Propofol Injection for Daily Headache

Chronic daily headache (CDH) is a common and extremely disabling pain syndrome with a recent estimated point prevalence of 2.98% of the general population1 Other studies estimate a prevalence of 4.1% and 4.7% with a higher prevalence in women of 9%.2,3 The condition is disabling and is a major cause of loss of function and work capacity resulting in significant personal and national economic loss4

Therapeutic options for treating chronic headache disorders are legion and diverse and there is a steady stream of publications advocating novel therapies. Pharmacological options that have been advocated include: sodium valproate, olanzapine, tricyclic and Selective Serotonin Reuptake Inhibitor antidepressants, tizanidine, butalbital, nefazodone, baclofen, prochlorperazine, intravenous lignocaine, dihydroergotamine, opioids, monoamine oxidase inhibitors, triptans, botulinum toxin, cryotherapy and analgesic withdrawal5-20 There is much debate as to why headache treatments fail and it is proposed that the choice of an inappropriate agent is a significant reason in some patients21

On the basis of a small number of studies intravenous propofol has now been advocated for the treatment of chronic daily headache22-24 It seems fair to conclude that the existing strife to promote propofol as an effective intervention for treating CDH in the absence of a randomized controlled trial merely reflects the lack of a single good treatment option for this condition. It is, furthermore, reasonable to speculate that the beneficial effect seen with propofol is a placebo response. Krusz et al reports the effects of subanesthetic doses of intravenous propofol on chronic migraine in a cohort of 77 subjects in an open-label study22 Propofol was administered by intermittent boluses; the average dose used was 110mg. It is reported that 63 out of 77 patients reported complete abolition of their headache with an average time of 20 - 30 minutes of propofol treatment. The range of propofol doses used is however not clearly stated and there was no assessment of the intervention beyond 24 hours. In this study lidocaine was mixed with propofol to reduce the pain of injection but it could not be said with any certainty that lidocaine itself has no effect on headache.

Mendes et al retrospectively reports 21 independent treatments of propofol in 18 patients suffering primary and secondary chronic daily headache over a 1 year period 23 Subanesthetic doses of propofol were administered by intermittent intravenous boluses 20 - 30mg every 3 to 5 minutes (range 60 - 380 mg). 6 patients reported complete relief of pain with the treatment, 1 patient had no pain relief; on average patients reported a 58.92% decrease in pain using a 0 - 10 scale. It is reported that 8 patients' headache cycles were broken when followed up several days later but no information is available on the longer term impact of these interventions. In these two studies there appears to have been good tolerance of propofol and no significant adverse events.

Drummond - Lewis reports 2 hospitalized patients with refractory migraine who were treated with anesthetic bolus doses of propofol (0.5 - 1mg / kg) with reported success 24. The authors comment there is uncertainty as to the most effective dose of propofol for this treatment.

Propofol is a safe modern, non-barbiturate intravenous anesthetic induction agent that was introduced into clinical anesthesia in 1977. It has established its place in practice because of its many ideal characteristics; it causes rapid and smooth loss of consciousness and rapid recovery, making it an ideal agent for day-case procedures25,26 There is extensive use of propofol as an intravenous sedative for procedures such as endoscopy, electroconvulsive therapy, cardioversion and for longer term sedation in the intensive care unit.27,28 Propofol has useful antiemetic properties and is known to be anti-epileptic and antipruritic.29 Although propofol can cause pain on intravenous injection there are several strategies to reduce this problem.30 There are reports of rare fatalities associated with high doses of propofol and reports of pancreatitis after single doses.31,32

Propofol (2,6 di-isopropylphenol) is available as an emulsion of 1 % propofol in 10% soyabean oil, 2% glycerol and 1.2% egg phosphatide. Doses up to 2.5mg / kg are used for induction of anesthesia and doses up to 200ug / kg /min are used for sedation purposes.33 It remains uncertain how propofol might exert its action in relieving headache. Propofol is known to stimulate gamma - aminobutyric acid- A (GABA-A) receptors by enhancing channel opening34 It has been shown that other GABA ergic drugs including sodium valproate, gabapentin and topiramate have a place in the treatment of migraine and it is speculated that propofol might be work in a similar way.35 Nonetheless, it is known that propofol has other important pharmacological activity including N-methyl-d-aspartate receptor antagonism, sympathetic inhibition, stimulation of nitric oxide production, calcium channel inhibition and modulation of prostaglandin E2 hyperalgesia.36-39 We propose that it is premature to speculate on the action of propofol in headache without clear evidence of efficacy.

