Diagnostics

Canadian Headache Society criteria for the diagnosis of acute migraine headache in the ED-do our patients meet these criteria?

Frederick W. Fiesseler DO^a,*, Renee L. Riggs DO^b, William Holubek MD^c,

Barnet Eskin MD, PhD^a, Peter B. Richman MD^d

^aDepartment of Emergency Medicine, Morristown Memorial Hospital, Morristown, NJ 07962, USA ^bDepartment of Emergency Medicine, Cooper Hospital/University Medical Center, Camden, NJ 08103, USA ^cDepartment of Emergency Medicine, Maimonides Medical Center, Brooklyn, NY 11219, USA

^dDepartment of Emergency Medicine, Mayo Clinic Hospital, Scottsdale, AZ 85259, USA

Received 16 December 2003; accepted 1 March 2004

Previous Presentation: ACEP Scientific Assembly, Boston 2003.

Abstract

Introduction: We previously reported that many patients who present to the ED with bmigraineQ head- ache do not meet the International Headache Society criteria (IHSC) for the diagnosis of acute migraine. Objective: The aim of the study was to compare the frequency for which ED patients with migraine headache meet the Canadian Headache Society criteria (CHSC) vs the IHSC.

Methods: This was a prospective, observational study, performed at a community ED. Consecutive patients who presented to study authors with a chief complaint of headache were enrolled. Historical/ clinical data were collected on a standardized form. Ninety-five percent confidence intervals (95% CIs) were calculated and Fisher exact test was used as appropriate.

Results: One hundred eighty-nine patients were enrolled in this study. Mean age was 38 years. Females comprised 69% of patients. Thirty-seven percent of patients had prior ED visits for headaches. A positive Family history of MIgraines was present in 35% of patients. Diagnostic imaging was previously performed in 44 of the enrollees to evaluate the cause of their headaches. A total of 43 (23%) patients had a prior diagnosis of migraine. Overall CHSC was met in 18% of patients, compared with 15% of patients who met IHSC. Discharge diagnosis of migraine was made in 41% of patients. Of these patients, 33% met CHSC and 28% met IHSC ( P = .30). For patients with discharge diagnosis of migraine, 33% of females and 36% of males fit CHSC ( P = .53), whereas 26% and 36% met IHSC ( P = .34), respectively. For patients with a prior diagnosis of migraine, 32% met CHSC and 26% met IHSC ( P = .24). Patients with a prior diagnosis of migraine and/or a discharge diagnosis of migraine met CHSC 31% (95% CI, 22%- 40%) of the time vs 25% for the IHSC (95% CI, 16%-34%) ( P = .26). Four patients without a discharge and/or previous diagnosis of migraine met CHSC; 3 met IHSC.

* Corresponding author. Tel.: +1 973 971 8919; fax: +1 973 290 7202.

E-mail address: [email protected] (F.W. Fiesseler).

0735-6757/$ - see front matter D 2005 doi:10.1016/j.ajem.2004.03.001

Conclusions: In our study population, only a minority of patients with headache who have prior diagnosis and/or ED diagnosis of migraine headache met CHSC. The utility of CHSC and/or IHSC to standardize ED patients for headache research may be limited.

D 2005

Introduction

migraine headaches are an incapacitating disease ac- counting for 2.8 million physician encounters each year [1]. Osterhaus et al [2] estimated that $2 to 17.2 billion dollars are lost in productivity each year secondary to this aliment. ED visits to alleviate such symptoms are common.

The recognized bgold standardQ for the diagnosis of a migraine remains the International Headache Society criteria (IHSC) (Fig. 1). Use of the IHSC in emergency medicine was recently endorsed by the American College of Emer- gency Physicians when it was suggested that emergency

physicians follow the guidelines set forth by the Headache Consortium in 2000 [3]. Use of the IHSC has standardized the diagnosis of migraines; however, its clinical relevance with respect to ED therapy remains questionable. Olesen and Lipton [4] notes that these criteria are not based on scientific literature. Furthermore, a recent study at our institution demonstrated that only a minority of patients discharged by emergency physicians with a previous and/ or discharge diagnosis of migraine met the IHSC [5].

