- Email: [email protected]
Treatment patterns of isolated benign headache in US emergency departments
ORIGINAL CONTRIBUTION
Treatment Patterns of Isolated Benign Headache in US Emergency Departments
From the Department of Emergency Medicine, The Permanente Medical Group, Roseville, CA.
David R. Vinson, MD
Author contribution is provided at the end of this article. Received for publication August 9, 2001. Revision received October 23, 2001. Accepted for publication October 30, 2001. Address for reprints: David R. Vinson, MD, Department of Emergency Medicine, Kaiser Permanente Medical Center, 1600 Eureka Road, Roseville, CA 95661-3027. Copyright © 2002 by the American College of Emergency Physicians. 0196-0644/2002/$35.00 + 0 47/1/121400 doi:10.1067/mem.2002.121400
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
See editorial, p. 334. Study objective: I sought to describe and analyze the treatment of a large representative sample of adult US emergency department patients with isolated primary headache. Methods: Information on adult patients with an isolated diagnosis of migraine headache or unspecified headache was extracted from the 100.4 million ED visits represented by the 1998 National Hospital Ambulatory Medical Care Survey. Demographic and clinical information are presented with descriptive statistics. The treatment of migraine headache was assessed in light of Canadian and US practice parameters. Results: The migraine headache and unspecified headache cohorts included 811,419 and 604,977 participants, respectively. The majority of patients were young, white, and female. Patients received a mean of 1.8 medications from a pharmacopoeia of 36 drugs. Most patients (84.8%) given a diagnosis of migraine headache received a parenteral agent. The most commonly used medications were meperidine (30.0%), ketorolac (21.4%), and prochlorperazine (16.7%). Adjunct antiemetics were commonly administered with parenteral opioids (89.8%). Promethazine and hydroxyzine, antiemetics without antiheadache effects, were used 6 times more commonly as adjuncts than the dopamine antagonists that have established antiheadache effects (ie, prochlorperazine, metoclopramide, droperidol; 78.0% versus 11.8%). The US and Canadian recommendations for the use of nonopioid abortive medications (dopamine-antagonist antiemetics, dihydroergotamine, and 5hydroxytrypamine1 [5-HT1] receptor agonists) are supported by strong evidence. However, parenterally treated patients with migraines received opioids as their only antiheadache medication more commonly than they received any of the aforementioned nonopioids in their regimen (45.7% versus 26.0%). Of all the opioid recipients, most (77%) did not receive any nonopioid
2 1 5
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
abortive headache medication. Meperidine was the most commonly administered opioid (70%). Conclusion: Polypharmacy and a broad pharmacopoeia characterize the US ED treatment of isolated benign headache. Opioid use, particularly meperidine, exceeds that of recommended nonopioid abortive migraine medications. [Vinson DR. Treatment patterns of isolated benign headache in US emergency departments. Ann Emerg Med. March 2002;39:215-222.] INTRODUCTION
Headache is a common condition in the United States for which emergency medical attention is frequently sought.1,2 At the physician’s disposal is an array of pharmacotherapeutic options. For mild-to-moderate headaches not associated with vomiting, a variety of oral agents and a large number of combinations are available. For moderate-to-severe headaches and those associated with vomiting, several categories of parenteral medications are in use, including oxygen, saline solution, nonsteroidal antiinflammatory drugs, dopamine-antagonist antiemetics, ergotamines, 5-hydroxytrypamine1 (5-HT1) receptor agonists, mixed opioid agonist-antagonists, and opioid agonists. The efficacies, hazards, and costs of these respective therapies have been carefully reviewed.3-6 For the management of acute migraine headache, guidelines exist that both rank the level of evidence supporting the use of each medication and recommend sensible treatment strategies.7,8 There is, however, little specific information available regarding current treatment patterns in the United States. The number and types of medications used in emergency medicine for the relief of primary headache disorders have not been examined. Thus, this study was undertaken to describe the treatment of emergency department patients given a diagnosis of benign headache by using a US government national database. M AT E R I A L S A N D M E T H O D S
Data for this study were obtained from the 1998 National Hospital Ambulatory Medical Care Survey (NHAMCS), a national probability sample survey conducted by the National Center for Health Statistics, Centers for Disease Control and Prevention.9 The NHAMCS studies patient visits to nonfederal, short-stay hospital EDs; chroniccare, military, and Veterans Administration hospitals are
2 1 6
excluded. Hospitals with average lengths of stay for all patients of less than 30 days and with specialties of “general” or “children’s general” are eligible for NHAMCS. Freestanding clinics are excluded. EDs are eligible only if they are open and staffed 24 hours a day. Before the survey period, NHAMCS study personnel met with hospital staff to review proper completion of the 1-page patient record form. Information collected for the survey was abstracted by hospital staff from the ED record at or near the time of the patient visit and included patient characteristics (age, sex, race, and ethnicity), provider characteristics (physician type), and visit characteristics, which included expected source of payment, patient’s principle reasons for the visit (to a maximum of 3), physician’s principle diagnoses (to a maximum of 3), and medication therapy (to a maximum of 6). NHAMCS used a 4-stage probability sample design to derive representative national estimates.9 This design involved selection of primary sampling units (which are geographically defined areas), hospitals within primary sampling units, EDs within hospitals, and patient visits within EDs. Visits were randomly selected from randomly assigned 4-week reporting periods. These reporting periods were selected throughout the year and covered all calendar months. Sample data were weighted to produce annual estimates. The representative sample for 1998 consisted of 488 hospitals. Of this group, 410 hospitals had EDs, and 398 (97%) of these participated in the survey. Data were recorded on 24,175 patient visits and weighted to represent an estimated 100.4 million visits to nonfederal, short-stay hospital EDs that year. From the NHAMCS public-use database, 2 categories of headache were isolated: migraine headache and unspecified headache. Diagnoses were determined by using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), listed in the diagnostic section of the NHAMCS database.9 The unspecified headache cohort includes all patients with an isolated, primary diagnosis of headache (ICD-9-CM code 784.0). The migraine headache cohort includes patients with an isolated diagnosis of migraine headache (ICD-9CM codes 346.0, 346.1, 346.2, 346.9) and patients with an isolated dual diagnosis of headache (ICD-9-CM code 784.0) and migraine headache (the aforementioned 346 series). Patients were excluded from the study if they were 15 years of age or younger or if the visit was related to an injury or poisoning. The analysis is restricted to patients with an exclusive diagnosis of headache as noted previously to avoid confounding diagnoses. Some patients with true migraine headache may have received the less-
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
specific diagnosis of unspecified headache (ICD-9-CM code 784.0), and vice versa, unspecified headache may have been mislabeled as migraine headache. Nonetheless, a distinction is maintained between the 2 cohorts throughout the analysis. The database reports only medications received during the ED visit, without specification of timing, repeat dosing, or sequence. Treatment self-administered before ED presentation, medication allergies, effectiveness of ED treatment in regard to symptom resolution, and relapse were not included in the data set. The analytic method is descriptive. The data are presented as percentages, and the results are stratified by headache classification. The pharmacotherapy used in the treatment of migraine headache is further analyzed in light of the recommendation by the Canadian Headache Society7 and the American Academy of Neurology US Headache Consortium.8 Demographic and clinical data that were unknown or missing from the patient record form are explicitly noted. None of the medication data entries for this cohort were illegible. Descriptive analyses were performed with statistical software (Microsoft Excel 2002, version 10.2; Microsoft Corporation, Redmond, WA). The public-use federal data used for this study excluded all patient identifiers, and therefore, the study was exempt from institutional review board approval. R E S U LT S
On the basis of weighted national data, approximately 100.4 million ED visits occurred in 1998. Nearly 5.2 million (5.1%) patients reported that headache or migraine headache was 1 of their 3 chief reasons for the ED visit. Of all the patients with a primary ED diagnosis of migraine headache, 811,419 (81%) patients met our study criteria. Only 604,977 (49%) patients with a primary ED diagnosis of unspecified headache met the inclusion criteria. The presence of coexisting diagnoses was the principal reason for exclusion in both the migraine and unspecified headache cohorts (16% and 31%, respectively), whereas young age (1% and 8%, respectively) and injury or poisoning (3% and 12%, respectively) were less common reasons. Demographic and clinical characteristics of both cohorts are listed in Table 1. The majority of patients were young, white, and female. Overall, pain was described as moderate or severe in 55.3% of the migraine headache cohort and 39.5% of the unspecified headache cohort. Only 12.5% were evaluated by residents or physician assistants, and few patients were hospitalized. Patients were administered from 0 to 5 medications during their
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
ED stay (Table 2). The migraine and the unspecified headache cohorts received a mean of 1.8 and 1.7 medications, respectively. A large variety of medications were administered, including 16 oral and 20 parenteral agents (Table 3). Of those patients from both cohorts who received parenteral opioids, nearly 90% were administered an antiemetic (Table 4). Of the migraine cohort, 688,294 (84.8%) received parenteral therapy. The US8 and Canadian7 recommendations (Table 5)7,8,10-19 for the use of nonopioid
Table 1.
