HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tsushima, Y. Articles by Endo, K. Search for Related Content PubMed PubMed Citation Articles by Tsushima, Y. Articles by Endo, K.

MR Imaging in the Evaluation of Chronic or Recurrent Headache¹

¹ From the Department of Radiology, Motojima General Hospital, 3–8 Nishi-Honcho, Ohta, Gunma 373-0033, Japan (Y.T.); and Department of Diagnostic Radiology and Nuclear Medicine, Gunma University Hospital, Gunma, Japan (K.E.). Received December 30, 2003; revision requested March 2, 2004; final revision received June 23; accepted July 26. Address correspondence to Y.T. (e-mail: yoshito@xa2.so-net.ne.jp).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To evaluate ability of magnetic resonance (MR) imaging to depict an abnormality in patients with chronic or recurrent headache without neurologic abnormality.

MATERIALS AND METHODS: Institutional review board approval and patient informed consent were not required. A total of 306 patients with normal neurologic findings and chronic or recurrent headache were examined with MR imaging. Patients were divided into three groups: those with no abnormality, those with minor abnormality, and those with clinically important intracranial abnormality, which may result in chronic or recurrent headache. Literature review was also performed. Upper 99.5% confidence bound for frequency of abnormal MR findings was calculated.

RESULTS: A total of 169 patients (55.2%) were placed in the first group, 135 (44.1%) were placed in the second group, and two (0.7%) were placed in the third group because they had a clinically important abnormality at MR imaging. Neither contrast material enhancement (n = 195) nor repeated MR imaging (n = 23) contributed to the diagnosis. Literature review revealed two previous studies concerning unspecified headache (in addition to the current study), including a total of 1036 MR imaging results and 22 (2.1%) clinically important results (upper 99.5% confidence bound, 3.4%). Twelve studies of migraine headache were found, with a total of 790 MR imaging examinations. Excluding the 19 patients with complicated migraine, the 99.5% confidence bound of the frequency of clinically important abnormality at MR imaging was estimated as 0.68%. Clinically important infarctions were noted on MR images in five (26.3%) of 19 patients with complicated migraine.

CONCLUSION: MR imaging is an unrewarding technique in the evaluation of patients with chronic or recurrent headache and normal neurologic findings. Neither contrast enhancement nor repeated MR imaging contributed to diagnosis, although the number of patients in the latter category was small.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Headache is one of the most common complaints that leads patients to see their physician. Although the majority of patients who present with chronic or recurrent headache have no neurologic abnormality, many patients undergo evaluation with computed tomography (CT) and, more recently, magnetic resonance (MR) imaging to exclude important abnormalities. Previous studies have demonstrated that CT is of extremely low yield in patients who undergo imaging for chronic headache without neurologic abnormality (1–3).

MR imaging is more sensitive than CT in the detection of an intracranial abnormality (4). Patients’ demand for thorough and high-tech evaluation and the increasing practice of definitive medicine seem to accelerate the use of MR imaging in the evaluation of headache. Despite this, limited data exist about the utility of head MR imaging in patients with chronic or recurrent headache with normal findings of a neurologic examination. Thus, the purpose of our study was to evaluate the ability of MR imaging to depict an abnormality in patients with chronic or recurrent headache without a neurologic abnormality.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Subjects
Of the head MR examinations performed at Motojima General Hospital, we retrospectively reviewed the clinical data and MR studies of 360 adult patients, all of whom were referred for MR evaluation of chronic or recurrent headache between September 2000 and August 2003. Patients’ charts and written reports of MR imaging findings were reviewed by one of the authors (Y.T.). Inclusion criteria were as follows: (a) chief complaint of chronic or recurrent headache with duration of 1 month or more, (b) no other neurologic symptoms or focal findings at physical examination, and (c) no prior head surgery, head trauma, or seizure. Of the 360 patients considered, 306 (136 men and 170 women) met these criteria, and the sample age range was 19–91 years (mean age ± standard deviation, 54.2 years ± 15.2). There was no difference in the age distribution according to sex (Table 1) (P = .58, ² test).

