Article, Emergency Medicine

A prospective randomized, double-dummy trial comparing IV push low dose ketamine to short infusion of low dose ketamine for treatment of pain in the ED

a b s t r a c t

Study objective: Compare adverse effects and Analgesic efficacy of low-dose ketamine for acute pain in the ED ad- ministered either by single Intravenous push or short infusion (SI).

Methods: Patients 18-65, presenting to ED with acute abdominal, flank, or Musculoskeletal pain with initial pain score >= 5, were randomized to ketamine 0.3 mg/kg by either IVP or SI with placebo double-dummy. Adverse ef- fects were evaluated by side effects Rating Scale for Dissociative Anesthetics (SERSDA) and Richmond Agitation- Sedation Scale (RASS) at 5, 15, 30, 60, 90, and 120 min post-administration; analgesic efficacy was evaluated by numerical rating scale (NRS).

Results: 48 patients enrolled in the study. IVP group had higher overall rates of feeling of unreality on SERSDA scale: 92% versus 54% (difference 37.5%; p = 0.008; 95% CI 9.3-59.5%). At 5 min median severity of feeling of un- reality was 3.0 for IVP versus 0.0 for SI (p = 0.001). IVP also showed greater rates of sedation on RASS scale at 5 min: median RASS -2.0 versus 0.0 (p = 0.01). Decrease in mean pain scores from baseline to 15 min was sim- ilar across groups: 5.2 +- 3.53 (95% CI 3.7-6.7) for IVP; 5.75 +- 3.48 (95% CI 4.3-7.2) for SI. There were no statis- tically significant differences with respect to changes in vital signs and need for Rescue medication.

Conclusion: Low-dose ketamine given as a short infusion is associated with significantly lower rates of feeling of unreality and sedation with no difference in analgesic efficacy in comparison to intravenous push.

(C) 2017

Introduction

Background

Ketamine’s role as a safe and effective modality for treating pain in the emergency department (ED) has recently been expanding [1]. Keta- mine possesses potent analgesic, amnestic, and anesthetic properties. It is a non-competitive N-methyl-D-aspartate (NMDA1) and glutamate re- ceptor antagonist that decreases central sensitization and “wind-up” phenomenon at the level of the spinal cord (dorsal ganglion) and cen- tral nervous system and provides anti-hyperalgesia, anti-allodynia, and anti-tolerance [2]. The NMDA receptor blockade leads to decreases in acute pain, opioid tolerance, opioid-induced hyperalgesia, as well as

* Corresponding author.

E-mail address: [email protected] (S. Motov).

decreases in persistent chronic (allodynia) and Neuropathic pain [2,3]. Ketamine given intravenously in low (sub-dissociative, analgesic) doses (0.1-0.3 mg/kg) provides effective analgesia with minimal effects on hemodynamics, cognition or consciousness [3].

Importance

A growing body of evidence from clinical trials conducted in the ED and pre-hospital settings supports the use of low-dose ketamine (LDK2) when administered as either an adjunct or single agent for a variety of acute and chronic painful conditions. These trials demonstrate signifi- cant opioid-sparing and analgesic efficacy that is comparable to opioids [4-8]. However, these studies also show significant rates (14%-80%) of bothersome adverse effects that include feeling of unreality, sedation,

1 Non-competitive N-methyl-D-aspartate. 2 Low-dose ketamine.

http://dx.doi.org/10.1016/j.ajem.2017.03.004

0735-6757/(C) 2017

nausea, vomiting, and dizziness. These adverse effects are apparent in the first several minutes after intravenous push (IVP3) administration and are typically short-lived. For that reason, at our institution, we mod- ified our protocol for the administration of LDK from an IVP over 3 to 5 min to a short infusion (SI4) over 10 to 15 min in order to mitigate these effects. To date, no trial has directly compared intravenous push dose to short infusion of LDK.

Goals of this investigation

The goal of our study was to evaluate the rates of adverse effects and the analgesic efficacy of intravenous push (IVP) dose of LDK in compar- ison to short infusion (SI) for adult patients presenting to the ED with a variety of acute painful conditions. We hypothesized that administra- tion of LDK by SI would be associated with decreased rates of adverse ef- fects and would have similar analgesic efficacy when compared to IVP dose.

