Share this article on:

Processed Electroencephalogram During Donation After Cardiac Death

Auyong, David B. MD*; Klein, Stephen M. MD; Gan, Tong J. MB, BS, FRCA; Roche, Anthony M. MB, ChB; Olson, DaiWai PhD, RN, CCRN; Habib, Ashraf S. MB, ChB, MSc, FRCA

doi: 10.1213/ANE.0b013e3181d27067
Neuroscience in Anesthesiology and Perioperative Medicine: Case Report

We present a case series of increased bispectral index values during donation after cardiac death (DCD). During the DCD process, a patient was monitored with processed electroencephalogram (EEG), which showed considerable changes traditionally associated with lighter planes of anesthesia immediately after withdrawal of care. Subsequently, to validate the findings of this case, processed EEG was recorded during 2 other cases in which care was withdrawn without the use of hypnotic or anesthetic drugs. We found that changes in processed EEG immediately after withdrawal of care were not only reproducible, but can happen in the absence of changes in major electromyographic or electrocardiographic artifact. It is well documented that processed EEG is prone to artifacts. However, in the setting of DCD, these changes in processed EEG deserve some consideration. If these changes are not due to artifact, dosing of hypnotic or anesthetic drugs might be warranted. Use of these drugs during DCD based primarily on processed EEG values has never been addressed.

Published ahead of print March 17, 2010 Supplemental Digital Content is available in the text.

From the *Department of Anesthesiology, Virginia Mason Medical Center, Seattle, Washington; and Departments of Anesthesiology, and Medicine-Neurology, Duke University Medical Center, Durham, North Carolina.

Supported by departmental funds, Duke University Medical Center.

Address correspondence and reprint requests to David B. Auyong, MD, Department of Anesthesiology, Virginia Mason Medical Center, 1100 Ninth Ave., Mailstop: B2-AN, Seattle, WA 98101. Address e-mail to

Accepted December 4, 2009

Published ahead of print March 17, 2010

Organ procurement and subsequent transplant can result in lifesaving treatment but can also be associated with ethical and legal controversy. In the United States, there are 2 main types of organ procurement, each with its own distinct clinical and ethical requirements. Currently, most organs are obtained from heart-beating donors, who either volunteer an organ or who are declared dead through brain death criteria (total and irreversible loss of function of the cortex and brainstem).1 If brain death is established, organ procurement can proceed while continuing cardiopulmonary support.

Donation after cardiac death (DCD), or non–heart-beating organ donation, is another option for procurement when severely ill patients who do not meet brain death criteria are declared dead by cardiopulmonary death criteria (absence of systemic blood flow, unresponsiveness, and apnea), before organ removal.2 DCD candidates are dependent on life-sustaining treatment, and the removal of those treatments should result in a rapid death. In controlled DCD, terminally ill (but not brain dead) patients are brought to the operating room, where medical care is withdrawn (halting artificial ventilation, IV infusions, and medications), but measures to hasten demise are prohibited. Sedatives and opioids should be used in accordance with palliative care guidelines, which permit their use, not to hasten demise, but rather to treat “discomfort or the appearance of discomfort.”3 Death is usually defined as arrest of blood flow for 2 to 5 minutes1 and death must occur within a specified length of time, usually 1 hour, to limit warm ischemic time of the vital organs.4 Organ procurement only proceeds after death is declared.

The DCD process places physicians in an unaccustomed role that conflicts with their usual practice and can require decisions based on ethical interpretation. These physicians must halt lifesaving care and yet, in the absence of obvious donor pain or suffering, withhold palliation for the same patient to avoid actions that hasten death, 2 roles that theoretically conflict with each other. Exacerbating these ethical dilemmas further, cognitive function is assumed to be absent and is almost impossible to assess. We present 3 cases of increased bispectral index (BIS) scores during the withdrawal of care during the DCD process and discuss possible explanations of this observation.

