`
`"""'**fi"'~ - “V v
`
`-- =A
`
`‘HEALTH SCIENCFS LIBRARY
`
`I
`
`/
`
`University of Wisconsin
`
`AUG 0 34-989 A
`
`1305 Linden 0}.
`Madison, Wis. 53766-
`
`A PUBLICATION OF THE SOCIETY OF BIOLOGICAL. PSYCHIATRY
`VOLUME 26, NUMBER 4
`AUGUST 1989‘
`Wockhardt 1032
`
`BIPCBF _26(4)1989 0
`ISSN 0006-3223
`
`’
`
`<
`
`0
`
`I ‘
`
`/"
`
`‘
`
`'
`
`,_.,
`
`Wockhardt 1032
`
`
`
`l\‘1'l\
`
`2 1l
`
`t1
`\|
`r
`u
`
`l
`
`4
`
`Biological Psychiatry is cited in Beck Medical Information, Biological Abstracts, Chemical Abstracts, urre/nt Contents, Excerpta Medica,
`Index Medicus, Mental Health Abstracts, Psychological Abstracts, Referativnyi Zhurnal, Science Citation Index, and Selected List of Tables
`of Contents of Psychiatric Periodicals.
`.
`.
`_
`'
`-
`
`Biological Psychiatry is published in two vdlumes mailed monthly and biweekly in January, February, March, and April
`by Elsevier Science Publishing Co, Inc, 655 Avenue of the Americas, New York, NY l0010. Institutional subscriptions:
`$448.00; individual subscn'f5tions: $218.00 and $135.00 for renewals; special in-training rate to residents, fellows, interns,
`and students: $64.00; society subscriptions: $25.00 ., Subscribers outside of the United States add $56.00 for surface postage
`and handling. Subscriptions outside of the United States a'.re~sent by air.
`
`Claims for missing issues ‘can be honored only up to three months from ate of issue. Duplicate copies will not be sent
`to replace ones undelivered through failure to notify'Elsevier of change of address.
`'
`Single copy and back volume information available upon request. Please direct orders for the journal, changes of address.
`and claims for missing issues to: Journals Fulfillment Department, Elsevier Science Publishing Co., lnc., 655 Avenue of
`the Americas, New York, NY 10010.
`'
`
`Please direct inquiries regarding the placement of advertising in this journal to: Synergistic Media Sales, 50 East 42nd
`Street, Suite 504, New York, NY 10017. (212) 682-2200.
`
`No responsibility is assumed by the Publisher or by the Society of Biological Psychiatry for any injury and/or damage 10
`persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any m€'Eh0d5=
`products,
`instructions or ideas contained in the material herein. No suggested test or procedure should be carried out
`unless, in the reader’s judgment, its risk is justified. Because of rapid advances in the medical sciences, we recommf‘-111.,
`that the independent verification of diagnoses and drug dosages should be made. Discussions, views and rec0mmendati0nS ,.._-,,.
`as to medical procedures, choice of drugs and drug dosages are the responsibility of the authors.
`
`I110‘
`© 1989 Society of Biological Psychiatry. This journal has been registered with the Copyright Clearance Center,
`Consent is given for copying of articles for personal or internal use, or for the personal or internal use of specific clis’-I115‘
`This consent does not extend to other kinds of copying, such as for general distribution, resale, advertising and promotional
`purposes, or for creating new collective works. For such purposes, consent will. be given on the condition that the C0P1fI—
`pay through the Center the per-copy fee stated in the code on the first page of each article for copying beyond that pefiilimd
`by the U.S. Copyright Law. If no code appears on an article, the author has not given broad consent to copy and perrnissiofl _V_V
`to copy must be obtained directly from the author.
`
`Second—class postage paid at New York, ;NY and additional mailing offices. Postmaster: send address changes to Biolvglcal
`Psychiatry, Elsevier Science Publishing Co., Inc., 655 Avenue of the Americas, New York, NY 10010.
`
`Manufactured in the U.S.A.
`
`Wockhardt 1032
`
`
`
`Wockhardt 1032
`
`
`
`This material may be protected by Copyright law (Title 17 U.S. Code)
`
`W”
`
`7
`
`.
