- Email: [email protected]
Clinical Outcomes of Children Treated with Intravenous Prochlorperazine for Migraine in a Pediatric Emergency Department
The Journal of Emergency Medicine, Vol. 39, No. 2, pp. 166 –173, 2010 Copyright © 2010 Elsevier Inc. Printed in the USA. All rights reserved 0736-4679/$–see front matter
doi:10.1016/j.jemermed.2008.08.012
Original Contributions
CLINICAL OUTCOMES OF CHILDREN TREATED WITH INTRAVENOUS PROCHLORPERAZINE FOR MIGRAINE IN A PEDIATRIC EMERGENCY DEPARTMENT Evelyne D. Trottier,
MD, FRCPC,*
Benoit Bailey, MD, MSC, FRCPC,*† Sabine Dauphin-Pierre,* and Jocelyn Gravel, MD, MSC, FRCPC*
*Division of Emergency Medicine, †Division of Clinical Pharmacology and Toxicology, Department of Pediatrics, CHU Sainte-Justine, Université de Montréal, Montréal, Quebec, Canada Reprint Address: Benoit Bailey, MD, MSC, FRCPC, Department of Pediatrics, CHU Ste-Justine, 3175 Chemin de la Côte-Ste-Catherine, Montréal, QC H3T 1C5, Canada
apy; and 3 patients returned to the ED within 48 h due to symptom recurrence. Conclusion: There was a treatment failure rate of 14% with the use of prochlorperazine in association with diphenhydramine for severe migraine in children seen in a pediatric ED. © 2010 Elsevier Inc.
e Abstract—Background: Prochlorperazine is the only treatment that has been studied so far in a randomized controlled trial and found to reduce pain at 1 h in children with migraine who presented to an emergency department (ED). Objective: To evaluate the rate of treatment failure associated with prochlorperazine used in children with severe migraine in a pediatric ED. Methods: This study was a retrospective chart review of patients < 18 years of age who visited the ED of a tertiary care pediatric hospital between November 2005 and June 2007. All patients diagnosed with migraine by the emergency physicians were included in the study. Charts were evaluated by a data abstractor blinded to the study hypothesis using a standardized datasheet. Inter-rater agreement was measured. Prochlorperazine treatment failure was defined as either administration of further rescue therapy, a hospitalization, or a return visit to the ED within 48 h for symptom recurrence or side effects from the medication. Results: Prochlorperazine was administered in 92 episodes of migraine, including 43 confirmed by a pediatric neurologist; all received diphenhydramine to prevent akathisia. A total of 13 (14%) of these patients had a treatment failure: 8 patients received one or more further rescue therapies after the administration of prochlorperazine; 5 patients were hospitalized, including 3 who had received further rescue ther-
e Keywords—prochlorperazine; migraine; children; emergency department
INTRODUCTION Migraine is the leading cause of acute and recurrent headache at all ages, and studies suggest that as many as 10% of children will eventually suffer from this condition (1). Physicians working in the emergency department (ED) regularly encounter this problem. Despite the frequency of this condition, treatment remains a challenge due to a lack of sufficient studies on migraine therapy in an emergency setting. This may explain the large variability in the clinical practice of emergency physicians for the treatment of migraine in children (2). Two recent meta-analyses that evaluated treatment of migraine and status migrainosus in children reported that only one randomized controlled trial (RCT) had been performed in the ED (3,4). In that RCT, prochlorperazine was better than ketorolac in decreasing pain by 50% or
Presented in part at the Canadian Association of Emergency Medicine annual meeting, Ottawa, Ontario, June 2008.
RECEIVED: 20 May 2008; FINAL ACCEPTED: 5 August 2008
SUBMISSION RECEIVED:
6 June 2008; 166
Failure with Prochlorperazine
providing complete relief by 1 h in children presenting to an ED with migraine (5). A retrospective study of 20 children treated with prochlorperazine but not in an ED, also has been reported and suggested the same efficacy (6). Despite these two studies, not much is known in terms of prochlorperazine efficacy for other important outcomes such as the need for further rescue therapy or the recurrence of symptoms after its administration, as well as its side effects in children (7). Most studies evaluating the treatment of migraine use a primary outcome of pain relief 1 h after treatment. Although this is an important outcome, there are many other clinical outcomes that are important to the patients. The International Headache Society (IHS) Clinical Trial Subcommittee mentions that important patient outcomes include the need for further rescue medication after the study medication and symptom recurrence (7). The objective of our study was to evaluate the treatment failure rate among children treated with a protocol using prochlorperazine as an adjunct to diphenhydramine in an ED. Treatment failure was defined as a patient needing further rescue medication, or being hospitalized, or needing to return to the ED for symptom recurrence or side effects from the medication within 48 h of the initial visit.
MATERIALS AND METHODS Study Design This study was a retrospective chart review of children diagnosed with migraine, as determined by the pediatric emergency physicians, and treated with intravenous prochlorperazine due to the intensity of the migraine.
Setting The cases reviewed were from the ED of a tertiary care pediatric hospital with an annual census of more than 60,000. The ED is staffed with pediatricians and pediatric emergency physicians supervising residents and medical students. Members of the ED use a standardized protocol for the treatment of migraine or status migrainosus.
Study Protocol This protocol suggests a treatment of prochlorperazine (0.15 mg/kg intravenous [i.v.]; maximal dose 10 mg) with diphenhydramine (0.5 mg/kg i.v.; maximal dose 25 mg) for patients over 8 years of age with migraine severe enough that the patient is unable to continue his or her normal
167
activities (8). Moreover, for migraine of moderate intensity, the protocol suggests oral naproxen or ibuprofen, and for migraine of mild intensity, it suggests oral ibuprofen. No additional i.v. fluid treatment is proposed in the protocol. The protocol does not recommend any particular rescue therapy if the primary treatment is not effective. Also, the length of stay in the ED after intravenous treatment with prochlorperazine is determined by the clinical response of the patient. It may vary with each case and each physician. Participants Patients ⬍ 18 years of age who visited the ED between November 2005 and June 2007 and had a final diagnosis of migraine made by the emergency physician were eligible to be included in the study if they also had received prochlorperazine. The ED computerized database was used to identify cases using the terms: Migraine, Migraine with aura, Classic migraine. This database does not include a diagnosis of vascular headache. When physicians are unsure about the diagnosis, they can assign a diagnosis of “headache” to the patient. Such patients were not reviewed. Further, study patients did not necessarily meet the IHS migraine criteria (9). These criteria have been shown to have modest sensitivity, and thus are more restrictive in comparison to expert clinical diagnosis (10). The authors wanted to use real life situations in which clinicians do not want to deny effective treatment to some of these patients, and thus, clinical diagnosis was selected as one of the inclusion criteria. The clinical diagnosis is made by a combination of the past medical history, the family medical history, and the actual symptoms of the patient (such as the presence or absence of throbbing headache, unilateral headache, photophobia, sonophobia, nausea, or vomiting). Some patients had a diagnosis confirmed by a pediatric neurologist from our hospital, which was evaluated by a review of the medical records in our hospital. Furthermore, return visits to the ED within 3 months after the initial diagnosis were identified and evaluated to rule out other possible causes of the patients’ headaches. For our study, all return visits for the same problem within a month of the initial visit were considered to be a single episode of migraine, and thus a recurrence. For the evaluation of prochlorperazine, only its first usage within the episode was evaluated. If a patient presented to the ED more than once but more than 1 month after the initial visit, they were considered as separate episodes. Record Review All identified charts were evaluated by a data abstractor blinded to the study hypothesis using a standardized