Methods

Study design and setting

Our study was a prospective, randomized, double-blind, double- dummy trial comparing safety and analgesic efficacy of intravenous LDK given as a push dose (over 5 min) versus given as a short infusion (over 15 min). The study was conducted at a 711-bed urban community teaching hospital with an annual ED census of N 120 000 visits. Patient screening, enrollment, and data collection were performed by study in- vestigators (SM, CF, JD, MM, MY, and IP).

This study was approved by the Maimonides Medical Center Institu- tional Review Board and registered with ClinicalTrials.gov (NCT02363270). Written and signed informed consent was obtained in accordance with institutional policy. We report this trial in accor- dance with the Consolidated Standards of Reporting Trials statement [9].

Selection of participants

Patients considered for enrollment into the study comprised adults aged 18 to 65 years who presented to the ED primarily for management of acute abdominal, flank, back, traumatic chest or musculoskeletal pain with an intensity of 5 or greater on a standard 0-10 Numeric pain rating scale (NRS5). Patients had to be awake, alert and oriented to person, place, and time, and able to demonstrate understanding of the informed consent process and content. Patients also had to demonstrate ability to verbalize the nature of any adverse effects they might experience as well as to express their severity of pain using the NRS. Exclusion criteria included pregnancy, breast-feeding, altered mental status, allergy to ke- tamine, weight b 46 kg or N 115 kg, unstable vital signs (systolic blood pressure b 90 or N 180 mm Hg, pulse rate b 50 or N 150 beats/min, and respiration rate b 10 or N 30 breaths/min), and medical history of acute head or eye injury, seizure, Intracranial hypertension, renal or hepatic insufficiency, alcohol or drug abuse, psychiatric illness, or recent (4 h before) analgesic use. For the purposes of this study, LDK was used without co-administration of any other analgesics, with the exception of rescue medication.

Enrollment of patients occurred between April 2015 and August 2016. Screening and enrollment took place Monday through Friday 8 AM to 8 PM when an ED pharmacist was available for blinded medica- tion preparation. Study investigators approached all potentially qualify- ing participants after they were evaluated by the treating emergency physician and determined to meet study inclusion criteria. All

3 Intravenous push.

4 Short infusion.

5 Numeric Pain Rating Scale.

participants provided written informed consent and Health Insurance Portability and Accountability Act authorization. For non-English speakers, a language appropriate consent form was used and non-inves- tigator, hospital-employed, trained interpreters assisted in acquisition of informed consent.

Interventions

Participants were allocated to two groups according to a predetermined randomization list that was generated using SPSS 20.0 by the research manager. Participants received intravenous LDK of

0.3 mg/kg by either (1) IVP given over 5 min or (2) mixed in 100 ml nor- mal saline solution and given via SI over 15 min. All participants re- ceived a corresponding placebo in order to maintain double-dummy design. Accordingly, all patients randomized to receive LDK via IVP also received 100 ml of normal saline via SI, and all patients randomized to receive LDK via SI also received 10 ml of normal saline via IVP. ED pharmacy investigators maintained the randomization list, prepared the medication, and delivered it in blinded fashion to the treating regis- tered nurse, who administered it to the study participant. The ED phar- macist set up the SI dose on an infusion pump with a 15 min run time and the IVP dose on a syringe pump with a 5 min run time. The two ad- ministration routes started simultaneously.

Methods and measurements

The preparing ED pharmacist, research manager, and statistician were the only ones with knowledge of the study arm to which each par- ticipant was randomized. Treating providers, participants, and the data collecting research team were blind to the medication route received. Study investigators consisted of two treating physicians who assisted in screening and supervised the research fellow and research coordina- tors who enrolled patients, recorded pain scores, vital signs, and adverse effects at baseline, 5, 15, 30, 60, 90, and 120 min. For subjects still desir- ing pain medication 30 min after study drug administration, investiga- tors offered Intravenous morphine at 0.1 mg/kg as a rescue analgesic.

