- Email: [email protected]
Droperidol vs. prochlorperazine for the treatment of acute headache
The Journal of Emergency Medicine, Vol. 26, No. 2, pp. 145–150, 2004 Copyright © 2004 Elsevier Inc. Printed in the USA. All rights reserved 0736-4679/04 $–see front matter
doi:10.1016/j.jemermed.2003.05.005
Original Contributions DROPERIDOL VS. PROCHLORPERAZINE FOR THE TREATMENT OF ACUTE HEADACHE Christopher S. Weaver, MD,* James B. Jones, MD, PharmD,* Carey D. Chisholm, MD,* Michael J. Foley, MD,* Beverly K. Giles, RN,* Geoffrey G. Somerville, BS, EMT-P,* Edward J. Brizendine, MS,† and William H. Cordell, MD *Department of Emergency Medicine and †Department of Biostatistics, Indiana University School of Medicine, Indianapolis, Indiana, USA Reprint Address: Christopher S. Weaver, MD, Department of Emergency Medicine, Indiana University School of Medicine OPW M200, 1001 W. Tenth Street, Indianapolis, IN 46202
e Abstract—To determine if droperidol i.v. is as effective as prochlorperazine i.v. in the emergency department (ED) treatment of uncomplicated headache, a randomized, controlled, blinded study was conducted in the Emergency Departments of two urban teaching hospitals. Patients > 18 years old with crescendo-onset headache were eligible for inclusion. Ninety-six patients (48 in each group) were randomized to receive droperidol 2.5 mg i.v. or prochlorperazine 10 mg i.v. Baseline characteristics were similar between the two study groups. For the main study outcome, 83.3% in the droperidol group and 72.3% in the prochlorperazine group reported 50% pain reduction at 30 min (p < .01; one-sided test of equivalence). The mean decrease in headache intensity was 79.1% (SD 28.5%) in the droperidol group and 72.1% (SD 28.0%) in the prochlorperazine group (p ⴝ .23). It is concluded that droperidol i.v. provided a similar reduction of headache as achieved with prochlorperazine i.v. with a similar incidence of akathisia. © 2004 Elsevier Inc.
seek care for headache in EDs because of the severity of the pain and associated symptoms, concerns about the etiology, lack of alternative venues of healthcare, and “last straw” desperation. The treatment options for the emergency treatment of headache greatly expanded after the 1986 study by Callaham and Raskin of dihydroergotamine therapy in headache that serendipitously demonstrated headache relief after pre-treatment with prochlorperazine (4). Jones et al. confirmed the efficacy of prochlorperazine i.v. in a randomized double-blind trial published in 1989 (5). Over the last 15 years, prochlorperazine has been accepted as a standard treatment for headache in the ED (6,7). Additional medications, including parenteral metoclopramide and the triptan drugs administered by various routes, have expanded the options for the emergency treatment of acute headache (8). Droperidol (Inapsine威) is a high potency, rapid-acting butyrophenone similar to haloperidol that has been used in the United States since approval by the Food and Drug Administration (FDA) in 1970. Droperidol exerts its mechanism of action via blockade of dopamine receptors in the subcortical, midbrain, and brainstem reticular formation and produces mild alpha-adrenergic blockade and peripheral vascular dilatation. Recommended doses for these indications range from 0.625 mg i.v./i.m. up to 10 mg, depending on the indication, although much larger doses have been used (9). Droperidol is used clinically for rapid tranquilization of violent patients, as an adjunct
e Keywords—headache; pain; droperidol; prochlorperazine
INTRODUCTION Headache is a common complaint in patients presenting to the Emergency Department (ED) (1–3). Many patients Presented at the Society for Academic Emergency Medicine Annual Meeting, St. Louis, MO, in May 2002. Currently, Dr. Jones works for Pharmacia Corporation, Peapack, New Jersey and Dr. Foley is an emergency physician in Madison, Wisconsin.
Original Contributions is coordinated by John Marx,
RECEIVED: 30 September 2002; FINAL ACCEPTED: 12 May 2003
SUBMISSION RECEIVED:
MD,
of Carolinas Medical Center, Charlotte, North Carolina
1 April 2003;
145
146
C. S. Weaver et al.
to anesthesia, and for the treatment of nausea and vomiting. More recently, anecdotal experience and several case series and clinical studies suggest the efficacy of droperidol in the acute treatment of headache (10 –13). We are unaware of any study that has compared droperidol i.v. to prochlorperazine i.v. in a randomized, double-blind fashion. We therefore conducted such a study to evaluate whether droperidol i.v. is as effective as prochlorperazine i.v. in the ED treatment of uncomplicated headache.