168
E. D. Trottier et al.
datasheet. The chart reviewer received 1 week of formal training by the primary investigator on records of patients seen from November 1–15, 2005. Inter-rater agreement between the data abstractor and the primary investigator was measured after the formal training. It was set, a priori, that a kappa score ⱖ 0.80 agreement would be needed for data abstraction to be acceptable. The performance of the data abstractor was evaluated by measuring inter-rater agreement in 10% of the remaining charts. The diagnosis of migraine was confirmed for every chart. The pain score (verbal pain scale from 0 to 10) was noted. The highest score noted by the physician or by the nurse at triage was recorded. In some files, the nurse recorded an interval of pain (0 –3, 4 –7, 8 –10) proposed by our computerized triage system. In those cases, we used the median pain score for the interval (1.5, 5.5, and 9, respectively). We also noted the pain score recorded after treatment in the ED. The need for other medication in the ED after the administration of prochlorperazine was evaluated. We classified the following medications as rescue therapies: oral doses of opiates or triptans, or any additional parenteral medications, including a second dose of prochlorperazine. The following medications were classified as a prophylactic treatment or minor analgesic: acetaminophen, non-steroidal anti-inflammatory drug, or amitriptyline. We also evaluated if the patient returned to the ED for headache within 48 h after treatment. We considered this a recurrence of symptoms after a responsive treatment (7). Finally, the medical charts were evaluated for adverse effects of prochlorperazine, such as akathisia, immediately after treatment during the ED stay and, if available, at any other visits. (No objective scale is used in our ED to evaluate akathisia.) Treatment failure was defined as the occurrence of either the administration of further rescue therapy, hospitalization, or return visits to the ED within 48 h for symptom recurrence or side effects.
would have received prochlorperazine for the treatment of migraine. The study was approved by our hospital’s Institutional Review Board.
Data Analysis
Rescue therapy. In the group of patients who had the diagnosis confirmed by a pediatric neurologist, 5 patients received one or more rescue therapies after the administration of prochlorperazine. Among these, 1 patient received a second dose of prochlorperazine, another patient received a second dose of prochlorperazine along with a dose of metoclopramide, and another patient received a dose of metoclopramide and a dose of opiate. These 3 patients were hospitalized. Two other patients received doses of opiates after prochlorperazine and were later discharged. In the group who did not have a diagnosis confirmed by a pediatric neurologist, 3 patients had a rescue therapy, 1 received a dose of opiate, and the other 2 received intravenous dimenhydrinate.
All data were entered in an Excel database (Microsoft Corporation, Redmond, WA). Statistical analysis was performed using SPSS v15.0 (SPSS Inc., Chicago, IL). Interrater agreement was measured using non-weighted kappa scores.
Sample Size We did not calculate a sample size, but we estimated that the evaluation of all the visits to the ED over an 18month period would provide more than 50 patients who
RESULTS From November 2005 to June 2007, there were 300 episodes of migraine in patients who presented to the ED. Among these patients, 2 were adults and were not included in the study. Another 12 charts from 12 different episodes could not be retrieved and assessed. Thus, over the 20-month period, there were 286 episodes (for a total of 349 visits) in 262 patients (153 girls and 109 boys) that were included in the study. Most patients received no treatment or were treated with medications that were administered orally. Prochlorperazine was administered in 99 episodes in patients with a diagnosis of migraine made by the emergency physician. Seven such episodes were later found not to be related to migraine and were excluded from the analysis. For the remaining 92 episodes occurring in patients between 7 and 17 years of age, 43 were patients who had a diagnosis of migraine confirmed by a pediatric neurologist in addition to the emergency physician. The diagnosis was not confirmed by a pediatric neurologist from our center for the remaining 49 episodes. However, none of these patients returned to the hospital for a reason that suggested an alternative diagnosis within 3 months of the initial visit. Demographic data of the patients included in the study are presented in Table 1.
Outcome of the Migraine A total of 13 (14%) patients had a treatment failure according to our definition (Table 2).
Failure with Prochlorperazine
169
Table 1. Demographic Data of Patients who Received Prochlorperazine in the ED According to the Confirmed or Unconfirmed Diagnosis of Migraine
Variable
Total (n ⫽ 92)
Mean age (SD) in 13.4 ⫾ 2.8 years Male (%) 27 (29) Mean initial pain 7.6 ⫾ 2.0 score (SD) out of 10 Head CT scan done 37 (40) Dose of 0.15 ⫾ 0.02 prochlorperazine (mg/kg) Received 92 (100) diphenhydramine (%) Median length of 6 (4–9) stay in the ED (IQR) in hours Medications at home before ED visit None (%) 17 (18) Acetaminophen 46 (50) (%) NSAID (%) 47 (51) Opiates (%) 6 (6) Prophylactic 10 (11) medications (%) Specific migraine 12 (13) medications (%)
Confirmed by a Pediatric Neurologist (n ⫽ 43)
Not Confirmed (n ⫽ 49)
13.0 ⫾ 2.4
13.7 ⫾ 3.7
17 (40) 7.2 ⫾ 2.4
10 (20) 7.9 ⫾ 1.7
24 (56) 0.15 ⫾ 0.02 43 (100) 5 (4–9.5)
13 (26) 0.15 ⫾ 0.01 49 (100)
[IQR] 5–18) days later in the group of patients who had a diagnosis confirmed. Five patients were not treated and the others received either prochlorperazine (n ⫽ 5), codeine (n ⫽ 2), metoclopramide (n ⫽ 1), or naproxen (n ⫽ 2) alone or in combination. In the unconfirmed diagnosis group, 3 patients returned to the ED a median of 6 (IQR 3–9) days later. One had no treatment. Two patients were treated with a dose of prochlorperazine and one of them was hospitalized. Pain score. There was a difference in pain score at time of discharge compared to pain on arrival for both the group of patients who had their diagnosis confirmed and the group that did not (Table 2). Overall, prochlorperazine decreased pain by a mean of 6.7 ⫾ 2.3 on a scale of 0 to 10.
6 (4–8.5)
Adverse Side Effects 9 (21) 21 (49)
8 (16) 25 (51)
22 (51) 4 (9) 5 (12)
25 (51) 2 (4) 5 (10)
9 (21)
3 (6)
SD ⫽ standard deviation; CT ⫽ computed tomography; ED ⫽ emergency department; IQR ⫽ interquartile range; NSAID ⫽ non-steroidal anti-inflammatory drug.