Demographic and clinical characteristics of adult US ED patients in 1998 with isolated benign headache. Migraine Headache (n=811,419) Age, y Mean Range Female sex Ethnicity Hispanic Not Hispanic Blank Race White Black Asian/Pacific Islander American Indian/Eskimo/Aleut Source of payment Private Medicare Medicaid Worker’s compensation Self-pay No charge Other Unknown or blank Health maintenance organization Yes No Unknown or blank Pain level at presentation None Mild Moderate Severe Unknown or blank Provider Staff physician Resident Physician assistant Nurse practitioner Admission
Unspecified Headache (n=604,977)
37.3±12.1 16–89 No. % 634,340 78.2
39.3±14.7 17–86 No. % 421,781 69.7
39,402 644,850 127,167
4.9 79.5 15.7
36,432 448,718 119,827
6.0 74.2 19.8
679,918 131,501 0 0
83.8 16.2 0.0 0.0
496,961 104,220 3,796 0
82.1 17.2 0.6 0.0
365,768 51,455 157,845 0 170,961 2,092 20,514 42,784
45.1 6.3 19.5 0.0 21.1 0.3 2.5 5.3
254,295 53,967 122,144 5,736 116,272 3,261 1,790 47,512
42.0 8.9 20.2 0.9 19.2 0.5 0.3 7.9
135,173 352,585 323,660
16.7 43.5 39.9
59,260 250,611 295,106
9.8 41.4 48.8
6,760 81,514 283,902 164,797 32,555
0.8 10.0 35.0 20.3 4.0
6,078 102,876 165,491 73,401 257,131
1.0 17.0 27.4 12.1 42.5
663,002 57,920 51,614 469 4,104
81.7 7.1 6.4 0.1 0.5
543,056 33,872 33,376 0 27,576
89.8 5.6 5.5 0.0 4.6
2 1 7
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
abortive medications (dopamine-antagonist antiemetics, dihydroergotamine, and 5-HT1 agonists) are supported by strong evidence. However, parenterally treated patients with migraines received opioids as their only abortive agent more commonly than they received any one of the aforementioned nonopioids in their regimen (45.7% versus 26.0%, Table 6). Of all the opioid recipients, most (77%) did not receive any nonopioid abortive medication. Meperidine was the most commonly administered opioid (70%). DISCUSSION
This study demonstrates that headache is a common complaint among US ED patients, many of whom report moderate-to-severe pain. Thirty-six drugs were used in 1998 in the abortive treatment of ED-diagnosed headache. Polypharmacy appears routine because most patients received at least 2 classes of medication. Parenteral agents were commonly used. The most frequently administered medications among all participants in this study were meperidine, ketorolac, and prochlorperazine, in that order. When parenteral opioids were used, adjunct antiemetics were nearly always given (Table 4). However, when pain is uncomplicated by nausea and vomiting, the routine administration of a prophylactic antiemetic may not be necessary.20-22 This is especially true if morphine is used because it causes significantly less drug-induced nausea than meperidine.21 Moderate-to-severe headache often is accompanied by nausea and vomiting, in which
Table 2.
Number of different medications administered to adult US ED patients with isolated benign headache. Migraine Headache (n=811,419) No. of Medications 0 1 2 3 4 5
2 1 8
Unspecified Headache (n=604,977)
No.
%
No.
%
45,783 208,812 428,100 98,538 24,879 5,307
5.6 25.7 52.8 12.1 3.1 0.7
82,127 134,442 275,260 89,994 23,154 0
13.6 22.2 45.5 14.9 3.8 0.0
Table 3.
Medications used in the treatment of adult US ED patients with isolated benign headache. Migraine Headache (n=811,419) Agents
No.
Oral agents Acetaminophen NSAIDs Ibuprofen Naproxen Ketorolac Bromfenac Aspirin Diclofenac Combination medications Vicodin or vicoprofen Fiorcet or fiorcet with codeine Tylenol with codeine Midrin Percocet or percodan Darvocet Zolmitriptan Tramadol Prednisone Parenteral agents Ketorolac Sumatriptan Dihydroergotamine Lidocaine Dexamethasone Antiemetics Antiheadache Prochlorperazine Metoclopramide Droperidol Chlorpromazine General Promethazine Hydroxyzine Diphenhydramine Benztropine Odansetron Opioids Mixed agonist-antagonists Nalbuphine Butorphanol Buprenorphan Agonists Meperidine Morphine Fentanyl
Unspecified Headache (n=604,977) %
No.
%
20,191
2.5
21,309
3.5
22,289 13,398 6,967 6,113 3,248 0
2.7 1.7 0.9 0.8 0.4 0.0
34,553 18,024 1,838 0 5,977 3,132
5.7 3.0 0.3 0.0 1.0 0.5
47,431 29,578
5.8 3.6
65,764 22,665
10.9 3.7
20,828 17,349 13,365 6,210 3,735 0 12,411
2.6 2.1 1.6 0.8 0.5 0.0 1.5
10,553 1,790 25,007 7,758 0 17,569 0
1.7 0.3 4.1 1.3 0.0 2.9 0.0
134,430 79,215 27,348 0 0
16.6 9.8 3.4 0.0 0.0
168,857 13,682 6,079 3,188 2,854
27.9 2.3 1.0 0.5 0.5
126,425 26,703 17,152 10,049
15.6 3.3 2.1 1.2
110,858 21,321 27,044 16,242
18.3 3.5 4.5 2.7
238,869 169,118 17,280 5,188 437 411,350*
29.4 20.8 2.1 0.6 0.1 50.7
119,238 76,463 50,467 0 0 195,328†
19.7 12.6 8.3 0.0 0.0 32.3
63,194 50,886 0
7.8 6.3 0.0
13,491 17,912 3,184
2.2 3.0 0.5
289,990 7,280 0
35.7 0.9 0.0
135,072 23,296 2,373
22.3 3.9 0.4
NSAIDs, Nonsteroidal anti-inflammatory drugs. * A total of 2,426 patients received >1 parenteral opioid (eg, morphine, meperidine). † A total of 8,276 patients received >1 parenteral opioid.
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
case an adjunct antiemetic is indicated. In this situation, an antiemetic with greater efficacy (eg, prochlorperazine> promethazine)23 is to be preferred, as is an antiemetic with demonstrated antiheadache effects (eg, prochlorperazine,7,8,10,11 droperidol,24 or metoclopramide7,8). Yet far more opioid recipients were administered promethazine or hydroxyzine instead of one of the superior antiemetics. If an adjunctive antiemetic is indicated, simply replacing hydroxyzine and promethazine with an antiheadache dopamine antagonist (eg, prochlorperazine or droperidol) would more effectively control the nausea and vomiting, as well as provide an additional therapeutic agent for the headache. Literature-based recommendations on the ED management of moderate-to-severe migraine headache are compromised by the quality of the evidence available.8,25 Although the American Academy of Neurology US Headache Consortium stated that there is insufficient evidence “to establish how to select one therapy over another,” they nevertheless put forth recommendations on the basis of available data, advocating antiemetics (particularly prochlorperazine) plus dihydroergotamine as the “therapy of choice” in the ED and giving high marks to sumatriptan.8 In a recent literature review of abortive migraine pharmacotherapy in emergency medicine, Kelly25 likewise concludes that “the most effective agents seem to be prochlorperazine, chlorpromazine, and sumatriptan, each of which in a number of studies, have achieved greater than 70% efficacy.” These preferred classes of medication are in
Table 4.
Antiemetics used in US ED patients who received parenteral opioids in the treatment of isolated benign headache.
Antiemetic General Promethazine Hydroxyzine Antiheadache Prochlorperazine Droperidol Metoclopramide None
MARCH 2002
39:3
Total (n=595,976)
Migraine Headache (n=408,924)
Unspecified Headache (n=187,052)
No.
%
No.
%
No.