MATERIALS AND METHODS We conducted a randomized, blinded, controlled trial in the EDs of Wishard Memorial and Methodist Hospital. Both are teaching hospitals located in urban Indianapolis, Indiana. The annual ED census was 105,000 at Wishard and 94,000 at Methodist at the time of the study. We enrolled all eligible patients in a convenience sample from August 7, 2000 to November 28, 2001 based on research coordinator coverage. Patients 18 years of age or older were eligible for participation in this study if they presented to ED triage with a crescendo-onset headache and had a normal neurological examination. A specific classification of headache (e.g., vascular vs. tension) was neither defined nor required for study entry in an effort to assess the medications’ effectiveness in treating all uncomplicated headaches (5). Patients were excluded if they were experiencing their first headache, had a temperature ⱖ 38°C (100.4°F), exhibited nuchal rigidity, or had thunderclap onset of the headache. Other exclusion criteria included self treatment with a pain medication or antiemetic in the 4 h before arrival; history of carbon monoxide exposure, peripheral vascular disease, cancer, or HIV infection; pregnancy; allergy to study medications; inability to speak or understand English; and lack of telephone for follow-up contact. The Institutional Review Boards at both institutions approved the study. All study participants gave written informed consent before enrollment. Study participants were instructed that they could receive rescue medications for either headache or nausea and vomiting at any time and that they could terminate their participation in the study at any time. This study was internally funded and none of the investigators has or had a significant financial interest in any of the manufacturers of the study medications. Patients were randomized by means of a computergenerated randomization table to receive droperidol 2.5 mg i.v. or prochlorperazine 10 mg i.v. followed by a 2 cc normal saline flush. A contract research pharmacy prepared the study medication kits. Each kit contained two
identical vials. The study drug kit contained one vial of droperidol 2.5 mg and one vial of normal saline. The control drug kit contained two vials, each containing prochlorperazine 5 mg. To maintain the blind, all vials contained 1 mL of solution and were identical in appearance. Instructions affixed to the kits directed the investigator to draw both vials into a single syringe and inject the contents over a 2-min period. The study participant, investigator, outcomes assessor, and data analysts were unaware of the randomization assignments. The study blind was maintained until data analysis was completed. Study participants did not receive any additional medication or i.v. fluids for headache, nausea, or vomiting during the first 30 min unless they requested to be withdrawn from the study. Rescue medications after 30 min were specified as meperidine 1 mg/kg i.v. for persisting headache, ondansetron 4 mg i.v. for persisting nausea or vomiting, and diphenhydramine hydrochloride 25–50 mg i.v. for extrapyramidal side effects. The main study outcome was the number of subjects in each group achieving at least 50% pain relief at 30 min. Secondary outcome measures included mean change in pain intensity, the proportion requiring rescue medications from 30 to 60 min, and the incidence of akathisia and other adverse events. Study participants were asked to record pain and nausea intensity and level of alertness on three separate 100-mm visual analog (VAS) scales bounded by the words “no pain”-“worst pain,” “no nausea”-“worst nausea,” and “sound asleep”“wide awake,” respectively, at baseline, 15 min, 30 min, 45 min, and 60 min (14). Our rating scale for akathisia was modified from the Prince Henry Hospital (PHH) Rating Scale of Akathisia and was employed in a previous study of prochlorperazine-induced akathisia by our research group (15,16). The rating scale included “subjective ratings” elicited from study participants by direct questioning and “objective ratings” assessed by investigator observation. Akathisia was defined as the occurrence of either or both of the following: spontaneous report or change in both the objective and subjective akathisia rating score compared to baseline. Other adverse events were recorded only if they were spontaneously reported at each of the time study intervals. Study participants were discharged from the ED at the discretion of the treating physician (a board certified emergency physician or a resident under the direction of a board certified emergency physician) and were contacted 24 h after discharge from the ED to assess the presence and intensity of headache and to ascertain if they had taken any additional medication. Participants rated their pain at 24 h on a verbal 0 to 10 pain scale. The study participants were also asked if they had returned to work or daily activities.
Droperidol vs. Prochlorperazine for Headache
147
Table 1. Baseline Characteristics of Trial Participants Total (n ⫽ 96) Female, number (%) Race/ethnicity, number (%) White Black Other Age, median (range), years Headache intensity, median (range), mm VAS Level of alertness, median (range), mm VAS Nausea intensity (for those reporting nausea), median (range), mm VAS
83 (86.5) 45 (46.9) 49 (51.0) 2 (2.1) 31 (18–68) 76.5 (18–100) 100 (88–100) 52 (5–95) n ⫽ 41
Sample size calculations were based on demonstrating equivalency between droperidol and prochlorperazine. This study was designed as a non-inferiority study rather than a superiority study using an equivalency window (⌬) of ⫺10%. Restated, equivalence would be demonstrated if the proportion of droperidol subjects experiencing at least a 50% reduction in headache intensity (the main study outcome) were not more than 10% less than the proportion of prochlorperazine subjects experiencing at least a 50% reduction. Collins et al. reported that 75% of the subjects receiving prochlorperazine i.v. experienced a 50% or more reduction in their headache intensity at 30 min (16). Based on this result and our equivalence definition, a sample of 230 subjects per group would have 80% power to demonstrate that droperidol is as effective as prochlorperazine in headache relief. Sample size calculations were based on using a one-sided test of equivalency with a significance level of ␣ ⫽ 0.05 and performed using nQuery Advisor version 4.0 (Statistical Solutions, Cork, Ireland). An intention-to-treat analysis was performed. For the primary outcome of the proportion of subjects experiencing at least a 50% reduction in pain at 30 min, a onesided test of equivalence was performed (17). The null hypothesis for this test was that the difference between the droperidol and prochlorperazine percentages was less than or equal to ⫺10%. A one-sided 95% confidence interval (CI) for the true difference in percentages was estimated. For all other test of hypotheses, a null hypothesis of no difference was assumed. To assess if the rate of pain relief was different between the two medications, a repeated measures analysis of variance (ANOVA) was performed. Generalized estimating equations were used to adjust parameter estimates for the repeated measurements made on each subject (18). Statistical analyses were performed using SAS version 8.2 (SAS Institute, Cary, NC) and S-Plus 2000 (MathSoft, Seattle, WA). Summary statistics for continuous data are presented as
Droperidol (n ⫽ 48) 44 (91.7) 22 (45.8) 26 (54.2) 0 (0.0) 30 (18–68) 68 (18–100) 100 (100–100) 50 (5–95) n ⫽ 19
Prochlorperazine (n ⫽ 48) 39 (81.3) 23 (47.9) 23 (47.9) 2 (4.2) 34 (19–64) 79 (21–100) 100 (88–100) 52 (23–86) n ⫽ 22
p value 0.14 0.33 0.27 0.07 0.33 0.37
the estimated median and range, whereas for categorical data, the frequency and percent are presented.
RESULTS We enrolled 96 non-consecutive patients between August 7, 2000 and December 10, 2001. Our study was terminated early due to chronic intermittent shortages of prochlorperazine as well as an FDA “black box” warning issued December 4, 2001 regarding QT prolongation and torsade de pointes in patients treated with droperidol (19,20). There were 48 study participants in each of the groups. All 96 participants remained in the study throughout the 60-min evaluation. There were two missing data points for headache intensity assessment— one in the droperidol group at 60 min and one in the prochlorperazine group at 30 min. Patient characteristics at baseline were similar between groups (Table 1). For the main study outcome, 83.3% in the droperidol group and 72.3% in the prochlorperazine group had 50% pain reduction at 30 min (Table 2). A one-sided test of equivalency yielded a p value of ⬍ 0.01, suggesting that droperidol was not inferior to prochlorperazine using an equivalency window (⌬) of ⫺10%. The 95% one-sided lower confidence limit for the difference was ⫺2.9%. This value is greater than the ⫺10% limit of equivalency. The mean decrease in headache intensity at 30 min was 79.1% (SD 28.5%) in the droperidol group and 72.1% (SD 28.0%) in the prochlorperazine group (p ⫽ 0.23; 95% CI ⫺4.6, 18.5). There was no significant difference in the rate of decrease in pain intensity between the two study groups over the first 60 min (p ⫽ 0.50) (Figure 1). Nor were there significant differences at 30 min for percent decrease in nausea (p ⫽ 0.22) or level of alertness (p ⫽ 0.98). Six study participants in each group required rescue analgesics (p ⫽ 1.0, 95% CI ⫺13.2, 13.2). A total of 14 (14.6%) study participants experienced akathisia during the first 60 min, 5 (10.5%) in the
