. Statistical analysis included multiple comparisons. Significance levels were corrected to adjusted for the errors multiple comparisons introduces.


Result   Anesthetic techniques were distributed as follows: spinal 10% (4,087 patients), epidural 6% (2,464 patients), and CSE 84% (34,301 patients). Patients who received an epidural were sicker overall at the time of admission. Factors out of the control of anesthesia; e.g. patient age and sex, ASA physical status, and surgical procedure were significant risk factors for many of the adverse outcomes. The incidence of cardiac, pulmonary, GI, renal, and thromboembolic complications was low and similar between anesthesia techniques. That said, the incidence of postoperative complications was slightly, but statistically significantly, lower in patients who had a spinal anesthetic. For example, 13% of spinal anesthetic patients were in the hospital for longer than four days compared to 20% and 22% respectively of CSE and epidural patients. (This was by far the largest difference of any of the adverse outcomes.)


Conclusion   The study suggests that a single shot spinal may have advantages over an epidural or CSE for total joint procedures including slightly lower incidences of several adverse events postoperatively. This may be due to the more complete, more dense block common to spinal anesthetics vs. epidural anesthesia.




This study suffers from a number of methodological and statistical issues, none of which is a fatal flaw by itself. But, when all of them are added up, I’m hesitant to give it too much weight. But, that said … the investigators get points for trying to answer the question, “is any neuraxial technique better than the others?” Because there is so little evidence available about outcomes following different neuraxial techniques for total joints it’s important not to ignore what we have, even if imperfect.


The message I’m willing to take home from this study is that spinals may … may be associated with slightly better outcomes than epidurals or combined spinal-epidural anesthesia for total joints. While a bit counterintuitive, another take away for me is that there isn’t really any compelling reason to choose one technique over another for the patient’s benefit. Because the outcomes were so similar we can choose the technique based upon surgeon’s preference or to fit in with our postoperative pain management plan without being conflicted about the possible impact on patient care. What we know with a high degree of certainty from other studies is that neuraxial anesthesia, with or without the addition of general anesthesia, offers multiple, highly clinically significant advantages for total joints.

Michael A. Fiedler, PhD, CRNA

© Copyright 2019 Anesthesia Abstracts · Volume 12 Number 14, January 22, 2019