Summary of evidence of the efficacy of zolpidem in brain damage:
September 2008


Ralf Clauss, MD, FRCP
Nuclear Medicine Department,
Royal Surrey County Hospital
Guildford. GU2 7XX
Andrew Sutton, MBBS, FF, MD (London)
c/o ReGen Therapeutics Plc
asutton@gcpl.co.uk

ralf.clauss@royalsurrey.nhs.uk

A summary of the use of zolpidem in patients
with brain injury.
INTRODUCTION
Zolpidem is a sedative medicine that has been routinely prescribed for insomnia for many years. In the USA it is the most used sedative of all and is known as Ambien®. It induces sleep by activating brain GABA(A) receptors which are inhibitory, macromolecular complexes that mediate sedative, anticonvulsant, anxiolytic and myo-relaxant effects. It is a chemically distinct imidazopiridine that works at the alpha sub-unit of these receptors, referred to as the benzodiazepine (omega) receptor. There are at least three omega receptor subtypes. Benzodiazepines bind non-selectively to these while zolpidem binds preferentially to omega-1 receptors.

1 LITERATURE ON ZOLPIDEM EFFICACY IN BRAIN INJURIES.
The first report in 20001 of unexpected improvements after zolpidem in a patient with impaired consciousness after traumatic brain injury, recorded that zolpidem can achieve this effect years after the injury. Other written reports continue to appear and are listed below, while individual anecdotal personal and internet communications are legion. The original centre in South Africa has now treated over 300 patients.

Table 1. Published References
Pathology Patients Reference
Anoxic Brain Damage
PVS/MCS 5 Clauss & Nel, 20001, 20042, 20063, Brefel-Courbon et al, 20074
Stroke 4 Clauss & Nel, 20042, 20055; Cohen et al, 20046
Near drowning 2 Clauss & Nel, 20042, 20063
Hypoxia cardiac arrest 2 Cohen SI and Duong 20087; Shadan et al, 20048, Shames & Ring 20089.
Other
Dementia 4 Jarry et al 200210, Clauss et al 20055
Bell’s Palsy 1 Clauss & Nel, 200511
Cerebellar ataxia 5 Clauss, Sathekge, Nel, 200412
Auditory impairment 2 Clauss et al, 20055

Collectively, the above evidence can be regarded as clear proof that the zolpidem effect on brain damage is genuine as it exceeds the criteria for proof of efficacy without a formal placebo-controlled clinical trial13. This is because the onset and depth of effect coincides with the absorption of zolpidem, the decline of effect coincides with its elimination and the effect is reproducible with every dose. The recent Cohen and Duong paper mentioned in the table above in particular describes how the patient relapsed dramatically when his zolpidem was stopped for 3 days7. This is effectively an “n-of-1” crossover study that meets the highest category of criteria as evidence of efficacy according to the working party report of the Evidence Based Medicines Group 200014.

Proportion of Responders

Responders do not appear to form the majority. Nel has found that the first dose response rate is around 10-15% in patients with marked neuro-disability, including PVS and MCS, while it may reach 30-60% in lesser disabilities. Further studies on response rate are required, especially in view of a recent patient who took 8 days’ treatment to respond.

Neurodormancy, the mechanism of effect
SPECT and PET brain imaging studies have shown that parts of damaged brain that are inactive and considered to be dead tissue until now, will start to function again after zolpidem 1, 2, 4, 6.

The inactive areas are now regarded as dormant instead. They show no typical location or distribution pattern and vary from patient to patient. Activation of these areas by zolpidem could be reproduced in animal studies and the effects blocked by flumazenil, a benzodiazepine antagonist15,16

The exact mechanism is not known but a recent magnetoenchalographic (MEG) investigation of a post-stroke subject at Aston University, Birmingham, UK has suggested that zolpidem (as opposed to placebo and another sedative zopiclone) seems to achieve its clinical effect by somehow reducing the pathological beta waves associated with dormant brain tissue and known to be associated with other neurological conditions17.

