Abstract
Background
Several treatment guidelines exist for cluster headache. However, it is not yet known how many cluster headache patients are treated according to these guidelines.
Methods
We enrolled 434 cluster headache patients with confirmed diagnosis referred to two tertiary pain centers. The history of treatment was registered and analyzed according to the treatment guidelines of the European Federation of Neurological Societies.
Results
Regarding acute attack treatment, 62.1% of the episodic and 71.0% of the chronic cluster headache patients were treated according to the guidelines. The efficacy rate was above 92% in both groups. Regarding prophylactic treatment, 31.3% of the episodic and 50.9% of the chronic cluster headache patients were treated according to the guidelines. The efficacy rate was 92.8% for episodic and 70.9% for chronic cluster headache.
Conclusion
The rate of guideline-adherent treatment in cluster headache is about 70% for acute treatment and about 35% for prophylactic treatment. The efficacy of this treatment is significantly higher than the efficacy of non-guideline-adherent treatment.
Introduction
Cluster headache as a very severe headache disorder has been difficult to treat, and many patients suffer extremely because of lack of treatment options and missing knowledge among physicians about the diagnosis and the treatment (1). However, in the past decades several clinical trials have been performed and efficacious treatment options have been established so that evidence-based treatment recommendations can be given. Consequently, some national and international guidelines on the treatment of cluster headache according to evidence-based medicine principles have been published (2–9).
These treatment guidelines on cluster headache have been published in scientific journals and on the Internet. Interestingly, these guidelines are very similar and do not show major differences. However, general practitioners who are often the first contact of patients with cluster headache and even several neurologists are not aware of these guidelines. Further, only little is known about the distribution of and the adherence to these guidelines among those neurologists who are aware of them.
Therefore, we conducted an exploratory study on the adherence of treatment strategies among cluster headache patients in Germany in order to evaluate whether they are treated according to the recommendations given in the guidelines and whether treatment according to the guidelines is efficacious. Since the national German guidelines on the treatment of cluster headache (2) and the European (European Federation of Neurological Societies (EFNS)) guidelines on the treatment of cluster headache (3) are nearly identical, we used the European guidelines and performed structured interviews with cluster headache patients to evaluate whether treatment according to the guidelines would result in a better subjective outcome.
Methods
We enrolled consecutive patients with the diagnosis of cluster headache who were referred by their general practitioner or by their neurologist to headache centers (Headache Outpatient Clinic at the Department of Neurology, University of Münster and Berlin Pain Centre) over a period of four years in order to obtain a second opinion on the diagnosis and/or on the treatment procedures. Other primary headache disorders were allowed but were not included in this study. The patients were interviewed always by the same investigator (VL) trained in the diagnosis and the treatment of cluster headache before the treatment eventually was changed. We excluded all patients who were under neurostimulation or botulinumtoxin treatment, because these interventions show some evidence of efficacy but are not yet included in guidelines. All patients gave informed consent prior to the interview. The study was approved by the Institutional Review Board of the Faculty of Medicine, University of Münster (date of approval March 5, 2009).
In the interview, the diagnosis of cluster headache was confirmed according to the criteria of the International Headache Society (IHS), second edition (10) or third edition (11). Only patients with confirmed cluster headache were included. We registered the baseline demographic and clinical characteristics of the patients and divided them into episodic and chronic cluster headache.
In the following interview, we asked the patients which drug treatments they have received in the past for cluster headache and how sufficient this treatment was for them. Efficacy of the treatment was rated by the patients as excellent/moderate/mild. They were asked for all treatments that are mentioned in the guidelines. In order to rate the efficacy from an objective point of view, we defined successful treatment of acute attack abortion by none or mild pain 15 minutes after drug intake and successful treatment of prevention by decrease of attack frequency by at least 50% over a period of at least two weeks. This is in concordance with the guidelines for trials on cluster headache (12).
Then, we evaluated whether this treatment was in concordance with the European guidelines for the treatment of cluster headache (3). We defined a treatment in concordance with these guidelines as a successful attack abortion (see above) when up to two first-line and one second-line drug were tried. Successful preventive treatment (see above) was in concordance with the guidelines if the patient had tried verapamil and/or at least two other drugs of the first- and second-line preventive drugs in the guidelines (in an adequate dose). For example, verapamil was considered guideline adherence if it was given in a dose of at least 240 mg daily, lithium had to show a serum level of 0.6 mmol/l, and valproate had to be given in a daily dose of at least 600 mg.
