Secondary Logo

Journal Logo

Anesthetic Management of a Patient With Multiple Previous Episodes of Postanesthesia Care Unit Delirium: A Case Report

Kreuzer, Matthias PhD*; Whalin, Matthew K. MD, PhD*; Hesse, September D. W. PhD*; Riso, Margaret A. MD; García, Paul S. MD, PhD*‡

doi: 10.1213/XAA.0000000000000497
Case Reports: Case Report

We report the case of a 37-year-old female patient who required 22 surgeries following a pedestrian versus car accident. She was enrolled in a clinical study investigating emergence from anesthesia. In 10 of her 22 surgeries, we assessed her cognitive status in the postanesthesia care unit (PACU) using the Confusion Assessment Method. We observed PACU delirium in all 4 cases in which the patient received sevoflurane, but only in 1 of 6 cases in which she received propofol. The patient showed EEG α-band activity similar to that of an elderly patient who may reflect a greater risk of PACU delirium.

From the *Department of Anesthesiology, Emory University, Atlanta, Georgia; Department of Anesthesiology & Perioperative Care, University of California, Irvine, California; and Department of Anesthesiology, Atlanta VA Medical Center, Decatur, Georgia.

Accepted for publication December 9, 2016.

Funding: Dr. Paul S. García’s research efforts are supported in part by a Career Development Award #BX00167 (PI: PS García, MD, PhD) from the United States Department of Veteran Affairs, Biomedical Laboratory Research and Development Service and the James S. McDonnell Foundation Grant #220023046 (PI: PS García, MD, PhD).

The authors declare no conflicts of interest.

Address correspondence to Paul S. García, MD, PhD, Department of Anesthesiology, Atlanta VA Medical Center, Mail Code 151, 1670 Clairmont Rd, Decatur, GA 30033. Address e-mail to

Although emergence from general anesthesia frequently is uneventful, acute delirium in the postanesthesia care unit (PACU-D) can occur in 10% to 45% of patients.1 PACU-D is characterized by deficits in attention and cognition, and it often is classified as hypoactive, hyperactive, or mixed.1 Hypoactive PACU-D is the most common type2 and often goes undetected because the patient is not disruptive. Although PACU-D often is transient, it may be associated with prolonged hospitalizations, increased costs, greater rates of mortality, and chronic brain dysfunction.3 This case report describes the course of a woman who had multiple surgeries over 6 months and developed PACU-D potentially related to the chosen anesthetic technique. In all 4 cases of sevoflurane anesthesia, the patient screened positive for PACU-D, but only in 1 of 6 cases with total intravenous anesthesia (TIVA) using propofol.

Although the intravenous drug propofol and the volatile anesthetic sevoflurane both produce a state of general anesthesia, subtle differences in the resulting physiology exist as the result of drug-specific pharmacologic interactions to slightly different sets of molecular targets. Propofol mainly acts on the GABAA receptor, whereas sevoflurane affects GABA-signaling as well as a larger number of other receptors.4,5 These differences lead to different spectral properties of the patient’s electroencephalogram (EEG) recorded during unconsciousness.6 Recently, a link between the patient’s age and drug-induced changes in the patient’s EEG spectrum has been described.7 Hence, we used the EEG power spectral density (PSD) as well as spectral coherence to investigate a potential relationship between the development of PACU-D and “abnormal” EEG observations. The patient provided written permission for publication of the case report.

Back to Top | Article Outline


A 36-year-old female patient with a history of cocaine use and depression with psychotic features presented with leg fractures after a pedestrian versus car accident. She had emergency surgery to repair a vascular injury and place an external fixator. She remained endotracheally intubated overnight but was extubated the next morning. She had 19 surgeries during her 2-month hospitalization and was discharged to a nursing home. Because of observed agitation after her second surgery and her history of depression with psychotic features, she received oral risperidone (Risperdal™; Janssen Pharmaceuticals, Inc, Titusville, NJ) 1 mg twice daily for the remainder of her hospitalization.

Throughout her hospitalization, the patient was stable and never required psychiatric consultation. The patient had no access to illicit substances during her hospital stay and subsequent ambulatory testing was negative for cocaine. Before her eighth surgery, she was enrolled in a prospective observational study investigating a potential link among preoperative comorbidities, anesthetic technique, and other intraoperative factors with adverse surgical outcomes. All local ethical considerations regarding research with human subjects have received approval by the institutional review board.

PACU-D assessment was performed after 10 of her 22 surgeries, that is, after surgeries 8 to 13, 15 to 17, and 22. We used the Confusion Assessment Method8 to screen for PACU-D 15 minutes after PACU admission and again at a time point 60 minutes after regaining response to verbal command. We have no PACU-D documentation for her previous surgeries, but per PACU nursing reports, it appears that she also had agitation during previous recoveries. She developed hyperactive PACU-D after her eighth surgery that resolved by 1 hour after emergence. For her ninth surgery, benzodiazepine premedication was avoided but transient PACU-D requiring haloperidol was observed.

