a b s t r a c t

Introduction: Intramuscular ketamine has become increasingly popular for prehospital chemical restraint of severely agitated or violent patients because of its favorable adverse effect profile, rapid onset, and wide therapeutic window. However, there is currently no literature quantifying the need for intubation or hospital admission for these patients once they reach the emergency department.

Methods: Medical records for patients receiving prehospital ketamine who were transported to a single level 1 trauma center were abstracted. Ketamine dose, patient weight, final disposition, and presence of intubation were recorded. Exclusion criteria were missing dose or weight and ketamine given for an indication other than chemical restraint. Statistical analysis was preformed with unadjusted Student t test. Statistical significance was defined as P b .05.

Results: A convenience sample of 51 consecutive patients was identified with 2 excluded because of missing data, leav- ing 49 for analysis. Ketamine dosing ranged from 2.25 to 9.42 mg/kg (mean, 5.26 +- 1.65 mg/kg). Significant differences were noted between those who required intubation (n = 14) and those who did not (n = 35) (6.16 +- 1.62 mg/kg vs 4.90 +- 1.54 mg/kg, P = .02). No patients were intubated prehospital. There was an increased dose in patients ad- mitted to a medical ward (57%, 28/49) that approached statistical significance (5.62 +- 1.80 vs 4.78 +- 1.31, P = .06). Conclusion: Intubation was observed in our emergency department in 29% of patients administered intramuscular ketamine for prehospital chemical restraint. There was a positive association between higher ketamine doses and both endotracheal intubation and hospital admission. Future research should aim to define the minimum effective ketamine dose for successful chemical restraint.

(C) 2014

Background and importance

intramuscular ketamine is gaining acceptance by Emergency Medical Services (EMS) medical directors for use in chemically

? Prior presentations: An abstract of this work was presented as a poster presentation at the NAEMSP annual meeting in January 2014.

?? Funding sources/disclosures: An internal, HealthPartners Institute for Education and

Research emergency medicine resident research grant in the form of $2000 was awarded to this project.

* Corresponding author: Regions Hospital, Department of Emergency Medical Services, 640 Jackson St, Mail Stop 13801B, St Paul, MN 55101. Tel.: +1 651 254 7781; fax: +1 651 778 3778.

E-mail address: Aaron.M.Burnett@healthpartners.com (A.M. Burnett).

¹ Primary manuscript writer, conception of study, data analysis.

² Data collection and analysis, manuscript review.

³ Data analysis and manuscript review.

⁴ Manuscript preparation and review.

⁵ Data abstraction and analysis. Received a $2000 internal resident research grant.

⁶ University of Kansas College of Medicine.

⁷ Data analysis and manuscript review. Mentoring of coauthors.

restraining agitated or violent patients that pose an immediate danger to EMS personnel in the prehospital setting [1-5]. One of the key values of ketamine as a chemical restraint in the prehospital setting, compared to more commonly used sedatives such as haloperidol and droperidol, is its rapid onset of action [6, 7]. The sedative effects of ketamine can be seen within 5 minutes of IM injection compared to more than 20 mi- nutes for haloperidol [6, 7]. Although the rapid onset of action may be favorable for the safety of the EMS providers, there are few reports quantifying adverse events and patient disposition after hospital arrival. In a small case series, hypoxia was one of the most clinically important complications that developed in the emergency department (ED) fol- lowing administration of ketamine in the prehospital setting (3/13 cases) [1]. Other studies have assessed the adverse reactions of keta- mine given for procedural sedation, as well as in populations of patients with suicidal ideation and severe depression [8-10]. The incidence of Respiratory complications in those studies was low and nonexistent, re- spectively. There is currently no literature examining the disposition of patients received in the ED who were administered ketamine as a chemical restraint by EMS prior to arrival.

http://dx.doi.org/10.1016/j.ajem.2014.10.016

0735-6757/(C) 2014

A.M. Burnett et al. / American Journal of Emergency Medicine 33 (2015) 76–79 77

To better understand the Risk-benefit ratio for using ketamine for prehospital chemical restraint for severely agitated or violent patients, we report a series of patients received at our ED who had been given IM ketamine by EMS for chemical restraint. The primary end point was to quantify the number of patients intubated in our ED after receiving ketamine by EMS for chemical restraint. Our secondary research question was to quantify the final disposition for all patients. We hypothesized that patients who were intubated or admitted to a medical bed would have received a higher average milligram per kilo- gram ketamine dose than those who were not.

