Classification of Headache
Jes Olesen
Richard B. Lipton
Undoubtedly, most physicians consider disease classification to be boring, difficult to learn, and clinically not very useful. In practice, most physicians do not memorize diagnostic criteria or take the time to look the criteria up except in special cases. However, clinical diagnosis always depends on a classification system and we all use formal or informal diagnostic criteria, often without knowing it.
A new classification system often finds its initial application in research. But, after it has been used for some years, all new research results in the particular field have been gathered using that system. To apply that information in practice, physicians must diagnose their patients in ways that are compatible with the new system. Through this process, the evidence-based medicine links a new classification system to clinical practice. Classification systems also have more direct effects. For example, the explicit diagnostic criteria for migraine without aura (7) request knowledge only about duration of attack, severity of pain, unilaterality, pulsating or no pulsating quality, and aggravation or no aggravation of pain by physical activity as well as knowledge about the associated symptoms of nausea, vomiting, and photo- and phonophobia. This serves as a clear guideline to the clinical interview and makes it unnecessary in most cases to ask many questions that physicians used in the past to characterize a headache.
The headache field has enjoyed a systematic hierarchical classification system and associated explicit (previously called operational), diagnostic criteria for more than 15 years after the publication of the first edition at the Classification of the International Headache Society (IHS) (ICHD-1) in 1988 (8). Building on ICHD-1, the second edition of the International Classification of Headache Disorders (ICHD-2) has recently been published (7). Like its predecessor, it will have a dominating influence on research and clinical practice for at least the next decade (Figs. 2-1 and 2-2).
GENERAL PRINCIPLES OF DISEASE CLASSIFICATION AND DIAGNOSTIC CRITERIA
To classify disease means to decide how many different entities that should be recognized and to order them in a logical system. There are two diametrically opposed forces in classification: the tendency to lump and the tendency to split. Lumping headache disorders leads in the extreme to one category: headache. To physicians interested in headache this sounds as a highly theoretical option, but it is the way many general physicians and the general population classify headache disorders. If they are a little bit more sophisticated, they recognize the difference between migraine and other headaches. When professionals talk about headache disorders, laypersons often believe that it does not include migraine and it is often useful for politicians and other decision makers to talk about “migraine and other headaches.” For extreme splitters, it is possible to diagnose an almost endless variety of headaches. Taken to the extreme this leads to the attitude that “There are no diseases, only patients.” Each patient has special characteristics and therefore merits his or her own category of subdiagnosis. Successful classifications balance these opposing tendencies by grouping similar patients together in ways that are useful for clinical practice. These similarities can be based on epidemiology, symptom profile, disease mechanism, and sometimes treatment response. Even sophisticated studies of brain blood flow, biochemistry, or genetics can be taken into account when constructing the classification system. Diagnostic criteria should, on the other hand, include only parameters that are available to the physician when diagnosing patients.
It is often suggested that there should be two different classifications: One for research, which can be detailed, and another for clinical practice, which should be simple and very easy to apply. It is absolutely hopeless, however, to have different classifications for research and clinical
practice because research results obtained with one kind of classification would be difficult to apply to patients diagnosed according to another system. The ICHD-1 and -2 use hierarchical classification to solve this problem (7,8). Patients can be diagnosed in groups and subgroups with various levels of diagnostic refinement. With such a system, patients can be diagnosed according to the first or second digit in general practice, and special clinics and researchers may diagnose to third or fourth digit. Ideally, a classification system should be based on etiology, but very
often the etiology is not known. The classification system then has to use clinical features and the results of laboratory tests, and thus is based on phenomenology. The latter applies to the primary headaches; etiology is used to classify the secondary headaches in ICHD-2.
practice because research results obtained with one kind of classification would be difficult to apply to patients diagnosed according to another system. The ICHD-1 and -2 use hierarchical classification to solve this problem (7,8). Patients can be diagnosed in groups and subgroups with various levels of diagnostic refinement. With such a system, patients can be diagnosed according to the first or second digit in general practice, and special clinics and researchers may diagnose to third or fourth digit. Ideally, a classification system should be based on etiology, but very
often the etiology is not known. The classification system then has to use clinical features and the results of laboratory tests, and thus is based on phenomenology. The latter applies to the primary headaches; etiology is used to classify the secondary headaches in ICHD-2.
 FIGURE 2-1. The Chairman of the First and Second Headache Classification Committee together with the secretary Shiela M. Westh, who did all the practical work on the first edition of the International Headache Classification. This photo was taken in 1988. |
 FIGURE 2-2. Second International Headache Classification Committee, at its last meeting in Copenhagen, March 2003. From left to right: Top row: Peter J. Goadsby, Guiseppe Nappi, Giorgio Sandrini, Tim Steiner. Middle row: Peer Tfelt-Hansen, James W. Lance, Miguel J.A. Lainez, Michael First. Front row: Hartmut Göbel, Fumihiko Sakai, Rigmor Jensen, Jes Olesen, Marie G. Bousser, Hans Christoph Diener. Missing from this picture are David Dodick, Richard B. Lipton, Jean Schoenen, and Stephen D. Silberstein, who were unable to attend the last meeting, but all did an equally good job to those present. Of these Peer Tfelt-Hansen, Giorgio Sandrini, and Rigmor Jensen were very active in subcommittees, but not members of the main committee. |
THE HISTORY OF HEADACHE CLASSIFICATION
The first attempt at creating a headache classification was made by an ad hoc committee of the National Institutes of Health in America. It published a paper entitled “Classification of Headache” in 1962 (1). This classification became quite widely used in North America and parts of Europe, but never gained true international acceptance or broad use. It recognized only a few headache disorders and its so-called definitions of different headache disorders were wide open for individual interpretation. Migraine, for example, was defined as “recurrent attacks of headache, widely varied in intensity, frequency, and duration. The attacks are commonly unilateral in onset; are usually associated with anorexia, and sometimes with nausea and vomiting; and some are proceeded by, or associated with, conspicuous sensory, motor, and mood disturbances; and are often familial” (1, p. XX). Some attempts were made to operationalize the diagnostic criteria for migraine (16,19). These criteria were never internationally accepted, but formed a valuable basis for the subsequent classification work. In 1985, the IHS formed a headache classification committee, which in 1988 published the first international classification of headache disorders (ICHD-1) (8). It contained operational (now called explicit) diagnostic criteria for all headache disorders. The system was endorsed by all national headache societies that were members of the IHS and by the World Federation of Neurology. The classification was translated into more than 20 different languages and has been used throughout the world as the only internationally accepted headache classification system. The World Health Organization (WHO) accepted the major principles of ICHD-1. After 15 years of service the ICHD-1 has now been replaced by ICHD-2, published in January 2004 following more than 4 years of intense preparations by a new international headache classification committee. The classification is available on the IHS web site (9). A slide kit is also available for downloading. A short version of the classification has been developed and is available on the IHS web site and as a printed pocket folder (10).
