Is intranasal midazolam an effective rescue medication in adolescents and adults with severe epilepsy?

Seizure 2000; 9: 417 422 doi: 10.1053/seiz.2000.0425, available online at http://www.idealibrary.com on Is intranasal midazolam an effective rescue medication in adolescents and adults with severe epilepsy? MARK SCHEEPERS, BRUCE SCHEEPERS, MICHAEL CLARKE, SUSAN COMISH & MATTHEW IBITOYE Heathfield, 30 Denmark Road, Gloucester, GL1 3HZ, UK; Lane, Alderley Edge, SK9 7UD, UK David Lewis Centre for Epilepsy, Mill Correspondence to: Mark Scheepers, Heathfield, 30 Denmark Road, Gloucester, GL1 3HZ, UK. E-mail: mscheeps@aol.com The aim of this study was to determine whether intranasal midazolam is a safe and effective rescue medication in adolescent and adult patients with severe epilepsy. This field trial was designed to test the feasibility of the use of intranasal midazolam as an alternative to rectal diazepam in a cohort of patients with severe epilepsy who require rescue medication as part of their treatment. A dose of intranasal midazolam (5 mg if the patient weighed less than 50 kg and 10 mg if the patient weighed over 50 kilograms) was prescribed for those who had previously responded to other rescue medication. Midazolam was prescribed buccally if excessive head movement accompanied seizures. The protocol reverted to the usual rescue medication if there was no response to midazolam within 10 minutes. Vital signs were monitored for half an hour following the administration of the treatment. Twenty-two patients received 84 treatment episodes and 79 of these were considered clinically effective. Five treatment failures were recorded, three due to poor technique in delivering the midazolam. Two patients were successfully retried on midazolam and a third is awaiting a retrial of this drug. The two other treatment failures received the drug buccally. In the first patient the clinical opinion was that this was possibly a psychogenic non-epileptic seizure. The other patient responded initially, but within an hour had another seizure requiring further rescue treatment. No significant adverse effects were reported. Our study shows that intranasal midazolam, when used appropriately, is an effective treatment in those who require rescue treatment. There are clear advantages in the use of midazolam over diazepam in the treatment of acute seizures. These include the favourable pharmacokinetic and pharmacodynamic properties of midazolam as well as the potential of a more acceptable and dignified administration route. c 2000 BEA Trading Ltd Key words: intranasal midazolam; rescue medication; epilepsy. INTRODUCTION Rescue medication plays an important role in the management of epilepsy. It is prescribed for patients with prolonged, repeated or cyanotic episodes during seizures and for those with severe post-ictal aggression or agitation. These patients usually receive rectal treatments including diazepam and paraldehyde. Rectal diazepam, although the current treatment of choice, has some significant drawbacks. The drug has a variable absorption, is highly lipophilic and therefore has a large volume of distribution leading to a half-life of 20 40 hours when administered rectally 1. The rectal route of administration has become unpopular because of the undignified nature of the procedure. Changing attitudes among some care organizations has complicated this with some school and social service employees in the UK reticent about administering a drug via the rectal route 2. Midazolam has recently been described as an alternative rescue medication in severe epilepsy 3 7. It has a unique chemical structure which allows it to be water soluble whilst in suspension and therefore readily absorbed via the intranasal and buccal mucosae. At physiological ph the ring structure closes and the drug become lipid soluble allowing it to readily cross the blood brain barrier. Metabolism is principally via hepatic enzymes and the elimination half-life 1059 1311/00/060417 + 06 $35.00/0 c 2000 BEA Trading Ltd

418 M. Scheepers et al. is 2 hours in healthy subjects 8. Previous studies using midazolam as an anaesthetic agent have concluded that the drug, when given intranasally, is an effective sedative and amnestic agent 12 22. These studies have improved our understanding of the drug, the pros and cons of various routes of administration and its pharmacodynamic and pharmacokinetic properties 12 22. There are few reports confirming the efficacy of midazolam in terminating seizures 4 7 and published data on the safety of midazolam in epilepsy, particularly in an adult population, is sparse 3 11. For the small number of people with epilepsy who require rescue medication it is a potentially life saving treatment that may significantly reduce morbidity and improve quality of life. A field trial was undertaken to determine the efficacy of intranasal midazolam as a rescue treatment in epilepsy. PATIENTS AND METHODS Patients at a specialist Epilepsy