There is ongoing discussion as to what is a good definition of chronic daily headache; this is a term that is not defined by the International Classification of Headache Disorders (ICHD)40 This is a potential impediment to research. It has been proposed that chronic daily headache can be subclassified into primary headache that includes transformed migraine (TM), chronic tension-type headache (CTTH), hemicrania continua (HC) and new daily persistent headache (NDPH)41 Secondary CDH has an identifiable underlying cause e.g. post traumatic, cervical spine disorders, vascular disorders, chronic meningitis, idiopathic intracranial hypertension, temporamandibular joint dysfunction and sinus disorders. Silberstein et al highlighted that not all chronic daily headaches can be classified according to the ICHD; however, in 1996 they managed to completely sub-classified a population of CDH sufferers into those with transformed migraine (TM) 78%, chronic tension-type headache (CTTH) 15.3% and "other" 6.7%.42 Silberstein defined chronic daily headache as headaches that last for more than 4 hours a day for more than 15 days a month and this definition is used in their later study of propofol and headache. This definition becomes perhaps less crucial when dealing with a population of patients who suffer intractable continuous headache; this tends to be the characteristic of patients who are already known to us at the University of Alberta Hospital Pain Centre.

Significance and importance Although it is reasonably argued that propofol sedation can be administered safely by non-anesthetists, there is still the possibility of serious airway complications and cardiorespiratory depression.43 The administration of propofol should not be undertaken lightly and certainly not used for an indication that has not been subject to rigorous trial. Previous studies of propofol and headache have only examined the short- term impact of this intervention on pain scores. The study proposed here looks at the effect of this treatment on pain and quality of life over a longer period - 30 days.

If subanesthetic doses of propofol are found to be effective in reducing chronic daily headache we become armed with a relatively safe non- narcotic intervention for this extremely disabling disorder that could be provided at a variety of centres e.g., emergency rooms, pain clinics, neurological facilities. In view of the reluctance by some emergency departments to administer narcotics for chronic headache we speculate that emergency departments may be more willing to provide this treatment than narcotics.44

Hypothesis A single subanesthetic dose of propofol will result in improved pain and quality of life for the next 30 days in persons with chronic daily headache (CDH)

Specific objectives To measure the effect of a single infusion of propofol at 40 mcg / kg / minute over 60 mins on headache-related quality of life (measured by the Headache Disability Index) and on headache severity (measured by the Headache Index) in subjects with chronic daily headache over 30 days45-47

Study Site

The Chronic Pain Centre 1A1 Walter Mackenzie Health Sciences Centre, University of Alberta, Edmonton. Alberta, T6G 2B7. Tel: 780 407 8638.

Methods Ethics approval for this study will be sought from the local institutional ethics board and Health Canada.

Inclusion criteria

- Adult subjects (18-65 years) with CDH by the criteria of Silberstein and Lipton.41 i.e. on the basis of pre-randomization history suffer chronic headaches lasting more than 4 hours a day for more than 15 days a month.

- Previously investigated to exclude serious treatable pathology.

- On the basis of the pre-randomization headache diary: the average pain intensity during each episode should be at least 5 /10 on the Numerical Rating Scale (NRS) on 4 days out of 7.

- Quality of life and function should be impaired as evidenced by a (pre-randomization) Headache Disability Index (HDI) of at least 40.