MR Imaging
All subjects underwent examination with the same 1.0-T MR imager (Magnex; Shimadzu, Kyoto, Japan), and transverse T1-weighted spin-echo (repetition time msec/echo time msec, 450/15) and proton-density-weighted and T2-weighted fast spin-echo (4000/20, 100; echo train length, eight) images were obtained. The matrix was 256 x 192, and the section thickness was 5 mm, with a gap of 2.5 mm for all sequences. Although we did not have a definite criterion for the indication of intravenous contrast material administration, contrast material–enhanced transverse T1-weighted images were obtained by using gadopentetate dimeglumine (Magnevist; Nihon Schering, Osaka, Japan) if a more detailed examination was either recommended by the radiologist or the patient’s physician or demanded by the patient. All MR images were interpreted by one of the authors (Y.T., with 15 years of experience as a general radiologist), and the images were not reinterpreted for the purpose of this study.

One of the authors (Y.T.) reviewed the reports of all head MR images, and the MR imaging results were divided into three groups: (a) those with no abnormality; (b) those with a minor abnormality, such as ischemic or atrophic changes, which neither explained the reason for headache nor changed the clinical or therapeutic approach; and (c) those with a clinically important intracranial abnormality, such as space-occupying lesions or bleeding, which may result in chronic or recurrent headache or change the clinical or therapeutic approach. We considered patients with venous angiomas to be in the second group.

Literature Review
One of the authors (Y.T.) searched MEDLINE for articles published between 1985 and 2003 by using the following key words: MRI and headache, MRI and migraine, MRI and tension-type headache, and Headache and diagnosis. Bibliographies from all articles were also reviewed to obtain any further relevant articles. All articles were reviewed in their entirety, although all case reports with a few patients or preliminary data were not included. Headache that was not described as migraine or tension-type was defined as unspecified. In most articles related to migraine, migraine was classified into more detailed categories such as common, classic, basilar, and complicated. In our literature review, complicated migraine was separately analyzed whenever possible. The word complicated was not always clearly defined in these articles, but it usually referred to attacks characterized by prolonged aura symptoms (5,6). A reported case was considered complicated migraine when the authors clearly mentioned that the case was complicated migraine or when the reported case had a "persistent" or "prolonged" aura.

Statistical Analysis
We calculated the upper 99.5% confidence bound for the frequency of a clinically important intracranial abnormality both in our patient group and in data from the literature reviews. We used the normal distribution to compute approximate confidence intervals for proportions from observations, so long as the sample size and the frequency of a positive test were large enough to make the approximation reasonably accurate. Usually, np and n(1 – p) must both exceed about 5, where p is the proportion of the observed sample having the attribute of interest, and n is the number observed. When it was not possible to use this approximation, we computed the upper 99.5% confidence bound on the basis of the binominal distribution (7).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Of the 306 patients, 169 were in the first group (no abnormality, 55.2%), 135 were in the second group (minor abnormality, 44.1%), and two (0.7%) were in the third group. These two patients had a clinically important intracranial abnormality; one had pituitary macroadenoma, and one had subdural hematoma. The upper 99.5% confidence bound for the frequency of clinically important abnormal MR findings in our series, which was calculated on the basis of binominal distribution, was 1.7%.

In 195 patients (63.7%), enhanced T1-weighted images were obtained, although contrast enhancement with gadopentetate dimeglumine did not result in detection of additional abnormal findings in any cases. Twenty-three patients underwent repeated head MR examinations, with an interval of 85–1039 days (mean ± standard deviation, 528 days ± 284), but no additional abnormality was identified in any patients. All repeated MR examinations were performed because of patients’ demands.

The literature review revealed two reports of unspecified headache (Table 2) (8,9), 12 reports of migraine headache (Table 3) (10–21), and one report of tension headache (Table 4) (20). All studies involved the use of methods similar to ours (ie, T1- and T2-weighted transverse images were obtained, and the MR imaging findings were divided into three groups according to clinical importance).

A total of 790 MR examinations were performed in patients with migraine (Table 3). Nineteen patients were reported to have complicated migraine, and five of these patients (26.3%) had clinically important infarctions (four cortical and one cerebellar) on MR images. By excluding these patients, however, only four of 771 patients (0.52%) with migraine but without a neurologic abnormality had clinically significant abnormalities on MR images (intracranial aneurysm, mesencephalic lipoma, epidermoid of cerebellopontine angle, and pituitary macroadenoma). All of these patients were reported in a single study by Pavese et al (19). The upper 99.5% confidence bound was calculated as 0.68%. In one study (14), four vascular malformations were identified in 58 patients with common migraine (migraine without aura). Since detailed information was not available in the original report and it was unknown whether these malformations were reasonably considered to result in chronic or recurrent headache, we did not include these four patients in the population with clinically important abnormalities. In another study (11), CT scans and MR images obtained in a 35-year-old woman with sudden onset of right homonymous hemianopsia demonstrated a large infarction of the left occipital lobe. This patient’s condition was defined as complicated migraine in our literature review.