Treatment Centre who require rescue medication as part of their epilepsy treatment were recruited to this study. The reasons for rescue medication were prolonged or repeated seizures, cyanotic events during seizures or disabling post-ictal aggression or agitation. Most patients usually received rectal diazepam although some were receiving other treatments including rectal paraldehyde or diazepam gel (for rectal use), administered buccally. All patients, their carers or parents were provided with an information leaflet detailing the unlicensed nature of this treatment and references to available, published data. Following discussions about the treatment, the patient or legal guardian signed a consent or assent to treatment form. Patients acted as their own controls as all had previously responded to an alternative rescue treatment. The protocol was designed so that intranasal midazolam was prescribed as first line treatment and previous rescue medication was available as a fallback. The need to use the previous rescue medication constituted a trial failure. Possible problems with intranasal delivery were discussed and where indicated, for example excessive head movements or upper respiratory tract infections, it was agreed that a similar dose of midazolam could be administered buccally. A qualified nurse administered the first dose of midazolam and monitored vital signs for 30 minutes, giving the fallback medication in the event of treatment failure. The delivery method altered slightly during the trial period. Initially the drug was drawn up with a syringe and needle and then administered via a neonatal feeding tube or the tubing of a butterfly needle with the needle removed. This method was found to be cumbersome and was further complicated by the presence of various sharp objects. The administration protocol was therefore modified to allow the use of a 1 ml Pasteur pipette (made of disposable soft plastic), the dose remained unchanged. The usual recommended dose for intravenous administration of midazolam in children is 0.2 mg kg 1. The proprietary vial of this preparation contains 2 ml of a 5 mg ml 1 concentration. In order to simplify the administration process, all patients weighing over 50 kg received 10 ml and those weighing less than 50 kg received 5 ml as a starting dose. Because of their physical characteristics, one patient received 20 mg after an initial treatment failure, this dose was effective and subsequent treatments with the Pasteur pipette were successful at a dose of 10 mg. The majority of patients received rescue medication for generalized tonic clonic seizures, but it was also used in some patients with complex partial or myoclonic seizures (refer to Table 1). RESULTS Twenty-two patients (13 male and 9 female), ranging in age from 12 72 (median 26) have received 84 doses of midazolam as rescue medication. Patients were experiencing between 1 and 250 seizures a month. There were five treatment failures in this study. Three treatment episodes in three different patients failed due to poor technique in delivering an adequate intranasal dose of the drug. Two of these patients have had more than one treatment and have subsequently responded using the new administration protocol whilst the other is awaiting an opportunity to be retried. All these received rectal diazepam after 10 minutes with good effect. Two treatments were given via the buccal route; both resulted in treatment failure. Both patients received this treatment because of excessive head movements during the seizure. In the first of these treatment failures, the seizure was retrospectively considered by staff to be non-epileptic and psychogenic in nature. The other responded initially with termination of the seizure, but recurrence within the hour required further rescue medication and the fallback treatment was given. There were no recorded abnormalities in blood pressure or pulse and no apnoea or diminished respiratory effort following the administration of midazolam. DISCUSSION Prolonged, repeated or cyanotic episodes during seizures were the most common reasons for prescribing rescue medication and in this study 58 of the

419 Table 1: Patients receiving intranasal or buccal midazolam. No Age Sex Classification/Syndrome Dose No of Reason for Failures (mg) doses treatment 1 12 M Lennox Gastaut 10 1 Prolonged 2 13 M Generalized 5 6 Cyanotic 3 14 F Localization related 5 9 Cyanotic 4 16 F Lennox Gastaut 10 2 Prolonged 5 16 M Generalized 10 1 Aggression 6 16 M Generalized 10 1 Clusters X 7 19 F Localization related 10 3 Clusters 8 20 M Localization related 10 a 13 Prolonged X 9 21 M Generalized 10 6 Cyanotic 10 25 M Localization related 10 1 Prolonged 11 25 M Generalized 10 3 Agitation X 12 27 F Localization related 10 2 Prolonged X b 13 31 M Localization related 10 2 Clusters 14 34 F Generalized 10 22 Agitation 15 35 M Localization related 10 1 Clusters 16 43 M Localization related 10 1 Prolonged 17 44 F Generalized 