Exclusion criteria

History of the following:

- Known or suspected allergy to propofol, intralipid or midazolam,

- In emergency or life-threatening situations,

- Those having language barriers (e.g. illiterate, not English-speaking, dysphasic),

- Known or suspected difficult airway or sleep apnea,

- Severe respiratory disease,

- Neuromuscular disease,

- Seizure disorder,

- Severe cardiac disease,

- Severe gastresophageal reflux disease,

- Pancreatitis,

- Lipid disorders,

- Receiving Total Parenteral Nutrition,

- Body mass index > 35, diabetes or major endocrine disorder,

- Hepatic or renal failure,

- Unstable psychiatric disorder,

- Known drug abuse,

- Pregnancy,

- Cognitively impaired.

Obtaining Consent Each subject will be given an information sheet describing the study procedures, offered the opportunity to ask any questions and must sign that they have understood the contents. A separate consent form will be signed by the participant. Any questions will be dealt with by an investigator. The headache diary and all information to be read by subjects will be designed to have a Flesch Kincaid reading age of less than or equal to grade 8. Subjects will be free to withdraw from the trial at any stage.

Pre-randomization - baseline data collection

- Demographic data to include: age, gender, education level (years), and socioeconomic status.48

- The features of the subject's headache will be recorded and the headache disorder will be classified according to the International Classification of Headache Disorders.40 The date of chronic headache onset will be recorded.

- A standard medical history will be recorded which will include details of current medication, (specifically headache medication) and stage of menstrual cycle.49

- A physical examination will be performed. Pulse, non-invasive blood pressure, respiratory rate, oxygen saturation from pulse oximetry, weight, height and temperature will be recorded.

- Subjects will then complete the Headache Disability Inventory (HDI).45

Headache Diary

• Participants will then be instructed in the completion of the daily headache diary (this is estimated to take 10 - 20 minutes). This diary will keep track of daily pain scores, activity, headache medication use and visits to the emergency room. Headache severity is recorded four times daily on an 11 point numerical rating scale (NRS) anchored at zero for no pain and 10 for the worst imaginable pain. Participants will be asked to keep this diary for 7 days pre-randomization and for 30 days commencing the day after the treatment / infusion.

Randomization and blinding Subjects that still meet the inclusion criteria will proceed to the treatment phase. Subjects will be assigned to either the propofol group or placebo group. Sealed envelopes containing a randomized assignment to either propofol or placebo group will be prepared by an individual unconnected with the study using a web-based random number generator, and balanced to ensure equal allocation to each group. The propofol group will receive propofol and a concurrent normal saline bolus. The placebo group will receive intralipid and an identical concurrent normal saline bolus that contains low dose midazolam. There is evidence that low dose midazolam (35 mcg / kg) is sufficient to produce adequate sedation but no evidence to support the use of single dose midazolam as a treatment for CDH; this makes midazolam a suitable choice as an active placebo.50 The use of an active placebo is to minimize treatment recognition and bias by patient and investigator. Identical, sequentially numbered syringes containing either propofol or intralipid will be prepared by the hospital pharmacy and these will be appropriately coupled with identical numbered minibags of saline containing either no drug or midazolam. Subjects will be randomized after inclusion and exclusion criteria have been applied. Investigators will use these syringes / minibags in numeric order. (The named contents of each syringe and minibag will be available to the investigators in a second sealed envelope to be opened only in case of emergency during the infusion). Subject data will be grouped by treatment allocation for analysis into 'Group A' and 'Group B', but the blind will not be fully broken until data analysis is complete. The subject's daily pain diaries will be coded with the corresponding syringe code.

Infusion procedure On the day of participation the participant will be required to be fasted from the previous midnight. The groups will receive the following treatments by electronic infusor. The dose of propofol selected is comparable with that of the previous studies.22-24

Active treatment

- Propofol infusion 40 mcg / kg / minute over 60 minutes coupled with a

- concurrent infusion of saline in a minibag to be delivered at 1ml / kg over first 10 mins.

Placebo group

- Intralipid by infusion, calculated to be delivered at the same rate as if it were propofol coupled with a

- concurrent infusion of saline in a minibag containing 35mcg / ml of midazolam to be delivered at 1ml / kg over the first 10 mins.

- To minimize pain on infusion (that may occur with either treatment) a large vein (preferably antecubital) will be cannulated.