These findings led Clauss to propose the concept of neurodormancy in 20042. It resembles the established diaschisis phenomenon, which is not unlike a form of hibernation where cells threatened with hypoxia shut down their metabolic processes to survive, but there are important differences, such as permanence, intralesional location and late onset. Neurodormancy appears to be a common denominator in patients who respond to zolpidem and explains the efficacy of zolpidem across a wide range of unrelated brain pathology.


2 SAFETY AND ADVERSE EFFECTS
Use in patients with brain damage

Clearly at higher doses zolpidem will sedate a patient. Any beneficial effect will then be masked by sedation. One report to this effect from Edinburgh has appeared recently.18 However, to date a higher dose does not appear to present a greater risk than normal sleep.

There are few reports of systematic trials of safety. Recently Krystal et al have reported that there were no serious adverse events in a thousand patients who took 12.5mg zolpidem for sleep at night for 6 months19. Sleep was maintained satisfactorily for the whole 6 months and there was no rebound effect on cessation.

Long term use
In longer term use as a sedative there are occasional reports of effects that disturb patients without having life threatening implications, ie: are recurrent cases of antegrade amnesia, hallucinations and sleepwalking. They have not been reported in patients with brain damage.

There has been no adverse implication for health in this new indication for the 300 or so patients that Dr Nel has now treated. One PVS patient has been treated every day for 8 years and he has steadily improved without any adverse effect apart from the expected sedation. Recently his SPECT scan was repeated prior to a dose and it was much improved compared with his first scan, indicating that reversal of dormancy has progressed pari passu with his clinical improvement. He still shows some benefit from an individual dose, in particular his IQ test result moves from 70 to 90 on a 5mg dose.

Overdose safety
In terms of acute overdose, zolpidem safety far exceeds other GABA agonist sedatives such as the benzodiazepines midazolam and triazolam. Wyss et al20 have published the only known survey of such overdose patients and reported that with forty fold the normal 10mg dose no severe symptoms occurred in patients with zolpidem single-drug poisonings, while coma was encountered in 4 cases (11%) with triazolam and 4 cases (10%)with midazolam.

Withdrawal effects
Abrupt withdrawal of high long-term doses has produced three letters citing epileptic seizures 21, 22, 23. They include a 50 year old woman who took normal doses daily for 5 years then due to tolerance the dose was increased to 450mg per day in divided doses21. She abstained for 12 hours and suffered an epileptic fit that she survived without permanent sequelae. Reports from Greece and India each cite an individual patient who had an epileptic seizure after long term high doses of zolpidem22, 23 and another report from France describes a dependence syndrome in two patients who were dependent on other drugs and were diagnosed with severe personality disorders24. However these are rare cases and the last two patients had a personality disorder.

To summarize safety, it is clear that the ratio of benefit to risk for this new usage is high due to the unique, life-changing nature of the benefit and the known tolerability of zolpidem. This safety record and the absence of reported toxicity, despite the very widespread use of zolpidem as the most prescribed sedative in the USA for example, justifies a trial of the treatment with laboratory safety testing approximately 6-monthly.


3 CONDUCTING AN OFF-LABEL TRIAL IN PATIENTS
While it must be emphasised that neither ReGen nor the authors of this summary are in favour of unauthorised clinical trials with novel medicines, the case of zolpidem resembles the common, so-called “off-label” use of a very well known medication. Doctors frequently and entirely legally prescribe medicines in patients where there is no official authorization to do so, paediatrics and geriatrics being common examples. The obligation on the prescriber is that such use is known to be in the best interest of the patient. In the case of zolpidem there is now strong evidence that a trial is worth undertaking even though the proportion that responds may be smaller than currently hoped. This is because the benefits to the patient, the carer and the health service can be extensive while zolpidem has been shown to present no predictable significant risk. It is also the only medication that has this beneficial effect in a predictable way with an observed mechanism of action, namely reversal of neurodormancy.
1. Identification of patients
Responders may be identified by a simple clinical test of daily zolpidem for up to two weeks with close clinical observation, or SPECT or PET brain imaging studies before and after a dose of zolpidem. These scans can be combined with CT or MRI.

