a b s t r a c t

Background: law enforcement restraint-related death is frequently associated with excited delirium syndrome (ExDS). Because such deaths are rare, the pathophysiology underlying ExDS deaths remains unknown, making identification of high-risk situations challenging. This study describes the medical conditions and situations surrounding restraint of individuals identified by law enforcement to be experiencing ExDS.

Methods: Individuals with ExDS as determined by Law enforcement officers during use of force encounters over a 3-year period were identified. For subjects who were brought to the emergency department after restraint, medical records and police narratives were reviewed to identify circumstances surrounding restraint, abnormalities found during evaluation, and final diagnoses.

Results: Sixty-six cases were identified, of which 43 had emergency department evaluation. On presentation, 36 (84%) were tachycardic and 3 (7%) were hyperthermic; 35 (77%) had toxicology studies positive for

stimulants; 2 (5%) had a pH level less than 7.2, and 5 (12%) had an elevated lactate; and 3 (7%) had a Creatinine kinase level higher than 1500 U/L. Two (5%) patients were admitted to the hospital for medical reasons: one had had a field pulseless electrical activity arrest prior to restraint and the other was admitted for rhabdomyolysis.

Conclusion: Officer-identified cases of ExDS infrequently involved individuals requiring extensive restraint or with medical conditions that objectively placed them at high risk for sudden death. The low specificity of this syndrome in predicting risk of sudden death may present a challenge to law enforcement and emergency physicians.

(C) 2014

Introduction

Law enforcement restraint-related sudden death has remained controversial over many decades despite an evolution of the restraint practices and weapons used. The controversy stems in great part because in many cases, no clear cause of death can be found [1-3].

This creates a challenge differentiating cases of potential excessive use of force (UOF) with situations where a patient is at an extremely high risk for death irrespective of the restraint practices used. Recently, there has been an increasing acceptance in the medical community to label individuals thought to be in the latter group as experiencing excited delirium syndrome (ExDS) [4].

? Index Medicus sub-headings: law enforcement officers, violence, delirium.

?? This work has not been presented elsewhere. There was no funding or financial support.

* Corresponding author. Division of Emergency Medicine, University of Washington Medical Center, Box 356123, 1959 NE Pacific St, Seattle, WA 98195. Tel.: +1 206 598

0103; fax: +1 206 598 4569.

E-mail address: [email protected] (J. Strote).

Excited delirium syndrome is a constellation of signs and symptoms including, among other things, Violent behavior, agitation, delirium, imperviousness to pain, and superhuman strength. It is a clinicopathologic diagnosis that is ultimately made only after death, based on autopsy, toxicology, and historical context. Potential pathophysiologic explanations include hypoxia, hyperkalemia, acido- sis, Autonomic dysfunction, and dopamine and catecholamine dysregulation, all of which could be impacted by struggle and restraint with law enforcement [3].

Although the role of ExDS in restraint-related death is often discussed, the inability to find a definitive pathophysiologic explanation for these events has made some physicians uncomfortable assigning ExDS as a primary cause of death [5]. Furthermore, police describe frequent encounters with individuals displaying the cardinal features of ExDS, but sudden death is only rarely an outcome in such incidents.

Efforts to improve restraint practices have been limited, therefore, by an inability to identify the combination of circumstances that might have a fatal outcome.

The goal of the current study is to evaluate the conditions of individuals identified by law enforcement to be experiencing

http://dx.doi.org/10.1016/j.ajem.2014.05.023

0735-6757/(C) 2014

1094 J. Strote et al. / American Journal of Emergency Medicine 32 (2014) 1093-1096

symptoms of ExDS. By examining law enforcement and medical records, we hoped to add to the data on the types and conditions of individuals who may be at risk for sudden death during restraint.

Methods

Study design, setting, and participants

The study uses a retrospective convenience sample cohort analysis of data covering a period from May 2008 through April 2011.

Inclusion criteria were individuals identified by law enforcement officers as experiencing signs of ExDS during UOF and other encounters. Cases were excluded if the subject was not brought to the emergency department (ED) for medical evaluation.

The cases occurred under the jurisdiction of one urban police department with a population of approximately 608600 [6]. Officer training about ExDS during the study period was department-wide and included both a yearly roll-call training video about the signs and symptoms of ExDS, a field reference guide summarizing the information and written protocols that reflected coordinated efforts with other emergency responders in the local fire agency and major trauma center.

Data collection

At regular intervals throughout the study period, incident reports were queried for the term excited delirium. For each identified case, incident and UOF reports were reviewed to identify race, sex, age, height, weight, weapon use, and officer injuries. Narratives were evaluated for the number of traits commonly accepted as associated with ExDS [7], including the following: bizarre behavior, violence, pain tolerance, unexpected strength, constant physical activity, blank stare, rapid breathing, sweating, inappropriate removal of clothing, hot to touch, attracted to glass or reflective surfaces, and speaking incoherently or making animal sounds.