Methods

This was a retrospective review of a convenience sample of consec- utive prehospital care reports and associated ED records for patients ad- ministered IM ketamine by EMS providers and transported to a single receiving hospital. Standard chart abstraction guidelines were followed [11]. These included training of abstractors, explicit inclusion and exclusion criteria, definitions of variables, use of abstraction forms, and periodic meetings between the 2 abstractors (BP, KG). As the pri- mary and secondary outcome variables were not subjective (intubation and hospital admission), there was no formal test of interrater reliability and the abstractors were not blinded to study hypothesis. The term medical bed refers to a nonpsychiatric inpatient hospital bed.

Setting

During the study period, prehospital administration of ketamine at a dose of 5 mg/kg via the IM route was authorized on standing order for adult patients (N 17 years old or, in those whose age was unknown, were showing signs of puberty) with severe agitation, those displaying active violence, and those who the treating paramedic felt were at risk of excited delirium. This IM ketamine dose has been advocated in the lit- erature for chemical restraint of patients in excited delirium [12]. The single receiving hospital for all patients in this study was an urban level 1 trauma center with annual ED volume of approximately 80000. This facility hosts an Accreditation Council for Graduate Medical Education-accredited emergency medicine residency and has emer- gency medicine board-certified/board-eligible supervising physicians present in the ED at all times.

Selection of participants

Inclusion criteria consisted of patients receiving ketamine by a single fire-based EMS service for chemical restraint and transported to the study hospital over a 2-year period. Patients were identified by searching the electronic prehospital care report (HealthEMS, Sansio Corporation, Duluth, Minnesota, Version 4) over a 24-month period and entering ketamine in the medications administered tab. The indica- tion for ketamine administration was considered chemical restraint if restraints was selected under treatments or if the narrative indicated that the patient was agitated, violent, or required sedation. Patients were excluded from analysis if the ketamine dose was not documented in the electronic prehospital care report, if the weight was not recorded in the hospital chart, or if ketamine was given for reasons other than chemical restraint (ie, analgesia).

Methods and measurements

Institutional review board approval was obtained for this study prior to data collection. Patient information obtained included the ketamine dose administered by EMS, weight recorded in the ED, whether or not the patient was intubated, indication for intubation, presence of ethanol, and ultimate disposition from the ED. The milligram per kilogram dose of ketamine administered by EMS was calculated.

Data analysis

Data were extracted from the electronic prehospital care report into Microsoft Excel software (Microsoft Corp, Redmond, WA), and descrip- tive and comparative statistics were calculated. A P value of b.05 was considered statistically significant. Patient demographic information and average milligram per kilogram ketamine dose were compared between groups using Student t test.

Results

Fifty-one patient encounters were identified between July 2011 and June 2013. Two patients were excluded because of missing documenta- tion (patient’s weight or administered ketamine dose), leaving a total of 49 patients for analysis. Of the 49 patients analyzed, 69% (34/49) were male and the mean age was 37.2 +- 13.3 years. Administered ketamine doses ranged from 2.25 to 9.42 mg/kg (mean dose, 5.26 +- 1.65 mg/kg). Fourteen patients (14/49, 29%; 95% confidence interval [CI], 17%- 43%) required intubation. No patients were intubated in the field. The 14 patients who required endotracheal intubation received a statistically significantly higher average ketamine dose compared to the 35 not re- quiring intubation (6.16 +- 1.62 mg/kg vs 4.90 +- 1.54 mg/kg, P = .02) (Fig. 1). The indication for intubation was listed as failure to protect air- way in 7, recurrent agitation with the need for additional sedation in 2, to facilitate emergent lumbar puncture in 1, hypoxia in 1, laryngospasm in

1, and unknown for 2.

A majority of patients required admission to the hospital (71%, 35/49), with medical admissions outnumbering psychiatric admissions (28 vs 7). The remaining 14 patients were discharged to home (14%, 7/49), jail (8%, 4/49), or a county detoxification facility (6%, 3/49) (Fig. 2). The difference in average ketamine dose for those admitted to a medical bed vs those discharged from the ED approached statistical significance (5.62 +- 1.80 medical admissions vs 4.78 +- 1.31 discharged, P = .06).