%
261,843 202,984
43.9 34.1
181,113 147,877
44.3 36.2
80,730 55,107
43.2 29.5
47,151 21,747 1,640 60,611
7.9 3.6 0.3 10.2
41,569 6,901 1,640 29,824
10.2 1.7 0.4 7.3
5,582 14,846 0 30,787
3.0 7.9 0.0 16.5
ANNALS OF EMERGENCY MEDICINE
step with the Canadian Headache Society’s algorithm for the treatment of severe migraine headache, in which dopamine-antagonist antiemetics are the agents of choice, to be followed by dihydroergotamine or sumatriptan if needed (Table 5).7 A similar sequence of parenteral medications had been advocated previously for the ED treatment of primary headache disorders.26 Although droperidol was not discussed in the US or Canadian recommendations because the supporting evidence postdates the guidelines, it may in fact be more effective than prochlorperazine,24 which currently is established as one of the best medications in the treatment of moderate-tosevere headache.7,8,10,11 The sedation caused by the dopamine-antagonist antiemetics may be useful in the treatment of headache,8 and the akathisia that commonly develops can be greatly reduced by adjunct diphenhydramine.27 Viewing the actual practice patterns of emergency physicians in light of the treatment recommendations, 2 areas of divergence are noted. On the one hand, only one fourth of patients with migraines who were treated with parenteral medications received one of the nonopioid agents as part of their drug regimen (Table 6). On the other hand, nearly half of parenterally medicated patients with migraines received opioids as their only abortive agent. In fact, opioids were more commonly used as upfront, first-line drugs rather than as second-line drugs reserved for the minority of patients who do not respond to effective nonopioid treatment. This study is not the first to notice such practices.28 Meperidine has been the opioid agonist most studied in the treatment of headache. There is no reason, however, to believe that it is the best of its class for this use. Some consider meperidine to be an inferior opioid in comparison with other less toxic, equianalgesic alternatives (eg, morphine sulfate).20,21 Given the duration of many migraine headaches, a case could be made for longer-acting opioids, like methadone. Although US emergency physicians appear to prefer meperidine over all other parenteral medications in the treatment of isolated benign headache, the evidence is not overwhelmingly in its favor. In their review of the literature through the year 1998, the multidisciplinary US Headache Consortium (endorsed by the American College of Emergency Physicians) found that “the results from the studies with meperidine showed that it was not superior to other effective medications (chlorpromazine IV, methotrimeprazine IM, ketorlac IM, dihydroergo-
2 1 9
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
tamine plus metoclopramide IV).”8 The Canadian Headache Society deemed meperidine “a last resort” in the treatment of moderate-to-severe migraine headache.7 Others agree, asserting that narcotics are “a suboptimal strategy and should not be first-line therapy” (emphasis original).29 Why then is meperidine the apparent drug of choice for headache therapy among US emergency physicians? The reasons for this practice have not been established but are likely to include a complex interaction between patient and physician factors (Figure).30,31 The strengths of this study include the representative nature of its database. Previous studies have suffered from limited external validity because they evaluated practice patterns only in a small number of EDs.32,33 By examining a carefully selected representative sample of 398 EDs across the country, the NHAMCS is able to address national practice patterns. This study also is strengthened by restricting its focus to patients most likely to be free of confounding diagnoses. Children, persons who were
injured or poisoned, and patients with any other discharge diagnosis were excluded from consideration. Moreover, this study presents real-world, provisional ED diagnoses that reflect the way emergency physicians practice day to day. This study has several weaknesses. The greatest shortcoming is the absence of desirable information in the NHAMCS database. It does not include medical therapy tried before the ED visit. Lacking this kind of data, one cannot assess the influence of failed outpatient therapy on the choice of ED treatment. Allergies and medication intolerance, other influential factors, are also absent. Likewise, although the NHAMCS database can report how patients were treated, it can make no comment regarding how well they were treated. Efficacy of therapy, medication side effects, incidence of rebound headache, revisits to the ED, and other outcome measures are unavailable. In addition, the sequential order in which the medications were administered cannot be ascertained with certainty.
Table 5.
Practice guidelines for the parenteral treatment of moderate-to-severe migraine headache. Level of Evidence8*
Medication 1. Effective nonopioid abortive agents 1A. Dopamine-antagonist antiemetics Prochlorperazine Or metoclopramide Or chlorpromazine Or droperidol† 1B. Migraine-specific agents DHE Or 5-HT1 receptor agonists (triptans) 1C. NSAIDs Ketorolac 2. Opioids 2A. Mixed opioid agonist-antagonists Butorphanol 2B. Opioid agonists Meperidine Methadone
1: IV; 2: IM, PR 2: IV 2: IV; 4: IM — 1 1 2
Comments From the US Headache Consortium (US)8 and the Canadian Headache Society (CAN)7
CAN: Recommended as first-line approach. US: Effective in the treatment of associated nausea and vomiting. In some patients with migraine, sedation may be useful. US: Consider as adjunct first-line agent in ED. Direct comparisons found that IV/IM prochlorperazine was significantly superior to IV/IM metoclopramide by the corresponding routes.10,11 US: No difference between IV metoclopramide and IV chlorpromazine.12 US: IV chlorpromazine was not different than IV DHE or IM ketorolac.13,14 — CAN: Consider adding if a 1A agent is ineffective at 30 minutes. US: Restricted to those with no contraindications for its use. US: IV DHE plus antiemetics may be used as therapy of choice in ED. SC DHE had lower incidence of headache recurrence compared with SC sumatriptan.15 SC/IM DHE has considerably less adverse effects than IV. Efficacy of nasal spray is supported by the highest quality of evidence. US: Efficacy supported by the highest quality of evidence. CAN: Suggested if 1A and 1B agents are ineffective. US: Consider for use in ED. CAN: Suggested only “when the preceding treatments are ineffective.”
1: intranasal; 2: IM US: Efficacy supported by the highest quality of evidence. Limit use because of rebound and overuse potential. CAN: Use only “as a last resort.” US: Address the abuse risk. 2 US: Not significantly different than IV chlorpromazine,16 IM ketorolac,17 IV DHE,18 DHE and metoclo pramide.19 Limit use because of headache rebound and dependency. 2: IM —
IV, Intravenous; IM, intramuscular; PR, per rectum; DHE, dihydroergotamine; SC, subcutaneous. * US Headache Consortium level of evidence: 1, proved pronounced statistical and clinical benefit (at least 2 double-blind, placebo-controlled studies and clinical impression of effect); 2, moderate statistical and clinical benefit (1 double-blind, placebo-controlled studies and clinical impression of effect); 3, proved statistically but not proved clinically OR clinically but not proved statistically effective (conflicting or inconsistent evidence); 4, proved to be statistically or clinically ineffective (failed efficacy versus placebo); 5, clinical and statistical benefits unknown (insufficient evidence available). † Data not available before publication of US and Canadian recommendations.
2 2 0
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
One cannot tell whether a patient who received prochlorperazine and meperidine may have received them together, may have required the prochlorperazine as a therapy for meperidine-induced nausea, or may have been given the opioid as a rescue agent after failure of the dopamine antagonist. Also, the database reports composite practice patterns. Variation in management is impossible to discern. Another weakness is the lack of formal, researchoriented diagnostic criteria, as revealed by the frequent diagnosis of unspecified headache, a diagnosis by definition without criteria.34 When compared with official criteria, real-life headache diagnoses commonly are misclassified, both in general practice35 and in emergency medicine.28,36 This may be because it is difficult and often unnecessary to establish a precise diagnosis in the ED2 and because the formal criteria may be in need of revision.37 Some patients given a diagnosis of unspecified headache in this study were believed by the physician to have a migraine headache because a portion of this cohort received migrainespecific medications (ie, either sumatriptan or dihydroer-
Table 6.
Parenteral treatment of US ED patients with migraine headache according to medication classes grouped in descending order.
gotamine). However, patients with migraines have been shown to respond to migraine-specific medication (sumatriptan), regardless of the type of disabling headache they may be experiencing, including tension-type headache.38 Despite the diagnostic inaccuracies inherent in the NHAMCS database, this study demonstrates how patients who are given a diagnosis of primary headache were treated, irrespective of whether their diagnoses were correct. Finally, tension-type headache is among the primary headache disorders but was not studied. The number of nonpoisoned adults who received an isolated diagnosis of nontraumatic tension-type headache was too small for analysis. In conclusion, headache and migraine headache are common complaints in US EDs (1 in 20 patients). The therapy of isolated primary headache is marked by polypharmacy and draws on 36 drugs, the most common of which are meperidine, ketorolac, and prochlorperazine, in that order. Promethazine and hydroxyzine are the antiemetics used more commonly in combination with opioids than superior antiemetics with established antiheadache effects (prochlorperazine, droperidol, and metoclopramide). Because of North American practice guidelines, treatment of isolated migraine headache included effective nonopioid abortive agents (dopamineantagonist antiemetics, dihydroergotamine, and 5-HT1 agonists) far less frequently than opioid agonists. Opioid
Parenterally Treated Patients With Migraine Headache (n=688,294) Drug Class Combinations* 1A/1B and combinations with descending drug classes Either 1A or 1B alone 1A plus 1B (1A or 1B) plus 1C (1A or 1B) plus (2A or 2B) 1C and combinations with descending drug classes 1C alone Plus (2A or 2B) 2A/2B and combination with any other drug class Either 2A or 2B alone (2A or 2B) plus (1A, 1B, or 1C)
No.
