Sleep, 9{1):285-289
`© 1986 Raven Press, New York
`
`Treatment of Narcolepsy with
`'Y-Hydroxybutyrate. A Review of Clinical and
`Sleep Laboratory Findings
`
`*Mortimer Marnelak, tMartin B. Scharf, and :j:Marcia Woods
`
`*Department of Psychiatry, University of Toronto,. Toronto, Ontario, Caooda; and tDepartment
`of Psychiatry, University of Cincinnati, School of Medicine, Cincinnati, Ohio, US.A.
`
`Summary: Previous studies on the effects of -y-hydroxybutyrate (GHB) on the
`sleep and clinical response of patients with narcolepsy are reviewed. New infor(cid:173)
`mation on 48 patients treated with GHB for as long as 9 years is presented. These
`studies indicate that 2.25 to 3.00 g of GHB, taken in conjunction with a low dose
`of a stimulant during the day, rapidly alleviate the symptoms of narcolepsy in most
`patients. Tolerance does not develop to this treatment regimen; neither have any
`patients discontinued the treatment because of side effects. In poor responders,
`daytime drowsiness and not cataplexy has been the most common residual symp(cid:173)
`tom. Sleep studies reveal that GHB induces REM followed by slow wave sleep.
`Although total sleep time at night may be unchanged, sleep is less fragmented.
`GHB appears to be effective because it can induce the symptoms of narcolepsy
`and contain them at night. It is noteworthy, therefore, that the central biochemical
`changes induced by GHB also appear comparable to those found naturally in
`narcolepsy. Key Words: Narcolepsy-Sleep.....---y-Hydroxybutyrate.
`
`-y-Hydroxybutyrate (GHB) is a naturally occurring metabolite of the human nervous
`system, where it is found in highest concentrations in the hypothalamus and basal ganglia
`( 1). The recent discovery of central recognition sites with high affinity for GHB suggests
`that GHB functions as a neurotransmitter or neuromodulator rather than as an incidental
`breakdown product of -y-aminobutyric acid metabolism (2).
`In healthy human volunteers, low doses ( ~ 30 mglkg) of GHB promote a normal sequence
`of NREM and REM sleep lasting ~2 to 3 h (3). Slow wave activity is increased and REM
`sleep appears after a normal latency. Previous work in animals had also demonstrated that
`GHB promoted both slow wave sleep (SWS) and REM sleep (4). In addition, the animal
`data indicated that tolerance did not develop to the hypnotic effects of GHB (5). These
`properties suggested that GHB might be a useful therapeutic agent for individuals who
`required long-term use of a hypnotic. A clinical and sleep laboratory trial of GHB was
`undertaken in a group of chronic insomniacs with histories of mental depression (6). As
`in the healthy volunteers, GHB acted for 2 to 3 h, increased SWS, and sustained REM
`sleep. However, GHB shifted REM sleep to the first third of the night, significantly
`
`Accepted for publication October 1985.
`Address correspondence and reprint requests to Dr. M. Mamelak, Sunnybrook Hospital,. 2075 Bayview
`Avenue, Toronto, Ontario, Canada, M4N 3M5.
`
`285
`
`Ranbaxy Ex. 10 11
`IPR Petition - USP 8, 772,306
`
`

`
`286
`
`M. MAMELAK ET AL.
`
`shortening the REM sleep latency and prolonging the first REM sleep period, In one
`subject, REM sleep occurred at sleep ons.et; in the morning the subject reported being
`unable to move for a short period before falling asleep. GHB had induced sleep paralysis.
`This finding led Broughton and Mamelak to examine the effects of GHB in narcoleptics
`(7). It was postulated that narcolepsy stemmed from a failure to consolidate sleep at night
`and that GHB given repeatedly during the night would help reintegrate sleep because of
`its unique facilitating actions on both NREM and REM sleep. It was hoped that this would
`alleviate the diurnal symptoms of the disease and, in the first trial of the compound, such
`alleviation occurred. More extensive clinical and sleep laboratory trials have since confirmed
`this observation.