Back to Top | Article Outline


Case 1

A 54-year-old man was transferred to our institution after presenting at an outside hospital with an initial Glasgow Coma Scale (GCS) score of 10T. Evaluation revealed a diffuse subarachnoid, intraventricular hemorrhage, and a 7 × 8 × 9 mm aneurysm along the anterior communicating artery. Although the patient presented in our emergency department with a GCS score of 7T, he became unresponsive to stimulation. Despite coiling of the aneurysm and placement of a ventriculostomy, he remained comatose over the next 8 days. His prognosis was deemed poor based on a GCS score of 3; flaccid paralysis on physical examination; increased intracranial pressures (>60 mm Hg); and repeated imaging with computed tomography showing increasing hemorrhage, edema, and bilateral anterior cerebral artery infarcts. During apnea testing, irregular but spontaneous respiration was present and transcranial Doppler revealed middle cerebral artery mean velocities of 83 cm/s on the right and 77 cm/s on the left (normal: 46–86 cm/s) with anterograde flow in all major cerebral arteries, eliminating the diagnosis of brain death. No further studies to assess neurologic function were performed. Withdrawal of care and then organ donation were discussed with the patient's family, who consented to both.

The patient was transferred to the operating room for withdrawal of care followed by organ donation after cardiac death. The patient was unresponsive to verbal commands, with an initial heart rate of 78 bpm, arterial blood pressure of 130/80 mm Hg, measured via a radial artery catheter, and peripheral oxygen saturation (SpO2) of 100% on FIO2 of 1.0. The patient arrived to the operating room with the trachea intubated and the lungs mechanically ventilated at a rate of 12 breaths/min with a tidal volume of 550 mL. Normal saline at 125 mL/h and phenylephrine at 1 μg/kg/min were infused through a left subclavian catheter. After surgical preparations were complete, including cannulation of the femoral vessels and sterile preparations, a BIS monitor (BIS 2000®, Aspect Medical Systems, Newton, MA) was placed on the patient's forehead according to the manufacturer's recommendations.5 Initial BIS values, obtained on arrival to the operating room, ranged between 1 and 20 (Fig. 1). No neuromuscular blockers, sedatives, or inhaled drugs were used. After preparations were complete, all circulatory and respiratory support was discontinued.

Figure 1

Figure 1

Within the first 5 minutes after withdrawing care, the BIS value increased from 1 to 92, without evidence of electromyographic (EMG) interference on the BIS monitor. For the first 25 minutes after withdrawal, the radial artery blood pressure remained >100/60 mm Hg and the heart rate, initially 74 bpm, gradually increased to 108 bpm. The SpO2 started at 100% and remained >90% for 8 minutes after withdrawal of care, at which point, it gradually decreased to <85% between 8 and 20 minutes after withdrawal. After 20 minutes, we were unable to obtain an SpO2 waveform. No gross movements of the chest or abdomen were observed that could be interpreted as spontaneous respirations, and the patient was not monitored with end-tidal CO2 at this time. The BIS values remained increased at 85 to 95 for 30 minutes after withdrawal of care. Twenty-five minutes after withdrawal of care, the patient's systolic blood pressure decreased from >100 mm Hg to <40 mm Hg and his heart rate decreased from 108 bpm to zero, while the BIS value also decreased to <4. The patient was pronounced dead by physical examination 5 minutes after the absence of any systemic anterograde blood flow (as monitored through the radial artery catheter), and 32 minutes after discontinuation of support. When the patient was pronounced dead, the BIS value was 0. Between withdrawal of care and pronouncement of death, the patient remained motionless and no operating room personnel were in physical or verbal contact with the patient.