148
C. S. Weaver et al.
Table 2. Categorical Relief of Headache Intensity at 30 Minutes
Percent Relief
Droperidol (n ⫽ 48)
Prochlorperazine (n ⫽ 47)
ⱖ 50% relief, number (%) ⱖ 75% relief, number (%) 100% relief, number (%)
40 (83.3) 31 (64.6) 26 (54.2)
34 (72.3) 26 (55.3) 18 (38.3)
p value* ⬍ 0.01 0.03 ⬍ 0.01
One-sided 95% CI for difference† ⫺2.9, 100 ⫺7.2, 100 ⫺0.7, 100
* Using a 1-sided test of equivalence (⌬ ⫽ ⫺10%). † Upper limit is reported as 100% because this is the maximum of all possible values of the difference of two proportions.
droperidol group and 9 (18.8%) in the prochlorperazine group (p ⫽ 0.25; 95% CI ⫺22.4, 5.7). No other adverse events were spontaneously reported in either group. We were able to contact 83 (86.5%) of the study participants 24 h after discharge (40 in the droperidol group and 43 in the prochlorperazine group). At 24 h follow-up, 27.5% (11/40) of the droperidol group reported headache compared to 34.8% (15/43) of the prochlorperazine group (p ⫽ 0.47; 95% CI ⫺27.2, 12.5). The mean headache intensity at 24 h for the droperidol group was 4.7 (SD 2.1) compared to 5.0 (SD 2.7) in the prochlorperazine group (p ⫽ 0.78) using a verbal 0 –10 pain intensity scale. In the droperidol group, 13 (32.5%) reported use of supplemental analgesics since ED discharge compared to 18 (41.9%) in the prochlorperazine group (p ⫽ 0.38). At 24 h after discharge, 2.5% (1/40) of the droperidol group versus 14% (6/43) of the prochlorperazine group had akathisia (p ⫽ 0.06; CI ⫺22.9%, 0.0%). It is important to emphasize that the 24-h akathisia follow-up was based only on the patient’s self-report and not on the investigator’s rating, as was done during the first 60 min. Twenty-seven (67.5%) of the droperidol group and 28 (65.1%) of the prochlorperazine group
Figure 1. Mean pain intensity after medication administration. Bars represent 95% Confidence Intervals.
reported they had returned to normal daily activities (p ⫽ 0.82). DISCUSSION Droperidol has been used clinically for three decades for tranquilization of violent patients, anesthesia and neuroleptanalgesia, treatment of nausea and vomiting, and, more recently, headache. The results of our study add to the growing evidence of the efficacy and safety of droperidol in treating headache. Our study demonstrates similar relief of headache intensity measured at 30 min and 24 h between study participants receiving droperidol 2.5 mg i.v. and those receiving prochlorperazine 10 mg i.v. The incidence of akathisia was also similar between the two groups. Wang et al. studied the use of droperidol 2.5 mg i.v. every 30 min up to three doses in 35 patients with status migrainosus or refractory migraine (10). The success rate (defined as headache-free or mild headache) was 88% of the 25 patients with status migrainosus and 100% of those with refractory migraine. Seven patients received one dose, 12 received two doses, and 16 received three doses (mean 5.6 mg) of droperidol i.v. This contrasts to our study where a single 2.5 mg dose of droperidol was administered i.v. In a retrospective study of 37 patients, Richman et al. reported symptomatic relief in 81% at 30 min after droperidol 2.5 mg i.m. (11). Richman et al. then conducted a randomized, blinded comparison of droperidol 2.5 i.m. and meperidine 1.5 mg/kg i.m. in treating 29 ED patients with acute migraine as defined by International Headache Society criteria (12). The two groups were similar for mean change in VAS score and proportion requesting rescue medication. Miner et al. conducted a prospective, single-blind comparison of droperidol (either 5 mg i.m. or 2.5 mg i.v.) and prochlorperazine (either 10 mg i.v. or i.m.) in the treatment of benign headaches (13). They reported that 60.9% of the 82 study participants receiving droperidol and 44.2% of the 86 receiving prochlorperazine achieved ⱖ 50% pain relief at 30 min. Twenty-six percent of those
Droperidol vs. Prochlorperazine for Headache
who received droperidol and 18.2% of those who received prochlorperazine had rebound headache. These results compare to our study where the proportion of subjects reporting headache at 24 h was 27.5% in the droperidol group and 34.8% in the prochlorperazine group (with a follow-up rate of 86.5%). Akathisia and drowsiness are the main adverse events after droperidol therapy for headache. Richman et al., in their retrospective study, reported an 8% incidence of mild akathisia (11). In their prospective study, Richman et al. reported an incidence of akathisia of 13.3% after droperidol 2.5 mg i.m. (12). The incidence of sedation was 6.7% in the droperidol i.m. group compared to 14.3% in the meperidine (1.5 mg/kg) i.m. group. Miner et al. reported an akathisia incidence of 6.1% in the droperidol group and 8.1% in the prochlorperazine group (13). Dystonic reactions were classified separately. These akathisia rates compare to a rate of 10.4% (5 of 48) for the droperidol group in our study and 18.8% (9/47) in the prochlorperazine group. Our research group has previously reported on the rates of akathisia after prochlorperazine as a 2-min bolus or 15-min infusion (16). In that study, in the group receiving prochlorperazine by i.v. injection over 2 min, 26.0% (13/50) had akathisia. The results of our study are subject to several limitations. First, we did not attempt to classify the type of headaches. Other investigators classify headaches using taxonomies such as the International Headache Society criteria (21). We believe, along with Jones et al., that such an attempt to divide headaches into the categories of vascular or tension cephalalgia may be unnecessary for the purposes of emergency treatment (5). Second, our sampling was non-consecutive, which may have introduced bias. Third, because this study was conducted at two academic centers, the results may not be generalizable to the community setting. Our study was terminated early because of chronic, intermittent shortages of prochlorperazine, as well as the Food and Drug Administration “black box” warning in December 2001 regarding droperidol. This warning read, “Cases of QT prolongation or torsades de pointes have been reported in patients receiving INAPSINE [droperidol] at doses at or below recommended doses. Some cases have occurred in patients with no known risk factors for QT prolongation and some cases have been fatal.” The Canadian Health Protection branch issued a similar warning and sales of droperidol were discontinued in the United Kingdom (22,23). Because the data and rationale that led to these warnings were unavailable to us to evaluate the potential hazard to study participants, enrollment was terminated. Droperidol has been used in EDs for years, not only for the treatment of headache, but for rapid tranquiliza-