`
`CLINICAL AND SLEEP LABORATORY FINDINGS
`
`This article summarizes the published clinical and sleep laboratory data on the use of
`GHB in narcolepsy and presents new clinical data on 48 patients who have been treated
`with GHB for as long as 9 years. In all the studies reviewed here, the diagnosis of narcolepsy
`was established on the basis of a medical history of daytime sleepiness and cataplexy and
`was confirmed by sleep laboratory data which demonstrated a sleep onset REM period at
`night or during at least one sleep latency test during the day. The patients in these studies
`all received ~2.25-3 g of GHB two or three times during the night. In all, they received
`5-7 g of GHB every night. Broughton and Mamelak have reported two trials of GHB, the
`first a preliminary trial on four patients (7), and the second a more elaborate trial in which
`continuous 48-h recordings were made on 14 narcoleptic patients before and after 7-10
`days of treatment with GHB (8,9). In these trials, patients were off all medication for at
`least 2 weeks before starting GHB. Scharf et al. (10) recently reported on the effects of
`GHB on 30 narcoleptic patients. These investigators conducted sleep laboratory studies on
`their patients before and after 4 weeks of treatment with this agent. During this interval,
`the patients who were using tricyclics were withdrawn from these drugs. Overall stimulant
`consumption in this patient group was also reduced. Twelve of the patients were again
`studied in the laboratory after 6 months,
`
`Clinical findings
`All studies agree that over the first few nights of treatment, GHB virtually eliminates
`nightmares and hallucinations. Sleep paralysis may be intensified on the first or second
`night but then disappears also. Dreaming persists, but loses its frightening quality. In the
`morning, most patients report having slept sounder and feeling more rested. Daytime attacks
`of sleep and cataplexy are slower to disappear, but nevertheless are significantly reduced
`in number after 1 week of treatment (10). Residual attacks of cataplexy are milder, shorter
`in duration, and easier to control, tending to occur late in the day when the patient is tired.
`The most refractory symptoms are daytime drowsiness and the need for sleep. Even with
`stimulants, these symptoms are not fully alleviated in some cases. Nevertheless, it is im(cid:173)
`portant to emphasize that in spite of the sleep latency data to be reviewed below, GHB
`does effect an improvement in daytime alertness. For example, Montplaisir and Barbe(cid:173)
`zieux (11) treated five nonapneic patients who had excessive daytime drowsiness with
`GHB .. None of their patients were given stimulants. Within weeks, all patients reported
`feeling more alert during the day. However, GHB must be used in repeated doses during
`the night, and symptoms usually recur the following day when treatment is stopped or
`when only one dose is used.
`
`Sleep. Vol. 9, No. J, 1986
`
`

`
`NARCOLEPSY AND GHB
`
`287
`
`Sleep laboratory findings
`The most constant effect observed in patients after GHB is administered is an increase
`in SWS and a decrease in the REM sleep latency. GHB characteristically induces a sleep(cid:173)
`onset REM period followed by a period of SWS, after which the patient often spontaneously
`awakes. This sequence takes -2 to 3 h. Decreased REM latency is a persistent effect of
`the drug and may be observed even after 6 months of treatment (10). Total nocturnal REM
`sleep duration may be increased but usually is unchanged. The number of REM sleep
`periods is unchanged, but the REM density is decreased. Total sleep time at night may be
`increased or unchanged. Patients develop no tolerance to the hypnotic actions of GHB over
`a 6-month period. GHB improves sleep continuity at night and significantly decreases the
`duration of REM sleep and SWS during the day. Daytime sleep periods > 45 min in
`duration decrease in frequency, but drowsy stage 1 sleep may even be increased in duration
`(9). No change occurs in the average sleep latency during the day after 4 weeks or even
`after 6 months on GHB (10). REM latency during the day, however, is significantly de(cid:173)
`creased. It should be noted that the sleep and REM latency tests during the day were
`conducted with some patients taking stimulants. Although the average sleep latency did
`not change, patients required less stimulant medication when taking GHB.
`
`LONG~TERM USE OF GHB
`
`The cases of 48 patients who have been taking GHB for 6 months to 9 years are now
`being followed in Toronto. The cases of other patients who were started on this treatment
`in Toronto are being followed by their physicians in other parts of Canada and the United
`States. These patients, 21 men and 27 women, range in age from 17 to 71 years. All combine
`stimulants during the day with GHB at night. The commonest schedule is GHB about 30
`mg/kg or 2.25-3 g twice each night and a single long-acting 15 mg dexedrine dospan in
`the morning. Patients are encouraged to nap late in the afternoon when the dexedrine is
`wearing off to produce a more alert evening, but many do not do so regularly. The use of
`GHB in this patient series ranges from 4.5 to 9 g/night. The use of dexedrine ranges from
`JO to 30 mg daily. Some patients prefer methylphenidate, but none uses more than 30 mg
`daily. As part of their treatment regimen, patients are advised to refrain from heavy meals
`and excessive quantities of carbohydrate-rich foods. Once the treatment regimen has been
`adjusted to achieve optimal levels of sleep at night and wakefulness during the day, little
`change is required. The development of drug tolerance has not been observed.