After transitioning to TIVA with propofol in ensuing cases, the recovery improved. She developed PACU-D in only 1 of 6 cases performed under TIVA with propofol. With sevoflurane, she showed signs of hyperactive PACU-D in all 4 cases. Table 1 displays the detailed characteristics of her study protocol anesthetics. Each episode of PACU-D was transient and had resolved completely before the next surgery. For several of these surgeries, we obtained frontal EEG data with a SEDLineTM monitor (Masimo, Irvine, CA) from positions FP1, FP2, FPz, FP7, and FP8 according to the international 10 to 20 system with the SEDLineTM EEG electrode strip during general anesthesia and the emergence period. EEG sampling rate was 250 Hz.

Table 1

Table 1

We conducted EEG processing and analysis with MATLAB (R2015a, The MathWorksTM, Natick, MA). We filtered the EEG to 0.1 to 40 Hz using the filtfilt function, performed down sampling to 125 Hz, and estimated the PSD with the pwelch function and the coherence with mscohere with NFFT set to 128, that is, a frequency resolution of 125/128 Hz. We assessed PSD and coherence for the entire recording using 10-second EEG windows and a 1-second shift. We used the episode from the start of surgery until termination of the anesthetic drug to calculate averaged PSD and average coherence under general anesthesia. In addition, we estimated the average θ-band (4–8 Hz) and α-band (8–12 Hz) coherence, to estimate her “brain age” according to Purdon et al7 and quantified her anesthesia emergence according to the EEG trajectories described by Chander et al.9 If the EEG trajectory during anesthesia emergence does not include a transient episode of nonslow-wave activity, the patient’s postanesthesia state may be of low quality, due to an abrupt transition from deep anesthesia.

Compared with the propofol cases, the sevoflurane cases (all with PACU-D development) lasted longer and greater doses of opioids were given. In the propofol case with PACU-D, the patient did receive a moderate dose of fentanyl (total 250 μg; 2.1 μg/kg). Retrospective analysis of the SEDLineTM Patient State Index showed no association between the index and PACU-D. Figure 1 shows representative spectrograms for a sevoflurane and a propofol case; each demonstrates the EEG frequency distribution specific for each substance as reported by Akeju et al.6 Sevoflurane caused a broad activation of low frequencies in the δ- (0.5–4 Hz), θ-, and α-range, whereas anesthetic doses of propofol led to a strong activation of δ- and α-frequencies, but not theta.

Figure 1

Figure 1

Figure 2 displays averaged PSD and coherence for the surgery period for the cases with recorded EEG. The single PSD as well as coherence analysis reflect the published spectral properties characteristic for either sevoflurane or propofol.6 The propofol cases show the distinct α peaks, whereas with sevoflurane, power in the θ-range is increased.6 In contrast to the representative PSD plots presented by Akeju et al,6 we could not observe a well-developed α peak for the sevoflurane cases. This missing peak may be an indicator of increased noxious stimulation or a more aroused brain.10,11

Figure 2

Figure 2

We did not find any PSD differences within the propofol cases to separate the PACU-D case from the others. The coherence analysis revealed that the difference between the θ and α coherence in the propofol group was greatest for the PACU-D case (Table 2). We further used the coherence analysis to estimate her “brain age” using research results from Purdon et al,7 showing α-band F7–F8 coherence decreases with age, to estimate her “real brain age.” The estimated α-band coherence during anesthesia maintenance for a 37-year-old woman is around 64% for propofol and around 70% for sevoflurane.7 Our patient’s α-band coherence ranged from 40% to 56% during propofol and from 44% to 53% during sevoflurane anesthesia (Table 2). According to the linear regression in the article of Purdon et al,7 this low coherence may signify that this patient’s brain was similar to that of a patient older than 90 years of age, and therefore at greater risk for PACU-D.

Table 2

Table 2

Table 2 contains the detailed values of analyses. Evaluation of EEG emergence trajectories showed that the patient woke up from anesthesia in a similar manner each time. She followed the trajectory of directly transitioning from δ-dominant anesthesia to wakefulness without dwelling in a low amplitude (nonslow-wave state), a trajectory that is associated with increased risk of pain and agitation in the PACU.9

Back to Top | Article Outline


Our patient had some underlying psychiatric disease and substance abuse (cocaine and marijuana) that may predispose her to delirium.12 Fortunately, her PACU-D episodes generally resolved before PACU discharge. The latest guidelines recommend patients without agitation to receive nonpharmacologic treatment such as structured orientation, sleep hygiene, and sensory optimization.13 Patients with agitation may benefit from antipsychotics. Although premedication with antipsychotics remains controversial, there is some evidence that high-risk patients may benefit.14,15 For our patient, intravenous anesthesia appeared to offer some protection against PACU-D, but further study is needed to determine whether this applies more broadly to other patients.

PACU-D occurred in all of our patient’s sevoflurane cases but only in 1 of 6 surgeries in which propofol TIVA techniques were used. The sevoflurane cases lasted longer, and the patient tended to receive a greater amount of opioid, which may have contributed to PACU-D. The EEG analysis revealed nonpharmacologic risks that may have predisposed the patient for PACU-D. Other factors of possible influence such as body temperature, respiratory management or type of surgery did not differ in the 7 cases with EEG recording. The patient’s body temperature on arrival to PACU ranged from 36.3°C to 36.9°C. In all except the final case, the patient breathed spontaneously through a supraglottic airway during the anesthetic procedure. The first 6 surgical procedures were incisions and drainages. Case 7 involved surgical fixation of the femur.