149
tion of violent patients, and treatment of nausea and vomiting. Because we have never observed any obvious untoward effects, apart from akathisia after parenteral droperidol administration, in years of clinical use, electrocardiographic monitoring of our study participants or screening for prolonged QT intervals before treatment was not considered. The FDA warning of December 4, 2001 further recommended that all patients receiving droperidol have a screening 12-lead electrocardiogram to assess the presence of prolonged QT interval and be monitored for 2–3 h afterward (19). Such screening and monitoring will add to the cost of ED treatment and further burden the already crowded emergency medical system. The warning also recommends that the drug should be reserved for use in patients who do not respond to or cannot tolerate alternate drugs (19). Unfortunately, the major alternative therapy of prochlorperazine, used as the control drug in our study, is no longer being manufactured (20). In summary, our study suggests that droperidol and prochlorperazine provide similar pain reduction for headaches in Emergency Department patients with a similar incidence of akathisia. Parenteral prochlorperazine is now considered a standard therapy for the emergency treatment of headache and our study adds to the growing evidence of the similar efficacy of droperidol for this indication. Unfortunately, chronic shortages of prochlorperazine and concerns about QT prolongation and torsades de pointes associated with droperidol therapy threaten the clinical use of these two time- and research-tested medications. This confluence of events may substantially reduce the therapeutic armamentarium available to clinicians for treating the common complaint of headache.
REFERENCES 1. Silberstein SD. Evaluation and emergency treatment of headache. Headache 1992;32:396 – 407. 2. Newman LC, Lipton RB. Emergency department evaluation of headache. Neurol Clin 1998;16:285–303. 3. Cordell WH, Keene KK, Giles BK, Jones JB, Jones JH, Brizendine EJ. The high prevalence of pain in emergency medical care. Am J Emerg Med 2002;20:165–9. 4. Callaham M, Raskin N. A controlled study of dihydroergotamine in the treatment of acute migraine headache. Headache 1986;26: 168 –71. 5. Jones J, Sklar D, Dougherty J, White W. Randomized double-blind trial of intravenous prochlorperazine for the treatment of acute headache. JAMA 1989;261:1174 – 6. 6. Thomas SH, Stone CK, Ray VG, Whitley TW. Intravenous versus rectal prochlorperazine in the treatment of benign vascular or tension headache: a randomized, prospective, double-blind trial. Ann Emerg Med 1994;24:923–7. 7. Seim MB, March JA, Dunn KA. Intravenous ketorolac vs intravenous prochlorperazine for the treatment of migraine headaches. Acad Emerg Med 1998;5:573– 6. 8. Coppola M, Yealy DM, Leibold RA. Randomized, placebo-con-
150
9.
10. 11. 12. 13. 14.
trolled evaluation of prochlorperazine versus metoclopramide for emergency department treatment of migraine headache. Ann Emerg Med 1995;26:541– 6. Baldessarini R. Drugs and the treatment of psychiatric disorders. In: Hardman J, Gilman A, Limbird L, eds. Goodman and Gilman’s the pharmacological basis of therapeutics, 9th edn. New York: The McGraw-Hill Companies, Inc.; 1996:399 – 422. Wang SJ, Silberstein SD, Young WB. Droperidol treatment of status migrainosus and refractory migraine. Headache 1997;37: 377– 82. Richman PB, Reischel U, Ostrow A, et al. Droperidol for acute migraine headache. Am J Emerg Med 1999;17:398 – 400. Richman PB, Allegra J, Eskin B, et al. A randomized clinical trial to assess the efficacy of intramuscular droperidol for the treatment of acute migraine headache. Am J Emerg Med 2002;20:39 – 42. Miner JR, Fish SJ, Smith SW, Biros MH. Droperidol vs. prochlorperazine for benign headaches in the emergency department. Acad Emerg Med 2001;8:873–9. Hall JE, Uhrich TD, Barney JA, Arain SR, Ebert TJ. Sedative, amnestic, and analgesic properties of small-dose dexmedetomidine infusions. Anesth Analg 2000;90:699 –705.
C. S. Weaver et al. 15. Sachdev P. Research diagnostic criteria for drug-induced akathisia: conceptualization, rationale and proposal. Psychopharmacology 1994;114:181– 6. 16. Collins RW, Jones JB, Walthall JD, et al. Intravenous administration of prochlorperazine by 15-minute infusion versus 2-minute bolus does not affect the incidence of akathisia: a prospective, randomized, controlled trial. Ann Emerg Med 2001;38:491– 6. 17. Blackwelder WC. “Proving the null hypothesis” in clinical trials. Control Clin Trials 1982;3:345–53. 18. Liang K, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13–22. 19. US Food and Drug Administration. Inapsine (droperidol). Available at: http://www.fda.gov/medwatch/SAFETY/2001/safety01.htm #inapsi. Accessed January 15, 2002. 20. Horowitz B, Bizovi K, Moreno R. Droperidol— behind the black box warning. Acad Emerg Med 2002;9:615– 8. 21. ICD-10 Guide for Headaches. International Headache Classification Committee. Cephalalgia 1997;17(Suppl 19):1– 82. 22. Arrhythmias from droperidol? Med Lett Drugs Ther 2002;44: 53– 4. 23. Droperidol to be discontinued in the UK. React Wkly 2001:835.