Acute ethanol intoxication was listed as a discharge diagnosis in 61% (17/28) admitted to a medical bed and 57% (8/14) who were intubated. The remaining 43% (6/14) of intubated patients in our data set did not have intoxication among their discharge diagnoses that included the following: EtOH withdrawal (1), seizure with postictal state (1), acute traumatic brain injury from a motor vehicle crash (1), acute psychotic episode (2), and altered mental status of unknown etiology (1).

Fig. 1. Box-whisker plot of ketamine dose by intubation status.

78 A.M. Burnett et al. / American Journal of Emergency Medicine 33 (2015) 76–79

between the prehospital ketamine administration and the decision to admit.

We observed a wide range of weight-based ketamine doses admin- istered by paramedics. One of the purported benefits of ketamine is its wide therapeutic window, which was an attractive pharmacokinetic property for Prehospital use. The majority of research on weight estima- tion has been conducted in the pediatric population. Using tools such as the Broselow tape to standardize weight estimation has been shown to improve paramedic Drug dosing [17]. However, other studies have questioned the accuracy of these same tools [18]. Compared to ED pro- viders (nurses, medical students, and resident and staff physicians), paramedics in one study were significantly worse at estimating the weight of supine adult ED patients clothed in a hospital gown [19]. By using visual estimation alone, paramedics have been shown to misjudge patient weight by an average of 9.8 kg [20]. The best estimator of patient weight is the patient themselves, although the ability or willingness of a

Fig. 2. Box-whisker plot of ketamine dose by disposition.

Discussion

Our data indicate that 29% (95% CI, 17%-43%) of patients who re- ceived IM ketamine for prehospital chemical restraint were ultimately intubated in our ED. None of the patients intubated in our series were intubated by EMS providers in the prehospital setting. The most com- monly documented indication for intubation was inability to protect airway (n = 7). This diagnosis has a significant subjective component and may vary between individual physicians based on their comfort treating patients in a dissociated state, the need for the patient to leave the ED to obtain computed tomography scans, and the ability to assign a dedicated nurse to monitor the dissociated patient. Only one patient had documented hypoxia prior to intubation, and this patient had a blood ethanol concentration of 0.22 g/dL.

The observation of endotracheal intubation after prehospital chemi- cal restraint with ketamine does not necessarily signify iatrogenesis from the medication. The Intubation rate of 29% in our case series is not unprecedented for studies examining chemical restraint by EMS. In a study examining midazolam and droperidol for prehospital chemi- cal restraint, Martel et al described an intubation rate of 9.7% (4/41; 95% CI, 3%-24%) for patients receiving droperidol and 40% (12/30; 95% CI, 23%-59%) for those receiving midazolam [13]. In our cohort, 61% (17/28) of the patients admitted to a hospital bed and 57% (8/14) of those re- quiring intubation were diagnosed with acute ethanol intoxication. Acute drug or ethanol intoxication is common in patients requiring chemical restraint and may increase the rates of adverse events in this population. Acute ethanol intoxication has been identified as an inde- pendent predictor of hypoxia in patients undergoing chemical restraint in the ED [14]. In a population of patients with undifferentiated agita- tion who required chemical restraint in an ED, 94% (135/144) were found to have acute alcohol intoxication with postsedation respiratory depression occurring in 40% (20/25) sedated with droperidol, 57% (26/46) for ziprazadone, and 50% (24/48) for midazolam [15]. In anoth- er study, 91% (10/11) of acutely agitated patients experiencing hypoxia or airway obstruction post chemical restraint with droperidol or mid- azolam had an elevated ethanol level [16].

The prehospital administration of ketamine for chemical restraint does not mandate admission to our hospital. The admission rate report- ed in our data is higher than that previously reported for agitated pa- tients in an ED sedated with droperidol (8%), ziprazadone (25%), or midazolam (18%) [15]. A majority of patients admitted from our ED were intoxicated with ethanol, but this is not a condition that typically independently justifies hospital admission to our institution. The indica- tions for admission in the ED providers’ documentation were examined, but these were often vague and referred to ongoing workup of altered mental status. We were not able to objectively determine causality

highly agitated patient to provide this information is uncertain [21]. Given the documented difficulty with paramedic weight estimation, along with the association between increased ketamine dosage and eventual intubation, EMS guidelines should consider novel ways to im- prove individualization of prehospital medication administration and weight estimation.