Outcomes

The main outcomes included: 1) the overall rates as well as the spe- cific severity levels of the side effects which were recorded in accor- dance with the Side Effects Rating Scale for Dissociative Anesthetics (SERSDA6) [10]. The SERSDA measures the severity of nine adverse ef- fects based on a five point scoring system from “0” (adverse effect ab- sent) to “4” (adverse effect is very bothersome); and 2) severity of agitation and/or sedation in accordance to the nine point Richmond Ag- itation-Sedation Scale (RASS7) [11] with scores ranging from “-4” (deeply sedated) to “0” (alert and calm) to “+4” (combative). Second- ary outcomes consisted of (1) a standard 11 point NRS score for self-re- ported Pain severity, (2) changes in vital signs, and (3) need for Rescue analgesia.

Data analysis

Research staff recorded all data on data collection sheets (separate from clinical data). These were entered into Microsoft Excel, and then imported into the programs used for statistical analysis. Baseline char- acteristics of patients in each treatment group were described in terms of mean +- standard deviation for normally distributed variables and frequency (percent) in the case of categorical variables. Student t– tests were used to compare simple group differences in terms of means (e.g., age), while chi square tests were used to look at differences in terms of percent rates (e.g., sex). A Generalized linear model with an

6 Side Effects Rating Scale for Dissociative Anesthetics.

7 Richmond Agitation-Sedation Scale.

underlying gamma distribution was used for the analysis over time of all ordinal (e.g., SERDSA unreality score) while mixed model regression was used for continuous outcomes (e.g., pain). This analysis allowed for inclusion of all patients regardless of any missing data under the as- sumption of data being missing at random. Only time points prior to when either the SERSDA or RAS scores had completely resolved to “0” were included for the generalized linear model analysis since otherwise lack of variability did not allow for model effects or test statistics to be estimated.

A bootstrap algorithm based on 1000 randomly selected samples from the data was used to estimate non-parametric 95% confidence limits for the median differences [12]. All statistical analyses were car- ried out using SAS 9.4 (SAS Inc., Cary, NC). Based on the study of Andolfatto et al. (2013) [13] of Intranasal ketamine usage in the ED, we estimated a standard deviation = 1.20 for severity of any adverse event on the SERSDA. Assuming at least a one unit difference between the two administration methods (on a scale from 0 to 4), we further es- timated an effect size d = 0.83. A sample size of 24 patients per group was calculated to be necessary for 80% power to detect this effect size

with alpha = 0.05. No adjustment of the significance level was made to control for multiple comparisons in order to maximize the power of detecting any adverse event.

Results

We enrolled 48 subjects (24 in the IVP group and 24 in the SI group) into our study (Fig. 1). The groups were similar in terms of demographic characteristics (Table 1). Mean ages were 42.2, 43.6 years old, respec- tively with 37.5% and 50.0% males, also respectively. Mean baseline NRS pain scores in the two groups were both N 8.0 and not significantly different from each other. The first dropout occurred at 60 min in the IVP group; 21 participants in each group were still observable at 120 min.

Main outcomes

As shown in Table 2, overall rates of adverse effects as measured on SERSDA scale (i.e., subject presented with the adverse effect at any time

Fig. 1. Study flow diagram for consented subjects.

Table 1

Baseline patient characteristics.

Characteristics

Push

Drip

p-Value

No. of subjects

24

24

Age

42.2 (15.1)a

43.6 (12.3)

0.67

Male sex

9 (37.5%)b

12 (50.0%)

0.56

Pain

8.92 (1.4)a

8.20 (1.4)

0.11

Source of pain Abdominal

8 (33.3%)b

17 (70.8%)

0.06

Back

4 (16.7%)

1 (4.2%)

Flank

8 (33.3%)

3 (12.5%)

Other

4 (16.7%)

3 (12.5%)

a Mean (standard deviation).

b Frequency (percent).

