The validity of the inclusion of “lupus headache” in the systemic lupus erythematosus disease activity index


The Validity of the Inclusion of “Lupus Headache” in the Systemic Lupus Erythematosus Disease Activity Index

To the Editor:

The Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) was first described in 1992 (1) and was updated in 2000 (2). It has since become a frequently used measure of disease activity in research studies of SLE. “Lupus headache” is included in the SLEDAI as a descriptor, defined as “severe, persistent headache; may be migrainous, but must be nonresponsive to narcotic analgesia” (1). Since the weighting for this item is 8, the highest of any descriptor, there is potential for the presence or absence of lupus headache to significantly impact the overall score. We contend that the definition given is insufficiently detailed to ensure adequate interrater reproducibility. In addition, the existence of lupus headache as a specific headache disorder is in doubt. For example, in a recent study by Hanly et al (3), the SLEDAI was used to describe a large group of patients with SLE, among whom headache disorders, including “nonspecific” headache, were present in more than one-third of cases. Whether or not any of the headache disorders documented were included in the SLEDAI descriptor lupus headache was not described in detail. However, since the mean SLEDAI score in the study was a moderate 5.8, small changes in rating of lupus headache by the study investigators could have made a relatively large difference in this score.

The term “lupus headache” has been used by a number of authors, but there is no consistent definition of this headache phenotype. Omdal et al (4) described lupus headache as “severe, disabling, persistent and not responsive to narcotic analgesics.” Brunner and colleagues (5) used an even less specific definition, which did not address response to narcotic analgesia, in their study of headache in pediatric SLE. If these descriptions are cross-checked with International Headache Society (IHS) classification criteria (6), a widely accepted set of criteria designed for use in research studies of headache disorders, it is possible to allocate the type of headache portrayed in these definitions to various categories (depending on the presence of certain additional features), including new daily persistent headache, chronic migraine, and medication-overuse headache developing in a patient with a background of migraine. Furthermore, lupus headache was not included in the American College of Rheumatology case definitions of neuropsychiatric syndromes in SLE, published in 1999 (7).

With reference to the SLEDAI definition of lupus headache, what duration defines “persistent”? Is this longer than 72 hours, the current upper limit of attack duration described in the IHS definition of migraine? Response to analgesia is not included in IHS definitions, and narcotic analgesia is not generally recommended for the treatment of migraine attacks.

Although there is little doubt that headache is a common symptom among patients with SLE, there is no evidence to support the existence of a headache phenotype that is specific to SLE. A recent meta-analysis by Mitsikostas et al (8) showed that the published data concerning headache disorders and SLE were generally retrospective and of poor quality. Pooled data from the IHS-based studies showed that 57% of SLE patients reported any type of headache disorder, a prevalence similar to that among control subjects. No evidence of a distinctive form of headache disorder attributable to SLE and no association between headache and SLE disease activity were found. Our own data derived from a cohort of 61 patients with SLE also failed to demonstrate any correlation between disease activity (ascertained using the SLEDAI) and headache type and frequency (Davey R, et al: unpublished observations).

In conclusion, there is considerable doubt that lupus headache exists as an independent entity in clinical practice, and we are concerned that this SLEDAI descriptor could be misused as a “catchall” item for patients describing a range of headache disorders. A large, prospective, appropriately controlled study of headache in SLE has yet to be carried out, but until these data are available we recommend that the descriptor lupus headache be removed from the SLEDAI.

Richard Davey MD*, John Bamford MD*, Paul Emery MD*, * Leeds General Infirmary, Leeds, UK.