%
178,842
26.0
142,386 17,951 18,505 75,919 99,135
20.7 2.6 2.7 11.0 14.4
80,834 18,301 408,924
11.7 2.7 59.4
314,704 94,220
45.7 13.7
*Drug
classes in descending order: 1. Nonopioid abortive migraine medications 1A. Dopamine-antagonist antiemetic 1B. Dihydroergotamine or 5-HT1 agonists 1C. Nonsteroidal anti-inflammatory drugs 2. Opioid analgesics 2A. Mixed agonist-antagonists 2B. Opioid agonists
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
Figure.
Author’s opinions as to why opioids might be used so commonly in the ED treatment of migraine headache. Patient factors 1. Failed to respond to optimal nonopioid outpatient polypharmacy. 2. Allergies, intolerance, or contraindications to nonopioid abortive agents. 3. Established unresponsiveness to all ED nonopioid headache treatments. 4. Requests what has been previously effective. May be unaware that nonopioid medications can be effective. 5. Demands opioids; refuses alternatives. Physician factors 1. Lack of awareness or familiarity with the literature. 2. Disagreement with the studies and the recommendations. May believe opioids should be first-line therapy for ED patients with moderate-tosevere benign headache. 3. Inability to overcome inertia of established individual or group practice patterns. 4. Prefers the expedience of providing an effective opioid with a single intramuscular shot. 5. Difficulty overcoming patient expectations for opioids. a. Avoidance of confrontation in ED. b. Avoidance of patient complaints to administration.
2 2 1
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
recipients often did not receive any nonopioid antiheadache medication during their ED stay. Author contribution: DRV is the sole author and takes responsibility for the paper as a whole.
20. Clark RF, Wei EM, Anderson PO. Meperidine: therapeutic use and toxicity. J Emerg Med. 1995;13:797-802. 21. Silverman ME, Shih RD. Morphine causes less nausea than meperidine when used as an analgesic in the ED [abstract]. Acad Emerg Med. 2001;8:430-431. 22. Talbot-Stern J, Paoloni R. Prophylactic metoclopramide is unnecessary with intravenous analgesia in the ED. Am J Emerg Med. 2000;18:653-657.
I thank Dr. John Edmeads for his critical review of an earlier version of the manuscript.
23. Ernst AA, Weiss SJ, Park S, et al. Prochlorperazine versus promethazine for uncomplicated nausea and vomiting in the emergency department: a randomized, double-blind clinical trial. Ann Emerg Med. 2000;36:89-94.
REFERENCES
24. Miner JR, Fish SJ, Smith SW, et al. Droperidol vs. prochlorperazine for benign headaches in the emergency department. Acad Emerg Med. 2001;8:873-879.
1. Linet MS, Stewart WF, Celentano DD, et al. An epidemiologic study of headache among adolescents and young adults. JAMA. 1989;261:2211-2216. 2. Field AG, Wang E. Evaluation of the patient with nontraumatic headache: an evidence based approach. Emerg Med Clin North Am. 1999;17:127-152. 3.
Bartleson JD. Treatment of migraine headaches. Mayo Clin Proc. 1999;74:702-708.
25. Kelly AM. Migraine: pharmacotherapy in the emergency department. J Accid Emerg Med. 2000;17:241-245. 26. Thomas SH, Stone CK. Emergency department treatment of migraine, tension, and mixedtype headache. J Emerg Med. 1994;12:657-664.
4. Gray RN, McCrory DC, Eberlein K, et al. Parenteral drug treatments for acute migraine headache. Technical review, 2.5. February 1999 (prepared for the Agency for Health Care Policy and Research under Contract No. 290-94-2025). Springfield, VA: National Technical Information Service; 1999. NTIS Accession No. 127862. May be ordered online: http://www.ntis.gov.
27. Vinson DR, Drotts DL. Diphenhydramine for the prevention of akathisia induced by prochlorperazine: a randomized, controlled trial. Ann Emerg Med. 2001;37:125-131.
5.
29. Ward TN, Levin M, Phillips JM. Evaluation and management of headache in the emergency department. Med Clin North Am. 2001;85:971-985.
Ferrari MD. Migraine. Lancet. 1998;351:1043-1051.
6. Capobianco DJ, Cheshire WP, Campbell JK. An overview of the diagnosis and pharmacologic treatment of migraine. Mayo Clin Proc. 1996;71:1055-1066. 7. Pryse-Phillips WE, Dodick DW, Edmeads JG, et al. Guidelines for the diagnosis and management of migraine in clinical practice. Canadian Headache Society. Can Med Assoc J. 1997;156:1273-1287. 8. Matchar DB, Young WB, Rosenberg JH, et al. Evidence-based guidelines for migraine headaches in the primary care setting: pharmacological management of acute attacks. Neurology 2000. American Academy of Neurology US Headache Consortium. Available at: http://www.aan.com/public/practiceguidelines. Accessed December 23, 2001. 9. Centers for Disease Control and Prevention. National Center for Health Statistics, Ambulatory Health Care Data, National Hospital Ambulatory Medical Care Survey, Public-use Data Files, 1998. Available at: http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm. Accessed December 23, 2001. 10. Coppola M, Yealy DM, Leibold RA. Randomized, placebo-controlled evaluation of prochlorperazine versus metoclopramide for emergency department treatment of migraine headache. Ann Emerg Med. 1995;26:541-546. 11. Jones J, Pack S, Chun E. Intramuscular prochlorperazine versus metoclopramide as singleagent therapy for the treatment of acute migraine headache. Am J Emerg Med. 1996;14:262264. 12. Cameron JD, Lane PL, Speechley M. Intravenous chlorpromazine vs intravenous metoclopramide in acute migraine headache. Acad Emerg Med. 1995;2:597-602. 13. Bell R, Montoya D, Shuaib A, et al. A comparative trial of three agents in the treatment of acute migraine headache. Ann Emerg Med. 1990;19:1079-1082.
28. Maizels M. Headache evaluation and treatment by primary care physicians in an emergency department in the era of triptans. Arch Intern Med. 2001;161:1969-1973.
30. Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA. 1999;282:1458-1465. 31. Wyszewianski L, Green LA. Strategies for changing clinicians’ practice patterns. A new perspective. J Fam Pract. 2000;49:461-464. 32. Salomone JA III, Thomas RW, Althoff JR, et al. An evaluation of the role of the ED in the management of migraine headaches. Am J Emerg Med. 1994;12:134-137. 33. Klapper JA, Stanton J. Current emergency treatment of severe migraine headaches. Headache. 1993;33:560-562. 34. The Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias, and facial pain. Cephalalgia. 1988;8(Suppl 7):1-96. 35. Solomon S. Diagnosis of primary headache disorders. Validity of the International Headache Society criteria in clinical practice. Neurol Clin. 1997;15:15-26. 36. Fiesseler FW, Richman PB, Mandell M, et al. The International Headache Society definition of acute migraine headache: do emergency department patients with headaches meet these criteria [abstract]? Acad Emerg Med. 2001;8:508-509. 37. Rains J, Penzien D, Lipchik G, et al. Diagnosis of migraine: empirical analysis of a large clinical sample of atypical migraine (IHS 1.7) patients and proposed revision of the IHS criteria. Cephalalgia. 2001;21:584-595. 38. Lipton RB, Stewart WF, Cady R, et al. Sumatriptan for the range of headaches in migraine sufferers: results of the Spectrum Study. Headache. 2000;40:783-791.
14. Shrestha M, Singh R, Moreden J, et al. Ketorolac vs chlorpromazine in the treatment of acute migraine without aura. A prospective, randomized, double-blind trial. Arch Intern Med. 1996;156:1725-1728. 15. Winner P, Ricalde O, Le Force B, et al. A double-blind study of subcutaneous dihydroergotamine vs subcutaneous sumatriptan in the treatment of acute migraine. Arch Neurol. 1996;53:180-184. 16. Lane PL, McLellan BA, Baggoley CJ. Comparative efficacy of chlorpromazine and meperidine with dimenhydrinate in migraine headache. Ann Emerg Med. 1989;18:360-365. 17. Larkin GL, Prescott JE. A randomized, double-blind, comparative study of the efficacy of ketorolac tromethamine versus meperidine in the treatment of severe migraine. Ann Emerg Med. 1992;21:919-924. 18. Belgrade MJ, Ling LJ, Schleevogt MB, et al. Comparison of single-dose meperidine, butorphanol, and dihydroergotamine in the treatment of vascular headache. Neurology. 1989;39:590592. 19. Scherl ER, Wilson JF. Comparison of dihydroergotamine with metoclopramide versus meperidine with promethazine in the treatment of acute migraine. Headache. 1995;35:256-259.