2. Initial Dosing
In patients with brain damage many regimes have now been tried but the most frequent has been morning dosing with sub-sedative doses, ie: 5mg in adults and 2.5 mg in children. These doses may have to be repeated at approximately 4 hourly intervals8. This compares with the UK recommended sedative dose of 10mg at night, although some adult patients do require 20mg for sleep induction.
However, such is the variation between patients that some have preferred a higher evening dose of 10mg so that patients obtain the full benefit in the evening before sleep and then have a small top-up dose of 2.5 mg once or twice during the day. This may be a logical approach to the first doses when the physician or the carers consider it best to give a full dose to maximize the chance of obtaining a beneficial response, but to do so when it is useful for the patient to go to sleep if the dose proves to be sedative but ineffectual.
Some patients have reported that the original brand marketed by Sanofi and called Stilnoct© in the UK is more effective than generic formulations. The reason for this is unknown, although different rates of absorption into the brain may be having an influence.



3. Longer term dosing
Patients may need several dose adjustments to achieve an optimal balance between effect and sedation. Dose requirements may increase periodically or reduce when a response has become established.
After treating over 300 patients Nel and Clauss have found that the optimal regimen is a sub-sedative morning dose 1 hour after breakfast. In adults 5 mg may be given, while in children half this dose has been used.

4.Responses
Response to the tablets may start within 30 minutes, peak by 1 hour and last 3-4 hours. The response may be subtle, such as saying a whole sentence for the first time and may be more difficult to detect if masked by sedation. One patient responded only after 8 days so it appears wise to continue for two weeks before pronouncing a patient totally unresponsive.

The longest therapy duration is over 8 years. The patient described by Clauss and Nel in 2000 began in a deeply unconscious state and is fully conscious now. His IQ improves from 70 – 90 on zolpidem. Full blood count, liver and renal functions remain within normal limits. ReGen Therapeutics Plc recommends repeating such tests 6 monthly because the safety of continuous, long term dosing has not been fully established, despite the well-known safety of zolpidem in its traditional use, or in acute overdose.

5. Age Range of patients
Responses occur at any age. The youngest patient was 2 years of age and the eldest over 80 years2. Some patients injured at birth have responded when in their twenties or thirties.

6. Clinical trials
In 2008 ReGen Therapeutics will be conducting a trial in 20 stroke patients using a sublingual fluid formulation because it will be absorbed faster and more consistently than tablets. This was shown in the first study that was conducted in 2007 in known responder patients, where a sublingual spray produced a peak effect at 15 minutes while a tablet did so at about 90 minutes.
7. Conclusion
The evidence for an important role of zolpidem in the treatment of brain damage becomes more compelling as more centres report their findings. SPECT, PET and more recently MEG scans show that the mechanism includes reversal of neurodormant brain areas that were hitherto considered beyond repair. A wide range of brain pathology has responded to zolpidem including hypoxia from all origins. The depth of injury also ranges from profoundly impaired consciousness (PVS/ MCS) to milder injuries such as damage to the origins of a cranial nerve.

8. References
(1.) Clauss R P, Güldenpfennig W M, Nel W H, Sathekge M M & Venkannagari R R.(2000)
Extraordinary arousal from semi-comatose state on zolpidem. S Afr Med J, 2 90 68.

(2.) Clauss RP, Nel HW (2004). The effect of zolpidem on brain injury and diaschisis as detected by
99mTc HMPAO Brain SPECT in humans. Arzneim.-Forsch./Drug Res, 54 641-646.

(3.) Clauss R P & Nel W H. (2006). Drug induced arousal from permanent vegetative state.
Neurorehabilitation, 21(1) 23-8.

(4.) Brefel-Courbon C et al (2007). Clinical and Imaging evidence of zolpidem effect in hypoxic
encephalopathy. Annals of Neurology, March.