In 1997, the Canadian Headache Society published a modified version of the IHSC [6] (Fig. 2). These criteria were created following a needs assessment performed by

Fig. 1 International Headache Society diagnostic criteria.

Fig. 2 Canadian Headache Society diagnostic criteria.

members of the Canadian Headache Society [6]. Adjust- ments to the diagnosis of migraine without aura included expanding Headache duration, more flexible pulsatile symptoms, and adding osmophobia as a possible diagnostic criterion. In addition to the changes made to the IHSC criteria, they suggest adding additional questions to improve pattern recognition of migraine [6].

We suspected that the Canadian Headache Society criteria (CHSC) might allow for broader standardization of patients with vascular-type headaches who present to the ED. The purpose of this study was to compare the frequency for which ED patients with migraine headache met the CHSC vs the IHSC.

Methods

Study design

This was a prospective observational study.

Setting

The study was conducted in the Department of Emer- gency Medicine at Morristown Memorial Hospital, a community-based tertiary care center in northern New Jersey between October 2001 and August 2002. The ED has an annual census of approximately 65000 visits and has an active academic program and emergency medicine residency. The institutional review board at our institution approved this study before patient enrollment.

Population

We enrolled a convenience sample of adult patients presenting to the ED with a chief complaint of headache. Each of these patients presented to 1 of 6 emergency physicians. Patients were excluded for any of the following reasons: (1) medically unstable as determined by the attend- ing physician; (2) patient found to be disoriented, intoxi-

cated, and/or with an altered mental status; (3) headache associated with head trauma. In addition, patients were also excluded if they declined permission to participate.

Study protocol

Informed consent was obtained from participating pa- tients before enrollment. Patients and physicians recorded relevant demographic and clinical variables on a standard- ized data collection form. Both the IHSC and the CHSC

Table 1 Characteristics of patients with headache enrolled in the study
	Overall	CHSC	IHSC
No. of patients	189	34 (18%)	28 (15%)
Age (mean, y)	38	35	38
Sex
Female	130 (69%)	27 (79%)	22 (78%)
Male	59 (31%)	7 (21%)	6 (22%)
Race
Asian	8 (4%)	1 (3%)	1 (3%)
Black	25 (13%)	3 (8%)	1 (3%)
Caucasian	125 (66%)	25 (74%)	21 (75%)
Hispanic	23 (12%)	5 (15%)	5 (18%)
Other	8 (4%)	0	0
Previous ED visits	70 (37%)	24 (71%)	19 (68%)
Neurologist following	45 (24%)	14 (41%)	13 (46%)
Family history	66 (35%)	18 (53%)	14 (50%)
Previous medications
None	39 (20%)	4 (12%)	2 (7%)
b-Blocker	2 (1%)	0	0
Ergotamine	1 (1%)	0	0
Narcotic	23 (12%)	4 (12%)	4 (14%)
NSAID	92 (49%)	13 (38%)	11 (40%)
Antidepressant	6 (3%)	2 (6%)	2 (7%)
Antiemetic	3 (2%)	0	0
Acetominophen	45 (24%)	4 (12%)	3 (11%)
Imitrex	22 (12%)	9 (24%)	7 (25%)
NSAID indicates nonsteroidal anti-inflammatory drug.

were not identified among the variables addressed on the instrument.

Statistical analysis

Data were entered into Microsoft Excel for Windows (Microsoft Corporation, Redmund, Wash). Categorical data were analyzed by Fisher exact tests. Ninety-five percent confidence intervals (95% CIs) were calculated as appro- priate. All tests were 2-tailed with a set at .05. The primary outcome parameter was the number of patients diagnosed with migraine upon discharge, and/or who had received a previous diagnosis of migraine, who met either the CHSC or the IHSC.

Results

Between October 2001 and August 2002, a total of 189 patients were enrolled. All eligible patients agreed to participate and were surveyed. Mean age of participants was 38 years (F13.7 SD); females comprised the majority of enrollees (69%) and 66% were Caucasian (Table 1).