2 2 2
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
Treatment Patterns of Isolated Benign Headache in US Emergency Departments
From the Department of Emergency Medicine, The Permanente Medical Group, Roseville, CA.
David R. Vinson, MD
Author contribution is provided at the end of this article. Received for publication August 9, 2001. Revision received October 23, 2001. Accepted for publication October 30, 2001. Address for reprints: David R. Vinson, MD, Department of Emergency Medicine, Kaiser Permanente Medical Center, 1600 Eureka Road, Roseville, CA 95661-3027. Copyright © 2002 by the American College of Emergency Physicians. 0196-0644/2002/$35.00 + 0 47/1/121400 doi:10.1067/mem.2002.121400
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
See editorial, p. 334. Study objective: I sought to describe and analyze the treatment of a large representative sample of adult US emergency department patients with isolated primary headache. Methods: Information on adult patients with an isolated diagnosis of migraine headache or unspecified headache was extracted from the 100.4 million ED visits represented by the 1998 National Hospital Ambulatory Medical Care Survey. Demographic and clinical information are presented with descriptive statistics. The treatment of migraine headache was assessed in light of Canadian and US practice parameters. Results: The migraine headache and unspecified headache cohorts included 811,419 and 604,977 participants, respectively. The majority of patients were young, white, and female. Patients received a mean of 1.8 medications from a pharmacopoeia of 36 drugs. Most patients (84.8%) given a diagnosis of migraine headache received a parenteral agent. The most commonly used medications were meperidine (30.0%), ketorolac (21.4%), and prochlorperazine (16.7%). Adjunct antiemetics were commonly administered with parenteral opioids (89.8%). Promethazine and hydroxyzine, antiemetics without antiheadache effects, were used 6 times more commonly as adjuncts than the dopamine antagonists that have established antiheadache effects (ie, prochlorperazine, metoclopramide, droperidol; 78.0% versus 11.8%). The US and Canadian recommendations for the use of nonopioid abortive medications (dopamine-antagonist antiemetics, dihydroergotamine, and 5hydroxytrypamine1 [5-HT1] receptor agonists) are supported by strong evidence. However, parenterally treated patients with migraines received opioids as their only antiheadache medication more commonly than they received any of the aforementioned nonopioids in their regimen (45.7% versus 26.0%). Of all the opioid recipients, most (77%) did not receive any nonopioid
2 1 5
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
abortive headache medication. Meperidine was the most commonly administered opioid (70%). Conclusion: Polypharmacy and a broad pharmacopoeia characterize the US ED treatment of isolated benign headache. Opioid use, particularly meperidine, exceeds that of recommended nonopioid abortive migraine medications. [Vinson DR. Treatment patterns of isolated benign headache in US emergency departments. Ann Emerg Med. March 2002;39:215-222.] INTRODUCTION
Headache is a common condition in the United States for which emergency medical attention is frequently sought.1,2 At the physician’s disposal is an array of pharmacotherapeutic options. For mild-to-moderate headaches not associated with vomiting, a variety of oral agents and a large number of combinations are available. For moderate-to-severe headaches and those associated with vomiting, several categories of parenteral medications are in use, including oxygen, saline solution, nonsteroidal antiinflammatory drugs, dopamine-antagonist antiemetics, ergotamines, 5-hydroxytrypamine1 (5-HT1) receptor agonists, mixed opioid agonist-antagonists, and opioid agonists. The efficacies, hazards, and costs of these respective therapies have been carefully reviewed.3-6 For the management of acute migraine headache, guidelines exist that both rank the level of evidence supporting the use of each medication and recommend sensible treatment strategies.7,8 There is, however, little specific information available regarding current treatment patterns in the United States. The number and types of medications used in emergency medicine for the relief of primary headache disorders have not been examined. Thus, this study was undertaken to describe the treatment of emergency department patients given a diagnosis of benign headache by using a US government national database. M AT E R I A L S A N D M E T H O D S
Data for this study were obtained from the 1998 National Hospital Ambulatory Medical Care Survey (NHAMCS), a national probability sample survey conducted by the National Center for Health Statistics, Centers for Disease Control and Prevention.9 The NHAMCS studies patient visits to nonfederal, short-stay hospital EDs; chroniccare, military, and Veterans Administration hospitals are
2 1 6
excluded. Hospitals with average lengths of stay for all patients of less than 30 days and with specialties of “general” or “children’s general” are eligible for NHAMCS. Freestanding clinics are excluded. EDs are eligible only if they are open and staffed 24 hours a day. Before the survey period, NHAMCS study personnel met with hospital staff to review proper completion of the 1-page patient record form. Information collected for the survey was abstracted by hospital staff from the ED record at or near the time of the patient visit and included patient characteristics (age, sex, race, and ethnicity), provider characteristics (physician type), and visit characteristics, which included expected source of payment, patient’s principle reasons for the visit (to a maximum of 3), physician’s principle diagnoses (to a maximum of 3), and medication therapy (to a maximum of 6). NHAMCS used a 4-stage probability sample design to derive representative national estimates.9 This design involved selection of primary sampling units (which are geographically defined areas), hospitals within primary sampling units, EDs within hospitals, and patient visits within EDs. Visits were randomly selected from randomly assigned 4-week reporting periods. These reporting periods were selected throughout the year and covered all calendar months. Sample data were weighted to produce annual estimates. The representative sample for 1998 consisted of 488 hospitals. Of this group, 410 hospitals had EDs, and 398 (97%) of these participated in the survey. Data were recorded on 24,175 patient visits and weighted to represent an estimated 100.4 million visits to nonfederal, short-stay hospital EDs that year. From the NHAMCS public-use database, 2 categories of headache were isolated: migraine headache and unspecified headache. Diagnoses were determined by using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), listed in the diagnostic section of the NHAMCS database.9 The unspecified headache cohort includes all patients with an isolated, primary diagnosis of headache (ICD-9-CM code 784.0). The migraine headache cohort includes patients with an isolated diagnosis of migraine headache (ICD-9CM codes 346.0, 346.1, 346.2, 346.9) and patients with an isolated dual diagnosis of headache (ICD-9-CM code 784.0) and migraine headache (the aforementioned 346 series). Patients were excluded from the study if they were 15 years of age or younger or if the visit was related to an injury or poisoning. The analysis is restricted to patients with an exclusive diagnosis of headache as noted previously to avoid confounding diagnoses. Some patients with true migraine headache may have received the less-
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
specific diagnosis of unspecified headache (ICD-9-CM code 784.0), and vice versa, unspecified headache may have been mislabeled as migraine headache. Nonetheless, a distinction is maintained between the 2 cohorts throughout the analysis. The database reports only medications received during the ED visit, without specification of timing, repeat dosing, or sequence. Treatment self-administered before ED presentation, medication allergies, effectiveness of ED treatment in regard to symptom resolution, and relapse were not included in the data set. The analytic method is descriptive. The data are presented as percentages, and the results are stratified by headache classification. The pharmacotherapy used in the treatment of migraine headache is further analyzed in light of the recommendation by the Canadian Headache Society7 and the American Academy of Neurology US Headache Consortium.8 Demographic and clinical data that were unknown or missing from the patient record form are explicitly noted. None of the medication data entries for this cohort were illegible. Descriptive analyses were performed with statistical software (Microsoft Excel 2002, version 10.2; Microsoft Corporation, Redmond, WA). The public-use federal data used for this study excluded all patient identifiers, and therefore, the study was exempt from institutional review board approval. R E S U LT S
On the basis of weighted national data, approximately 100.4 million ED visits occurred in 1998. Nearly 5.2 million (5.1%) patients reported that headache or migraine headache was 1 of their 3 chief reasons for the ED visit. Of all the patients with a primary ED diagnosis of migraine headache, 811,419 (81%) patients met our study criteria. Only 604,977 (49%) patients with a primary ED diagnosis of unspecified headache met the inclusion criteria. The presence of coexisting diagnoses was the principal reason for exclusion in both the migraine and unspecified headache cohorts (16% and 31%, respectively), whereas young age (1% and 8%, respectively) and injury or poisoning (3% and 12%, respectively) were less common reasons. Demographic and clinical characteristics of both cohorts are listed in Table 1. The majority of patients were young, white, and female. Overall, pain was described as moderate or severe in 55.3% of the migraine headache cohort and 39.5% of the unspecified headache cohort. Only 12.5% were evaluated by residents or physician assistants, and few patients were hospitalized. Patients were administered from 0 to 5 medications during their
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
ED stay (Table 2). The migraine and the unspecified headache cohorts received a mean of 1.8 and 1.7 medications, respectively. A large variety of medications were administered, including 16 oral and 20 parenteral agents (Table 3). Of those patients from both cohorts who received parenteral opioids, nearly 90% were administered an antiemetic (Table 4). Of the migraine cohort, 688,294 (84.8%) received parenteral therapy. The US8 and Canadian7 recommendations (Table 5)7,8,10-19 for the use of nonopioid
Table 1.