We found only one study of head MR imaging in patients with tension headache and normal findings of neurologic examinations (Table 4) (20). In the study, no clinically important abnormalities were detected in any patients, but the number of patients evaluated was very small (n = 35).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In 1994, The Quality Standards Subcommittee of the American Academy of Neurology published the following practice guideline (1,2):

In adult patients with recurrent headaches that have been defined as migraine, including those with visual aura, with no recent change in pattern, no history of seizure, and no other focal neurologic signs or symptoms, the routine use of neuroimaging is not warranted. In patients with atypical headache patterns, a history of seizures, or focal neurologic signs or symptoms, CT or MRI may be indicated.

This issue was readdressed in a report of the U.S. Headache Consortium (22), with similar recommendations. However, a large study that includes more than 1000 patients was recommended to help clinicians and patients make better-informed decisions, since there has been insufficient evidence to define the roles of CT and MR imaging in the evaluation of patients with unspecified chronic or recurrent headache without neurologic abnormality (1,2,22). In particular, many studies used CT (2,3), and a limited number of studies were performed to evaluate the role of MR imaging (8,9). MR imaging may be more sensitive than CT in the identification of various intracranial abnormalities, but this may be of little clinical importance in the evaluation of patients with headache and no neurologic signs or symptoms (22). Cost is another major consideration when analyzing a relatively innocuous procedure in relation to a large group of symptomatic individuals.

We found only two reports in which the authors looked at the use of MR imaging in patients with unspecified headache, and we combined the data from the present study and two previous studies (8,9). The numbers of patients in each study may not be large enough to give precise estimates to allow general applicability. By combining studies, however, the results of the individual studies can be strengthened, and overall estimates for items of interest can be provided (23). The populations in these studies were reasonably similar, and the actual frequency of clinically important abnormalities on MR images ranged from 0.7% to 3.7%. The upper 99.5% confidence bound was calculated as 3.4%. According to a meta-analysis (22), important intracranial abnormalities detected on CT scans in patients with unspecified headache ranged from 0.0% to 6.7% in 10 studies.

There were many studies of the MR findings in patients with migraine, although each study had a relatively small number of patients. The estimated 99.5% confidence bound of clinically important results of each study ranged from 2.9% to 26.8%. This wide range means that too few patients were included in each study to give precise estimates.

In our literature review, the frequency of clinically important abnormalities in patients with complicated migraine was high (26.3%), although the number of patients was small. The word complicated was not always clearly defined in these articles, but it usually refers to attacks characterized by prolonged aura (5,6). Such prolonged neurologic symptoms are difficult to distinguish from those due to cerebral vascular accidents; thus, it is reasonable to use MR imaging in the evaluation of patients. We found only one study of MR findings in patients with tension headaches, and only a small number of patients were evaluated. Further study of the usefulness of MR imaging in patients with tension headaches is necessary.

The major positive outcome derived from MR imaging in patients with headache is identification of a treatable lesion, which may improve the quality of life or survival rate. According to the data of previous and current studies, the important abnormalities include brain tumor, hydrocephalus, arteriovenous malformations, and subdural hematoma. Assessing patient health outcomes in the early diagnosis of the previously mentioned lesions is not easy, and the changes in patient care by failing to diagnose some of these lesions prior to the occurrence of objective findings may be small. The identification of these lesions, however, might lead to surgical cure. In fact, in our series, the patients with pituitary macroadenoma and subdural hematoma were successfully cured with surgical interventions; thus, the MR examinations were clearly beneficial for these patients. Unfortunately, there is not enough information in the reviewed literature to help determine the value of early detection of these intracranial lesions prior to the onset of neurologic deficit in patients with chronic or recurrent headache.

Another important outcome is the relief from anxiety that accompanies the absolute knowledge that a patient does not harbor a brain tumor or other intracranial disease. Many patients seek attention for their headaches because they have a real fear that a brain tumor is causing them (2); however, anxiety relief is subjective and difficult to measure.