Statistical analysis was performed with SPSS version 18.0. Data are presented as arithmetic mean with standrd deviation or as a percentage. Statistical comparisons were made by non-parametric testing using chi2-test for qualitative data. Significance level was set at p = 0.05.
Results
Demographic data of all cluster headache patients included (n = 434).
Subjective efficacy as rated by the patients for cluster headache medication.
sc: subcutaneous.
Response to treatment according or not according to guidelines. For definition of successful or unsuccessful see text. All data are given as percentage. For further statistical analysis see text.
Comparing episodic cluster headache versus chronic cluster headache by chi2-test.
Discussion
We present a sample of cluster headache patients who were referred to a secondary headache/pain center in order to receive a second opinion on diagnosis and treatment. Only patients with a confirmed diagnosis of cluster headache were enrolled. The characteristics of our sample are very similar to those of other published larger cluster headache samples and to population-based epidemiological data (13–17), although we had a large proportion of chronic cluster headache patients with 26.3%. Altogether, we conclude that our sample is representative of cluster headache patients affected moderately to severely.
We evaluated which drugs have been tried by the patients and how successful these different treatments were according to the patients’ impression. The majority of the patients had tried oxygen and verapamil as first-line drugs according to the European guideline for the treatment of cluster headache, and a majority had experience with sumatriptan (subcutaneous, nasal, and oral) and with steroids. However, all other possible drugs were tried only by a minority. The ratings of subjective efficacy were similar to the data from clinical trials, e.g. an excellent efficacy of subcutaneous sumatriptan (18) and a moderate to excellent efficacy of nasal sumatriptan (19), which is not approved in Germany for cluster headache, or zolmitriptan (20). With respect to preventive treatment, only verapamil and steroids received sufficient efficacy. All other preventive drugs were rated as poor by about half of the patients who tried this drug.
In general, patients with chronic cluster headache were more difficult to treat than patients with episodic cluster headache. This difference was consistent except for the acute attack treatment not according to the guideline. This is easy to explain since the only effective ways to treat acute cluster headache attacks are limited and all ways are listed as first- or second-line in the guideline. So, if somebody is not treating an acute cluster headache attack, there is no choice of real effective treatments both for episodic and chronic cluster headache.
When separating our sample into those patients treated according to the guidelines and those not treated according to the guidelines, we observed a clear, significant difference. The rates of successful treatment were higher both for acute and preventive treatment when patients were treated according to the guidelines.
Our study has some limitations that must be considered when interpreting the results. First, the study was not prospective and therefore not randomized. However, it would be very difficult to design a prospective randomized study on this issue, since it would be unethical to enroll cluster headache patients in a prospective study for at least some weeks and to withhold from them evidence-based treatment procedures. Furthermore, it was not our intention to compare treatment according to guidelines and not according to guidelines in a prospective way; this would be a clinical trial with a different design. Second, we had a high proportion of severely affected cluster headache patients. This is a disadvantage since the sample is not representative of all cluster headache patients. However, it can be seen as a strength as patients seen by such centers are probably the ones who are not easily treated to start with. A representative sample for all cluster headache patients might have resulted in lower differences between guideline accordant and non-accordant treatments. However, a representative sample for all cluster headache patients would be possible only on a population-based (or at least general practitioner-based) level. In fact, our study is not evaluating the adherence of the patients but the efficacy of treatment in accordance with guidelines. Third, the data from the patients were in part not based on headache diaries. Some patients were keeping headache diaries (n = 134), but this was not a requirement to be included. We know that headache severity data from the memory of the patients tend to describe the headache more severe than when evaluating headache diaries (21). However, since this was not a prospective study, enrollment only of patients with a headache diary would have been a major bias. Interestingly, when analyzing both patient groups separately, we did not detect a significant difference in their guideline adherence.
In conclusion, our data show that patients with the correct diagnosis of cluster headache and treated according to evidence-based guidelines can be treated very efficaciously with a clearly improved quality of life. This suggests that the use of evidence-based cluster headache guidelines results in clearly less suffering from cluster headache. It would be ideal to confirm these results in a prospective study in order to recommend the use of the guidelines on more evidence. Nevertheless, the distribution of evidence-based guidelines on the treatment of cluster headache in all countries and also for all neurologists is highly recommended.
Clinical implications
Of patients with episodic cluster headache, 62.1% are treated according to acute attack treatment guidelines. Only 31.3% of patients with episodic cluster headache are treated according to treatment guidelines with respect to prophylactic treatment. Patients with chronic cluster headache show higher adherence to treatment guidelines than patients with episodic cluster headache. Treatment according to guidelines is much more efficacious both in episodic and chronic cluster headache.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.