Demographic and clinical characteristics of adult US ED patients in 1998 with isolated benign headache. Migraine Headache (n=811,419) Age, y Mean Range Female sex Ethnicity Hispanic Not Hispanic Blank Race White Black Asian/Pacific Islander American Indian/Eskimo/Aleut Source of payment Private Medicare Medicaid Worker’s compensation Self-pay No charge Other Unknown or blank Health maintenance organization Yes No Unknown or blank Pain level at presentation None Mild Moderate Severe Unknown or blank Provider Staff physician Resident Physician assistant Nurse practitioner Admission
Unspecified Headache (n=604,977)
37.3±12.1 16–89 No. % 634,340 78.2
39.3±14.7 17–86 No. % 421,781 69.7
39,402 644,850 127,167
4.9 79.5 15.7
36,432 448,718 119,827
6.0 74.2 19.8
679,918 131,501 0 0
83.8 16.2 0.0 0.0
496,961 104,220 3,796 0
82.1 17.2 0.6 0.0
365,768 51,455 157,845 0 170,961 2,092 20,514 42,784
45.1 6.3 19.5 0.0 21.1 0.3 2.5 5.3
254,295 53,967 122,144 5,736 116,272 3,261 1,790 47,512
42.0 8.9 20.2 0.9 19.2 0.5 0.3 7.9
135,173 352,585 323,660
16.7 43.5 39.9
59,260 250,611 295,106
9.8 41.4 48.8
6,760 81,514 283,902 164,797 32,555
0.8 10.0 35.0 20.3 4.0
6,078 102,876 165,491 73,401 257,131
1.0 17.0 27.4 12.1 42.5
663,002 57,920 51,614 469 4,104
81.7 7.1 6.4 0.1 0.5
543,056 33,872 33,376 0 27,576
89.8 5.6 5.5 0.0 4.6
2 1 7
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
abortive medications (dopamine-antagonist antiemetics, dihydroergotamine, and 5-HT1 agonists) are supported by strong evidence. However, parenterally treated patients with migraines received opioids as their only abortive agent more commonly than they received any one of the aforementioned nonopioids in their regimen (45.7% versus 26.0%, Table 6). Of all the opioid recipients, most (77%) did not receive any nonopioid abortive medication. Meperidine was the most commonly administered opioid (70%). DISCUSSION
This study demonstrates that headache is a common complaint among US ED patients, many of whom report moderate-to-severe pain. Thirty-six drugs were used in 1998 in the abortive treatment of ED-diagnosed headache. Polypharmacy appears routine because most patients received at least 2 classes of medication. Parenteral agents were commonly used. The most frequently administered medications among all participants in this study were meperidine, ketorolac, and prochlorperazine, in that order. When parenteral opioids were used, adjunct antiemetics were nearly always given (Table 4). However, when pain is uncomplicated by nausea and vomiting, the routine administration of a prophylactic antiemetic may not be necessary.20-22 This is especially true if morphine is used because it causes significantly less drug-induced nausea than meperidine.21 Moderate-to-severe headache often is accompanied by nausea and vomiting, in which
Table 2.
Number of different medications administered to adult US ED patients with isolated benign headache. Migraine Headache (n=811,419) No. of Medications 0 1 2 3 4 5
2 1 8
Unspecified Headache (n=604,977)
No.
%
No.
%
45,783 208,812 428,100 98,538 24,879 5,307
5.6 25.7 52.8 12.1 3.1 0.7
82,127 134,442 275,260 89,994 23,154 0
13.6 22.2 45.5 14.9 3.8 0.0
Table 3.
Medications used in the treatment of adult US ED patients with isolated benign headache. Migraine Headache (n=811,419) Agents
No.
Oral agents Acetaminophen NSAIDs Ibuprofen Naproxen Ketorolac Bromfenac Aspirin Diclofenac Combination medications Vicodin or vicoprofen Fiorcet or fiorcet with codeine Tylenol with codeine Midrin Percocet or percodan Darvocet Zolmitriptan Tramadol Prednisone Parenteral agents Ketorolac Sumatriptan Dihydroergotamine Lidocaine Dexamethasone Antiemetics Antiheadache Prochlorperazine Metoclopramide Droperidol Chlorpromazine General Promethazine Hydroxyzine Diphenhydramine Benztropine Odansetron Opioids Mixed agonist-antagonists Nalbuphine Butorphanol Buprenorphan Agonists Meperidine Morphine Fentanyl
Unspecified Headache (n=604,977) %
No.
%
20,191
2.5
21,309
3.5
22,289 13,398 6,967 6,113 3,248 0
2.7 1.7 0.9 0.8 0.4 0.0
34,553 18,024 1,838 0 5,977 3,132
5.7 3.0 0.3 0.0 1.0 0.5
47,431 29,578
5.8 3.6
65,764 22,665
10.9 3.7
20,828 17,349 13,365 6,210 3,735 0 12,411
2.6 2.1 1.6 0.8 0.5 0.0 1.5
10,553 1,790 25,007 7,758 0 17,569 0
1.7 0.3 4.1 1.3 0.0 2.9 0.0
134,430 79,215 27,348 0 0
16.6 9.8 3.4 0.0 0.0
168,857 13,682 6,079 3,188 2,854
27.9 2.3 1.0 0.5 0.5
126,425 26,703 17,152 10,049
15.6 3.3 2.1 1.2
110,858 21,321 27,044 16,242
18.3 3.5 4.5 2.7
238,869 169,118 17,280 5,188 437 411,350*
29.4 20.8 2.1 0.6 0.1 50.7
119,238 76,463 50,467 0 0 195,328†
19.7 12.6 8.3 0.0 0.0 32.3
63,194 50,886 0
7.8 6.3 0.0
13,491 17,912 3,184
2.2 3.0 0.5
289,990 7,280 0
35.7 0.9 0.0
135,072 23,296 2,373
22.3 3.9 0.4
NSAIDs, Nonsteroidal anti-inflammatory drugs. * A total of 2,426 patients received >1 parenteral opioid (eg, morphine, meperidine). † A total of 8,276 patients received >1 parenteral opioid.
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
case an adjunct antiemetic is indicated. In this situation, an antiemetic with greater efficacy (eg, prochlorperazine> promethazine)23 is to be preferred, as is an antiemetic with demonstrated antiheadache effects (eg, prochlorperazine,7,8,10,11 droperidol,24 or metoclopramide7,8). Yet far more opioid recipients were administered promethazine or hydroxyzine instead of one of the superior antiemetics. If an adjunctive antiemetic is indicated, simply replacing hydroxyzine and promethazine with an antiheadache dopamine antagonist (eg, prochlorperazine or droperidol) would more effectively control the nausea and vomiting, as well as provide an additional therapeutic agent for the headache. Literature-based recommendations on the ED management of moderate-to-severe migraine headache are compromised by the quality of the evidence available.8,25 Although the American Academy of Neurology US Headache Consortium stated that there is insufficient evidence “to establish how to select one therapy over another,” they nevertheless put forth recommendations on the basis of available data, advocating antiemetics (particularly prochlorperazine) plus dihydroergotamine as the “therapy of choice” in the ED and giving high marks to sumatriptan.8 In a recent literature review of abortive migraine pharmacotherapy in emergency medicine, Kelly25 likewise concludes that “the most effective agents seem to be prochlorperazine, chlorpromazine, and sumatriptan, each of which in a number of studies, have achieved greater than 70% efficacy.” These preferred classes of medication are in
Table 4.
Antiemetics used in US ED patients who received parenteral opioids in the treatment of isolated benign headache.
Antiemetic General Promethazine Hydroxyzine Antiheadache Prochlorperazine Droperidol Metoclopramide None
MARCH 2002
39:3
Total (n=595,976)
Migraine Headache (n=408,924)
Unspecified Headache (n=187,052)
No.
%
No.
%
No.