doi:10.1016/j.jemermed.2003.05.005
Original Contributions DROPERIDOL VS. PROCHLORPERAZINE FOR THE TREATMENT OF ACUTE HEADACHE Christopher S. Weaver, MD,* James B. Jones, MD, PharmD,* Carey D. Chisholm, MD,* Michael J. Foley, MD,* Beverly K. Giles, RN,* Geoffrey G. Somerville, BS, EMT-P,* Edward J. Brizendine, MS,† and William H. Cordell, MD *Department of Emergency Medicine and †Department of Biostatistics, Indiana University School of Medicine, Indianapolis, Indiana, USA Reprint Address: Christopher S. Weaver, MD, Department of Emergency Medicine, Indiana University School of Medicine OPW M200, 1001 W. Tenth Street, Indianapolis, IN 46202
e Abstract—To determine if droperidol i.v. is as effective as prochlorperazine i.v. in the emergency department (ED) treatment of uncomplicated headache, a randomized, controlled, blinded study was conducted in the Emergency Departments of two urban teaching hospitals. Patients > 18 years old with crescendo-onset headache were eligible for inclusion. Ninety-six patients (48 in each group) were randomized to receive droperidol 2.5 mg i.v. or prochlorperazine 10 mg i.v. Baseline characteristics were similar between the two study groups. For the main study outcome, 83.3% in the droperidol group and 72.3% in the prochlorperazine group reported 50% pain reduction at 30 min (p < .01; one-sided test of equivalence). The mean decrease in headache intensity was 79.1% (SD 28.5%) in the droperidol group and 72.1% (SD 28.0%) in the prochlorperazine group (p ⴝ .23). It is concluded that droperidol i.v. provided a similar reduction of headache as achieved with prochlorperazine i.v. with a similar incidence of akathisia. © 2004 Elsevier Inc.
seek care for headache in EDs because of the severity of the pain and associated symptoms, concerns about the etiology, lack of alternative venues of healthcare, and “last straw” desperation. The treatment options for the emergency treatment of headache greatly expanded after the 1986 study by Callaham and Raskin of dihydroergotamine therapy in headache that serendipitously demonstrated headache relief after pre-treatment with prochlorperazine (4). Jones et al. confirmed the efficacy of prochlorperazine i.v. in a randomized double-blind trial published in 1989 (5). Over the last 15 years, prochlorperazine has been accepted as a standard treatment for headache in the ED (6,7). Additional medications, including parenteral metoclopramide and the triptan drugs administered by various routes, have expanded the options for the emergency treatment of acute headache (8). Droperidol (Inapsine威) is a high potency, rapid-acting butyrophenone similar to haloperidol that has been used in the United States since approval by the Food and Drug Administration (FDA) in 1970. Droperidol exerts its mechanism of action via blockade of dopamine receptors in the subcortical, midbrain, and brainstem reticular formation and produces mild alpha-adrenergic blockade and peripheral vascular dilatation. Recommended doses for these indications range from 0.625 mg i.v./i.m. up to 10 mg, depending on the indication, although much larger doses have been used (9). Droperidol is used clinically for rapid tranquilization of violent patients, as an adjunct
e Keywords—headache; pain; droperidol; prochlorperazine
INTRODUCTION Headache is a common complaint in patients presenting to the Emergency Department (ED) (1–3). Many patients Presented at the Society for Academic Emergency Medicine Annual Meeting, St. Louis, MO, in May 2002. Currently, Dr. Jones works for Pharmacia Corporation, Peapack, New Jersey and Dr. Foley is an emergency physician in Madison, Wisconsin.
Original Contributions is coordinated by John Marx,
RECEIVED: 30 September 2002; FINAL ACCEPTED: 12 May 2003
SUBMISSION RECEIVED:
MD,
of Carolinas Medical Center, Charlotte, North Carolina
1 April 2003;
145
146
C. S. Weaver et al.
to anesthesia, and for the treatment of nausea and vomiting. More recently, anecdotal experience and several case series and clinical studies suggest the efficacy of droperidol in the acute treatment of headache (10 –13). We are unaware of any study that has compared droperidol i.v. to prochlorperazine i.v. in a randomized, double-blind fashion. We therefore conducted such a study to evaluate whether droperidol i.v. is as effective as prochlorperazine i.v. in the ED treatment of uncomplicated headache.