`Thirty-six patients, 13 men and 23 women, are virtually symptom-free. They are able
`to function satisfactorily at work or school and are not embarrassed by their illness. The
`remainder are symptomatic to varying degrees. As in the earlier studies, daytime drowsiness
`and the need for sleep are the most common residual symptoms. Cataplexy is rarely a
`serious concern. Poor nocturnal sleep of patients who are taking GHB appears to be one
`factor that predicts a poor response. Patients who relate the development of their illness to
`irregular work hours or to a head injury also tend to respond less well. This largely accounts
`for the disparity in the response observed between the men and women. Nevertheless, it
`should be noted that some patients who have had these predisposing factors have responded
`welL Cataplexy can be difficult to control in patients who have been withdrawn from high
`doses of tricyclics. At present only 1 patient remains on 10 mg chlorimipramine daily after
`nearly 3 years on GHB. Symptoms can intensify in all patients, even. those who have
`responded well, during periods of stress. Similarly, drowsiness can prevail during long
`periods of monotonous activity.
`
`Slup, Vol. 9, No. 1, /986
`
`

`
`288
`
`M. MAMELAK ET AL.
`
`Few adverse effects have been observed. All patients have been followed with serial
`liver, renal and blood studies, periodic chest x-rays, and electrocardiograms; no abnor(cid:173)
`malities have been noted. On the first few nights of treatment with GHB, two patients had
`enuresis. Scharf et al. ( 10) reported one such incident in their series. Patients who resist
`the sleep-inducing properties of the drug may become confused and emotionally labile.
`When treatment with GHB is first started, patients may experience sleep paralysis or
`discover that they are cataplectic if they try to walk after taking this agent. Price et aL
`(12) reported cataplexy and confusion in narcoleptic patients given GHB intravenously
`during the day. Nevertheless, three patients in the Toronto series have reported intermittent
`sleepwalking while on GHB. This is a more persistent adverse effect which may appear
`after a period of treatment with GHB. It has been satisfactorily controlled with 5-10 mg
`of methtrimeprazine at bedtime. Weight loss has been an unexpected benefit for a number
`of obese patients. GHB is also being used without adverse effects in one narcoleptic patient
`with central sleep apnea (13).
`Five patients have discontinued treatment with GHB. One did so because he found the
`treatment regimen inconvenient. A young woman planned to become pregnant and feared
`the potential teratogenic effects of GHB. The other 3 patients did not feel that they were
`being helped.
`Mechanism of action
`The effectiveness of GHB can be attributed to its capacity to induce the major symptoms
`of narcolepsy, that is, sleep and the motor inhibitory phenomena associated with REM
`sleep, and to contain them at night. Thus, the reliable induction of REM and NREM sleep
`at night, coupled with the prevention of daytime sleep by stimulants, gradually recruits and
`consolidates sleep at night, and eliminates it during the day. Sleep paralysis and nocturnal
`hallucinations disappear with this recruitment. The treatment, however, is palliative. Short(cid:173)
`latency nocturnal REM sleep periods and daytime REM sleep betray the persistent dis(cid:173)
`sociation of sleep.
`Could an abnormality in endogenous GHB metabolism be a factor in the development
`of narcolepsy? Early REM sleep periods occur with some consistency in two conditions,
`narcolepsy and depression. In both of these conditions, the early REM sleep periods are
`thought to reflect a metabolic shift towards increased cholinergic and decreased catecho(cid:173)
`laminergic neurotransmission (14). In narcoleptic dogs, for example, increased numbers of
`muscarinic cholinergic receptors have been described in the pontine region (15). Dopamine
`utilization is decreased in the brains of these animals, aUhough dopamine levels are increased
`(16). GHB produces a comparable metabolic shift. It increases brain dopamine levels but
`inhibits dopamine release (17), and it enhances acetylcholine release, at least in the striatal
`region where this has been measured (18). GHB can induce early REM sleep periods in
`cats (4),. but it does not do so reliably in humans except in depression and narcolepsy. The
`nervous system in these conditions appears particularly sensitive to the actions of GHB
`which, in such states, can provoke not only sleep onset REM periods, but dissociated
`episodes of motor inhibition in the form of sleep paralysis and cataplexy. GHB receptors
`are found in highest concentrations in nerve ending fractions rich in acetylcholine (2), a
`neurotransmitter closely implicated in the induction of REM sleep (14). It wculd be in(cid:173)
`teresting to know if there are any changes in the sensitivity of these receptors in narcolepsy.