%
261,843 202,984
43.9 34.1
181,113 147,877
44.3 36.2
80,730 55,107
43.2 29.5
47,151 21,747 1,640 60,611
7.9 3.6 0.3 10.2
41,569 6,901 1,640 29,824
10.2 1.7 0.4 7.3
5,582 14,846 0 30,787
3.0 7.9 0.0 16.5
ANNALS OF EMERGENCY MEDICINE
step with the Canadian Headache Society’s algorithm for the treatment of severe migraine headache, in which dopamine-antagonist antiemetics are the agents of choice, to be followed by dihydroergotamine or sumatriptan if needed (Table 5).7 A similar sequence of parenteral medications had been advocated previously for the ED treatment of primary headache disorders.26 Although droperidol was not discussed in the US or Canadian recommendations because the supporting evidence postdates the guidelines, it may in fact be more effective than prochlorperazine,24 which currently is established as one of the best medications in the treatment of moderate-tosevere headache.7,8,10,11 The sedation caused by the dopamine-antagonist antiemetics may be useful in the treatment of headache,8 and the akathisia that commonly develops can be greatly reduced by adjunct diphenhydramine.27 Viewing the actual practice patterns of emergency physicians in light of the treatment recommendations, 2 areas of divergence are noted. On the one hand, only one fourth of patients with migraines who were treated with parenteral medications received one of the nonopioid agents as part of their drug regimen (Table 6). On the other hand, nearly half of parenterally medicated patients with migraines received opioids as their only abortive agent. In fact, opioids were more commonly used as upfront, first-line drugs rather than as second-line drugs reserved for the minority of patients who do not respond to effective nonopioid treatment. This study is not the first to notice such practices.28 Meperidine has been the opioid agonist most studied in the treatment of headache. There is no reason, however, to believe that it is the best of its class for this use. Some consider meperidine to be an inferior opioid in comparison with other less toxic, equianalgesic alternatives (eg, morphine sulfate).20,21 Given the duration of many migraine headaches, a case could be made for longer-acting opioids, like methadone. Although US emergency physicians appear to prefer meperidine over all other parenteral medications in the treatment of isolated benign headache, the evidence is not overwhelmingly in its favor. In their review of the literature through the year 1998, the multidisciplinary US Headache Consortium (endorsed by the American College of Emergency Physicians) found that “the results from the studies with meperidine showed that it was not superior to other effective medications (chlorpromazine IV, methotrimeprazine IM, ketorlac IM, dihydroergo-
2 1 9
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
tamine plus metoclopramide IV).”8 The Canadian Headache Society deemed meperidine “a last resort” in the treatment of moderate-to-severe migraine headache.7 Others agree, asserting that narcotics are “a suboptimal strategy and should not be first-line therapy” (emphasis original).29 Why then is meperidine the apparent drug of choice for headache therapy among US emergency physicians? The reasons for this practice have not been established but are likely to include a complex interaction between patient and physician factors (Figure).30,31 The strengths of this study include the representative nature of its database. Previous studies have suffered from limited external validity because they evaluated practice patterns only in a small number of EDs.32,33 By examining a carefully selected representative sample of 398 EDs across the country, the NHAMCS is able to address national practice patterns. This study also is strengthened by restricting its focus to patients most likely to be free of confounding diagnoses. Children, persons who were
injured or poisoned, and patients with any other discharge diagnosis were excluded from consideration. Moreover, this study presents real-world, provisional ED diagnoses that reflect the way emergency physicians practice day to day. This study has several weaknesses. The greatest shortcoming is the absence of desirable information in the NHAMCS database. It does not include medical therapy tried before the ED visit. Lacking this kind of data, one cannot assess the influence of failed outpatient therapy on the choice of ED treatment. Allergies and medication intolerance, other influential factors, are also absent. Likewise, although the NHAMCS database can report how patients were treated, it can make no comment regarding how well they were treated. Efficacy of therapy, medication side effects, incidence of rebound headache, revisits to the ED, and other outcome measures are unavailable. In addition, the sequential order in which the medications were administered cannot be ascertained with certainty.
Table 5.
Practice guidelines for the parenteral treatment of moderate-to-severe migraine headache. Level of Evidence8*
Medication 1. Effective nonopioid abortive agents 1A. Dopamine-antagonist antiemetics Prochlorperazine Or metoclopramide Or chlorpromazine Or droperidol† 1B. Migraine-specific agents DHE Or 5-HT1 receptor agonists (triptans) 1C. NSAIDs Ketorolac 2. Opioids 2A. Mixed opioid agonist-antagonists Butorphanol 2B. Opioid agonists Meperidine Methadone
1: IV; 2: IM, PR 2: IV 2: IV; 4: IM — 1 1 2
Comments From the US Headache Consortium (US)8 and the Canadian Headache Society (CAN)7
CAN: Recommended as first-line approach. US: Effective in the treatment of associated nausea and vomiting. In some patients with migraine, sedation may be useful. US: Consider as adjunct first-line agent in ED. Direct comparisons found that IV/IM prochlorperazine was significantly superior to IV/IM metoclopramide by the corresponding routes.10,11 US: No difference between IV metoclopramide and IV chlorpromazine.12 US: IV chlorpromazine was not different than IV DHE or IM ketorolac.13,14 — CAN: Consider adding if a 1A agent is ineffective at 30 minutes. US: Restricted to those with no contraindications for its use. US: IV DHE plus antiemetics may be used as therapy of choice in ED. SC DHE had lower incidence of headache recurrence compared with SC sumatriptan.15 SC/IM DHE has considerably less adverse effects than IV. Efficacy of nasal spray is supported by the highest quality of evidence. US: Efficacy supported by the highest quality of evidence. CAN: Suggested if 1A and 1B agents are ineffective. US: Consider for use in ED. CAN: Suggested only “when the preceding treatments are ineffective.”
1: intranasal; 2: IM US: Efficacy supported by the highest quality of evidence. Limit use because of rebound and overuse potential. CAN: Use only “as a last resort.” US: Address the abuse risk. 2 US: Not significantly different than IV chlorpromazine,16 IM ketorolac,17 IV DHE,18 DHE and metoclo pramide.19 Limit use because of headache rebound and dependency. 2: IM —
IV, Intravenous; IM, intramuscular; PR, per rectum; DHE, dihydroergotamine; SC, subcutaneous. * US Headache Consortium level of evidence: 1, proved pronounced statistical and clinical benefit (at least 2 double-blind, placebo-controlled studies and clinical impression of effect); 2, moderate statistical and clinical benefit (1 double-blind, placebo-controlled studies and clinical impression of effect); 3, proved statistically but not proved clinically OR clinically but not proved statistically effective (conflicting or inconsistent evidence); 4, proved to be statistically or clinically ineffective (failed efficacy versus placebo); 5, clinical and statistical benefits unknown (insufficient evidence available). † Data not available before publication of US and Canadian recommendations.
2 2 0
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
One cannot tell whether a patient who received prochlorperazine and meperidine may have received them together, may have required the prochlorperazine as a therapy for meperidine-induced nausea, or may have been given the opioid as a rescue agent after failure of the dopamine antagonist. Also, the database reports composite practice patterns. Variation in management is impossible to discern. Another weakness is the lack of formal, researchoriented diagnostic criteria, as revealed by the frequent diagnosis of unspecified headache, a diagnosis by definition without criteria.34 When compared with official criteria, real-life headache diagnoses commonly are misclassified, both in general practice35 and in emergency medicine.28,36 This may be because it is difficult and often unnecessary to establish a precise diagnosis in the ED2 and because the formal criteria may be in need of revision.37 Some patients given a diagnosis of unspecified headache in this study were believed by the physician to have a migraine headache because a portion of this cohort received migrainespecific medications (ie, either sumatriptan or dihydroer-
Table 6.
Parenteral treatment of US ED patients with migraine headache according to medication classes grouped in descending order.
gotamine). However, patients with migraines have been shown to respond to migraine-specific medication (sumatriptan), regardless of the type of disabling headache they may be experiencing, including tension-type headache.38 Despite the diagnostic inaccuracies inherent in the NHAMCS database, this study demonstrates how patients who are given a diagnosis of primary headache were treated, irrespective of whether their diagnoses were correct. Finally, tension-type headache is among the primary headache disorders but was not studied. The number of nonpoisoned adults who received an isolated diagnosis of nontraumatic tension-type headache was too small for analysis. In conclusion, headache and migraine headache are common complaints in US EDs (1 in 20 patients). The therapy of isolated primary headache is marked by polypharmacy and draws on 36 drugs, the most common of which are meperidine, ketorolac, and prochlorperazine, in that order. Promethazine and hydroxyzine are the antiemetics used more commonly in combination with opioids than superior antiemetics with established antiheadache effects (prochlorperazine, droperidol, and metoclopramide). Because of North American practice guidelines, treatment of isolated migraine headache included effective nonopioid abortive agents (dopamineantagonist antiemetics, dihydroergotamine, and 5-HT1 agonists) far less frequently than opioid agonists. Opioid
Parenterally Treated Patients With Migraine Headache (n=688,294) Drug Class Combinations* 1A/1B and combinations with descending drug classes Either 1A or 1B alone 1A plus 1B (1A or 1B) plus 1C (1A or 1B) plus (2A or 2B) 1C and combinations with descending drug classes 1C alone Plus (2A or 2B) 2A/2B and combination with any other drug class Either 2A or 2B alone (2A or 2B) plus (1A, 1B, or 1C)
No.