point during the study period) were similar between the two groups with the exception of overall feeling of unreality, which was significant- ly higher in the IVP group: 91.7% for IVP versus 54.2% for SI (percent dif- ference 37.5%; p = 0. 008). The median SERSDA scores for unreality were likewise significantly higher for the IVP group up to 30 min (p = 0.004). Looking more closely at severity of feelings of unreality over time, this difference was most marked at 5 and 15 min, as shown in Fig. 2. At 5 min, the median for feeling unreality score was 3.0 for the IVP group versus 0 for the SI group (median difference = 3; (p = 0.001); 95% CI 1 to 4). In the IVP group, 2 (8%) patients reported no ad- verse effects (0 score) while 11 (46%) reported the feeling of unreality to be very bothersome (score 4); in contrast, 13 (54%) of the SI groups reported no change while 4 (17%) reported their symptoms as very bothersome. The IVP group was still elevated with regard to the feeling of unreality at 15 min although the difference was no longer significant. The median score was 2 for the IVP group versus 0 for SI (median differ- ence = 2; p = 0.14; 95% CI -1 to 3).

The IVP group also showed a significantly greater degree of sedation

at 5 min (p = 0.01) as measured by the Richmond Agitation-Sedation Scale (RASS). As shown in Fig. 3, the median RASS score was – 2 for the IVP group versus a median RASS of 0 for the SI group (median differ- ence = – 2; 95% CI -2 to – 0.5). In the IVP group at 5 min, 13 (54%) showed light to deep sedation (scores <=- 2) and 6 (25%) were alert and calm (score 0). No patients in the SI group had scores <=-2, while 6 (25%) were drowsy (score -1) and 15 (62%) were alert and calm.

Secondary outcomes

From baseline to 15 min (Fig. 4), both groups showed similar signif- icant decreases in pain scores as measured by NRS. Mean pain decreased by 5.17 +- 3.53 (95% CI 3.67 to 6.66) in the IVP group and by 5.75 +- 3.48 (95% CI 4.28 to 7.22) in the SI group; p-value for overall difference =

0.026. There were no differences in analgesic efficacy between the treat- ment groups at any time point in the study (p = 0.14). Likewise, there were no statistically significant differences between the groups with

Table 2

Rates of SERDSA adverse effects encountered in each group at any point throughout the study period

Adverse effects

Group

p-Value

Push

Drip

Fatigue

1 (4.2%)a

2 (8.3%)

0.55

Dizziness

16 (66.7%)

18 (75%)

0.75

Headache

4 (16.7%)

4 (16.7%)

1.00

Unreality

22 (91.7%)

13 (54.2%)

0.008

Hearing

0

0

NA

Vision

6 (25%)

9 (37.5%)

0.53

Mood change

3 (12.5%)

2 (8.3%)

0.64

Discomfort

6 (25%)

4 (16.7%)

0.72

Hallucination

2 (8.3%)

3 (12.5%)

0.64

a Frequency (percent).

Fig. 2. Severity of feeling of unreality (according to SERSDA severity score).1

respect to the use of rescue morphine analgesia (Table 3) or changes in vital signs.

Limitations

This was a single center study in which subjects were enrolled as a convenience sample based on availability of members of both the ED re- search and pharmacy teams. This may have led to selection bias or under-representation of patients who may present to the ED late at night. Our stringent exclusion criteria and sample size of 48 subjects were inadequate to assess variance in safety of the two different admin- istration routes of study medication.

Fig. 3. Agitation (+) versus sedation (-) according to Richmond Agitation-Sedation Scale.1

Fig. 4. Boxplots of pain for two methods of ketamine administration.