Because of the very low frequency of positive MR examinations, it is important to define the clinical characteristics of chronic headaches that would help identify those patients at higher risk for intracranial disease and to evaluate the role of repeated neuroimaging in patients with previous negative findings. Literature abounds concerning those specifics of the headache history that increase suspicion for intracranial abnormalities, such as a change in headache pattern, side-locked headaches (unilateral headaches that never change sides), and new onset of chronic daily headache (2). However, these specifics have not been analyzed in detail in any large series. In our literature review, the frequency of clinically important abnormalities in patients with complicated migraine was high (26.3%), although the number of patients was small. The presence of prolonged aura in patients with migraine may be an important clinical sign that aids in the identification of patients who should undergo an MR examination. In two reports, the authors tried to define the clinical specifics that would help identify those patients on the basis of findings on MR images (8,9). Wang et al (9) showed a higher number of major abnormalities in patients who have atypical headaches and those who are middle-aged. Jordan et al (8) evaluated the referral patterns for imaging headache but concluded that the frequency of clinically important positive studies was unrelated to referring specialty or experience.

We tried to evaluate the role of repeated MR imaging examinations in patients with previous negative studies. In our series, only 23 patients underwent repeated head MR examinations, and no additional abnormality was identified in any patient. Since the number of patients was small and some patients may visit other hospitals to obtain a second opinion, we cannot conclude that the frequency of major abnormalities found with repeated MR imaging is low. Further studies are necessary to clarify the role of repeated MR examinations in patients with chronic or recurrent headaches.

Contrast enhancement might help to define intracranial abnormalities that may be unclear on conventional T1- and T2-weighted MR images. Although the use of contrast material results in increased cost and a raised risk of adverse reaction, little objective data exist to define the usefulness of contrast material in MR examination of patients with unspecified chronic or recurrent headache. In our series, no additional information was obtained on the enhanced T1-weighted transverse MR images in any patient. Previous studies (8,16,18) also demonstrated that enhancement with gadopentetate dimeglumine resulted in no additional findings in patients with unspecified or migraine headaches. Although contrast enhancement may contribute to the diagnosis in the presence of a suspicious lesion on an unenhanced MR image, contrast enhancement did not contribute to the diagnosis in all other instances (24).

This study had several limitations: First, the number of patients with complicated migraine was exceedingly small when compared with the number of patients with uncomplicated migraine or unspecified headache. This probably reflected the prevalence of complicated migraine, which is a rare condition, at least when compared with the other form of headache. In addition, as described previously, the word complicated was not always clearly defined in the articles. Thus, it was possible that the actual frequency of clinically important findings in patients with complicated migraine might be much higher or much lower if a representative number of patients with this condition had been included in the analysis. Second, in our literature review, it may not be appropriate to combine data from different articles. Since the most popular meta-analytic techniques, such as the Mantel-Haenszel test and Peto method, need a data matrix that can be entered into a 2 x 2 matrix, these methods cannot be applied to our data. Thus, there may be some bias that occurs with pooling the data.

In conclusion, MR imaging is an unrewarding technique in the evaluation of patients with chronic or recurrent headache and normal neurologic findings. However, our literature review suggests that MR imaging may be warranted in patients with complicated migraine. Neither contrast enhancement nor repeat MR imaging seems to contribute to the diagnosis, although the number of patients who underwent repeat MR examinations in our study was small.