For each case, medical records from associated ED visits and hospital admissions were reviewed for medical and psychiatric history, vital signs, study results, disposition, hospital course when admitted, and discharge diagnoses.

The study was approved with a waiver of consent by the lead author’s home institution’s Human Subjects Division.

Data analysis

All data were entered into a spreadsheet (Microsoft Excel X for Mac, Redmond, WA) and analyzed using descriptive statistics.

Results

During the study period, there were 66 cases identified. Of those, 21 were not brought to the ED and 2 had no ED records available, leaving 43 cases for review.

Subjects ranged in age from 21 to 56 years, with a median of 30 years. Most were men (41; 95%). Race and weight information was not identified for 9 subjects (21%). The most common race identified was African American (19; 56%), followed by white (12; 35%) and Asian American (3; 9%). Median subject weight was 81.8 kg, with a range of 64 to 132 kg. Median body mass index was 26.6 kg/m², with a range of 20.8 to 40.2 kg/m².

All law enforcement descriptions included at least 1 sign associated with ExDS: 5 (12%) had less than 3, 19 (44%) had 3 to 5,

and 19 (44%) had more than 5.

Weapons were used by officers in 13 (30%) cases. Weapons used included the following: conducted electrical weapon (12; 33%), capsicum spray (1; 2%), and baton (1;2 %). No canine, flashlight, or firearm use was reported. When conducted electrical weapons were

used, 1 (25%) discharge occurred in 3 cases, 2 (17%) in 2 cases, and more than 2 (58%) in 7 of cases. Officers were injured in 3 (7%) of cases. All officer injuries were limited to minor abrasions and contusions.

More than one-third (15; 35%) of the subjects received diazepam by emergency medical services in the field.

On arrival to the ED, the majority (36; 84%) were tachycardic (16 [37%)]had a pulse greater than 120), 3 (7%) were hypertensive, 2

were hypotensive (5%), and 3 (7%) were hyperthermic. Emergency department laboratory results are reported in Table. Two patients were severely acidotic, and 1 had Severe rhabdomyolysis. No patients had a positive troponin or hyperkalemia. Urine toxicology was positive for 26 (79%) of those tested and had more than 1 substance in 18 (55%). Blood alcohol levels were positive for 9 (23%) of those tested.

Electrocardiograms were performed on 23 (54%) patients. Most (17;74%) showed sinus tachycardia only. Significant abnormal- ities included 1 (4%) ECG with a long QTc interval and 1 (4%) ECG with a previously undocumented Left bundle-branch block that was later shown not to be due to an acute myocardial infarction.

During evaluation in the ED, 20 (47%) were noted to have a psychiatric history and 40 (93%) were documented to have a Substance abuse history. Cardiovascular disease was noted in 4 subjects (9%); all 4 had hypertension only.

Only 4 (9%) subjects were diagnosed as having trauma from law enforcement restraint. These included a zygomatic arch fracture, hand laceration, head laceration, and closed head injury.

Disposition from the ED included 7 (16%) medical/surgical admissions, 6 (14%) psychiatric admissions, 23 (54%) discharges to self-care, and 7 (16%) discharge to police custody. Two patients eloped prior to formal discharge. Of those admitted for medical/ surgical reasons, diagnoses included persistent altered mental status (2), rhabdomyolysis, seizure, hand trauma, and pulseless electrical activity arrest. No deaths occurred.

Most ED diagnoses were altered mental status, delirium, psychosis, or acute toxic encephalopathy (32; 81%); 2 (5%) patients’ diagnoses included the term excited delirium.

Table

ED laboratory evaluation

n = 43

Toxicology n = 33 (77%)

Positive	26 (79%)
Multipositive	18 (55%)
Cocaine	12 (36%)
Amphetamine	8 (24%)
Phencyclidine	8 (24%)
Opiates	3 (9%)
Marijuana	14 (42%)
Blood alcohol	n = 39 (91%)
Positive	9 (23%)
N 100 mg/dL	4 (10%)
pH	n = 6 (14%)
b7.2	2 (33%)
Bicarbonate	n = 40 (93%)
b20 mEq/L	10 (25.0%)
b10 mEq/L	3 (7.5%)
Lactate	n = 5 (12%)
N 10 mmol/L	2 (40%)
Creatinine kinase	n = 24 (56%)
N 400 U/L	9 (38%)
N 1000 U/L	1 (4.2%)
Potassium	n = 40 (93%)
N 5.4 mEq/L	0 (0%)
Creatinine	n = 40 (93%)
N 1.4 mg/dL	18 (45%)
Troponin	n = 12 (28%)
Positive	0 (0%)

Results from laboratory evaluations of individuals brought to the ED after being identified as experiencing excited delirium by law enforcement officers in the field.