Discussion

Sub-dissociative dose ketamine analgesia in the ED is gaining recog- nition as a viable adjunct and alternative to opioid analgesics for manag- ing a variety of acute and chronic painful conditions. To date, the high incidence of unpleasant (bothersome) adverse effects associated with ketamine administration has proven to be a significant barrier to the promotion of more widespread and well-tolerated utilization of this an- algesic modality. It has remained unclear if the rate of intravenous ad- ministration influences the Adverse effect profile of LDK. Several clinical trials evaluated the role of short-term (over 10 min) and contin- uous LDK infusion on frequency of adverse effects and analgesic efficacy. Goltser et al. utilized a short infusion of LDK analgesia in 14 ED patients with acute and chronic painful conditions by administering 0.3 mg/kg over 10 min and demonstrated acceptable pain relief in 11 patients (NRS reduction of N 3) and minor side effects in only two patients (diz- ziness and tinnitus) [14]. Similarly, Ahern et al. prospectively adminis- tered 15 mg of intravenous ketamine that was immediately followed by a continuous infusion of 20 mg/h for 1 h to 38 patients with acute pain. At the 10 min mark, 7 patients were pain-free and 25 and 26 pa- tients had significant pain relief (NRS reduction N 3) at 60 and 120 min, respectively. However, 87% of patients experienced adverse ef- fects of nausea, fatigue, headache, and feeling of unreality [15]. We pos- tulated that the high observed rates of bothersome adverse effects from LDK administration may be mediated by the rate of initial bolus dose administration.

Multiple studies have demonstrated a clear correlation of low-dose ketamine-induced side effects, particularly feeling of unreality and diz- ziness, with rapid rates of infusion. This phenomenon can be explained

Table 3

Rates of rescue morphine over time by number of patients/group

Time of rescue med.

Group

p-Value

Push

Drip

30 mina

1/24 (4.2%)b

1/24 (4.2%)

1.00

60 min

4/23 (17.4%)

2/24 (8.3%)

0.42

90 min

2/22 (9.1%)

4/23 (17.4%)

0.66

120 min

1/21 (4.8%)

0/21 (0.0%)

1.00

a No rescue morphine given before 30 min.

b Frequency (percent).

by the high lipophilicity of ketamine which allows for rapid penetration of the blood-brain barrier and immediate saturation of the NMDA/gluta- mate receptor complex. In contrast, a slow (over 10-15 min) infusion of the same dose of ketamine leads to a steady saturation of the receptors with a decrease in neuropsychiatric side effects.

In order to test this hypothesis, we conceived a novel double-blind, double-dummy randomized clinical trial that directly compares rates of adverse effects of sub-dissociative dose ketamine given via intrave- nous push to those when given by short infusion. The trial results dem- onstrate that administration of LDK by SI rather than IVP significantly reduces the two major bothersome adverse effects of feeling of unreal- ity and sedation. These results validate our site’s recent LDK protocol change from IVP to SI administration.

One of the possible barriers to short infusion of low-dose ketamine is a practical one that requires availability of the infusion pump which must be programmed by a nurse. However, modern infusion pumps that are widely available in EDs have the capability of being pre-pro- grammed for commonly used medications, including ketamine, and re- quire only patients’ weights to be entered, which reduces the possibility of drug and dosing pump errors. In addition, these pumps have the ad- vantage of great ease and speed of set-up without delaying provision of analgesia. Based on our experience, formal monitoring of patients re- ceiving low-dose ketamine infusion is not necessary for patient safety. We do formally monitor patients for research purposes, but in our clin- ical practice we do not routinely monitor patients in our ED when short infusion of low-dose ketamine is being administered.

Our study demonstrated significantly higher rates of feeling of unre- ality and sedation when LDK was administered as an intravenous push. While these side effects are not dangerous to patients, they are very un- pleasant and in some cases exquisitely bothersome which can lead to patients requesting termination of therapy. Therefore it has become our practice to administer low-dose ketamine in pre-programmed infu- sion pumps. These pumps have the advantage of great ease and speed of set up without any added cost, as infusions up to 15 min in duration are billed the same as intravenous push doses.

In conclusion, intravenous sub-dissociative dose ketamine given as a short infusion significantly decreases rates of feeling of unreality and se- dation in the first 15 min of administration without sacrificing analgesic efficacy. These results support more widespread and well-tolerated use of ketamine as an adjunct or alternative to opioid analgesics for treating pain in the emergency department.

Additional contributions

John Marshall, MD for his support and guidance; Maryam Zaeem, PharmD, Russell Bardsley, PharmD, and Nechama Rothberger, PharmD for medication administration to study patients, and all the volunteers for their assistance. Authors acknowledge and thank all the ED nurses for their tireless help and support of this project.