	FOOTNOTES

 
Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, Y.T., K.E.; study concepts, Y.T., K.E.; study design, Y.T.; literature research, Y.T.; clinical studies, Y.T.; data acquisition and analysis/interpretation, Y.T.; statistical analysis, Y.T.; manuscript preparation, Y.T.; manuscript definition of intellectual content and editing, Y.T., K.E.; manuscript revision/review, K.E.; manuscript final version approval, Y.T., K.E.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Practice parameter: the utility of neuroimaging in the evaluation of headache in patients with normal neurological examinations (summary statement). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 1994; 44:1353-1354.[Free Full Text]
2. Frishberg BM. The utility of neuroimaging in the evaluation of headache in patients with normal neurologic examinations. Neurology 1994; 44:1191-1197.[Free Full Text]
3. Dumas MD, Pexman JH, Kreeft JH. Computed tomography evaluation of patients with chronic headache [published correction appears in Can Med Assoc J. 1995; 152:158]. CMAJ 1994; 151:1447-1452.
4. Haughton VM, Rimm AA, Sobocinski KA, et al. A blinded clinical comparison of MR imaging and CT in neuroradiology. Radiology 1986; 160:751-755.[Abstract/Free Full Text]
5. 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.
6. International Headache Classification Committee. ICD-10 Guide for Headaches. Cephalalgia 1997; 17(suppl 19):1-82.
7. Rümke CL. Uncertainty as to the acceptance or rejection of the presence of an effect in relation to the number of observations in an experiment. Triangle 1968; 8:284-289.[Medline]
8. Jordan JE, Ramirez GF, Bradley WG, Chen DY, Lightfoote JB, Song A. Economic and outcomes assessment of magnetic resonance imaging in the evaluation of headache. J Natl Med Assoc 2000; 92:573-578.[Medline]
9. Wang HZ, Simonson TM, Greco WR, Yuh WT. Brain MR imaging in the evaluation of chronic headache in patients without other neurologic symptoms. Acad Radiol 2001; 8:405-408.[CrossRef][Medline]
10. Soges LJ, Cacayorin ED, Petro GR, Ramachandran TS. Migraine: evaluation by MR. AJNR Am J Neuroradiol 1988; 9:425-429.[Abstract]
11. Kuhn MJ, Shekar PC. A comparative study of magnetic resonance imaging and computed tomography in the evaluation of migraine. Comput Med Imaging Graph 1990; 14:149-152.[CrossRef][Medline]
12. Jacome DE, Leborgne J. MRI studies in basilar artery migraine. Headache 1990; 30:88-90.[CrossRef][Medline]
13. Igarashi H, Sakai F, Kan S, Okada J, Tazaki Y. Magnetic resonance imaging of the brain in patients with migraine. Cephalalgia 1991; 11:69-74.[CrossRef][Medline]
14. Prager JM, Rosenblum J, Mikulis DJ, Diamond S, Freitag FG. Evaluation of headache patients by MRI. Headache Q 1991; 2:192-196.
15. Ziegler DK, Batnitzky S, Barter R, McMillan JH. Magnetic resonance image abnormality in migraine with aura. Cephalalgia 1991; 11:147-150.[CrossRef][Medline]
16. Osborn RE, Alder DC, Mitchell CS. MR imaging of the brain in patients with migraine headaches. AJNR Am J Neuroradiol 1991; 12:521-524.[Abstract]
17. Fazekas F, Koch M, Schmidt R, et al. The prevalence of cerebral damage varies with migraine type: a MRI study. Headache 1992; 32:287-291.[CrossRef][Medline]
18. Robbins L, Friedman H. MRI in migraineurs. Headache 1992; 32:507-508.[CrossRef][Medline]
19. Pavese N, Canapicchi R, Nuti A, et al. White matter MRI hyperintensities in a hundred and twenty-nine consecutive migraine patients. Cephalalgia 1994; 14:342-345.[CrossRef][Medline]
20. De Benedittis G, Lorenzetti A, Sina C, Bernarsconi V. Magnetic resonance imaging in migraine and tension-type headache. Headache 1995; 35:264-268.[CrossRef][Medline]
21. Cooney BS, Grossman RI, Farber RE, Goin JE, Galetta SL. Frequency of magnetic resonance imaging abnormalities in patients with migraine. Headache 1996; 36:616-621.[CrossRef][Medline]
22. Frishberg BM, Rosenberg JH, Matchar DB, et al. Evidence-based guideline in the primary care setting: neuroimaging in patients with nonacute headache. U.S. Headache Consortium Web site. Available at: www.aan.com. Accessed November 2003.
23. Irwig L, Tosteson AN, Gatsonis C, et al. Guidelines for meta-analyses evaluating diagnostic tests. Ann Intern Med 1994; 120:667-676.[Abstract/Free Full Text]
24. Demaerel P, Wilms G, Baert AL. Is contrast enhancement in cranial computed tomography necessary? (letter). Lancet 1995; 345:389.

This article has been cited by other articles:

				J. E. Jordan and for the Expert Panel on Neurologic Imaging Headache AJNR Am. J. Neuroradiol., October 1, 2007; 28(9): 1824 - 1826. [Full Text] [PDF]

				V. K. Gupta, Y. Tsushima, and K. Endo MR Imaging in Primary Headaches Radiology, February 1, 2006; 238(2): 754 - 755. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tsushima, Y. Articles by Endo, K. Search for Related Content PubMed PubMed Citation Articles by Tsushima, Y. Articles by Endo, K.