With respect to historical features, 70 (37%) of patients had previous ED visits for similar headaches. Imaging modalities (ie, head computed tomography scan and/or magnetic resonance imaging) had been performed in 44% of enrollees previously to determine the cause of their head-

aches. With respect to family history, 35% of patients had a family member with a Chronic headache pattern. Physicians had previously provided a diagnosis of the headache as migrainous in 81 (43%) patients.

Only a minority of patients met either the CHSC or the IHSC. Of the total patients enrolled in our study, only 34 (18%) met the CHSC compared with 28 (15%) meeting the IHSC for the diagnosis of migraine. Discharge diagnosis of migraine was made by the caring emergency physician in 41% of patients overall (Fig. 3). Of these patients, no statistical improvement was demonstrated with regard to either criterion. Of those patients discharged with the diagnosis of migraine, 33% (26 of 78) met CHSC and 28% (22 of 78) met IHSC ( P = .30). For patients with a prior diagnosis of migraine, 32% (26 of 81) fit CHSC and 26% (21 of 81) fulfilled IHSC ( P = .24). No significant differences existed in relation to sex. Only 33% (21 of 64) of females and 36% (5 of 14) of males met CHSC ( P = .53), whereas 26%

(17 of 64) and 36% (5 of 14) met IHSC ( P = .34),

respectively. Patients with a prior diagnosis of migraine and/or a discharge diagnosis of migraine met CHSC 31% (29 of 95) of the time (95% CI, 22%- 40%) vs 25% (24 of 95) (95% CI, 16%-34%) for the IHSC ( P = .26). Uniformly, no differences occurred with regard to either criteria used. Four patients without a discharge or previous bmigraineQ diagnosis met CHSC. This also occurred in 3 patients who fulfilled migraine diagnosis using the IHSC.

Fig. 3 Discharge diagnosis by number of patients.

Analysis of the individual modifications in the CHSC revealed no significant changes from the IHCS. Addition of pulsatile nature or osmophobia did not increase the number of patients who met diagnostic criteria for migraine. Six more patients were diagnosed with migraine, solely based on increase in headache duration.

Discussion

Members of the Canadian Headache Society used a multidisciplinary group reviewing literature and comparing alternative clinical pathways to develop their standard definition of migraine headache. Modifications of the IHSC were intended on improving the reliability of interpretation of the individual diagnostic criteria. The group based their recommendations on the most informative, statistically appropriate references that were available at the time [6].

The first modification pertains to duration. The IHSC requires a duration of 4 to 72 hours. The CHSC expands on this to include those headaches from 2 hours up until 72 hours. With this modification alone, there was not statisti- cally significant increase in patients who met a standard definition of migraine; only 6 more patients were included based on this criterion. Second, the CHSC inclusion of osmophobia (fear of odors) as a feature of migraine did not result in any additional patients who could be considered as having a migraine. Overall, osmophobia was found in 29% of all patients with headache presenting to the ED. Likewise, the addition of pulsating quality at any phase of the attack failed to add additional patients to the diagnosis of migraine by CHSC. Further analysis of this symptom demonstrated that no patient diagnosis was included or excluded based on pulsating quality for the IHSC or CHSC. Overall, our results demonstrate that most patients in our ED population, who were diagnosed with migraine head- aches previously and/or in the ED, did not fit the CHSC. These results mirror those of our previous study, in which we found that most patients with migraine evaluated in the ED did not meet the IHSC [5]. In our current study, the modifications of the CHSC did not provide for the opportunity to standardize additional patients as having

migraine headaches when compared with the IHSC.

The utility of strict inclusion criteria for migraines has been questioned. In a prospective observational study of migraine patients, Ducharme et al [7] demonstrated that at

24 to 72 hours after discharge, no significant difference occurred in migraine recurrence between those meeting the IHSC and those diagnosed at the discretion of the ED attending. Analysis using the CHSC was not performed, but the results of this study lead us to conclude that the results would be similar.