There is currently no universally accepted standard for prehospital chemical restraint. The ideal agent would have rapid and predictable onset via the IM route, have no synergistic effects with ethanol, not cause respiratory depression, have no proarrhythmic Cardiac effects, and not precipitate hypotension. Benzodiazepines may potentiate the respiratory depressant effects of alcohol, and midazolam has an FDA black box warning for respiratory arrest [22]. The administration of midazolam by EMS to patients in physical restraints resulted in de- creased agitation scores compared to patients transported with physical restraints alone [23]. However, lorazepam was shown to require multi- ple doses to achieve adequate sedation [24, 25]. The onset of sedation with IM benzodiazepines varies significantly between agents and be- tween patients given the same agent. The time to sedation can be 30 +- 20 minutes with 2 mg of IM lorazepam and 18 +- 14 minutes with 5 mg of IM midazolam [26]. Butyrophenones, including haloperidol and droperidol, have been used with success in the sedation of violent patients. The onset of sedation when given IM is typically within 15 to 30 minutes, and droperidol has a more rapid onset of action than halo- peridol [7]. Both of these agents cause QTc prolongation and have been associated with death from Torsades de pointes. As of August 2014, there is a national shortage of droperidol; and this agent is not currently available in our service area. Of the medications reported in the literature as potential options for chemical restraint, benzodiazepines, droperidol, and now ketamine have been associated with intubation.

Limitations

This was a retrospective chart review and reports only descriptive statistics of the group of patients that met inclusion and had the absence of exclusion criteria. Of the 51 patients who were identified from EMS records as having received ketamine for chemical restraint, 2 were ex- cluded because of missing data. Although we do not have information for patients transferred by EMS to other receiving Tertiary care hospitals in our area, the study hospital is the only designated level 1 trauma cen- ter. This may have introducED referral bias that we believe would have biased study patients toward higher acuity, which could falsely elevate the rates of intubation and hospital admission. In addition, the authors recognize that performing regression analysis to account for other po- tential confounders (alcohol intoxication, other ingestions, baseline co- morbidities, etc) would be preferable but was not possible given the only moderate numbered retrospective series studied here. There was no clear documentation of the estimated weight determined by the treating paramedics upon which their 5-mg/kg IM ketamine dose calcu- lation was made. We believe doses of greater than 5 mg/kg of IM

A.M. Burnett et al. / American Journal of Emergency Medicine 33 (2015) 76–79 79

ketamine were the result of incorrect weight estimation, but correct weight estimation with incorrect calculation of milligram per kilogram dose cannot be ruled out. Although we describe an association between ketamine dose, intubation, and hospital admission, our data do not establish causality.

Conclusion

Intubation was observed in our ED in 29% of patients administered IM ketamine for prehospital chemical restraint. There was a positive association between higher ketamine doses and both endotracheal intubation and hospital admission. Future research should aim to define the minimum effective ketamine dose for successful chemical restraint.

Acknowledgments

We would like to thank the Regions Hospital Critical Care Research Center for assistance with statistical analysis, data collection, and manuscript preparation.