Hospitalization. In the group of patients who had the diagnosis confirmed by a pediatric neurologist, 5 patients were hospitalized after prochlorperazine administration, including 3 who had also received a rescue therapy. The patients were hospitalized to treat their symptoms or for further investigations. In the group who did not have the diagnosis confirmed by a pediatric neurologist, no patient was hospitalized. Return visit to the ED. After treatment with prochlorperazine and subsequent discharge, one patient in the group of patients who had the diagnosis confirmed by a pediatric neurologist returned to the ED within 48 h. This patient was not treated. In the group of patients who did not have the diagnosis confirmed, 2 patients returned to the ED within 48 h; these patients received prochlorperazine at their second visit and were discharged. Besides the patients who returned to the ED within 48 h and were considered treatment failures, 13 patients returned to the ED within the first month after receiving prochlorperazine, a median of 7 (interquartile range
Akathisia was diagnosed in 1 patient, with the migraine diagnosis confirmed by the pediatric neurologist, despite pretreatment with diphenhydramine (0.5 mg/kg i.v.—the recommended dose in the ED protocol). This patient was successfully treated with a second dose of diphenhydramine. In addition, akathisia was suspected, based on descriptions by nurses and physicians in the charts, in another five cases (three in the confirmed group and two in the
Table 2. Outcome of Patients who Received Prochlorperazine in the ED According to the Confirmed or Unconfirmed Diagnosis of Migraine
Variable
Total (n ⫽ 92)
Total failure (%) 13 (14) Rescue therapy (%)* 8 (9) Hospitalization (%) 5 (5)† Return visit to the ED 3 (3) within 48 h (%) Pain intensity Decrease in pain 6.7 ⫾ 2.3 score (SD) out of 10 Decrease in pain 88 ⫾ 19 score in percent (SD) Decrease of at 74/79 (94) least 50% in pain score (%) Pain free (%) 54/87 (62)
Confirmed by a Pediatric Not Neurologist Confirmed (n ⫽ 43) (n ⫽ 49) 8 (19) 5 (12) 5 (12)† 1 (2)
5 (10) 3 (6) 0 (0) 2 (4)
6.3 ⫾ 2.4
7.0 ⫾ 2.3
87 ⫾ 17
88 ⫾ 21
35/37 (95)
39/42 (92)
23/40 (58)
31/47 (66)
* Rescue therapy ⫽ oral doses of opiates or triptans or any other parenteral medications including a second dose of prochlorperazine. † Three of these patients also had received rescue therapy. ED ⫽ emergency department; SD ⫽ standard deviation.
170
E. D. Trottier et al.
unconfirmed group) due to the presence of symptoms of restlessness. Those cases were not treated. One patient presented with a rash with pruritus on the arm and on the face after prochlorperazine administration. She was successfully treated with a second dose of diphenhydramine. No patients were hospitalized or returned after discharge from the ED solely for akathisia or any other adverse effects.
Discharge After receiving prochlorperazine, patients were discharged on various medications (Table 3). The patient in the confirmed diagnosis group who returned to the ED within 48 h was discharged on naproxen and amitryptiline. In the unconfirmed diagnosis group, one patient who returned to the ED within 48 h had been discharged on naproxen and on amitryptiline, and another such patient on naproxen alone.
Inter-rater Agreement The inter-rater agreement for the 51 charts evaluated twice was excellent: kappa scores of 1.0 for hospitalization status, 0.94 ⫾ 0.11 for the return to the ED, and 0.81 ⫾ 0.25 for the use of a rescue therapy.
DISCUSSION This study reports the clinical outcomes of children who received a combination of prochlorperazine and diphenhydramine for the treatment of migraine in a pediatric ED. The results suggest that 14% of such patients present
Table 3. Medications Received by Patients at the Time of Discharge According to the Confirmed or Unconfirmed Diagnosis of Migraine
Medications No treatment Naproxen Ibuprofen Amitryptiline Codeine Acetaminophen Sumatriptan Metoclopramide Valproic acid
Total by a Pediatric Not Neurologist Confirmed Confirmed (n ⫽ 92) (n ⫽ 43) (n ⫽ 49) (%) (%) (%) 16 (17) 54 (59) 22 (24) 7 (8) 4 (4) 2 (2) 2 (2) 1 (1) 1 (1)
7 (16) 25 (58) 14 (33) 5 (12) 1 (2) 0 (0) 1 (2) 0 (0) 1 (0)
9 (18) 29 (59) 8 (16) 2 (4) 3 (6) 2 (4) 1 (2) 1 (2) 0 (0)
a treatment failure as defined by the need of a second rescue therapy, hospitalization, or a return to the ED within 48 h. The efficacy of prochlorperazine for the treatment of migraine and status migrainosus in children seen in a pediatric ED has been evaluated in only one RCT (5). In that study, prochlorperazine was found to be better than ketorolac for the primary outcome (decrease in the intensity of the migraine by 50% or complete relief by 1 h): 28/33 (85%) vs. 16/29 (55%), respectively, a difference of 30% (95% confidence interval [CI] 8 –52) (5). The results of this study are confirmed by results from a cohort of children with intractable headache or status migrainosus seen in a headache center (6). Fifteen of 20 (75%) patients had a reduction of their pain intensity by at least 50% by 1 h after administration of prochlorperazine, and 19/20 patients by 3 h (6). Another way of measuring the efficacy of a study drug for migraine is to measure the use of rescue medications. In the RCT of Brousseau et al., 4/33 (12%) of the children who received prochlorperazine subsequently needed a rescue medication (i.e., ketorolac) compared to 12/29 (41%) of those initially treated by ketorolac, a difference of ⫺29% (95% CI ⫺49 –7%) (5). In the cohort study, only 1/20 (5%) children received a rescue therapy after prochlorperazine administration (6). In our study, the emergency physicians gave a rescue medication after prochlorperazine in 8/92 (9%) episodes. Another important factor to consider in the efficacy of migraine treatment is the recurrence of symptoms (7). In the RCT by Brousseau et al., recurrences of headache within 48 h after discharge from the ED in children who received only one treatment in the ED, either prochlorperazine or ketorolac, were 7/26 (27%) and 4/13 (31%), respectively, a difference of ⫺4%, (95% CI ⫺34 –27%) (5). Overall, the recurrence was 13/50 (26%), including children who received more than one treatment in the ED (5). In the cohort study by Kabbouche et al., only 2/20 (10%) patients at 24-h follow-up were not pain free (6). In another study published only as an abstract, 18/51 (35%) patients who received prochlorperazine had a recurrence of headache at 24-h telephone follow-up (11). In our study, only 3/92 (3%) patients returned to the ED within 48 h for some recurrence of their symptoms. This does not exclude the possibility that other patients had a recurrence of symptoms. It does suggest that either their symptoms were not severe enough to bring them to the ED again or they may have presented to another center. In our study, 5 (5%) patients were hospitalized after administration of prochlorperazine, suggesting failure to respond; all these cases had a diagnosis of migraine confirmed by a pediatric neurologist. Prochlorperazine, a dopamine receptor antagonist, can precipitate extrapyramidal symptoms, especially akathi-
Failure with Prochlorperazine
sia. In adults, 36% to 44% of patients receiving prochlorperazine present with akathisia (12–15). This percentage decreases to 14% if diphenhydramine is used as an adjunct therapy (13). One study evaluated the incidence of akathisia in children after a combination of prochlorperazine and diphenhydramine (11). The authors reported that 12% of the children were reported to have developed akathisia at a 24-h telephone follow-up. The incidence in the ED was not evaluated. In the RCT by Brousseau et al., one child who initially received ketorolac appeared to have developed agitation in the ED 35 min after administration of prochlorperazine (5). His symptoms responded to one dose of diphenhydramine. At the 48-h follow-up, one child was described as having muscle stiffness after discharge that had resolved spontaneously within a few hours. In the other study of prochlorperazine, no side effects were reported among the 20 patients treated (6). In our study, only one patient had a definitive diagnosis of akathisia, but 5 others were suspected of having it, based on the description by the nurses or the physicians, but apparently it was not recognized as such or not considered severe enough to require treatment. These numbers probably do not represent the true incidence of akathisia. Our hypothesis is that akathisia is probably under-diagnosed by pediatric emergency physicians because they do not look for it or it is not recognized by the nurses. Nurses working in a pediatric ED may not be familiar with this symptom. In our study, if symptoms of akathisia occurred after discharge from the ED, they were not severe enough to require a return visit to the ED.