(5.) Clauss RP, Jayarajan V, Nel HW, Saunders E. (2005).Evidence for zolpidem efficacy in auditory
impairment. 9th European Federation of Neurological Societies Congress, Athens, Greece, September 7-20

(6.) Cohen L et al. (2004). l Transient Improvement of Aphasia with Zolpidem
N. Engl. J. Med., 350(9):949-950.

(7.) Cohen SI and Duong TT (2008). Sleep, February: Am J Phys Med Rehabil. Increased Arousal in a Patient with Anoxic Brain Injury After Administration of Zolpidem

(8.) Shadan F F et al (2004) Zolpidem for Postanoxic Spasticity
Southern Med J., 97(8):791-792.

(9.) Shames JL & Ring, H (2008). Transient reversal of anoxic brain injury-related Minimally Conscious State after zolpidem administration: A case report. Arch Phys Med Rehabil Vol 89 Feb 386-388.

(10.) Jarry C, Fontenas JP, Jonville-Bera AP, Autret-Leca E (2002). Beneficial effect of zolpidem for
dementia. Ann. Pharmacoter. Nov; 36(11):1808.

(11.) Clauss RP, Nel HW (2005). Evidence for zolpidem efficacy in brain damage. SA Fam Pract, 47(3): 49-50.

12). Clauss RP, Sathekge MM, Nel HW (2004). Transient improvement of Spinocerebellar Ataxia with
Zolpidem. N. Engl. J. Med. 351, 511-512.

13.) Glasziou P, Chalmers I, Rawlins M, McCulloch P (2007). When are randomised trials unnecessary?
Picking signal from noise. Centre for Evidence-Based Medicine, Department of Primary Health Care, University of Oxford, Oxford OX3 7LF. BMJ, 334:349-351 (17 February).

14.) Guyatt GH, Keller JL Jaeschke R, et al (2000). The n-of-1 randomised controlled trial: clinical
usefulness. Report of the Evidence-Based Medicine Working Group. JAMA. Sep 13; 284(10):1290-6.

15.) Clauss RP, Dormehl IC, Kilian E, et al (2001). Measurement of cerebral perfusion after zolpidem
administration in the baboon. Arzneim.-Forsch./Drug Res. 51, 619-622.

16.) Clauss RP, Dormehl IC, Kilian E, et al (2002). Cerebral blood perfusion after treatment with
omega receptor drugs, zolpidem and flumazenil in the baboon.
Arzneim.-Forsch./Drug Res.52, 740- 744.

17) Stephen D. Hall et al. Desynchronization of pathological low-frequency brain activity by
the hypnotic drug zolpidem (2008). Nature Precedings : hdl:10101/npre.2008.1966.1 : Posted 11 Jun


18.) Singh R et al. (2008). Zolpidem in a minimally conscious state. Brain Injury, Volume 22, Issue 1, pages 103 - 106

19.) Krystal AD et al (2007). Long-Term Efficacy and Safety of Zolpidem Extended-Release 12.5 mg, Administered 3 to 7 Nights Per Week for 24 Weeks, in Patients With Chronic Primary Insomnia: A 6-Month, Randomized, Double-Blind, Placebo-Controlled, Parallel-Group, Multicenter Study. Sleep, Vol 31, 1, 79-90

20.) Wyss PA, Radavanovic D, Meier-Abt PJ (1996),. Acute overdose of zolpidem(Stilmox®).
Schweiz Med Wochenschr. May 4: 126(18):750-6. [Article in German.]

21.) Barrero-Hernández FJ, et al. (2002) Epileptic seizures as a sign of abstinence from chronic consumption of zolpidem. Rev Neurol. Feb 1-15; 34(3):253-6.

22.) Tripodianakis J, et al. (2003). Zolpidem-related epileptic seizures: a case report.
Eur Psychiatry. May; 18(3): 140-1.

23.) Sethi PK & Khandelwal DC. (2005). Zolpidem at supra-therapeutic doses can cause drug
abuse, dependence & withdrawal seizure. J Ass Physicians India.Feb; 53:139-40

24.) Boulanger-Rostowsky L, et al. (2004) Dependence on zolpidem: a report of two cases Encephale. Mar-Apr; 30(2):153-5.