`
`« 1’
`
`_
`
`331
`
`BIOL PSYCHIATRY
`1989;26i331—343
`4
`
`
`
`_AEfficacy of‘Gamma-Hydroxybut-yrate versus
`lcfilacebo in Treating Narcolepsy—CatapleXy:
`* Double-Blind Subjective Measures
`,
`V
`,
`
`1gLaiWrenceiScrima, Paul
`Charles Hiller
`
`Hartman, Frank» H. Johnson, Jr., and
`
`S
`
`The efficacy of gamma-hydroxybutyratel (GHB) versus placebo for treating narcolepsy
`' was evaluated in 20 patients with narcolepsy, 10 men and 10 women, using a double-
`blind counterbalanced crossover design. Each patient completed a daily sleep—wak_e log
`and questionnaire during a 14-day baseline, a 29-day placebo ‘period, a 29-day GHB
`period ,(50 mg GHB/kg/night. given.25 mg/kg h.s. and 25 mg/kg 3 hrxlater), and a 6-day
`washout period after each treatment. Cataplexy frequency was significantly lower during
`GHB treatment than during placebo treatment (p = 0.022). Compared to baseline values,
`the number ofcataplexy attacks per day declined by 52 %. and 69% during GHB treatment
`weeks I and 4, respectively. The number of subjective arousals from sleep was less with
`GHB than with placebo (p = 0.035), and the number ofsleep attacks was not significantly
`diflerent during GHB versus placebo treatment. GHB did not have a significant eflect
`on subjective estimates of sleep onset latency, total sleep time, Stanford Sleepiness Scale
`ratings at morning ,wake—up, methylphenidate usage, or the number of naps per day. The
`results indicate that'GHB is efiicacious for reducing the frequency of cjataplexy attacks
`and subjective nocturnal arousals in patients with narcolepsy within the first 4 weeks of
`treatment.
`'
`V
`
`.;..._.._£2.
`
`._.e~..._.v
`
`.,..t.t..(..._-..._y~.
`
`._....__../\_...i..
`
`in
`i
`
`7'
`
`i
`“
`ii
`“
`
`_;‘~
`15,1‘,
`‘M
`
`
`
`1,
`ii i
`1
`9:
`i
`
` -
`
`
`9 Introduction
`
`Narcolepsy isa chronic, incurable disorder characterized by intermittent excessive daytime
`sleepiness and abnormal rapid eye moven1ent;__(REM)_ sleep manifestations, such as s1eep-
`V onset REM periods, cataplexy, sleep paralysis, and"/or hypnagogic hallucinations (Associ—
`ation of Sleep Disorders Centers 1979). Cataplexy is a sudden loss of muscle tone that occurs
`primarily during emotional arousal; sleep paralysis is an inability to move upon first lying
`down or upon waking; and hypnagogic hallucinations are dream-like hallucinations that
`occur at sleep onset. Most patients with narcolepsy also have disrupted nocturnal sleep
`(Montplaisir 1976). Narcolepsy is generally treated with a central nervous system stimulant
`
`
`
`From the Sleep Disorders Center, Pulmonary Division, Department of Medicine, University of Arkansas for Medical Sciences,
`Little Rock, AR.
`_
`,
`Supported in part by Orphan Products Grant FD—R-000115 from the Food and Drug Administration.
`Address reprint requests to Dr. L. Scrima, Sleep Disorders Cener, Slot 594, University of Arkansas for Medical Sciences,
`4301 West Markham-Street, Little Rock, AR 72205.
`‘
`‘
`~
`.
`»
`-
`Received July 29, 1988; revised December 14, 1988.
`
`© 1989 Society ofwgeggféfatchiqiw
`
`« 0006.-3223/89/$03 .50
`
`Wockhardt 1032
`
`
`
`I 3 1 i
`
`i.
`1
`
`
`
`n,_M_,4,,4_
`
`.1P
`
`(e. g. , amphetamine, methylphenidate, or pemoline) to reduce excessive daytime sleepiness
`(EDS) (Parkes 1976) and an antidepressant (e.g., imipramine or protriptyline) to contro1
`cataplexy and other REM sleep-related symptoms (Takahashi 19]6). Stimulants, however,
`do not fully control EDS in most narcolepsy patients (Parkes 1976) and can have undesirable
`side effects on the cardiovascular, gastrointestinal, and central nervous systems (Gilman et .