Limitations The main limitation of the present study concerned the diagnosis of migraine. It was not always clear from the chart if patients with a clinical diagnosis of migraine made by the pediatric emergency physician also had the IHS criteria. The modest sensitivity of these criteria has been discussed in comparison to clinical diagnosis (10,16,17). We elected to report all patients with a clinical diagnosis of migraine who received prochlorperazine to reflect everyday practice in which the emergency physician wants to assign a diagnosis to start an effective treatment as soon as possible (10). In our study, 43/92 (47%) patients had the diagnosis confirmed by a pediatric neurologist. The other patients did not return to our ED with an alternate diagnosis within 3 months and did not appear to behave differently. We cannot exclude the possibility that some patients may have presented to another center and had a different diagnosis made there. Another potential limitation is that this was a retrospective chart review. There is always the possibility that
171
something was done but not recorded in the chart. However, our primary outcomes were robust and usually are required to be written in the chart (order of medications, hospitalization, and return visit to the ED). Despite this, however, we again cannot exclude the possibility that patients subsequently presented to another ED or to their family physician due to a recurrence of symptoms. Many studies have shown a high placebo effect with migraine treatment (18,19). Our study had no control group to evaluate the placebo effect. Our results show, however, that prochlorperazine is effective when used in a real-life situation. Doing so, our study design minimized the well-known bias (Hawthorne effect) generated by participating in a clinical trial (20). Despite the fact that a verbal pain score was recorded at triage in most cases, it was not always easy to find a pain score recorded after prochlorperazine administration or any other medication. In most cases, a pain score was recorded at the time of discharge and may reflect the overall efficacy of the treatment in patients who received more than one medication in the ED. However, our study confirmed that prochlorperazine decreases pain significantly; measured either by a decrease of at least 50% in pain intensity or by pain free status at discharge. In fact, prochlorperazine, in some cases with the addition of some form of rescue therapy, was found to decrease pain by a mean of 88%. This is an impressive number, considering that for most patients this was not their first treatment, and that the emergency physicians considered their symptoms to be severe. However, a potential limitation of these results is that the verbal pain scale has not been validated in children in the ED (21). Furthermore, these assessments were not done at standardized times and, thus, are difficult to compare to those from other studies. This is the reason why we decided not to include them in our definition of treatment failure. No objective scale of akathisia, such as the Prince Henry Hospital Rating Scale or the recently simplified version adapted for the ED setting, was used in our ED (22). This certainly limits the findings of our study regarding the rate of akathisia observed in children, especially considering the number of adults who develop akathisia after administration of prochlorperazine (12– 15). If it did appear in more than 1 patient, none was severe enough to require a return visit to our ED.
CONCLUSION In conclusion, our study demonstrates that treatment failure associated with the use of prochlorperazine for severe migraine in children seen in a pediatric ED seems to be 14%. It also seems that prochlorperazine is well tolerated, despite the possible occurrence of akathisia
172
E. D. Trottier et al.
even when diphenhydramine is used to prevent it. Further studies are needed to assess alternative medications in case of treatment failure with prochlorperazine, as well as to identify the optimal medication to prescribe at discharge for children who have a migraine severe enough to require prochlorperazine.
REFERENCES 1. Winner P, Martinez W, Mate L, Bello L. Classification of pediatric migraine: proposed revisions to the IHS criteria. Headache 1995; 35:407–10. 2. Richer L, Graham L, Klassen T, Rowe B. Emergency department management of acute migraine in children in Canada: a practice variation study. Headache 2007;47:703–10. 3. Damen L, Bruijn JKJ, Verhagen AP, et al. Symptomatic treatment of migraine in children: a systematic review of medication trials. Pediatrics 2005;116:295–302. 4. Bailey B, McManus BC. Treatment of children with migraine in the emergency department: a qualitative systematic review. Pediatr Emerg Care 2008;24:321–30. 5. Brousseau DC, Duffy SJ, Anderson AC, Linakis JG. Treatment of pediatric migraine headaches: a randomized, double-blind trial of prochlorperazine versus ketorolac. Ann Emerg Med 2004;43:256 – 62. 6. Kabbouche MA, Vockell AL, LeCates SL, Powers SW, Hershey AD. Tolerability and effectiveness of prochlorperazine for intractable migraine in children. Pediatrics 2001;107:e62. 7. International Headache Society Clinical Trials Subcommittee. Guidelines for controlled trials of drugs in migraine second edition. Cephalalgia 2000;20:765– 86. 8. Pryse-Phillips WEM, Dodick DW, Edmeads JG, et al. Guidelines for the diagnosis and management of migraine in clinical practice. CMAJ 1997;156:1273– 87.
9. Olesen J; Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders. Cephalalgia 2004;24(Suppl 1).9 –160. 10. Winner P. How do we diagnose migraine and childhood syndromes? Curr Pain Head Rep 2005;9:345–50. 11. Cummins McManus B, Bailey B. Status migrainosus in children: a 3 year experience with prochlorperazine [abstract]. Pediatr Res 2004;55:108A. 12. Drotts DL, Vinson DR. Prochlorperazine induces akathisia in emergency patients. Ann Emerg Med 1999;34:469 –75. 13. Vinson DR, Drotts DL. Diphenhydramine for prevention of akathisia induced by prochlorperazine: a randomized, controlled trial. Ann Emerg Med 2001;37:125–31. 14. Vinson DR, Migala AF, Quesenberry CP. Slow infusion for the prevention of akathisia induced by prochlorperazine: a randomized controlled trial. J Emerg Med 2001;20:113–9. 15. Vinson DR. Diphenhydramine in the treatment of akathisia induced by prochlorperazine. J Emerg Med 2004;26:265–70. 16. Winner P, Martinez W, Mate L, Bello L. Classification of pediatric migraine: proposed revisions to the IHS criteria. Headache 1995; 35:407–10. 17. Maytal J, Young M, Shechter A, Lipton RB. Pediatric migraine and the International Headache Society (IHS) criteria. Neurology 1997;4:602–7. 18. Antonaci F, Chimento P, Diener HC, Sances G, Bono G. Lessons from placebo effects in migraine treatment. J Headache Pain 2007; 8:63– 6. 19. Loder E, Goldstein R, Biondi D. Placebo effects in oral triptan trials: the scientific and ethical rational for continued use of placebo controls. Cephalalgia 2005;25:124 –31. 20. Braunholtz DA, Edwards SJ, Lilford RJ. Are randomized clinical trials good for us in the short term? Evidence for a “trial effect”. J Clin Epidemiol 2001;54:217–24. 21. Bailey B, Bergeron S, Gravel J, Daoust R. Comparison of four pain scales in children with acute abdominal pain in a pediatric emergency department. Ann Emerg Med 2007;50:379 – 83. 22. Vinson DR. Development of a simplified instrument for the diagnosis and grading of akathisia in a cohort of patients receiving prochlorperazine. J Emerg Med 2006;31:139 – 45.
Failure with Prochlorperazine
ARTICLE SUMMARY 1. Why is this topic important? Most studies evaluating the treatment of migraine use a primary outcome of pain relief 1 h after treatment. Although this is an important outcome, there are many other clinical outcomes that are important to patients. 2. What does this study attempt to show? The study attempts to evaluate the efficacy of prochlorperazine in terms of treatment failure defined as a patient needing further rescue medication, or being hospitalized, or needing to return to the emergency department (ED) for symptom recurrence or side effects within 48 h of the initial visit. 3. What are the key findings? There seems to be a treatment failure rate of 14% with the use of prochlorperazine in association with diphenhydramine for severe migraine in children seen in a pediatric ED. 4. How is patient care impacted? To be successfully treated for a severe migraine with prochlorperazine, children may need to receive another rescue medication in the ED before discharge. Also, a recurrence of symptoms may appear within the first 48 h after discharge. Thus, patients should be discharged with analgesics.