MATERIALS AND METHODS We conducted a randomized, blinded, controlled trial in the EDs of Wishard Memorial and Methodist Hospital. Both are teaching hospitals located in urban Indianapolis, Indiana. The annual ED census was 105,000 at Wishard and 94,000 at Methodist at the time of the study. We enrolled all eligible patients in a convenience sample from August 7, 2000 to November 28, 2001 based on research coordinator coverage. Patients 18 years of age or older were eligible for participation in this study if they presented to ED triage with a crescendo-onset headache and had a normal neurological examination. A specific classification of headache (e.g., vascular vs. tension) was neither defined nor required for study entry in an effort to assess the medications’ effectiveness in treating all uncomplicated headaches (5). Patients were excluded if they were experiencing their first headache, had a temperature ⱖ 38°C (100.4°F), exhibited nuchal rigidity, or had thunderclap onset of the headache. Other exclusion criteria included self treatment with a pain medication or antiemetic in the 4 h before arrival; history of carbon monoxide exposure, peripheral vascular disease, cancer, or HIV infection; pregnancy; allergy to study medications; inability to speak or understand English; and lack of telephone for follow-up contact. The Institutional Review Boards at both institutions approved the study. All study participants gave written informed consent before enrollment. Study participants were instructed that they could receive rescue medications for either headache or nausea and vomiting at any time and that they could terminate their participation in the study at any time. This study was internally funded and none of the investigators has or had a significant financial interest in any of the manufacturers of the study medications. Patients were randomized by means of a computergenerated randomization table to receive droperidol 2.5 mg i.v. or prochlorperazine 10 mg i.v. followed by a 2 cc normal saline flush. A contract research pharmacy prepared the study medication kits. Each kit contained two
identical vials. The study drug kit contained one vial of droperidol 2.5 mg and one vial of normal saline. The control drug kit contained two vials, each containing prochlorperazine 5 mg. To maintain the blind, all vials contained 1 mL of solution and were identical in appearance. Instructions affixed to the kits directed the investigator to draw both vials into a single syringe and inject the contents over a 2-min period. The study participant, investigator, outcomes assessor, and data analysts were unaware of the randomization assignments. The study blind was maintained until data analysis was completed. Study participants did not receive any additional medication or i.v. fluids for headache, nausea, or vomiting during the first 30 min unless they requested to be withdrawn from the study. Rescue medications after 30 min were specified as meperidine 1 mg/kg i.v. for persisting headache, ondansetron 4 mg i.v. for persisting nausea or vomiting, and diphenhydramine hydrochloride 25–50 mg i.v. for extrapyramidal side effects. The main study outcome was the number of subjects in each group achieving at least 50% pain relief at 30 min. Secondary outcome measures included mean change in pain intensity, the proportion requiring rescue medications from 30 to 60 min, and the incidence of akathisia and other adverse events. Study participants were asked to record pain and nausea intensity and level of alertness on three separate 100-mm visual analog (VAS) scales bounded by the words “no pain”-“worst pain,” “no nausea”-“worst nausea,” and “sound asleep”“wide awake,” respectively, at baseline, 15 min, 30 min, 45 min, and 60 min (14). Our rating scale for akathisia was modified from the Prince Henry Hospital (PHH) Rating Scale of Akathisia and was employed in a previous study of prochlorperazine-induced akathisia by our research group (15,16). The rating scale included “subjective ratings” elicited from study participants by direct questioning and “objective ratings” assessed by investigator observation. Akathisia was defined as the occurrence of either or both of the following: spontaneous report or change in both the objective and subjective akathisia rating score compared to baseline. Other adverse events were recorded only if they were spontaneously reported at each of the time study intervals. Study participants were discharged from the ED at the discretion of the treating physician (a board certified emergency physician or a resident under the direction of a board certified emergency physician) and were contacted 24 h after discharge from the ED to assess the presence and intensity of headache and to ascertain if they had taken any additional medication. Participants rated their pain at 24 h on a verbal 0 to 10 pain scale. The study participants were also asked if they had returned to work or daily activities.
Droperidol vs. Prochlorperazine for Headache
147
Table 1. Baseline Characteristics of Trial Participants Total (n ⫽ 96) Female, number (%) Race/ethnicity, number (%) White Black Other Age, median (range), years Headache intensity, median (range), mm VAS Level of alertness, median (range), mm VAS Nausea intensity (for those reporting nausea), median (range), mm VAS
83 (86.5) 45 (46.9) 49 (51.0) 2 (2.1) 31 (18–68) 76.5 (18–100) 100 (88–100) 52 (5–95) n ⫽ 41
Sample size calculations were based on demonstrating equivalency between droperidol and prochlorperazine. This study was designed as a non-inferiority study rather than a superiority study using an equivalency window (⌬) of ⫺10%. Restated, equivalence would be demonstrated if the proportion of droperidol subjects experiencing at least a 50% reduction in headache intensity (the main study outcome) were not more than 10% less than the proportion of prochlorperazine subjects experiencing at least a 50% reduction. Collins et al. reported that 75% of the subjects receiving prochlorperazine i.v. experienced a 50% or more reduction in their headache intensity at 30 min (16). Based on this result and our equivalence definition, a sample of 230 subjects per group would have 80% power to demonstrate that droperidol is as effective as prochlorperazine in headache relief. Sample size calculations were based on using a one-sided test of equivalency with a significance level of ␣ ⫽ 0.05 and performed using nQuery Advisor version 4.0 (Statistical Solutions, Cork, Ireland). An intention-to-treat analysis was performed. For the primary outcome of the proportion of subjects experiencing at least a 50% reduction in pain at 30 min, a onesided test of equivalence was performed (17). The null hypothesis for this test was that the difference between the droperidol and prochlorperazine percentages was less than or equal to ⫺10%. A one-sided 95% confidence interval (CI) for the true difference in percentages was estimated. For all other test of hypotheses, a null hypothesis of no difference was assumed. To assess if the rate of pain relief was different between the two medications, a repeated measures analysis of variance (ANOVA) was performed. Generalized estimating equations were used to adjust parameter estimates for the repeated measurements made on each subject (18). Statistical analyses were performed using SAS version 8.2 (SAS Institute, Cary, NC) and S-Plus 2000 (MathSoft, Seattle, WA). Summary statistics for continuous data are presented as
Droperidol (n ⫽ 48) 44 (91.7) 22 (45.8) 26 (54.2) 0 (0.0) 30 (18–68) 68 (18–100) 100 (100–100) 50 (5–95) n ⫽ 19
Prochlorperazine (n ⫽ 48) 39 (81.3) 23 (47.9) 23 (47.9) 2 (4.2) 34 (19–64) 79 (21–100) 100 (88–100) 52 (23–86) n ⫽ 22
p value 0.14 0.33 0.27 0.07 0.33 0.37
the estimated median and range, whereas for categorical data, the frequency and percent are presented.
RESULTS We enrolled 96 non-consecutive patients between August 7, 2000 and December 10, 2001. Our study was terminated early due to chronic intermittent shortages of prochlorperazine as well as an FDA “black box” warning issued December 4, 2001 regarding QT prolongation and torsade de pointes in patients treated with droperidol (19,20). There were 48 study participants in each of the groups. All 96 participants remained in the study throughout the 60-min evaluation. There were two missing data points for headache intensity assessment— one in the droperidol group at 60 min and one in the prochlorperazine group at 30 min. Patient characteristics at baseline were similar between groups (Table 1). For the main study outcome, 83.3% in the droperidol group and 72.3% in the prochlorperazine group had 50% pain reduction at 30 min (Table 2). A one-sided test of equivalency yielded a p value of ⬍ 0.01, suggesting that droperidol was not inferior to prochlorperazine using an equivalency window (⌬) of ⫺10%. The 95% one-sided lower confidence limit for the difference was ⫺2.9%. This value is greater than the ⫺10% limit of equivalency. The mean decrease in headache intensity at 30 min was 79.1% (SD 28.5%) in the droperidol group and 72.1% (SD 28.0%) in the prochlorperazine group (p ⫽ 0.23; 95% CI ⫺4.6, 18.5). There was no significant difference in the rate of decrease in pain intensity between the two study groups over the first 60 min (p ⫽ 0.50) (Figure 1). Nor were there significant differences at 30 min for percent decrease in nausea (p ⫽ 0.22) or level of alertness (p ⫽ 0.98). Six study participants in each group required rescue analgesics (p ⫽ 1.0, 95% CI ⫺13.2, 13.2). A total of 14 (14.6%) study participants experienced akathisia during the first 60 min, 5 (10.5%) in the
