• Number of days with headache from week 2 after injection to week 7.
  

• No. of days until the patient has got > 50% of the pain level he or she had before the injection.
  
• No. of drop-outs due to long-lasting improvement after first injection
  
• No. of hours with headache from week 2 to week 7
  
• Average headache intensity from week 2 to week 7.
  
• Headache index (number of hours with headache times the number of pain intensity)
  
• Number of days with neck pain from week 2 to week 7.
  
• Number of days with shoulder or arm pain from week 2 to week 7.
  
• Number of doses with analgesics from week 2 to week 7.
  
• Number of days with sick-leave from week 2 to week 7.
  

Cervicogenic headache is a unilateral headache stemming from the neck. Usually, there are no pathological findings on x-ray or MRI of the neck. It is supposed that pain may stem from various structures in the upper part of the cervical spine. Regardless of the source, it is often believed that the neck muscles may be involved in the pain generation, either primarily or secondarily.

Treatment of cervicogenic headache is often difficult. The effect of drugs is usually limited. Various surgical techniques such as radiofrequency generation of the facet joints in the neck have been tried, but with little success (4).

Botulinum toxin injection in muscles have for several years been used in conditions with pathologically increased muscle activity, such as spasticity and dystonias. Gradually, it has also been used in many pain conditions, among them headaches. One case history (5) and a randomized placebo controlled, double-blind study (6), have shown effect in cervicogenic headache. The latter study had some methodological weaknesses, since it was small, only 26 patients, and the placebo group had prior to treatment only half as much pain as the group receiving botulinum toxin treatment. In addition, pain was not registered daily, but only before (prior) treatment and after 3 or 4 weeks. A review considers the documentation on treatment with botulinum toxin in idiopathic and cervicogenic headaches to be inconclusive (7).

In our Department, we have tried this treatment on a few patients with typical cervicogenic headache with excellent effect and without side-effects. Therefore, it would be of considerable interest to perform a larger study with good scientific quality and a higher statistical power than the above-mentioned one.

As the basis for our study we would adopt a conservative hypothesis (H0): Botox injections in cervical muscles is not superior to placebo in alleviating pain in unilateral cervicogenic headache.

Inclusion Criteria: 1. Headache should be strictly unilateral without side-shift 2. The patients should exhibit traits indicating involvement of neck structures, fulfilling at least 2 of the following: I. Reduced neck range of motion II. Typical headache can be elicited by palpation or pressure on muscles of the occiput or in the neck.

III. Typical headache can be precipitated by certain neck movements or by adopting a certain position of the head.

IV. There is a radiation of pain to the ipsilateral shoulder or arm.

3. There is a positive effect of blockades. In this study we will only perform a blockade of the greater occipital nerve.

4. The condition should have lasted for more than 1 year. 5. Age between 18 and 70 years. 6. The headache should be present more than 15 days a month, and the frequency should be based on a headache diary filled in the last month before inclusion.

Exclusion Criteria: 1. Cervical spinal stenosis, cervical disc herniation, or other neck pathology that should be treated surgically.

2. Malignant disorder. 3. Rheumatic disorder or other disorders which induces analgesic use. 4. Patients with other clinically relevant disease (liver, kidney, endocrinological, metabolical, systemic, psychiatric).

5. Frequent other headache type that cannot be distinguished from cervicogenic headache.

6. Pending or ongoing litigation for head- or neck trauma.