173
doi:10.1016/j.jemermed.2008.08.012
Original Contributions
CLINICAL OUTCOMES OF CHILDREN TREATED WITH INTRAVENOUS PROCHLORPERAZINE FOR MIGRAINE IN A PEDIATRIC EMERGENCY DEPARTMENT Evelyne D. Trottier,
MD, FRCPC,*
Benoit Bailey, MD, MSC, FRCPC,*† Sabine Dauphin-Pierre,* and Jocelyn Gravel, MD, MSC, FRCPC*
*Division of Emergency Medicine, †Division of Clinical Pharmacology and Toxicology, Department of Pediatrics, CHU Sainte-Justine, Université de Montréal, Montréal, Quebec, Canada Reprint Address: Benoit Bailey, MD, MSC, FRCPC, Department of Pediatrics, CHU Ste-Justine, 3175 Chemin de la Côte-Ste-Catherine, Montréal, QC H3T 1C5, Canada
apy; and 3 patients returned to the ED within 48 h due to symptom recurrence. Conclusion: There was a treatment failure rate of 14% with the use of prochlorperazine in association with diphenhydramine for severe migraine in children seen in a pediatric ED. © 2010 Elsevier Inc.
e Abstract—Background: Prochlorperazine is the only treatment that has been studied so far in a randomized controlled trial and found to reduce pain at 1 h in children with migraine who presented to an emergency department (ED). Objective: To evaluate the rate of treatment failure associated with prochlorperazine used in children with severe migraine in a pediatric ED. Methods: This study was a retrospective chart review of patients < 18 years of age who visited the ED of a tertiary care pediatric hospital between November 2005 and June 2007. All patients diagnosed with migraine by the emergency physicians were included in the study. Charts were evaluated by a data abstractor blinded to the study hypothesis using a standardized datasheet. Inter-rater agreement was measured. Prochlorperazine treatment failure was defined as either administration of further rescue therapy, a hospitalization, or a return visit to the ED within 48 h for symptom recurrence or side effects from the medication. Results: Prochlorperazine was administered in 92 episodes of migraine, including 43 confirmed by a pediatric neurologist; all received diphenhydramine to prevent akathisia. A total of 13 (14%) of these patients had a treatment failure: 8 patients received one or more further rescue therapies after the administration of prochlorperazine; 5 patients were hospitalized, including 3 who had received further rescue ther-
e Keywords—prochlorperazine; migraine; children; emergency department
INTRODUCTION Migraine is the leading cause of acute and recurrent headache at all ages, and studies suggest that as many as 10% of children will eventually suffer from this condition (1). Physicians working in the emergency department (ED) regularly encounter this problem. Despite the frequency of this condition, treatment remains a challenge due to a lack of sufficient studies on migraine therapy in an emergency setting. This may explain the large variability in the clinical practice of emergency physicians for the treatment of migraine in children (2). Two recent meta-analyses that evaluated treatment of migraine and status migrainosus in children reported that only one randomized controlled trial (RCT) had been performed in the ED (3,4). In that RCT, prochlorperazine was better than ketorolac in decreasing pain by 50% or
Presented in part at the Canadian Association of Emergency Medicine annual meeting, Ottawa, Ontario, June 2008.
RECEIVED: 20 May 2008; FINAL ACCEPTED: 5 August 2008
SUBMISSION RECEIVED:
6 June 2008; 166
Failure with Prochlorperazine
providing complete relief by 1 h in children presenting to an ED with migraine (5). A retrospective study of 20 children treated with prochlorperazine but not in an ED, also has been reported and suggested the same efficacy (6). Despite these two studies, not much is known in terms of prochlorperazine efficacy for other important outcomes such as the need for further rescue therapy or the recurrence of symptoms after its administration, as well as its side effects in children (7). Most studies evaluating the treatment of migraine use a primary outcome of pain relief 1 h after treatment. Although this is an important outcome, there are many other clinical outcomes that are important to the patients. The International Headache Society (IHS) Clinical Trial Subcommittee mentions that important patient outcomes include the need for further rescue medication after the study medication and symptom recurrence (7). The objective of our study was to evaluate the treatment failure rate among children treated with a protocol using prochlorperazine as an adjunct to diphenhydramine in an ED. Treatment failure was defined as a patient needing further rescue medication, or being hospitalized, or needing to return to the ED for symptom recurrence or side effects from the medication within 48 h of the initial visit.
MATERIALS AND METHODS Study Design This study was a retrospective chart review of children diagnosed with migraine, as determined by the pediatric emergency physicians, and treated with intravenous prochlorperazine due to the intensity of the migraine.
Setting The cases reviewed were from the ED of a tertiary care pediatric hospital with an annual census of more than 60,000. The ED is staffed with pediatricians and pediatric emergency physicians supervising residents and medical students. Members of the ED use a standardized protocol for the treatment of migraine or status migrainosus.
Study Protocol This protocol suggests a treatment of prochlorperazine (0.15 mg/kg intravenous [i.v.]; maximal dose 10 mg) with diphenhydramine (0.5 mg/kg i.v.; maximal dose 25 mg) for patients over 8 years of age with migraine severe enough that the patient is unable to continue his or her normal
167
activities (8). Moreover, for migraine of moderate intensity, the protocol suggests oral naproxen or ibuprofen, and for migraine of mild intensity, it suggests oral ibuprofen. No additional i.v. fluid treatment is proposed in the protocol. The protocol does not recommend any particular rescue therapy if the primary treatment is not effective. Also, the length of stay in the ED after intravenous treatment with prochlorperazine is determined by the clinical response of the patient. It may vary with each case and each physician. Participants Patients ⬍ 18 years of age who visited the ED between November 2005 and June 2007 and had a final diagnosis of migraine made by the emergency physician were eligible to be included in the study if they also had received prochlorperazine. The ED computerized database was used to identify cases using the terms: Migraine, Migraine with aura, Classic migraine. This database does not include a diagnosis of vascular headache. When physicians are unsure about the diagnosis, they can assign a diagnosis of “headache” to the patient. Such patients were not reviewed. Further, study patients did not necessarily meet the IHS migraine criteria (9). These criteria have been shown to have modest sensitivity, and thus are more restrictive in comparison to expert clinical diagnosis (10). The authors wanted to use real life situations in which clinicians do not want to deny effective treatment to some of these patients, and thus, clinical diagnosis was selected as one of the inclusion criteria. The clinical diagnosis is made by a combination of the past medical history, the family medical history, and the actual symptoms of the patient (such as the presence or absence of throbbing headache, unilateral headache, photophobia, sonophobia, nausea, or vomiting). Some patients had a diagnosis confirmed by a pediatric neurologist from our hospital, which was evaluated by a review of the medical records in our hospital. Furthermore, return visits to the ED within 3 months after the initial diagnosis were identified and evaluated to rule out other possible causes of the patients’ headaches. For our study, all return visits for the same problem within a month of the initial visit were considered to be a single episode of migraine, and thus a recurrence. For the evaluation of prochlorperazine, only its first usage within the episode was evaluated. If a patient presented to the ED more than once but more than 1 month after the initial visit, they were considered as separate episodes. Record Review All identified charts were evaluated by a data abstractor blinded to the study hypothesis using a standardized