148
C. S. Weaver et al.
Table 2. Categorical Relief of Headache Intensity at 30 Minutes
Percent Relief
Droperidol (n ⫽ 48)
Prochlorperazine (n ⫽ 47)
ⱖ 50% relief, number (%) ⱖ 75% relief, number (%) 100% relief, number (%)
40 (83.3) 31 (64.6) 26 (54.2)
34 (72.3) 26 (55.3) 18 (38.3)
p value* ⬍ 0.01 0.03 ⬍ 0.01
One-sided 95% CI for difference† ⫺2.9, 100 ⫺7.2, 100 ⫺0.7, 100
* Using a 1-sided test of equivalence (⌬ ⫽ ⫺10%). † Upper limit is reported as 100% because this is the maximum of all possible values of the difference of two proportions.
droperidol group and 9 (18.8%) in the prochlorperazine group (p ⫽ 0.25; 95% CI ⫺22.4, 5.7). No other adverse events were spontaneously reported in either group. We were able to contact 83 (86.5%) of the study participants 24 h after discharge (40 in the droperidol group and 43 in the prochlorperazine group). At 24 h follow-up, 27.5% (11/40) of the droperidol group reported headache compared to 34.8% (15/43) of the prochlorperazine group (p ⫽ 0.47; 95% CI ⫺27.2, 12.5). The mean headache intensity at 24 h for the droperidol group was 4.7 (SD 2.1) compared to 5.0 (SD 2.7) in the prochlorperazine group (p ⫽ 0.78) using a verbal 0 –10 pain intensity scale. In the droperidol group, 13 (32.5%) reported use of supplemental analgesics since ED discharge compared to 18 (41.9%) in the prochlorperazine group (p ⫽ 0.38). At 24 h after discharge, 2.5% (1/40) of the droperidol group versus 14% (6/43) of the prochlorperazine group had akathisia (p ⫽ 0.06; CI ⫺22.9%, 0.0%). It is important to emphasize that the 24-h akathisia follow-up was based only on the patient’s self-report and not on the investigator’s rating, as was done during the first 60 min. Twenty-seven (67.5%) of the droperidol group and 28 (65.1%) of the prochlorperazine group
Figure 1. Mean pain intensity after medication administration. Bars represent 95% Confidence Intervals.
reported they had returned to normal daily activities (p ⫽ 0.82). DISCUSSION Droperidol has been used clinically for three decades for tranquilization of violent patients, anesthesia and neuroleptanalgesia, treatment of nausea and vomiting, and, more recently, headache. The results of our study add to the growing evidence of the efficacy and safety of droperidol in treating headache. Our study demonstrates similar relief of headache intensity measured at 30 min and 24 h between study participants receiving droperidol 2.5 mg i.v. and those receiving prochlorperazine 10 mg i.v. The incidence of akathisia was also similar between the two groups. Wang et al. studied the use of droperidol 2.5 mg i.v. every 30 min up to three doses in 35 patients with status migrainosus or refractory migraine (10). The success rate (defined as headache-free or mild headache) was 88% of the 25 patients with status migrainosus and 100% of those with refractory migraine. Seven patients received one dose, 12 received two doses, and 16 received three doses (mean 5.6 mg) of droperidol i.v. This contrasts to our study where a single 2.5 mg dose of droperidol was administered i.v. In a retrospective study of 37 patients, Richman et al. reported symptomatic relief in 81% at 30 min after droperidol 2.5 mg i.m. (11). Richman et al. then conducted a randomized, blinded comparison of droperidol 2.5 i.m. and meperidine 1.5 mg/kg i.m. in treating 29 ED patients with acute migraine as defined by International Headache Society criteria (12). The two groups were similar for mean change in VAS score and proportion requesting rescue medication. Miner et al. conducted a prospective, single-blind comparison of droperidol (either 5 mg i.m. or 2.5 mg i.v.) and prochlorperazine (either 10 mg i.v. or i.m.) in the treatment of benign headaches (13). They reported that 60.9% of the 82 study participants receiving droperidol and 44.2% of the 86 receiving prochlorperazine achieved ⱖ 50% pain relief at 30 min. Twenty-six percent of those
Droperidol vs. Prochlorperazine for Headache
who received droperidol and 18.2% of those who received prochlorperazine had rebound headache. These results compare to our study where the proportion of subjects reporting headache at 24 h was 27.5% in the droperidol group and 34.8% in the prochlorperazine group (with a follow-up rate of 86.5%). Akathisia and drowsiness are the main adverse events after droperidol therapy for headache. Richman et al., in their retrospective study, reported an 8% incidence of mild akathisia (11). In their prospective study, Richman et al. reported an incidence of akathisia of 13.3% after droperidol 2.5 mg i.m. (12). The incidence of sedation was 6.7% in the droperidol i.m. group compared to 14.3% in the meperidine (1.5 mg/kg) i.m. group. Miner et al. reported an akathisia incidence of 6.1% in the droperidol group and 8.1% in the prochlorperazine group (13). Dystonic reactions were classified separately. These akathisia rates compare to a rate of 10.4% (5 of 48) for the droperidol group in our study and 18.8% (9/47) in the prochlorperazine group. Our research group has previously reported on the rates of akathisia after prochlorperazine as a 2-min bolus or 15-min infusion (16). In that study, in the group receiving prochlorperazine by i.v. injection over 2 min, 26.0% (13/50) had akathisia. The results of our study are subject to several limitations. First, we did not attempt to classify the type of headaches. Other investigators classify headaches using taxonomies such as the International Headache Society criteria (21). We believe, along with Jones et al., that such an attempt to divide headaches into the categories of vascular or tension cephalalgia may be unnecessary for the purposes of emergency treatment (5). Second, our sampling was non-consecutive, which may have introduced bias. Third, because this study was conducted at two academic centers, the results may not be generalizable to the community setting. Our study was terminated early because of chronic, intermittent shortages of prochlorperazine, as well as the Food and Drug Administration “black box” warning in December 2001 regarding droperidol. This warning read, “Cases of QT prolongation or torsades de pointes have been reported in patients receiving INAPSINE [droperidol] at doses at or below recommended doses. Some cases have occurred in patients with no known risk factors for QT prolongation and some cases have been fatal.” The Canadian Health Protection branch issued a similar warning and sales of droperidol were discontinued in the United Kingdom (22,23). Because the data and rationale that led to these warnings were unavailable to us to evaluate the potential hazard to study participants, enrollment was terminated. Droperidol has been used in EDs for years, not only for the treatment of headache, but for rapid tranquiliza-