J. Strote et al. / American Journal of Emergency Medicine 32 (2014) 1093-1096 1095

Discussion

The current study finds that subjects identified by law enforcement as experiencing ExDS rarely required significant UOF to achieve restraint or had major traumatic injury during restraint. Similarly, very few of the identified subjects had significant Vital sign abnormalities, acidosis, rhabdomyolysis, preexisting cardiovascular disease, or other conditions associated with or postulated to be a cause of restraint-related death [7-9], even when more significant UOF occurred.

The true number of benign incidents is likely underestimated in our study: more than one-third of cases were excluded because the subject was not felt to need medical evaluation. This is in contrast to how ExDS is commonly described. The rarity of restraint-related death has precluded good incidence estimates, but ExDS is commonly described as a “medical emergency,” [10] with mortality estimates ranging from 8% to as high as 66% of cases [11-13].

Large numbers of individuals demonstrating the traits of ExDS without necessarily being at risk create a problem for law enforce- ment, as identification of ExDS has been touted as one key to in- custody death prevention [14]. Overidentification may lead to unnecessary de-escalation of force, jeopardizing officer safety. Under- identification may endanger those being restrained. It is not clear how to balance these competing values, but the implications are large given that up to 16% of UOF encounters may involve subjects with 3 or more features of ExDS [14].

If the specificity of ExDS symptoms for predicting risk of sudden death is potentially very low, it also raises the question of its use in medicine generally. First described well over a century ago, it remains a diagnosis of exclusion when symptoms are present and no other cause of death can be found [1]. Because there is no clear pathophysiology in patients who die with associated ExDS, its only current use is as a potential identifier of high-risk patients or as a cause of death for Medical Examiners who would otherwise have to label a case “unknown.” To that end, it has been noted elsewhere as a “political diagnosis:” some argue that it protects officers falsely accused of causing deaths that were inevitable; others argue that it protects officers who have used fatal excessive force from facing responsibility [5]. The term is accepted by both the National Board of Medical Examiners and the American College of Emergency Physicians but not the American Medical Association or the American Psychiatric Association. A continued uncertainty behind this diagnosis may be reflected in this study by rare physician use.

Until more is known about ExDS, law enforcement should continue to be taught to identify and respond to dangerous encounters. Common sense backed up by recent studies suggests that avoiding resistance during restraint situations is the key to preventing injuries to both officers and those they are trying to restrain [15]. Extensive officer training on ExDS and force de- escalation as well as coordinated protocol development with other emergency responders in the jurisdiction studied may have contrib- uted to the limited UOF and injury findings here. Although the use of chemical restraint to diminish resistance cannot frequently be administered prior to initial control of a subject, the high rate of prehospital diazepam use in this study may have also contributed to the limited vital sign and laboratory abnormalities found. A potential use of chemical restraints has been suggested by others [13,16].

In addition to the low rate of life-threatening medical conditions in patients identified as having ExDS, we found that most cases had at least mild renal insufficiency. Although not fatal by itself, this could be relevant to further study given the postulated pathophysiology of ExDS- related death involving acidosis, rhabdomyolysis, and/or hyperkalemia [17]. Consistent with prior studies on law enforcement UOF, there were high rates of drug and alcohol use as well as psychiatric disease. The significance of this is unclear because most patients with psychiatric- or drug-induced delirium, even when requiring law enforcement UOF, do not have bad medical outcomes [18].

Two of the cases described here had potentially life-threatening acidosis. One patient had a witnessed seizure on scene with normalization of pH quickly after intravenous hydration, suggesting that the acidosis was due to the seizure. The other had a field pulseless electrical activity arrest. This patient was noted by Prehospital providers to have collapsed prior to any attempt at restraint. Two rounds of cardiopulmonary resuscitation were performed with spontaneous return of circulation. On arrival in the ED, the patient had a pH of 7.0 and a lactate of 18 mmol/L, with traces of cocaine and marijuana in his urine. He did not have rhabdomyolysis. The patient was discharged after 29 days with a diagnosis of Anoxic brain injury. Neither case suggests that restraint or UOF had any effect on the subjects’ acidosis or ultimate outcome, and the latter is consistent with an argument that certain patients are at risk for sudden death irrespective of law enforcement intervention.