Back to Top | Article Outline

Case 2

A 69-year-old man was found unresponsive on the second postoperative day after multilevel spinal fusion and decompression. He was diagnosed with a large posterior intracerebral hemorrhage and underwent evacuation. His prognosis was deemed poor and his family consented to DCD 4 days later. Medications before withdrawal included nicardipine infusion (2.5 mg/h), mannitol, and metoprolol. No sedatives, muscle relaxants, or analgesics were given to the patient during his 4-day intensive care unit course. His vital signs just before withdrawal of care were arterial blood pressure 142/59 mm Hg, heart rate 98 bpm, respiratory rate 32 breaths/min, and SpO2 100% on 30% FIO2. During the DCD process, the mean arterial blood pressure decreased to <50 mm Hg 20 minutes after withdrawal, and heart rate decreased from 98 bpm to zero over 37 minutes. Cardiopulmonary arrest occurred 37 minutes after withdrawal of care. BIS values, EMG, suppression ratio (SR), and signal quality index (SQI) were captured (Fig. 2). BIS values increased after withdrawal to a maximum of 84 (median 64, range 46–84), whereas EMG values did increase slightly from a median of 36 (range 31–42) before withdrawal to 43 (range 36–55) at the time of maximal BIS value increase. SR appropriately increased to 100 after withdrawal. SQI decreased after withdrawal, but this did not correlate with the increase in BIS values.

Figure 2

Figure 2

Back to Top | Article Outline

Case 3

A 79-year-old man receiving chronic anticoagulation for atrial fibrillation presented to an outside hospital with aphasia and vomiting. A left frontal intracerebral hemorrhage was diagnosed and the patient rapidly deteriorated requiring tracheal intubation and mechanical ventilation. Physical examination revealed a decerebrate posture, 2-mm fixed pupils bilaterally, but intact corneal reflex. The family requested withdrawal of care. This patient did not qualify for organ donation and withdrawal of care was performed without monitoring. Before withdrawal of care, his arterial blood pressure was 144/78 mm Hg, heart rate was 104 bpm, and SpO2 was 98% on 40% FIO2. BIS values, EMG, SR, SQI, and raw electroencephalogram (EEG) were captured all via the BIS machine (Fig. 3). BIS values increased after withdrawal to a median of 76 (range 60–83) from a baseline median of 37 (range 28–46), whereas EMG values decreased slightly from a median of 51 (range 43–58) before withdrawal to 49 (range 40–54) at the time of maximal BIS value increase. As with case 2, SR appropriately increased to 100 after withdrawal. SQI had a small decrease after withdrawal that did correlate with the increase in BIS values. Raw EEG recordings revealed patterns of higher frequency with decreased amplitude after withdrawal of care (Fig. 4).

Figure 3

Figure 3

Figure 4

Figure 4

Back to Top | Article Outline


These cases demonstrate the assessment of brain function in 3 patients during the withdrawal of care process using BIS, a form of processed EEG. BIS is frequently used in the operating room and critical care units to assess depth of anesthesia via analysis of EEG. Because none of the patients met brain death criteria, we expected the BIS readings not to be zero before withdrawal of care, and indeed, they were not. By using an alternative method to assess cortical function during the dying process, we hoped to have some evaluation of brain function during death. The high values obtained suggest that artifact or a distinct change in cortical electrical activity may have occurred during DCD. The appropriate treatments for this situation, as well as its ethical implications, have not been adequately addressed in current literature.

Despite these patients' poor prognoses, the observed increase of BIS values took place during a period of hemodynamic stability (0–25 minutes after withdrawal of care). Accompanied by the increase in BIS value was a concomitant increase in heart rate that increased 34 bpm after withdrawal of care in case 1. This increase in heart rate could be secondary to increased carbon dioxide tension during this period of apnea. It is unclear, however, what response should be undertaken by the physician in this circumstance whereby acute changes in 2 separate variables are taking place. Treatment with a hypnotic drug or analgesic would be an appropriate response to these changes in standard anesthetic practice. Similar treatment in the setting of the DCD process, based primarily on processed EEG values, has never been addressed and highlights some of the dilemmas presented in the operating room during DCD.