Thus, the paradox for investigators in the area of ED headache research remains with respect to classification. Attempts to define migraines by rigid criteria such as the IHSC or CHSC will inevitably result in exclusion of the majority of patients who present to the ED with a bbenignQ headache. On the other hand, failure to standardize patients in headache research will result in inclusion of patients with different etiologies who may respond differently to various therapies. In the study of Vinson et al [8] of 490 benign headaches, it was found that more than 40% of patients were discharged with the diagnosis of unspecified headache. Similarly, 37% of patients in both our previous and present study had the nebulous diagnosis of undifferentiated headache upon discharge [5].

We suggest further modification of the IHSC for the purpose of ED research. For example, criteria for number of previous episodes could be decreased, but not eliminated. It is important to recognize the recurrence of this aliment. Limitations based on duration have minimal clinical relevance. People often have difficulty estimating this, leading to inaccuracy and questionable validity. The expanded duration used by the CHSC seems appropriate. Osmophobia and pulsatile quality should be excluded from diagnostic criteria as both appear to add little to standard- izing patients. It is anticipated that these modifications would be as sensitive and less onerous without a significant loss of specificity (Fig. 4). Implementation, along with validation, of these modifications should ultimately be correlated with treatment modalities.

The utility of diagnostic criteria is important, not only for research but also to allow physicians to standardize diagnostic and treatment modalities. Stringent diagnostic criteria theoretically can assist in eliminating other concerning underlying diseases and further eliminate

Fig. 4 Modified IHSC.

needless diagnostic testing. These, in essence, can be used as a clinical guideline to help streamline physician’s diagnostic modalities, limiting resources, while saving time, and money. All of which is paramount in medicine today. Is it imperative that we make an accurate diagnosis of migraine and that clinical criteria are both validated and used.

Limitations and future questions

The major limitation of this study is the potential for selection bias. Enrollment of patients was based on a convenience sample, not a true consecutive series. While on duty, study emergency physician investigators enrolled consecutive patients who were eligible for the study. Work schedules of each varied, including mornings, evenings, and overnight shifts. Therefore, we suspect that our study population sample represented our diverse patient popula- tion with headache as a whole.

This study leaves multiple questions for further research and debate in view of our findings. Only a minority of patients fit either the CHSC or the IHSC. Should we continue to use these as guides for future research? What diagnostic criteria should further migraine studies use? Does the use of the IHSC /CHSC, which excludes a large segment of the headache population, render this research, in many ways, less useful? Should further modification of these criteria be considered? What is the Treatment modality of choice for bundifferentiated headaches,Q and will those headaches thought to be migrainous respond to similar therapeutic modalities of analgesic control?

Conclusion

In our study population, less than half of patients with headache who have prior diagnosis and/or ED diagnosis of migraine headache met CHSC. The utility of CHSC and IHSC to standardize ED patients for headache research may be limited. Further modification of the CHSC/IHSC for emergency medicine research should be considered.

References

1. Schappert SM, Nelson C. National Ambulatory Medical Care Survey, 1995-1996 summary. National center for health statistics. Vital Health Stat 1999;13:142.
2. Osterhaus JT, Gutterman JT, Plachetka JR. Health care resources and lost labor costs of migraine headaches in the United States. Pharmacoeconomics 1992;2:67 - 76.
3. Pietrzak MP. ACEP endorses headache guidelines. ACEP News 2000;9:5.
4. Olesen J, Lipton RB. Migraine classification and diagnosis. Neurology 1994;4(Suppl 4):6 - 10.
5. Fiesseler FW, Kec R, Mandell M, et al. The international headache society definition of acute migraine headache-do emergency depart- ment patients with headaches meet these criteria? Am J Emerg Med 2002;20:618 - 23.
6. Pryse-Phillips WE, Dodick DW, Edmeads JG, et al. Guidelines for the diagnosis and management of migraines in clinical practice. CMAJ 1997;156:1273 - 87.
7. Ducharme J, Beveridge RC, Lee JS, Beaulieu S. Emergency management of migraine: is the headache really over? Acad Emerg Med 1998;5:899 - 905.
8. Vinson DR, Hurtado TR, Vandenberg JT. Banwart. Variations among emergency departments in the treatment of benign headaches. Ann Emerg Med 2003;41:90 - 8.