Meetings

The study was accepted for oral presentation at the May 2017 Soci- ety for Academic Emergency Medicine Conference in Orlando, Florida.

Grant support

None.

Conflicts of interest

All authors have completed and submitted the ICMJE Form for Dis- closure of Potential Conflicts of Interest. The authors have no indepen- dent disclosures or conflicts of interest.

Author contributions

A. Likourezos and P. Homel had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Motov, Fromm.

Acquisition, analysis, or interpretation of data: All authors.

Statistical analysis: Likourezos, Homel.

Drafting of the manuscript: Drapkin.

Critical revision of the manuscript for important intellectual content: Motov, Homel, Fromm.

Obtained funding: Fromm.

Study supervision: Motov, Fromm.

References

  1. Gao M, Rejaei D, Liu H. Ketamine use in current clinical practice. Acta Pharmacol Sin 2016;37(7):865-72.
  2. Gorlin AW, Rosenfeld DM, Ramakrishna H. Intravenous sub-anesthetic ketamine for perioperative analgesia. J Anaesthesiol Clin Pharmacol 2016;32(2):160-7.
  3. Kurdi MS, Theerth KA, Deva RS. Ketamine: current applications in anesthesia, pain, and critical care. Anesth Essays Res 2014;8(3):283-9.
  4. Johansson P, Kongstad P, Johansson A. The effect of combined treatment with mor- phine sulphate and low-dose ketamine in a prehospital setting. Scand J Trauma Resusc Emerg Med 2009;17:61.
  5. Galinski M, Dolveck F, Combes X, Limoges V, Smail N, Pommier V, et al. Management of severe acute pain in emergency settings: ketamine reduces morphine consump- tion. Am J Emerg Med 2007;May;25(4):385-90.
  6. Ahern TL, Herring AA, Stone MB, Frazee BW. Effective analgesia with low-dose keta- mine and reduced dose hydromorphone in ED patients with severe pain. Am J Emerg Med 2013;31(5):847-51.
  7. Beaudoin FL, Lin C, Guan W, Merchant RC. Low-dose ketamine improves pain relief in patients receiving Intravenous opioids for acute pain in the emergency depart- ment: results of a randomized, double-blind, clinical trial. Acad Emerg Med 2014; 21(11):1193-202.
  8. Motov S, Rockoff B, Cohen V, Pushkar I, Likourezos A, McKay C, et al. Intravenous sub-dissociative-dose ketamine versus morphine for analgesia in the emergency de- partment: a randomized controlled trial. Ann Emerg Med 2015;22(3):251-7.
  9. CONSORT. Statement http://www.consort-statement.org/consort-2010. (Accessed on 9/25/16).
  10. Eide Stubhaug A, Oye I, Breivik H. Continuous subcutaneous administration of the N methyl-D-aspartic acid (NMDA) receptor antagonist ketamine in the treatment of post-herpetic neuralgia. Pain 1995;61(2):221-8.
  11. Ely EW, Truman B, Shintani A, Thomason JW, Wheeler AP, Gordon S, et al. Monitor- ing sedation status over time in ICU patients: reliability and validity of the Richmond Agitation-Sedation Scale (RASS). JAMA 2003;289:2983-91.
  12. Efron B, Tibshirani R. An introduction to the bootstrap. Boca Raton, FL: Chapman & Hall/CRC; 1998.
  13. Andolfatto G, Willman E, Joo D, Miller P, Wong WB, Koehn M, et al. Intranasal keta- mine for analgesia in the emergency department: a prospective observational series. Acad Emerg Med 2013;20(10):1050-4.
  14. Goltser A, Soleyman-Zomalan E, Kresch F, Motov S. Short (low-dose) ketamine infu- sion for managing acute pain in the ED: case-report series. Am J Emerg Med 2015; 33(4):601.e5-7.
  15. Ahern TL, Herring AA, Miller S, Frazee BW. Low-dose ketamine infusion for emer- gency department patients with severe pain. Pain Med 2015;16(7):1402-9.

Leave a Reply

Your email address will not be published. Required fields are marked *