168
E. D. Trottier et al.
datasheet. The chart reviewer received 1 week of formal training by the primary investigator on records of patients seen from November 1–15, 2005. Inter-rater agreement between the data abstractor and the primary investigator was measured after the formal training. It was set, a priori, that a kappa score ⱖ 0.80 agreement would be needed for data abstraction to be acceptable. The performance of the data abstractor was evaluated by measuring inter-rater agreement in 10% of the remaining charts. The diagnosis of migraine was confirmed for every chart. The pain score (verbal pain scale from 0 to 10) was noted. The highest score noted by the physician or by the nurse at triage was recorded. In some files, the nurse recorded an interval of pain (0 –3, 4 –7, 8 –10) proposed by our computerized triage system. In those cases, we used the median pain score for the interval (1.5, 5.5, and 9, respectively). We also noted the pain score recorded after treatment in the ED. The need for other medication in the ED after the administration of prochlorperazine was evaluated. We classified the following medications as rescue therapies: oral doses of opiates or triptans, or any additional parenteral medications, including a second dose of prochlorperazine. The following medications were classified as a prophylactic treatment or minor analgesic: acetaminophen, non-steroidal anti-inflammatory drug, or amitriptyline. We also evaluated if the patient returned to the ED for headache within 48 h after treatment. We considered this a recurrence of symptoms after a responsive treatment (7). Finally, the medical charts were evaluated for adverse effects of prochlorperazine, such as akathisia, immediately after treatment during the ED stay and, if available, at any other visits. (No objective scale is used in our ED to evaluate akathisia.) Treatment failure was defined as the occurrence of either the administration of further rescue therapy, hospitalization, or return visits to the ED within 48 h for symptom recurrence or side effects.
would have received prochlorperazine for the treatment of migraine. The study was approved by our hospital’s Institutional Review Board.
Data Analysis
Rescue therapy. In the group of patients who had the diagnosis confirmed by a pediatric neurologist, 5 patients received one or more rescue therapies after the administration of prochlorperazine. Among these, 1 patient received a second dose of prochlorperazine, another patient received a second dose of prochlorperazine along with a dose of metoclopramide, and another patient received a dose of metoclopramide and a dose of opiate. These 3 patients were hospitalized. Two other patients received doses of opiates after prochlorperazine and were later discharged. In the group who did not have a diagnosis confirmed by a pediatric neurologist, 3 patients had a rescue therapy, 1 received a dose of opiate, and the other 2 received intravenous dimenhydrinate.
All data were entered in an Excel database (Microsoft Corporation, Redmond, WA). Statistical analysis was performed using SPSS v15.0 (SPSS Inc., Chicago, IL). Interrater agreement was measured using non-weighted kappa scores.
Sample Size We did not calculate a sample size, but we estimated that the evaluation of all the visits to the ED over an 18month period would provide more than 50 patients who
RESULTS From November 2005 to June 2007, there were 300 episodes of migraine in patients who presented to the ED. Among these patients, 2 were adults and were not included in the study. Another 12 charts from 12 different episodes could not be retrieved and assessed. Thus, over the 20-month period, there were 286 episodes (for a total of 349 visits) in 262 patients (153 girls and 109 boys) that were included in the study. Most patients received no treatment or were treated with medications that were administered orally. Prochlorperazine was administered in 99 episodes in patients with a diagnosis of migraine made by the emergency physician. Seven such episodes were later found not to be related to migraine and were excluded from the analysis. For the remaining 92 episodes occurring in patients between 7 and 17 years of age, 43 were patients who had a diagnosis of migraine confirmed by a pediatric neurologist in addition to the emergency physician. The diagnosis was not confirmed by a pediatric neurologist from our center for the remaining 49 episodes. However, none of these patients returned to the hospital for a reason that suggested an alternative diagnosis within 3 months of the initial visit. Demographic data of the patients included in the study are presented in Table 1.
Outcome of the Migraine A total of 13 (14%) patients had a treatment failure according to our definition (Table 2).
Failure with Prochlorperazine
169
Table 1. Demographic Data of Patients who Received Prochlorperazine in the ED According to the Confirmed or Unconfirmed Diagnosis of Migraine
Variable
Total (n ⫽ 92)
Mean age (SD) in 13.4 ⫾ 2.8 years Male (%) 27 (29) Mean initial pain 7.6 ⫾ 2.0 score (SD) out of 10 Head CT scan done 37 (40) Dose of 0.15 ⫾ 0.02 prochlorperazine (mg/kg) Received 92 (100) diphenhydramine (%) Median length of 6 (4–9) stay in the ED (IQR) in hours Medications at home before ED visit None (%) 17 (18) Acetaminophen 46 (50) (%) NSAID (%) 47 (51) Opiates (%) 6 (6) Prophylactic 10 (11) medications (%) Specific migraine 12 (13) medications (%)
Confirmed by a Pediatric Neurologist (n ⫽ 43)
Not Confirmed (n ⫽ 49)
13.0 ⫾ 2.4
13.7 ⫾ 3.7
17 (40) 7.2 ⫾ 2.4
10 (20) 7.9 ⫾ 1.7
24 (56) 0.15 ⫾ 0.02 43 (100) 5 (4–9.5)
13 (26) 0.15 ⫾ 0.01 49 (100)
[IQR] 5–18) days later in the group of patients who had a diagnosis confirmed. Five patients were not treated and the others received either prochlorperazine (n ⫽ 5), codeine (n ⫽ 2), metoclopramide (n ⫽ 1), or naproxen (n ⫽ 2) alone or in combination. In the unconfirmed diagnosis group, 3 patients returned to the ED a median of 6 (IQR 3–9) days later. One had no treatment. Two patients were treated with a dose of prochlorperazine and one of them was hospitalized. Pain score. There was a difference in pain score at time of discharge compared to pain on arrival for both the group of patients who had their diagnosis confirmed and the group that did not (Table 2). Overall, prochlorperazine decreased pain by a mean of 6.7 ⫾ 2.3 on a scale of 0 to 10.
6 (4–8.5)
Adverse Side Effects 9 (21) 21 (49)
8 (16) 25 (51)
22 (51) 4 (9) 5 (12)
25 (51) 2 (4) 5 (10)
9 (21)
3 (6)
SD ⫽ standard deviation; CT ⫽ computed tomography; ED ⫽ emergency department; IQR ⫽ interquartile range; NSAID ⫽ non-steroidal anti-inflammatory drug.
Hospitalization. In the group of patients who had the diagnosis confirmed by a pediatric neurologist, 5 patients were hospitalized after prochlorperazine administration, including 3 who had also received a rescue therapy. The patients were hospitalized to treat their symptoms or for further investigations. In the group who did not have the diagnosis confirmed by a pediatric neurologist, no patient was hospitalized. Return visit to the ED. After treatment with prochlorperazine and subsequent discharge, one patient in the group of patients who had the diagnosis confirmed by a pediatric neurologist returned to the ED within 48 h. This patient was not treated. In the group of patients who did not have the diagnosis confirmed, 2 patients returned to the ED within 48 h; these patients received prochlorperazine at their second visit and were discharged. Besides the patients who returned to the ED within 48 h and were considered treatment failures, 13 patients returned to the ED within the first month after receiving prochlorperazine, a median of 7 (interquartile range
Akathisia was diagnosed in 1 patient, with the migraine diagnosis confirmed by the pediatric neurologist, despite pretreatment with diphenhydramine (0.5 mg/kg i.v.—the recommended dose in the ED protocol). This patient was successfully treated with a second dose of diphenhydramine. In addition, akathisia was suspected, based on descriptions by nurses and physicians in the charts, in another five cases (three in the confirmed group and two in the
Table 2. Outcome of Patients who Received Prochlorperazine in the ED According to the Confirmed or Unconfirmed Diagnosis of Migraine
Variable
Total (n ⫽ 92)
Total failure (%) 13 (14) Rescue therapy (%)* 8 (9) Hospitalization (%) 5 (5)† Return visit to the ED 3 (3) within 48 h (%) Pain intensity Decrease in pain 6.7 ⫾ 2.3 score (SD) out of 10 Decrease in pain 88 ⫾ 19 score in percent (SD) Decrease of at 74/79 (94) least 50% in pain score (%) Pain free (%) 54/87 (62)
Confirmed by a Pediatric Not Neurologist Confirmed (n ⫽ 43) (n ⫽ 49) 8 (19) 5 (12) 5 (12)† 1 (2)
5 (10) 3 (6) 0 (0) 2 (4)
6.3 ⫾ 2.4
7.0 ⫾ 2.3
87 ⫾ 17
88 ⫾ 21
35/37 (95)
39/42 (92)
23/40 (58)
31/47 (66)
* Rescue therapy ⫽ oral doses of opiates or triptans or any other parenteral medications including a second dose of prochlorperazine. † Three of these patients also had received rescue therapy. ED ⫽ emergency department; SD ⫽ standard deviation.