This study suggests that most law enforcement cases identified as ExDS may not involve individuals at high risk for sudden death during restraint or restraint-related injury. If this is found to be true in larger studies, it reinforces the need for further research on UOF, ExDS, and law enforcement restraint and specifically what combination of these conditions creates higher risk situations. As has been noted elsewhere, the rarity of lethal cases makes such research extremely difficult and a national registry that includes medical information could prove invaluable. Until more specific high-risk situations are described or the pathophysiology behind such deaths can be identified, a potential for misuse of the term will remain and controversy surrounding the diagnosis and associated deaths will continue.

Limitations

This study had multiple limitations that may affect its results and conclusions.

Importantly, the study was limited in both how data were collected and the sample size. Cases were chosen where officers used the term “excited delirium” in their reports and may have missed a skewed portion of police encounters where ExDS occurred. Thus, the cases sampled may not have represented accurately the true scope of cases even within the study setting. Similarly, only a small number of cases were collected and one-third of the cases were excluded, as there was no record of a medical evaluation, leaving a small sample size and again increasing the risk of bias. It is possible that the cases for which the medical evaluation was missing could have skewed the results specifically as well, although most likely by underestimating the benign nature of most encounters.

It is possible that the officers in this study misidentified subjects experiencing ExDS. Experienced practitioners in emergency medical services, emergency medicine, and law enforcement, however, are well aware of the high frequency of encounters with individuals having ExDS symptoms that do not lead to death. A recent study also confirmed that law enforcement officers can successfully identify such individuals as well as the fact that subsequent death is uncommon. [14]

Furthermore, the data presented here are for one city only with a unique law enforcement training program, well-developed joint law enforcement/emergency responder field protocols, and general popula- tion. Therefore, the data may not reflect populations in other communities.

References

1. Kodikara S, Cunningham K, Pollanen MS. “Excited delirium syndrome”: is it a cause of death? Legal Med 2012;14:252-4.
2. Grant JR, Southall PE, Fowler DR, Mealey J, Thomas EJ, Kinlock TW. Death in custody: a historical analysis. J Forensic Sci 2007;52:1177-81.
3. Gill JR. The syndrome of excited delirium. Forensic science, medicine, and pathology; 2014.
4. DeBard ML. ACEP Excited Delirium Task Force: White Paper Report on Excited Delirium Syndrome; 2009.
5. Ranson D. Excited delirium syndrome: a political diagnosis? J Law Med 2012;19:667-72.

   1096 J. Strote et al. / American Journal of Emergency Medicine 32 (2014) 1093-1096

   Bureau UC. US Census Bureau Dicennial Census; 2010.

6. Vilke GM, DeBard ML, Chan TC, Ho JD, Dawes DM, Hall C, et al. Excited delirium syndrome (ExDS): defining based on a review of the literature. J Emerg Med 2012;43:897-905.
7. Stratton SJ, Rogers C, Brickett K, Gruzinski G. Factors associated with sudden death of individuals requiring restraint for excited delirium. Am J Emerg Med 2001;19:187-91.
8. Strote J, Range Hutson H. Taser use in restraint-related deaths. Prehosp Emerg Care 2006;10:447-50.
9. Ho JD, Smith SW, Nystrom PC, Dawes DM, Orozco BS, Cole JB, et al. Successful management of excited delirium syndrome with prehospital ketamine: two case examples. Prehosp Emerg Care 2013;17:274-9.
10. Takeuchi A, Ahern TL, Henderson SO. Excited delirium. West J Emerg Med 2011;12:77-83.
11. Vilke GM. Pathophysiologic changes due to TASER devices versus excited delirium: potential relevance to deaths-in-custody? J Forensic Leg Med 2011;18:291.
12. Vilke GM, Bozeman WP, Dawes DM, Demers G, Wilson MP. Excited delirium syndrome (ExDS): treatment options and considerations. J Forensic Leg Med 2012;19:117-21.
13. Hall CA, Kader AS, Danielle McHale AM, Stewart L, Fick GH, Vilke GM. Frequency of signs of excited delirium syndrome in subjects undergoing police use of force: descriptive evaluation of a prospective, consecutive cohort. J Forensic Leg Med 2013;20:102-7.
14. Castillo EM, Prabhakar N, Luu B. Factors associated with law enforcement-related use-of-force injury. Am J Emerg Med 2012;30:526-31.
15. Weiss S, Peterson K, Cheney P, Froman P, Ernst A, Campbell M. The use of chemical restraints reduces agitation in patients transported by emergency medical services. J Emerg Med 2012;43:820-8.
16. Maher PJ, Walsh M, Burns T, Strote J. prehospital resuscitation of a man with excited delirium and cardiopulmonary arrest. CJEM 2013;15:1-4.
17. Strote J, Verzemnieks E, Walsh M, Hutson HR. Use of force by law enforcement: an evaluation of safety and injury. J Trauma 2010;69:1288-93.