10 1 Prolonged 18 47 F Generalized 10 2 Clusters 19 47 M Generalized 10 1 Clusters 20 52 M Localization related 10 1 Prolonged 21 56 F Localization related 10 4 Prolonged X b 22 72 F Generalized 10 1 Cyanotic a Has also received one dose of 20 mg. b Received buccal midazolam. Table 2: Reason for treatment and outcome. Reason for treatment Number of patients Number of treatments Failures in each group Cyanotic episodes 4 22 0 Clusters of seizures 6 10 1 Prolonged seizures 9 26 3 Post-ictal events 3 26 1 84 treatments (69%) fell within this group. There were no treatment failures in the group with cyanotic episodes. The three treatment failures in the prolonged seizure group comprised two where the failure was attributed to poor technique and one due to a psychogenic non-epileptic attack. The authors were concerned that the short duration of action of midazolam would make it an inappropriate drug for the treatment of seizure clusters (repeated seizures) where there is full recovery between seizures. No defining criteria were set for seizure clustering other than a past history and routine use of rescue medication to prevent this occurring. Apart from one treatment episode, this medication appeared to successfully prevent seizure clustering in patients with a history of these episodes. It may be argued that in some patients rescue medication may not have been required, but for those patients included in the study, clustering remained a typical occurrence and rescue medication part of their routine treatment protocol. The protocol for most patients required rescue treatment after the second seizure. The one treatment failure in this group may indicate that in the event of breakthrough seizures within an hour, alternative longer-acting medication may be required. A suggested protocol would be to treat the first episode with midazolam and to then use an oral dose of a longer-acting benzodiazepine if the patient was to regularly experience further seizures. Some patients with short-lived major seizures may experience severe post-ictal aggression or agitation that requires medication. Three individuals who suffered from debilitating post-ictal events received 26 of the 84 treatments (31%). There was one treatment failure in this group believed to be due to poor technique. The rapid action of midazolam without prolonged effect has resulted in a successful alternative to the more sedating and longer-acting oral or rectal medication previously prescribed. Patients in this group are often extremely embarrassed or remorseful when they discover the results of their behaviour. They often have shorter seizures, but a protracted post-ictal period. It is essential to deliver medication very rapidly and for the medication to have

420 M. Scheepers et al. Fig. 1: Original administration devices: 2 ml syringe, neonatal feeding tube and butterfly needle tubing with the needle removed. Fig. 2: Pasteur pipette (1 ml) with midazolam vial (2 ml). a rapid onset of effect. Our experience is that this treatment has been extremely successful in preventing such episodes (in a limited number of patients with a long history of this disorder) with a dramatic improvement in post-ictal behaviour. Midazolam has an acidic ph (3.28) and therefore is a potential irritant, which it is suggested may cause discomfort when used intranasally 23. We can report that in our cohort of patients, who are aged 12 years and older, on questioning none have experienced or complained of debilitating effects. Some patients who experience simple or complex partial seizures have had interictal trials of intranasal saline solution in order to acclimatize them to this route of delivery. Early reports suggest that they have tolerated this route of administration and can possibly be given intranasal midazolam despite no loss of awareness during partial seizures. This raises the possibility of self-administration of midazolam in order to prevent seizure progression for some patients. The patients included in this study are now all prescribed midazolam as first line rescue treatment on a named patient basis. Having improved the delivery method via a 1 ml disposable plastic Pasteur pipette, non-professionally trained care staff and parents have successfully used this treatment method. Rescue medication has an important place in the treatment of severe epilepsy. Intranasal midazolam appears to be a safe and effective rescue treatment that has a number of advantages over rectal diazepam. Nasal treatment is more dignified and more acceptable to both patients and carers. Midazolam has more favourable pharmacokinetic and pharmacodynamic properties resulting in a rapid onset of action without accumulation and subsequent hangover effect. Further benefit has been derived for the patients in this study through the apparent lessening of handicap by improving access to education, employment and social opportunities due to the rapid onset of action and quick recovery. This field trail confirms that intranasal midazolam is effective at terminating prolonged seizures, preventing seizure clustering and avoiding post-ictal behavioural complications in adolescent and adult patients with severe epilepsy. The intranasal route has been demonstrated to be accessible, acceptable and apparently free