`
`REFERENCES
`1. Snead OC, Morley BJ. Ontogeny of gammahydroxybutyric acid. I. Regional concentration in developing
`rat, monkey and human brain. Brain Res 1981;227:579-89.
`
`Sleep. Vol. 9, No. 1. 1986
`
`

`
`NARCOLEPSY AND GHB
`
`289
`
`2. Maitre M, Rumigny JF, Cash C, Mandel P. Subcellular distribution of gammahydroxybutyrate binding
`sites in rat brain. Principal localization in the synaptosomal fraction. Biochem Biophys Res Commun
`1983;110:262-5.
`3. Yamada Y, Yamamoto J, Fujiki A, Hishikawa Y, Kanedo Z. Effect of butyrolactone and gammahydroxy(cid:173)
`butyrate on the EEG and sleep cycle in man. Electroenceph Clin Neurophysiol 1967;22:558--62.
`4. Matsuzaki M, Takagi H, Tokizane T. Paradoxical phase of sleep: its artificial induction in the cat by
`sodium butyrate. Science 1964;146:1328--9.
`S. Vickers MD. Gammahydroxybutyric acid. Intern Aneszhesiol Clin 1969:75-89.
`6. Mamelak M, Escriu JM, Stokan 0. The effects of gammahydroxybutyrate on sleep. Biol Psychiatry
`1977;12:273-88.
`7. Broughton R, Mamelak M. Gammahydroxybutyrate in the treatment of compound narcolepsy: a preliminary
`report. In: Guilleminault A, Dement We, Passouant P ( eds ). Narcolepsy. Spectrum, New York, 1976:659--
`67.
`8. Broughton R, Mamelak M. The treatment of narcolep5y cataplexy with nocturnal gammahydroxybutyrate.
`Can J Neurol Sci 1979;6:1-6.
`9. Broughton R, Mamelak M. Effects of nocturnal gammahydroxybutyrate on sleep waking patterns in
`narcolepsy-cataplexy. Can J Neurol Sci 1980;7:23-3 L
`10. Scharf MB, Brown D, Woods M, Brown L, Hirschowitz J. The effects and effectiveness of gammahy(cid:173)
`droxybutyrate in patients with narcolepsy. J Clin Psychiatry 1985;46:222-5.
`11. Montplaisir J, Barbezieux M. Gammahydroxybutyrate de sodium (GHB) dans le traitment de l'hypersomnie
`essentielle. Can J Psychiatry 26: I 62-6.
`12. Price PA, Schacter M, Smith SJ, Baxter eH, Parkes JD. Gammahydroxybutyratc in narcolepsy. Ann
`Neurol 1981;9:198.
`13. Mamelak M, Webster P. Treatment of narcolepsy and sleep apnea with gamma-hydroxybutyrate: a clinical
`and polysomnographic case study. Sleep 1981;4:105-11.
`14. Gillin JC, Sitaram N. Rapid eye movement sleep: cholinergic mechanisms. Psycho! Med 1984;14:501-6.
`15. Boehme RE, Baker TL, Mefford IM, Barchas ID, Dement We, Ciarenella RD. Narcolepsy: cholinergic
`receptor changes in an animal model. Life Sci 1984;34: 1825-8.
`16. Mefford IN, Baker TL, BoehmeR, et al. Narcolepsy: biogenic amine deficits in an animal model. Science
`1 983;220:629--32.
`17. Roth RH, Walrers JR, Aghajanlan GK. Effect of impulse flow on the release and synthesis of dopamine
`in the rat striatum. In: Usdin E, Snyder SH (ed~). Frontiers in catecholamine research. Pergamon Press,
`Oxford, 1973:567-74.
`18. Stadler H, Lloyd K, Bartholine E. Dopaminergic inhibition of striatal cholinergic neurons: synergistic
`blocking action of gamma-butyrolactonc and neuroleptic drugs. Naunyn Schmiedebergs Arch Pharmacol
`1974;283:129-34.
`
`Sleep, Vol. 9, No. J, 1986