`al. ‘1985). Treatment of cataplexy with antidepressants has been reported to be successful in
`most cases (Takahashi 1976; Billiard et al. 1983), but these drugs have adverse effects,
`including (1) prolonged cardiac conduction times that may promote dangerous ventricular
`arrhythmias; (2) postural hypotension; (3) anticholinergic effects, such asblurred vision,
`dry mouth, and impotence (Gilman et al. 1985); (4) suppression of REM sleep (Zung 1969;
`Cadilhac 1976); and (5) increased nocturnal myoclonus (Guilleminault et al. 1976). The
`anticholinergic side effects often result in patient self-withdrawal from antidepressants, which
`- is usually followed by an increase in the frequency and severity of cataplexy events (Scrima
`1981; Scharf and Fletcher 1988). Both stimulants and anticataplexy drugs may become less
`effective as tolerance increases (Parkes 1976; Broughton and Mamelak 1979).
`.
`Gamma—hydroxybutyrate (GHB) is a four—carbon fatty acid that occurs naturally in the
`mammalian central nervous system (Muyard and Laborit 1977) and has been termed a
`“putative neurotransmitter” (Mandel et al. 1987). GHB_ wasreported to induce anesthesia
`at 60—70 mg/kg (Vickers 1969), but the report does not make it clear whether the doses
`were given orally or intravenously. Lower oral doses of GHB were reported to induce
`sleep in psychiatric patients (Mamelak etal. 1977), but the minimum GHB dose that
`will induce sleep ha“s\not been systematically determined; Unlike other hypnotics, GHB
`given orally‘ induces and maintains sleep withoutsuppressing REM or delta stages of
`sleep (Mamelak et_ al. 1977). It was first reported in 1976 (Broughton and Mamelak) that
`GHB , given orally h. s. and two to three additional times during the sleep period, improved
`nighttime sleep and reduced cataplexy and sleep attacks in patients with narcolepsy.
`Subsequent studies confirmed that most narcolepsy patients had moderate to large re-
`ductions in cataplexy frequency and daytime sleepiness, as well as reduced sleep dis—
`ruption, hypnagogic hallucinations, and sleep paralysis after taking GHB in divided dose,
`i.e. , a dose h.s. and one to two additional times during the night (Broughton and Mamelak
`1979, 1980; Scharf et al. 1985). Polysomnographic recordings indicated that narcolepsy
`patients taking a divided dose of GHB had increased sleep continuity, decreased REM
`fragmentation, and increased amounts of delta sleep (Broughton and Mamelak 1980;
`Scharf et al; 1985). However, 1 month of oral administration of a single h.s. dose of
`GHB improved daytime sleepiness in only 39% of patients with narcolepsy,
`though '
`cataplexy frequency was reduced in 83% of the patients (Montplaisir and Godbout
`“W
`Tolerance to GHB has not been found to develop, even after daily use by patients With
`-
`narcolepsy for as long as 9 years (Mamelak et al."l986). Adverse side _effects._have~b¢6I1"
`infrequent, mild, and have occurred mainly during the first few days of treatment (Brough'
`ton and Mamelak 1979; Scharf et al. 1985; Mamelak et al. 1986).
`This report describes the results of the first double—blind study of the effects 0f:.GH—B_g.,_ _
`. on subjective symptoms of narcolepsy as compared to those of a placebo.
`
`Methods
`
`Subjects
`Ten women and 10 men with narcolepsy, diagnosed at the accredited Sleep Disufders
`Center (SDC) of the University of Arkansas for Medical Sciences (UAMS), partic1Pated
`
`Wockhardt 1032
`
`|
`
`
`
`
`
`Wockhardt 1032
`
`
`
`; Efficacy of Gamma-Hydroxybutyrate .