421 of adverse experiences and complications. Furthermore, non-trained care staff have successfully administered the treatment. Although no adverse experiences have been reported to date, the sample sizes in this and other studies remain too small to be certain of safety especially with prolonged, repeated use. Further multicentre studies are required in order to reach consensus for the future use of what is currently an unlicensed delivery route and treatment for epilepsy. REFERENCES 1. Betts, T. Epilepsy, Psychiatry and Learning Difficulty. Martin Dunitz, 1998. 2. Bird, J. Epilepsy and learning disabilities. In: Seminars in the Psychiatry of Learning Disabilities (Ed. O. Russell). Gaskell, London, The Royal College of Psychiatrists, 1997: pp. 223 245. 3. Scheepers, M., Scheepers, B. and Clough, P. Midazolam via the intranasal route: an effective rescue medication for severe epilepsy in adults with a learning disability. Seizure 1998; 6: 509 513. 4. Scott, R., Besag, F. and Neville, B. Buccal midazolam and rectal diazepam for treatment of prolonged seizures in childhood and adolescence: a randomised trial. Lancet 1999; 353: 623 626. 5. Scheepers, M., Comish, S., Cordes, L., Clough, P. and Scheepers, B. Buccal midazolam and rectal diazepam for epilepsy. Lancet 1999; 353: 1797. 6. Chattopadhay, A., Morris, B., Blackburn, L., Wassmer, E. and Whitehouse, W. Buccal midazolam and rectal diazepam for epilepsy. Lancet 1999; 353: 1797. 7. Lahat, E., Goldman, M., Barr, J., Eshel, G. and Berkovitch, M. Intranasal midazolam for childhood seizures. Lancet 1998; 352: 620. 8. O Regan, M., Brown, J. and Clarke, M. Nasal rather than rectal benzodiazepines in the management of acute childhood seizures? Developmental Medicine and Child Neurology 1996; 38: 1037 1041. 9. Scott, R., Neville, B., Besag, F. and Boyd, S. Nasal rather than rectal benzodiazepines in the treatment of acute childhood seizures? Developmental Medicine and Child Neurology 1997; 39: 137. 10. Wallace, S. Nasal benzodiazepines for management of acute childhood seizures? Lancet 1997; 349: 222. 11. Kendall, J., Reynolds, M. and Goldberg, R. Intranasal midazolam in patients with status epilepticus. Annals of Emergency Medicine 1997; 29: 415 417. 12. Zandsberg, S. and Rosenblum, M. Nonconventional drug administration in anaesthesia. Anaesthesiology Clinics of North America 1994; 12: 17 38. 13. Streisand, J. B. and Stanley, T. H. Newer drug delivery systems. Current Anaesthesia and Critical Care 1995; 6: 113 120. 14. Malinovsky, J.-M., Lejus, C., Servin, F. et al. Plasma concentrations of Midazolam after I.V., nasal or rectal administration in children. British Journal of Anaesthesia 1993; 70: 617 620. 15. Fukuta, O., Braham, R. L., Yanase, H. and Kuroso, K. Intranasal administration of Midazolam: pharmacokinetic and pharmacodynamic properties and sedative potential. Journal of Dentistry for Children 1997; 64: 89 98. 16. Rey, E., Delaunay, L., Pons, G., Murat, I., Richard, M. O., Saint-Maurice, C. and Olive, G. Pharmacokinetics of midazolam in children: comparative study of intranasal and intravenous administration. European Journal of Clinical Pharmacology 1991; 41: 355 357. 17. Wilton, N. C. T., Leigh, J., Rosen, D. R. and Pandit, U. A. Preanaesthetic sedation of preschool children using intranasal midazolam. Anaesthesiology 1988; 69: 972 975. 18. Harcke, H. T., Grissom, L. E. and Meister, M. A. Sedation in paediatric imaging using intranasal midazolam. Pediatric Radiology 1995; 25: 341 343. 19. Hogberg, L., Nordvall, M., Tjellstrom, B. and Stenhammar, L. Intranasal versus intravenous administration of midazolam to children undergoing small bowel biopsy. Acta Paediatrica 1995; 84: 1429 1431. 20. Fukuta, O., Braham, R. L., Yanase, H. and Kuroso, K. The sedative effects of intranasal midazolam administration in the dental treatment of patients with mental disabilities Part 2: Optimal concentration of intranasal midazolam. Journal of Clinical Pediatric Dentistry 1994; 18: 259 265. 21. Malinkovsky, J.-M., Populaire, C., Cozian, A., Lepage, J.-Y., Lejus, C. and Pinaud, M. Premedication with midazolam in children. Effect of intranasal, rectal and oral routes on plasma midazolam concentrations. Anaesthesia 1995; 50: 351 354. 22. Kaufman, E., Davidson, E., Sheinkman, Z. and Magora, F. Comparison between intranasal and intravenous midazolam sedation (with or without patient control) in a dental phobia clinic. Journal of Oral and Maxillofacial Surgery 1994; 52: 840 843. 23. Karl, H. W., Rosengerger, J. L., Larach, M. G. and Ruffle, J. M. Transmucosal administration of midazolam for premedication of paediatric patients. Anaesthesiology 1993; 78: 885 891.