170
E. D. Trottier et al.
unconfirmed group) due to the presence of symptoms of restlessness. Those cases were not treated. One patient presented with a rash with pruritus on the arm and on the face after prochlorperazine administration. She was successfully treated with a second dose of diphenhydramine. No patients were hospitalized or returned after discharge from the ED solely for akathisia or any other adverse effects.
Discharge After receiving prochlorperazine, patients were discharged on various medications (Table 3). The patient in the confirmed diagnosis group who returned to the ED within 48 h was discharged on naproxen and amitryptiline. In the unconfirmed diagnosis group, one patient who returned to the ED within 48 h had been discharged on naproxen and on amitryptiline, and another such patient on naproxen alone.
Inter-rater Agreement The inter-rater agreement for the 51 charts evaluated twice was excellent: kappa scores of 1.0 for hospitalization status, 0.94 ⫾ 0.11 for the return to the ED, and 0.81 ⫾ 0.25 for the use of a rescue therapy.
DISCUSSION This study reports the clinical outcomes of children who received a combination of prochlorperazine and diphenhydramine for the treatment of migraine in a pediatric ED. The results suggest that 14% of such patients present
Table 3. Medications Received by Patients at the Time of Discharge According to the Confirmed or Unconfirmed Diagnosis of Migraine
Medications No treatment Naproxen Ibuprofen Amitryptiline Codeine Acetaminophen Sumatriptan Metoclopramide Valproic acid
Total by a Pediatric Not Neurologist Confirmed Confirmed (n ⫽ 92) (n ⫽ 43) (n ⫽ 49) (%) (%) (%) 16 (17) 54 (59) 22 (24) 7 (8) 4 (4) 2 (2) 2 (2) 1 (1) 1 (1)
7 (16) 25 (58) 14 (33) 5 (12) 1 (2) 0 (0) 1 (2) 0 (0) 1 (0)
9 (18) 29 (59) 8 (16) 2 (4) 3 (6) 2 (4) 1 (2) 1 (2) 0 (0)
a treatment failure as defined by the need of a second rescue therapy, hospitalization, or a return to the ED within 48 h. The efficacy of prochlorperazine for the treatment of migraine and status migrainosus in children seen in a pediatric ED has been evaluated in only one RCT (5). In that study, prochlorperazine was found to be better than ketorolac for the primary outcome (decrease in the intensity of the migraine by 50% or complete relief by 1 h): 28/33 (85%) vs. 16/29 (55%), respectively, a difference of 30% (95% confidence interval [CI] 8 –52) (5). The results of this study are confirmed by results from a cohort of children with intractable headache or status migrainosus seen in a headache center (6). Fifteen of 20 (75%) patients had a reduction of their pain intensity by at least 50% by 1 h after administration of prochlorperazine, and 19/20 patients by 3 h (6). Another way of measuring the efficacy of a study drug for migraine is to measure the use of rescue medications. In the RCT of Brousseau et al., 4/33 (12%) of the children who received prochlorperazine subsequently needed a rescue medication (i.e., ketorolac) compared to 12/29 (41%) of those initially treated by ketorolac, a difference of ⫺29% (95% CI ⫺49 –7%) (5). In the cohort study, only 1/20 (5%) children received a rescue therapy after prochlorperazine administration (6). In our study, the emergency physicians gave a rescue medication after prochlorperazine in 8/92 (9%) episodes. Another important factor to consider in the efficacy of migraine treatment is the recurrence of symptoms (7). In the RCT by Brousseau et al., recurrences of headache within 48 h after discharge from the ED in children who received only one treatment in the ED, either prochlorperazine or ketorolac, were 7/26 (27%) and 4/13 (31%), respectively, a difference of ⫺4%, (95% CI ⫺34 –27%) (5). Overall, the recurrence was 13/50 (26%), including children who received more than one treatment in the ED (5). In the cohort study by Kabbouche et al., only 2/20 (10%) patients at 24-h follow-up were not pain free (6). In another study published only as an abstract, 18/51 (35%) patients who received prochlorperazine had a recurrence of headache at 24-h telephone follow-up (11). In our study, only 3/92 (3%) patients returned to the ED within 48 h for some recurrence of their symptoms. This does not exclude the possibility that other patients had a recurrence of symptoms. It does suggest that either their symptoms were not severe enough to bring them to the ED again or they may have presented to another center. In our study, 5 (5%) patients were hospitalized after administration of prochlorperazine, suggesting failure to respond; all these cases had a diagnosis of migraine confirmed by a pediatric neurologist. Prochlorperazine, a dopamine receptor antagonist, can precipitate extrapyramidal symptoms, especially akathi-
Failure with Prochlorperazine
sia. In adults, 36% to 44% of patients receiving prochlorperazine present with akathisia (12–15). This percentage decreases to 14% if diphenhydramine is used as an adjunct therapy (13). One study evaluated the incidence of akathisia in children after a combination of prochlorperazine and diphenhydramine (11). The authors reported that 12% of the children were reported to have developed akathisia at a 24-h telephone follow-up. The incidence in the ED was not evaluated. In the RCT by Brousseau et al., one child who initially received ketorolac appeared to have developed agitation in the ED 35 min after administration of prochlorperazine (5). His symptoms responded to one dose of diphenhydramine. At the 48-h follow-up, one child was described as having muscle stiffness after discharge that had resolved spontaneously within a few hours. In the other study of prochlorperazine, no side effects were reported among the 20 patients treated (6). In our study, only one patient had a definitive diagnosis of akathisia, but 5 others were suspected of having it, based on the description by the nurses or the physicians, but apparently it was not recognized as such or not considered severe enough to require treatment. These numbers probably do not represent the true incidence of akathisia. Our hypothesis is that akathisia is probably under-diagnosed by pediatric emergency physicians because they do not look for it or it is not recognized by the nurses. Nurses working in a pediatric ED may not be familiar with this symptom. In our study, if symptoms of akathisia occurred after discharge from the ED, they were not severe enough to require a return visit to the ED.