References

1. Burnett AM, Salzman JG, Griffith KR, Kroeer B, Frascone RJ. Emergency department experience with prehpostial ketamine: a case series of 13 patients. Prehosp Emerg Care 2012;16(4):553-9.
2. Ho JD, Smith SW, Nystrom PC, Dawes DM, Orozco BS, Cole JB, et al. Successful man- agement of excited delirium syndrome with prehospital ketamine: two case exam- ples. Prehosp Emerg Care 2013;17:274-9.
3. Svenson JE, Abernathy MK. Ketamine for prehospital use: new look at an old drug. Am J Emerg Med 2007;25:977-80.
4. Le Cong M, Gynther B, Hunter E, Schuller P. ketamine sedation for patients with acute agitation and psychiatric illness requiring aeromedical retrieval. Emerg Med J 2012;29:335-7.
5. Burnett AM, Watters BJ, Barringer KW, Griffith KR, Frascone RJ. Laryngospasm and hypoxia after intramuscular administration of ketamine to a patient in excited delir- ium. Prehosp Emerg Care 2012;16:412-4.
6. White PF, Way WL, Trevor AJ. Ketamine–its pharmacology and therapeutic uses. Anesthesiology 1982;56:119-36.
7. Thomas Jr H, Schwartz E, Petrilli R. Droperidol versus haloperidol for chemical re- straint of agitated and combative patients. Ann Emerg Med 1992;21:407-13.
8. Green SM, Rothrock SG, Lynch EL, Ho M, Harris T, Hestdalen R, et al. Intramuscular ketamine for pediatric sedation in the emergency department: safety profile in 1,022 cases. Ann Emerg Med 1998;31:688-97.
9. Strayer RJ, Nelson LS. Adverse events associated with ketamine for procedural seda- tion in adults. Am J Emerg Med 2008;26:985-1028.
10. Larkin GL, Beautrais AL, Powsner SM, Sanacore G, Krystal JH, Turelli RR, et al. A pro- spective open label trial of low dose ketamine for acute suicidal states in the emer- gency department. Ann Emerg Med 2010;56(3):53.
11. Gilbert EH, Lowenstein SR, Koziol-McLain J, Barta DC, Steiner J. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med 1996; 27:305-8.
12. Vilke GM, DeBard ML, Chan TC, Ho JD, Dawes DM, Hall C, et al. Excited delirium syn- drome (ExDS): defining based on a review of the literature. J Emerg Med 2012;43: 897-905.
13. Martel M, Miner J, Fringer R, Sufka K, Miamen A, Ho J, et al. Discontinuation of droperidol for the control of acutely agitated out-of-hospital patients. Prehosp Emerg Care 2005;9:44-8.
14. Wilson MP, MacDonald K, Vilke GM, Ronquillo L, Feifel D. Intramuscular ziprasidone: influence of alcohol and benzodiazepines on vital signs in the emergency setting. J Emerg Med 2013;45(6):901-8.
15. Martel M, Sterzinger A, Miner J, Clinton J, Biros M. Management of acute undifferen- tiated agitation in the emergency department: a randomized double-blind trial of droperidol, ziprasidone, and midazolam. Acad Emerg Med 2005;12:1167-72.
16. Isbister GK, Calver LA, Page CB, Stokes B, Bryant JL, Downes MA. Randomized con- trolled trial of Intramuscular droperidol versus midazolam for violence and acute behavioral disturbance: the DORM study. Ann Emerg Med 2010;56:392-401.
17. Kaji AH, Gausche-Hill M, Conrad H, Young KD, Koenig WJ, Dorsey E, et al. Emergency medical services system changes reduce pediatric epinephrine dosing errors in the prehospital setting. Pediatrics 2006;118:1493-500.
18. Nieman CT, Manacci CF, Super DM, Mancuso C, Fallon WF, et al. Use of the broselow tape may result in the underresuscitation of children. Acad Emerg Med 2006;13: 1011-9.
19. Hall II WL, Larkin GL, Trujillo MJ, Hinds JL, Delaney KD. Errors in weight estimation in the emergency department: comparing performance by providers and patients. J Emerg Med 2004;27:219-24.
20. Anglemyer BL, Hernandez C, Brice JH, Zou B. The accuracy of visual estimation of body weight in the ED. Am J Emerg Med 2004;22:526-9.
21. Fernandes CM, Clark S, Price A, Innes G. How accurately do we estimate patients’ weight in emergency departments? Can Fam Physician 1999;45:2373-6.
22. Brice JH, Pirrallo RG, Racht E, Zachariah BS, Krohmer J. Management of the violent patient. Prehosp Emerg Care 2003;7:48-55.
23. 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 service. J Emerg Med 2012;43(5):820-8.
24. Battaglia J, Moss S, Rush J, Kang J, Mendoza R, Leedom L, et al. Haloperidol, loraze- pam, or both for psychotic agitation? A multicenter, prospective, double-blind, emergency department study. Am J Emerg Med 1997;15:335-40.
25. Richards JohnR, Derlet RobertW, Duncan DavidR. Chemical restraint for the agitated patient in the emergency department: lorazepam versus droperidol. J Emerg Med 1998;16(4):567-73.
26. Nobay F, Simon BC, Levitt MA, Dresden GM. A prospective, double-blind, random- ized trial of midazolam versus haloperidol versus lorazepam in the chemical re- straint of violent and severely agitated patients. Acad Emerg Med 2004;11:744-9.