ICHD-2 Classification and Terminology
Like it predecessor, ICHD-2 is hierarchical using up to four digits to code for all varieties of headache disorders. These are now organized in 14 major groups. Groups 1 to 4 cover the primary headaches (Table 2-1), groups 4 to 12 the secondary headaches, group 13 the cranial neuralgias and facial pain, and group 14 other headaches, cranial neuralgias, or primary facial pain (Table 2-2). New groups in ICHD-2 are group 10 headache, attributed to disorder of homeostasis, and group 13 headache, attributed to psychiatric disorder. Also new is the subdivision of group 14 into headache not elsewhere classified and headache where insufficient information is available to classify. New entities have been included within most groups but particularly so in Chapter 4, which was previously called miscellaneous primary headaches now “Other Primary Headaches.” Major reorganization has also taken place within the chapter on headache attributed to substance use, especially the clear recognition of medication overuse headache. An appendix has been added that includes research criteria for new types of headache not yet sufficiently validated.
The taxonomy of headache disorders was quite markedly changed in ICHD-1, notably by the introduction of the migraine with aura and migraine without aura to replace common migraine and classical migraine, respectively, as well as tension-type headache replacing tension headache or muscle contraction headache. This change of taxonomy has largely been successful. Therefore, few changes have been made to the taxonomy of ICHD-2. Previously used terms are presented whenever such terms have existed, just like in ICHD-1.
ICHD-2 Diagnostic Criteria
The obvious difficulty in constructing diagnostic criteria for primary headache disorders is the lack of a definitive gold standard for diagnosis. For the primary headaches, there is no clear biochemical or diagnostic marker, no valid external diagnostic touchstone. ICHD-1 criteria were therefore constructed using clinical features and easily available results of laboratory investigations. ICHD-2 has continued this tradition. The diagnostic criteria previously called operational are now called explicit diagnostic criteria. This means that terms such as sometimes, usually, and often are avoided and instead numerical figures are given. Sometimes criteria are monothetic, requiring the presence or absence of a single characteristic. At other times the criteria are polythetic, requiring for example two out of four characteristics. These kinds of criteria allow use of characteristics that occur in, for example, 50% of the patients, such as the pain criteria of migraine without aura (see Table 2-2). The basic system from DSM-4 (3) used in ICHD-1 is still used in ICHD-2. Thus, the criteria contain several letter headings: A, B, and C, and each letter heading has to be fulfilled to come to a diagnosis.
ICHD-2 and WHO Classifications
The International Classification of Diseases, 9th edition (ICD-9) contained only a few headache entities and very important syndromes such as tension-type headache
were not included under neurology. However, the much more detailed classification and the diagnostic criteria of ICHD-1 were largely accepted by the WHO and included in ICD-10 (21). This meant that all primary headache syndromes were now grouped under neurology. Unfortunately, the United States has not yet adopted the ICD-10, which has greatly hampered a reasonable reimbursement of headache experts and therefore also the development of services to headache patients. From ICD-10 a neurologic adaptation (ICD-10 NA) was developed (22), which is more detailed and includes most of the headache disorders classified in ICHD-1. A further document, “ICD-10
guide for headaches,” was published to provide a cross-way between ICD-10 (11), ICD NA, and ICHD-1. In ICHD-2, it was decided to include both the IHS and ICD-10 NA code numbers. The latter system is less detailed. Different ICHD diagnoses may therefore often receive the same ICD-10 NA diagnosis. Table 2-1 gives ICHD-2 codes and diagnoses as well as ICD-10 NA codes.
were not included under neurology. However, the much more detailed classification and the diagnostic criteria of ICHD-1 were largely accepted by the WHO and included in ICD-10 (21). This meant that all primary headache syndromes were now grouped under neurology. Unfortunately, the United States has not yet adopted the ICD-10, which has greatly hampered a reasonable reimbursement of headache experts and therefore also the development of services to headache patients. From ICD-10 a neurologic adaptation (ICD-10 NA) was developed (22), which is more detailed and includes most of the headache disorders classified in ICHD-1. A further document, “ICD-10
guide for headaches,” was published to provide a cross-way between ICD-10 (11), ICD NA, and ICHD-1. In ICHD-2, it was decided to include both the IHS and ICD-10 NA code numbers. The latter system is less detailed. Different ICHD diagnoses may therefore often receive the same ICD-10 NA diagnosis. Table 2-1 gives ICHD-2 codes and diagnoses as well as ICD-10 NA codes.
Related posts:
Full access? Get Clinical Tree