Environment and Treatment The location of the study will be the treatment suite of the University of Alberta Hospital Pain Centre. At least one anesthesiologist will be on site and a post anesthesia recovery nurse. A peripheral intravenous cannula will be inserted into each participant and intravenous normal saline 250 cc will be commenced to run over one hour in conjunction with the treatments to flush the intravenous line. The patient will be placed on a stretcher and made comfortable within the privacy of the suite. The lighting may be dimmed but not to a level that would compromise monitoring and safety. Noise will be kept to a minimum.

Monitoring and Safety Non-invasive blood pressure will be recorded pre-infusion and then at 5 minute intervals in conjunction with continuous pulse oximetry and respiratory rate. Oxygen via nasal cannulae to keep oxygen saturations at or above 94% will be commenced. Basic and advanced life support equipment will be available in the suite. An investigator, a trained anesthesiologist with extensive experience in the administration of propofol and qualified in resuscitation, will be in constant attendance.

The infusion will be discontinued if the patient develops signs of respiratory depression (< 8 breaths / min), requires more than simple maneuvers to maintain the airway, develops oversedation or delerium - where a patient is not easily roused by verbal command, sustained hypotension, oxygen desaturation (below 94%), cardiac dysrhythmia or new medical symptoms. In the event of the development of any adverse reaction a referral to the relevant specialist service within the University of Alberta Hospital will be available. Patients will be observed for a period of two hours post infusion and will be free to leave the suite when their vital signs are stable, there are no concerns about level of consciousness and the patient meets the University of Alberta Hospital criteria for discharge form day- surgery. Full recovery from either infusion is anticipated to take less than two hours.51 The patient must be escorted home and must be in the company of a carer for 24 hours. The patient will be instructed not drive or operate heavy machinery for 24 hours post infusion.

Data Collection. Baseline data will be recorded as above (including HDI and Headache Diary). Pain, nausea and sedation scores will be recorded pre-infusion and then immediately post infusion Each subject will be given a daily headache diary to keep for the 30 days following the infusion. Subjects will be contacted by telephone at 24 and 72 hours post infusion and weekly thereafter to remind them to complete their headache diaries. At 30 days post-infusion the patients will complete a follow-up HDI by telephone or mail and their diaries collected.

Outcome measures. The primary study outcome will be Headache Disability Index and the Headache Index45-47 The baseline Headache Index is the mean of all 28 NRS pain scores (four measures daily for seven days) and the post-treatment Headache Index is the mean of all 120 NRS pain scores (four measures daily for thirty days) and provides a valid summary measure of overall headache activity.

Secondary outcome measures will include:

- Number of days with a headache of at least moderate severity defined as pain at least 5 on the Numerical Rating Scale (NRS).52

- Peak daily headache severity according to the NRS.

- Analgesic medication consumption will be measured in terms of morphine equivalents of narcotic medication and number of tablets /doses of acetaminophen / non-steroidal anti-inflammatory (NSAID) containing analgesics.53

- Visits to the Emergency Room for headache treatment.

- Sedation and nausea scores in the treatment and placebo groups.

- Ability of participants in each of the groups to correctly guess their treatment assignment.

- Adverse effects related to the infusion.

Power calculation

Mean HDI measurements of 60 (+/- 20) for severe CDH (the population toward which our clinic population is biased) would be expected. A clinically significant outcome would be a 29 point reduction of HDI. A sample size of 20 subjects will give 80% power to detect this difference with an alpha error rate of 0.05.

Analysis Between group demographic factors that will be examined and controlled for will include age, gender, education level (years), and socioeconomic status using independent samples t-tests.48 All analyses will be conducted on an intention to treat basis. The average absolute difference between the baseline and post-treatment HDI scores for the two infusion groups will be compared using analysis of variance (ANOVA). We will compare the changes in NRS scores between the two infusions groups over time using repeated measures ANOVA. Any demographic factors found to be significantly different between groups will be included as covariates in this analysis. All t-tests will use Bonferroni corrections to control for the inflation of alpha.