Using hypnotic drugs (benzodiazepines, barbiturates, volatile anesthetics, etc.) can lead to respiratory depression and systemic hypotension. Respiratory depression and hypotension, in the DCD process, can decrease oxygen delivery and perfusion to vital organs, potentially hastening death. Although an intervention to hasten death could potentially result in improved posttransplant organ function by limiting warm ischemic time of the donor organ,6 any such measure is unacceptable in the DCD process. An intervention whose primary intention is to hasten death, in the absence of outright donor discomfort, is not allowed during DCD donation.1 However, interventions to treat pain or suffering have a moral and compassionate indication. When used to treat pain, opioids in dying patients have been validated,7 and treatment with a hypnotic or anesthetic drug could be appropriate when a physician is trying to relieve patient suffering regardless of organ donor status.1 This apparent contradiction is explained by the rule of double effect that allows a primary intention (opioids for analgesia) even if a secondary outcome (respiratory depression) is potentially harmful.8 Whether this is indicated in the operating room during the DCD process based on processed EEG values has not been determined and may require further validation of processed EEG in end of life situations.

BIS is a depth of hypnosis monitor that uses processed EEG to generate a number between 0 (isoelectric EEG) and 100 (fully awake). The BIS value has been shown to be an effective monitor of hypnosis and anesthetic depth.9,10 Processed EEG monitoring is common in the operating room, offering several advantages over more formal neurologic testing including ease of monitor placement and relatively basic interpretation. Fifty-six critically ill or brain-dead patients were evaluated with BIS monitoring revealing consistent correlation between decreasing BIS values, onset of brain death, and brain death.11 Others have found BIS monitoring to be a useful tool to assess neurologic status and sedation in critically ill or brain-injured patients.1214

Despite validation in critically ill patients, there have also been several accounts of potential artifacts in BIS monitoring. In the cases presented herein, apnea with the resulting changes in cerebral blood flow caused by hypercapnia could affect BIS values. In addition, motion, subclinical seizures, EMG, states of low-voltage EEG, and electrocardiogram (ECG) interference have been attributed to incorrectly increase values even in organ donors meeting brain-death criteria.1517 Because the original raw EEG and EMG were not available in the first case, BIS values from 2 additional cases of withdrawal of care without sedatives or analgesics were recorded to evaluate the role that changes in EMG or superposition of ECG artifacts played. In cases 2 and 3, some interference from EMG was recorded as BIS values increased soon after withdrawal of care. It remains difficult to attribute all BIS changes to EMG interference. In addition, case 3 reveals a decreased amplitude and increased frequency of raw EEG recordings between pre- and postwithdrawal, with no apparent ECG artifacts, revealing a distinct difference in EEG recordings that are likely reflected by the BIS number. Similar processed and raw EEG changes have recently been reported, revealing that EEG changes are reproducible in end of life situations.18 Because the BIS algorithm by which raw EEG data are processed into an integer is proprietary and derived from studies of thousands of patients, definitive conclusions about the significance of the BIS readings in a single patient are not possible.19 In addition, because BIS was developed in the setting of sedation, its applicability to other circumstances such as patients with ischemic states or severe structural damage is relatively untested. In mild and moderate head injury, BIS values showed significant correlation to GCS score; however, individual BIS scores could not accurately predict depth of coma.20

Dismissing the increased BIS values or changes in raw EEG as artifact during withdrawal of care may not be accurate. In case 1, the prognosis was dire because of the devastating neurologic pathology, but the patient still had cerebral blood flow and brain function before withdrawal of care, leaving the unlikely possibility that there was cortical brain function during the dying process.

The controlled DCD process has the potential to provide valuable transplant organs but requires physicians to manage medical care with certain constraints. It seems that these changes in processed EEG are reproducible in end of life situations, and the appropriate response to these changes is unclear. If these increased BIS values reflect physiologic changes during hypercarbia, low-voltage EEG states, EMG, or other artifact, then dosing anesthetic drugs would not be indicated. However, if these changes are not artifact, perhaps dosing hypnotic or analgesic drugs is appropriate. Because of this controversy, additional data on raw and processed EEGs from many more patients during the dying process are warranted.