Our patient was 37 years old, but her history of underlying psychiatric disease and substance abuse may have contributed to an acceleration of brain aging. Based on our EEG analysis, her brain age resembled a geriatric patient. For all recorded EEG cases, our patient followed an emergence EEG trajectory that did not dwell in a REM-like low amplitude (nonslow-wave) state that stands in contrast to the archetypal emergence trajectory from a slow-wave-dominant pattern to nonslow-wave anesthesia before finally leading to wakefulness.9 Our patient’s trajectory appears to favor the development of negative outcomes after anesthesia, including greater pain and agitation in recovery.9 All these findings point toward a patient with high risk of possible adverse outcomes. In her case, she developed the adverse outcome PACU-D with sevoflurane almost exclusively. We observed a less-developed average α peak during her time receiving sevoflurane anesthesia than has been reported by Akeju et al.6 Reduced α peaks may correlate with a more aroused state or greater noxious stimulation during anesthesia.10,11 In her only PACU-D case with propofol, the difference between θ and α coherence was large, but the relevance of this finding should be a matter of future research.

On the basis of our findings, we conclude that our patient combined a number of risk factors to develop PACU-D. Her comorbidities and her history of substance abuse may have placed her brain in a “frail” state, and her EEG followed an unfavorable trajectory during emergence that may have stressed her brain further. We cannot generalize that propofol is protective against delirium, but this case illustrates the need to carefully consider the anesthetic management of patients who are at risk for delirium.

Back to Top | Article Outline


Name: Matthias Kreuzer, PhD.

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

Name: Matthew K. Whalin, MD, PhD

Contribution: This author helped collect the data and write the manuscript.

Name: September D. W. Hesse, PhD

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

Name: Margaret A. Riso, MD

Contribution: This author helped collect the data and write the manuscript.

Name: Paul S. García, MD, PhD

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

This manuscript was handled by: Hans-Joachim Priebe, MD, FRCA, FCAI.

Back to Top | Article Outline


1. Whitlock EL, Vannucci A, Avidan MS. Postoperative delirium. Minerva Anestesiol. 2011;77:448–456.
2. Robinson TN, Raeburn CD, Tran ZV, Angles EM, Brenner LA, Moss M. Postoperative delirium in the elderly: risk factors and outcomes. Ann Surg. 2009;249:173–178.
3. Franco K, Litaker D, Locala J, Bronson D. The cost of delirium in the surgical patient. Psychosomatics. 2001;42:68–73.
4. Alkire MT, Hudetz AG, Tononi G. Consciousness and anesthesia. Science. 2008;322:876–880.
5. Rudolph U, Antkowiak B. Molecular and neuronal substrates for general anaesthetics. Nat Rev Neurosci. 2004;5:709–720.
6. Akeju O, Westover MB, Pavone KJ, et al. Effects of sevoflurane and propofol on frontal electroencephalogram power and coherence. Anesthesiology. 2014;121:990–998.
7. Purdon PL, Pavone KJ, Akeju O, et al. The ageing brain: age-dependent changes in the electroencephalogram during propofol and sevoflurane general anaesthesia. Br J Anaesth. 2015;115(suppl 1):i46–i57.
8. Inouye SK. Delirium in older persons. N Engl J Med. 2006;354:1157–1165.
9. Chander D, García PS, MacColl JN, Illing S, Sleigh JW. Electroencephalographic variation during end maintenance and emergence from surgical anesthesia. PLoS One. 2014;9:e106291.
10. Kochs E, Bischoff P, Pichlmeier U, Schulte am Esch J. Surgical stimulation induces changes in brain electrical activity during isoflurane/nitrous oxide anesthesia. A topographic electroencephalographic analysis. Anesthesiology. 1994;80:1026–1034.
11. MacKay EC, Sleigh JW, Voss LJ, Barnard JP. Episodic waveforms in the electroencephalogram during general anaesthesia: a study of patterns of response to noxious stimuli. Anaesth Intensive Care. 2010;38:102–112.
12. Fricchione GL, Nejad SH, Esses JA, et al. Postoperative delirium. Am J Psychiatry. 2008;165:803–812.
13. The American Geriatrics Society Expert Panel on Postoperative Delirium in Older Adults. American Geriatrics Society abstracted clinical practice guideline for postoperative delirium in older adults. J Am Geriatr Soc. 2015;63:142–150.
14. Aizawa K, Kanai T, Saikawa Y, et al. A novel approach to the prevention of postoperative delirium in the elderly after gastrointestinal surgery. Surg Today. 2002;32:310–314.
15. Kalisvaart KJ, de Jonghe JF, Bogaards MJ, et al. Haloperidol prophylaxis for elderly hip-surgery patients at risk for delirium: a randomized placebo-controlled study. J Am Geriatr Soc. 2005;53:1658–1666.
© 2017 International Anesthesia Research Society