%
178,842
26.0
142,386 17,951 18,505 75,919 99,135
20.7 2.6 2.7 11.0 14.4
80,834 18,301 408,924
11.7 2.7 59.4
314,704 94,220
45.7 13.7
*Drug
classes in descending order: 1. Nonopioid abortive migraine medications 1A. Dopamine-antagonist antiemetic 1B. Dihydroergotamine or 5-HT1 agonists 1C. Nonsteroidal anti-inflammatory drugs 2. Opioid analgesics 2A. Mixed agonist-antagonists 2B. Opioid agonists
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
Figure.
Author’s opinions as to why opioids might be used so commonly in the ED treatment of migraine headache. Patient factors 1. Failed to respond to optimal nonopioid outpatient polypharmacy. 2. Allergies, intolerance, or contraindications to nonopioid abortive agents. 3. Established unresponsiveness to all ED nonopioid headache treatments. 4. Requests what has been previously effective. May be unaware that nonopioid medications can be effective. 5. Demands opioids; refuses alternatives. Physician factors 1. Lack of awareness or familiarity with the literature. 2. Disagreement with the studies and the recommendations. May believe opioids should be first-line therapy for ED patients with moderate-tosevere benign headache. 3. Inability to overcome inertia of established individual or group practice patterns. 4. Prefers the expedience of providing an effective opioid with a single intramuscular shot. 5. Difficulty overcoming patient expectations for opioids. a. Avoidance of confrontation in ED. b. Avoidance of patient complaints to administration.
2 2 1
TREATMENT PATTERNS OF ISOLATED BENIGN HEADACHE Vinson
recipients often did not receive any nonopioid antiheadache medication during their ED stay. Author contribution: DRV is the sole author and takes responsibility for the paper as a whole.
20. Clark RF, Wei EM, Anderson PO. Meperidine: therapeutic use and toxicity. J Emerg Med. 1995;13:797-802. 21. Silverman ME, Shih RD. Morphine causes less nausea than meperidine when used as an analgesic in the ED [abstract]. Acad Emerg Med. 2001;8:430-431. 22. Talbot-Stern J, Paoloni R. Prophylactic metoclopramide is unnecessary with intravenous analgesia in the ED. Am J Emerg Med. 2000;18:653-657.
I thank Dr. John Edmeads for his critical review of an earlier version of the manuscript.
23. Ernst AA, Weiss SJ, Park S, et al. Prochlorperazine versus promethazine for uncomplicated nausea and vomiting in the emergency department: a randomized, double-blind clinical trial. Ann Emerg Med. 2000;36:89-94.
REFERENCES
24. Miner JR, Fish SJ, Smith SW, et al. Droperidol vs. prochlorperazine for benign headaches in the emergency department. Acad Emerg Med. 2001;8:873-879.
1. Linet MS, Stewart WF, Celentano DD, et al. An epidemiologic study of headache among adolescents and young adults. JAMA. 1989;261:2211-2216. 2. Field AG, Wang E. Evaluation of the patient with nontraumatic headache: an evidence based approach. Emerg Med Clin North Am. 1999;17:127-152. 3.
Bartleson JD. Treatment of migraine headaches. Mayo Clin Proc. 1999;74:702-708.
25. Kelly AM. Migraine: pharmacotherapy in the emergency department. J Accid Emerg Med. 2000;17:241-245. 26. Thomas SH, Stone CK. Emergency department treatment of migraine, tension, and mixedtype headache. J Emerg Med. 1994;12:657-664.
4. Gray RN, McCrory DC, Eberlein K, et al. Parenteral drug treatments for acute migraine headache. Technical review, 2.5. February 1999 (prepared for the Agency for Health Care Policy and Research under Contract No. 290-94-2025). Springfield, VA: National Technical Information Service; 1999. NTIS Accession No. 127862. May be ordered online: http://www.ntis.gov.
27. Vinson DR, Drotts DL. Diphenhydramine for the prevention of akathisia induced by prochlorperazine: a randomized, controlled trial. Ann Emerg Med. 2001;37:125-131.
5.
29. Ward TN, Levin M, Phillips JM. Evaluation and management of headache in the emergency department. Med Clin North Am. 2001;85:971-985.
Ferrari MD. Migraine. Lancet. 1998;351:1043-1051.
6. Capobianco DJ, Cheshire WP, Campbell JK. An overview of the diagnosis and pharmacologic treatment of migraine. Mayo Clin Proc. 1996;71:1055-1066. 7. Pryse-Phillips WE, Dodick DW, Edmeads JG, et al. Guidelines for the diagnosis and management of migraine in clinical practice. Canadian Headache Society. Can Med Assoc J. 1997;156:1273-1287. 8. Matchar DB, Young WB, Rosenberg JH, et al. Evidence-based guidelines for migraine headaches in the primary care setting: pharmacological management of acute attacks. Neurology 2000. American Academy of Neurology US Headache Consortium. Available at: http://www.aan.com/public/practiceguidelines. Accessed December 23, 2001. 9. Centers for Disease Control and Prevention. National Center for Health Statistics, Ambulatory Health Care Data, National Hospital Ambulatory Medical Care Survey, Public-use Data Files, 1998. Available at: http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm. Accessed December 23, 2001. 10. Coppola M, Yealy DM, Leibold RA. Randomized, placebo-controlled evaluation of prochlorperazine versus metoclopramide for emergency department treatment of migraine headache. Ann Emerg Med. 1995;26:541-546. 11. Jones J, Pack S, Chun E. Intramuscular prochlorperazine versus metoclopramide as singleagent therapy for the treatment of acute migraine headache. Am J Emerg Med. 1996;14:262264. 12. Cameron JD, Lane PL, Speechley M. Intravenous chlorpromazine vs intravenous metoclopramide in acute migraine headache. Acad Emerg Med. 1995;2:597-602. 13. Bell R, Montoya D, Shuaib A, et al. A comparative trial of three agents in the treatment of acute migraine headache. Ann Emerg Med. 1990;19:1079-1082.
28. Maizels M. Headache evaluation and treatment by primary care physicians in an emergency department in the era of triptans. Arch Intern Med. 2001;161:1969-1973.
30. Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA. 1999;282:1458-1465. 31. Wyszewianski L, Green LA. Strategies for changing clinicians’ practice patterns. A new perspective. J Fam Pract. 2000;49:461-464. 32. Salomone JA III, Thomas RW, Althoff JR, et al. An evaluation of the role of the ED in the management of migraine headaches. Am J Emerg Med. 1994;12:134-137. 33. Klapper JA, Stanton J. Current emergency treatment of severe migraine headaches. Headache. 1993;33:560-562. 34. The Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias, and facial pain. Cephalalgia. 1988;8(Suppl 7):1-96. 35. Solomon S. Diagnosis of primary headache disorders. Validity of the International Headache Society criteria in clinical practice. Neurol Clin. 1997;15:15-26. 36. Fiesseler FW, Richman PB, Mandell M, et al. The International Headache Society definition of acute migraine headache: do emergency department patients with headaches meet these criteria [abstract]? Acad Emerg Med. 2001;8:508-509. 37. Rains J, Penzien D, Lipchik G, et al. Diagnosis of migraine: empirical analysis of a large clinical sample of atypical migraine (IHS 1.7) patients and proposed revision of the IHS criteria. Cephalalgia. 2001;21:584-595. 38. Lipton RB, Stewart WF, Cady R, et al. Sumatriptan for the range of headaches in migraine sufferers: results of the Spectrum Study. Headache. 2000;40:783-791.
14. Shrestha M, Singh R, Moreden J, et al. Ketorolac vs chlorpromazine in the treatment of acute migraine without aura. A prospective, randomized, double-blind trial. Arch Intern Med. 1996;156:1725-1728. 15. Winner P, Ricalde O, Le Force B, et al. A double-blind study of subcutaneous dihydroergotamine vs subcutaneous sumatriptan in the treatment of acute migraine. Arch Neurol. 1996;53:180-184. 16. Lane PL, McLellan BA, Baggoley CJ. Comparative efficacy of chlorpromazine and meperidine with dimenhydrinate in migraine headache. Ann Emerg Med. 1989;18:360-365. 17. Larkin GL, Prescott JE. A randomized, double-blind, comparative study of the efficacy of ketorolac tromethamine versus meperidine in the treatment of severe migraine. Ann Emerg Med. 1992;21:919-924. 18. Belgrade MJ, Ling LJ, Schleevogt MB, et al. Comparison of single-dose meperidine, butorphanol, and dihydroergotamine in the treatment of vascular headache. Neurology. 1989;39:590592. 19. Scherl ER, Wilson JF. Comparison of dihydroergotamine with metoclopramide versus meperidine with promethazine in the treatment of acute migraine. Headache. 1995;35:256-259.
2 2 2
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002