Back to Top | Article Outline


1. Recommendations for nonheartbeating organ donation. A position paper by the Ethics Committee, American College of Critical Care Medicine, Society of Critical Care Medicine. Crit Care Med 2001;29:1826–31
2. Van Norman GA. Another matter of life and death: what every anesthesiologist should know about the ethical, legal, and policy implications of the non-heart-beating cadaver organ donor. Anesthesiology 2003;98:763–73
3. Bernat JL, D'Alessandro AM, Port FK, Bleck TP, Heard SO, Medina J, Rosenbaum SH, Devita MA, Gaston RS, Merion RM, Barr ML, Marks WH, Nathan H, O'connor K, Rudow DL, Leichtman AB, Schwab P, Ascher NL, Metzger RA, Mc Bride V, Graham W, Wagner D, Warren J, Delmonico FL. Report of a National Conference on Donation after cardiac death. Am J Transplant 2006;6:281–91
4. Lewis J, Peltier J, Nelson H, et al. Development of the University of Wisconsin donation after cardiac death evaluation tool. Prog Transplant 2003;13:265–73
5. Kelley SD. Monitoring Level of Consciousness During Anesthesia and Sedation. A Clinician's Guide to the Bispectral Index. Newton, MA: Aspect Medical Systems, Inc., 2003
6. Foley DP, Fernandez LA, Leverson G, Chin LT, Krieger N, Cooper JT, Shames BD, Becker YT, Odorico JS, Knechtle SJ, Sollinger HW, Kalayoglu M, D'Alessandro AM. Donation after cardiac death: the University of Wisconsin experience with liver transplantation. Ann Surg 2005;242:724–31
7. Faber-Langendoen K, Lanken PN. Dying patients in the intensive care unit: forgoing treatment, maintaining care. Ann Intern Med 2000;133:886–93
8. Sulmasy DP, Pellegrino ED. The rule of double effect: clearing up the double talk. Arch Intern Med 1999;159:545–50
9. Liu J, Singh H, White PF. Electroencephalographic bispectral index correlates with intraoperative recall and depth of propofol-induced sedation. Anesth Analg 1997;84:185–9
10. Liu J, Singh H, White PF. Electroencephalogram bispectral analysis predicts the depth of midazolam-induced sedation. Anesthesiology 1996;84:64–9
11. Vivien B, Paqueron X, Le Cosquer P, Langeron O, Coriat P, Riou B. Detection of brain death onset using the bispectral index in severely comatose patients. Intensive Care Med 2002;28:419–25
12. Gilbert TT, Wagner MR, Halukurike V, Paz HL, Garland A. Use of bispectral electroencephalogram monitoring to assess neurologic status in unsedated, critically ill patients. Crit Care Med 2001;29:1996–2000
13. Deogaonkar A, Gupta R, DeGeorgia M, Sabharwal V, Gopakumaran B, Schubert A, Provencio JJ. Bispectral Index monitoring correlates with sedation scales in brain-injured patients. Crit Care Med 2004;32:2403–6
14. Escudero D, Otero J, Muniz G, Gonzalo JA, Calleja C, González A, Martínez A, Parra D, Yano R, Taboada F. The Bispectral Index Scale: its use in the detection of brain death. Transplant Proc 2005;37:3661–3
15. Petrie J, Rampil IJ. Electromyographic artifact confounds BIS monitoring in organ donors. Anesthesiology 2003;99:A348
16. Myles PS, Cairo S. Artifact in the bispectral index in a patient with severe ischemic brain injury. Anesth Analg 2004;98:706–7
17. Gaszynski T, Wieczorek A, Krupowczyk W, Gaszynski W. BIS for recognition of brain-death in potential organ donors. Crit Care 2002;6:P52
18. Chawla LS, Akst S, Junker C, Jacobs B, Seneff MG. Surges of electroencephalogram activity at the time of death: a case series. J Palliat Med 2009;12:1095–100
19. Ruskin KJ, Shelley KH. Patent medicine and the “black box.” Anesth Analg 2005;100:1361–2
20. Paul DB, Umamaheswara Rao GS. Correlation of bispectral index with Glasgow Coma Scale in mild and moderate head injuries. J Clin Monit Comput 2006;20:399–404
© 2010 International Anesthesia Research Society