Limitations The main limitation of the present study concerned the diagnosis of migraine. It was not always clear from the chart if patients with a clinical diagnosis of migraine made by the pediatric emergency physician also had the IHS criteria. The modest sensitivity of these criteria has been discussed in comparison to clinical diagnosis (10,16,17). We elected to report all patients with a clinical diagnosis of migraine who received prochlorperazine to reflect everyday practice in which the emergency physician wants to assign a diagnosis to start an effective treatment as soon as possible (10). In our study, 43/92 (47%) patients had the diagnosis confirmed by a pediatric neurologist. The other patients did not return to our ED with an alternate diagnosis within 3 months and did not appear to behave differently. We cannot exclude the possibility that some patients may have presented to another center and had a different diagnosis made there. Another potential limitation is that this was a retrospective chart review. There is always the possibility that
171
something was done but not recorded in the chart. However, our primary outcomes were robust and usually are required to be written in the chart (order of medications, hospitalization, and return visit to the ED). Despite this, however, we again cannot exclude the possibility that patients subsequently presented to another ED or to their family physician due to a recurrence of symptoms. Many studies have shown a high placebo effect with migraine treatment (18,19). Our study had no control group to evaluate the placebo effect. Our results show, however, that prochlorperazine is effective when used in a real-life situation. Doing so, our study design minimized the well-known bias (Hawthorne effect) generated by participating in a clinical trial (20). Despite the fact that a verbal pain score was recorded at triage in most cases, it was not always easy to find a pain score recorded after prochlorperazine administration or any other medication. In most cases, a pain score was recorded at the time of discharge and may reflect the overall efficacy of the treatment in patients who received more than one medication in the ED. However, our study confirmed that prochlorperazine decreases pain significantly; measured either by a decrease of at least 50% in pain intensity or by pain free status at discharge. In fact, prochlorperazine, in some cases with the addition of some form of rescue therapy, was found to decrease pain by a mean of 88%. This is an impressive number, considering that for most patients this was not their first treatment, and that the emergency physicians considered their symptoms to be severe. However, a potential limitation of these results is that the verbal pain scale has not been validated in children in the ED (21). Furthermore, these assessments were not done at standardized times and, thus, are difficult to compare to those from other studies. This is the reason why we decided not to include them in our definition of treatment failure. No objective scale of akathisia, such as the Prince Henry Hospital Rating Scale or the recently simplified version adapted for the ED setting, was used in our ED (22). This certainly limits the findings of our study regarding the rate of akathisia observed in children, especially considering the number of adults who develop akathisia after administration of prochlorperazine (12– 15). If it did appear in more than 1 patient, none was severe enough to require a return visit to our ED.
CONCLUSION In conclusion, our study demonstrates that treatment failure associated with the use of prochlorperazine for severe migraine in children seen in a pediatric ED seems to be 14%. It also seems that prochlorperazine is well tolerated, despite the possible occurrence of akathisia
172
E. D. Trottier et al.
even when diphenhydramine is used to prevent it. Further studies are needed to assess alternative medications in case of treatment failure with prochlorperazine, as well as to identify the optimal medication to prescribe at discharge for children who have a migraine severe enough to require prochlorperazine.
REFERENCES 1. Winner P, Martinez W, Mate L, Bello L. Classification of pediatric migraine: proposed revisions to the IHS criteria. Headache 1995; 35:407–10. 2. Richer L, Graham L, Klassen T, Rowe B. Emergency department management of acute migraine in children in Canada: a practice variation study. Headache 2007;47:703–10. 3. Damen L, Bruijn JKJ, Verhagen AP, et al. Symptomatic treatment of migraine in children: a systematic review of medication trials. Pediatrics 2005;116:295–302. 4. Bailey B, McManus BC. Treatment of children with migraine in the emergency department: a qualitative systematic review. Pediatr Emerg Care 2008;24:321–30. 5. Brousseau DC, Duffy SJ, Anderson AC, Linakis JG. Treatment of pediatric migraine headaches: a randomized, double-blind trial of prochlorperazine versus ketorolac. Ann Emerg Med 2004;43:256 – 62. 6. Kabbouche MA, Vockell AL, LeCates SL, Powers SW, Hershey AD. Tolerability and effectiveness of prochlorperazine for intractable migraine in children. Pediatrics 2001;107:e62. 7. International Headache Society Clinical Trials Subcommittee. Guidelines for controlled trials of drugs in migraine second edition. Cephalalgia 2000;20:765– 86. 8. Pryse-Phillips WEM, Dodick DW, Edmeads JG, et al. Guidelines for the diagnosis and management of migraine in clinical practice. CMAJ 1997;156:1273– 87.
9. Olesen J; Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders. Cephalalgia 2004;24(Suppl 1).9 –160. 10. Winner P. How do we diagnose migraine and childhood syndromes? Curr Pain Head Rep 2005;9:345–50. 11. Cummins McManus B, Bailey B. Status migrainosus in children: a 3 year experience with prochlorperazine [abstract]. Pediatr Res 2004;55:108A. 12. Drotts DL, Vinson DR. Prochlorperazine induces akathisia in emergency patients. Ann Emerg Med 1999;34:469 –75. 13. Vinson DR, Drotts DL. Diphenhydramine for prevention of akathisia induced by prochlorperazine: a randomized, controlled trial. Ann Emerg Med 2001;37:125–31. 14. Vinson DR, Migala AF, Quesenberry CP. Slow infusion for the prevention of akathisia induced by prochlorperazine: a randomized controlled trial. J Emerg Med 2001;20:113–9. 15. Vinson DR. Diphenhydramine in the treatment of akathisia induced by prochlorperazine. J Emerg Med 2004;26:265–70. 16. Winner P, Martinez W, Mate L, Bello L. Classification of pediatric migraine: proposed revisions to the IHS criteria. Headache 1995; 35:407–10. 17. Maytal J, Young M, Shechter A, Lipton RB. Pediatric migraine and the International Headache Society (IHS) criteria. Neurology 1997;4:602–7. 18. Antonaci F, Chimento P, Diener HC, Sances G, Bono G. Lessons from placebo effects in migraine treatment. J Headache Pain 2007; 8:63– 6. 19. Loder E, Goldstein R, Biondi D. Placebo effects in oral triptan trials: the scientific and ethical rational for continued use of placebo controls. Cephalalgia 2005;25:124 –31. 20. Braunholtz DA, Edwards SJ, Lilford RJ. Are randomized clinical trials good for us in the short term? Evidence for a “trial effect”. J Clin Epidemiol 2001;54:217–24. 21. Bailey B, Bergeron S, Gravel J, Daoust R. Comparison of four pain scales in children with acute abdominal pain in a pediatric emergency department. Ann Emerg Med 2007;50:379 – 83. 22. Vinson DR. Development of a simplified instrument for the diagnosis and grading of akathisia in a cohort of patients receiving prochlorperazine. J Emerg Med 2006;31:139 – 45.
Failure with Prochlorperazine
ARTICLE SUMMARY 1. Why is this topic important? Most studies evaluating the treatment of migraine use a primary outcome of pain relief 1 h after treatment. Although this is an important outcome, there are many other clinical outcomes that are important to patients. 2. What does this study attempt to show? The study attempts to evaluate the efficacy of prochlorperazine in terms of treatment failure defined as a patient needing further rescue medication, or being hospitalized, or needing to return to the emergency department (ED) for symptom recurrence or side effects within 48 h of the initial visit. 3. What are the key findings? There seems to be a treatment failure rate of 14% with the use of prochlorperazine in association with diphenhydramine for severe migraine in children seen in a pediatric ED. 4. How is patient care impacted? To be successfully treated for a severe migraine with prochlorperazine, children may need to receive another rescue medication in the ED before discharge. Also, a recurrence of symptoms may appear within the first 48 h after discharge. Thus, patients should be discharged with analgesics.
173