149
tion of violent patients, and treatment of nausea and vomiting. Because we have never observed any obvious untoward effects, apart from akathisia after parenteral droperidol administration, in years of clinical use, electrocardiographic monitoring of our study participants or screening for prolonged QT intervals before treatment was not considered. The FDA warning of December 4, 2001 further recommended that all patients receiving droperidol have a screening 12-lead electrocardiogram to assess the presence of prolonged QT interval and be monitored for 2–3 h afterward (19). Such screening and monitoring will add to the cost of ED treatment and further burden the already crowded emergency medical system. The warning also recommends that the drug should be reserved for use in patients who do not respond to or cannot tolerate alternate drugs (19). Unfortunately, the major alternative therapy of prochlorperazine, used as the control drug in our study, is no longer being manufactured (20). In summary, our study suggests that droperidol and prochlorperazine provide similar pain reduction for headaches in Emergency Department patients with a similar incidence of akathisia. Parenteral prochlorperazine is now considered a standard therapy for the emergency treatment of headache and our study adds to the growing evidence of the similar efficacy of droperidol for this indication. Unfortunately, chronic shortages of prochlorperazine and concerns about QT prolongation and torsades de pointes associated with droperidol therapy threaten the clinical use of these two time- and research-tested medications. This confluence of events may substantially reduce the therapeutic armamentarium available to clinicians for treating the common complaint of headache.
REFERENCES 1. Silberstein SD. Evaluation and emergency treatment of headache. Headache 1992;32:396 – 407. 2. Newman LC, Lipton RB. Emergency department evaluation of headache. Neurol Clin 1998;16:285–303. 3. Cordell WH, Keene KK, Giles BK, Jones JB, Jones JH, Brizendine EJ. The high prevalence of pain in emergency medical care. Am J Emerg Med 2002;20:165–9. 4. Callaham M, Raskin N. A controlled study of dihydroergotamine in the treatment of acute migraine headache. Headache 1986;26: 168 –71. 5. Jones J, Sklar D, Dougherty J, White W. Randomized double-blind trial of intravenous prochlorperazine for the treatment of acute headache. JAMA 1989;261:1174 – 6. 6. Thomas SH, Stone CK, Ray VG, Whitley TW. Intravenous versus rectal prochlorperazine in the treatment of benign vascular or tension headache: a randomized, prospective, double-blind trial. Ann Emerg Med 1994;24:923–7. 7. Seim MB, March JA, Dunn KA. Intravenous ketorolac vs intravenous prochlorperazine for the treatment of migraine headaches. Acad Emerg Med 1998;5:573– 6. 8. Coppola M, Yealy DM, Leibold RA. Randomized, placebo-con-
150
9.
10. 11. 12. 13. 14.
trolled evaluation of prochlorperazine versus metoclopramide for emergency department treatment of migraine headache. Ann Emerg Med 1995;26:541– 6. Baldessarini R. Drugs and the treatment of psychiatric disorders. In: Hardman J, Gilman A, Limbird L, eds. Goodman and Gilman’s the pharmacological basis of therapeutics, 9th edn. New York: The McGraw-Hill Companies, Inc.; 1996:399 – 422. Wang SJ, Silberstein SD, Young WB. Droperidol treatment of status migrainosus and refractory migraine. Headache 1997;37: 377– 82. Richman PB, Reischel U, Ostrow A, et al. Droperidol for acute migraine headache. Am J Emerg Med 1999;17:398 – 400. Richman PB, Allegra J, Eskin B, et al. A randomized clinical trial to assess the efficacy of intramuscular droperidol for the treatment of acute migraine headache. Am J Emerg Med 2002;20:39 – 42. Miner JR, Fish SJ, Smith SW, Biros MH. Droperidol vs. prochlorperazine for benign headaches in the emergency department. Acad Emerg Med 2001;8:873–9. Hall JE, Uhrich TD, Barney JA, Arain SR, Ebert TJ. Sedative, amnestic, and analgesic properties of small-dose dexmedetomidine infusions. Anesth Analg 2000;90:699 –705.
C. S. Weaver et al. 15. Sachdev P. Research diagnostic criteria for drug-induced akathisia: conceptualization, rationale and proposal. Psychopharmacology 1994;114:181– 6. 16. Collins RW, Jones JB, Walthall JD, et al. Intravenous administration of prochlorperazine by 15-minute infusion versus 2-minute bolus does not affect the incidence of akathisia: a prospective, randomized, controlled trial. Ann Emerg Med 2001;38:491– 6. 17. Blackwelder WC. “Proving the null hypothesis” in clinical trials. Control Clin Trials 1982;3:345–53. 18. Liang K, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13–22. 19. US Food and Drug Administration. Inapsine (droperidol). Available at: http://www.fda.gov/medwatch/SAFETY/2001/safety01.htm #inapsi. Accessed January 15, 2002. 20. Horowitz B, Bizovi K, Moreno R. Droperidol— behind the black box warning. Acad Emerg Med 2002;9:615– 8. 21. ICD-10 Guide for Headaches. International Headache Classification Committee. Cephalalgia 1997;17(Suppl 19):1– 82. 22. Arrhythmias from droperidol? Med Lett Drugs Ther 2002;44: 53– 4. 23. Droperidol to be discontinued in the UK. React Wkly 2001:835.