`
`113gg:2%s%c1:n31A43rRY
`
`333
`
`3
`
`’}f:_}»
`
`
`
`
`
`
`
`J'~‘~_.;._...._...r-__.._._.~‘~‘_...:........._.r./\__..__....___r~‘,__..-...._4;
`
`in the study; All patients were interviewed by an accredited clinical polysomnographer, '
`were given a physical examination by a physician, and had sleep disorders diagnostic
`tests’ (Guilleminault 1982). The diagnostic tests included an overnight polysomnogram
`(PSG) and an evaluation of their daytime sleepiness with the multiple sleep latency test
`‘ (MSLT). The criteria for inclusion in this study were: (1) a history of excessive daytime
`sleepiness and cataplexy, (2) >2 REM onsets on the MSLT, (3) a sleepiness index of
`275 on the MSLT, (4) at least 10 cataplexy attacks subjectively reported on a daily log
`during a 2-week baseline period, and (5) age between 16 and 65 years. Patients were
`excluded if they had other major health problems; were fertile women who were not
`practicing birth control; were nursing mothers; or had previously taken GHB or had other
`sleep disorders, with the exception of those commonly associated with narcolepsy, such
`as sleep paralysis, mild to moderate sleep apnea (arterial oxygen saturation 280%), and
`nocturnal myoclonus.
`_
`, The age mean i standard error was 45.9’ it 4.6 years (range l6—64) for the women
`and 49.1 i 4.0 (21-64) for the men. Weight (kg) mean : SE was 85.1>: 5.2 (range
`57-113) for women and 80.4 : 3.6 (54—90) for (men. Body mass index was 31.8 ' :
`' 2.5 (17.6—45.4) for Women and 26.2 i 0.9 (20.3—29.l) for men. Prior to the study, 7
`patients were on stimulants alone (methylphenidate, pernoline, or dextroamphetamine),
`11 were on a combination of stimulants and anticataplexy medications (imiprarnine or
`protriptyline),‘and 2 patients were not taking stimulants or anticataplexy medications.
`Patients who were taking stimulants" other than methylphenidate switched to methyl-
`phenidate (S30 mg/day) for the duration of the study. Seven patients were withdrawn
`from anticataplexy medications at least 2 weeks before the baseline period, and the
`remaining 4 patients on anticataplexy medications were withdrawn from imipramine 6,
`6, 5 , and 3 days "prior to baseline, respectively. Only the patient who stopped imiprarnine
`5 days prior to. baseline appeared to have elevated amounts of cataplexy events during
`the first 2 days of the baseline (18 and 12events, respectively; mean and SD for the rest
`of the baseline period: 5.7 i 3.5 events); these 2idays were excluded from the analysis.
`One patient continued taking propranolol to control hypertension throughout the study at
`a dose (40 mg/day) that was half the lowest dose of propranolol (80—480 ‘mg/day) reported
`to reduce narcolepsy symptoms for some ‘patients (Kales et al. 1979; Meier—Ewert et al.
`1985). This patient was included in the analysis of results, but analysis of the data with
`this patient excluded yielded the same pattern of statistically significant results.
`
`Procedures
`
`A double—blind, crossover design, ‘with order of treatment counterbalanced and randomly
`assigned, was utilized. Thus, each subject provided data for all phases of the study:
`baseline (14 days‘), first treatment (29 days), firstlwashout (6 days), second treatment (29
`days), and second washout~(6 days). Order of treatment was randomly assigned by the
`UAMS pharmacy, so that half of the men and half of the women received GHB in the
`first treatment period and placebo in the second, whereas the remaining subjects received
`placebo first and‘GHB second. All SDC staff were blind to the order of treatment for
`subjects. During GHB treatment, subjects received 58 bottles prepared by the pharmacy,
`each containing 25 mg GHB/kg body weight, mixed with distilled water and syrup of
`orange. During the placebo treatment, subjects received 58 identical bottles with an
`equivalent amount of fluid, consisting of syrup of orange in distilled water. During each
`treatment period, subjects were instructed orally and in writing to (1) ‘refrigerate, but not
`
`_
`
`‘
`
`'
`
`.
`
`1.
`
`11
`
`1411
`
`if
`if?
`
`ll;
`
`if
`53}?
`
`ii
`
`1
`
`é
`
`-
`
`1
`
`
`
`Wockhardt 1032
`
`Wockhardt 1032
`
`
`
` 1 5
`
`.
`
`_
`
`freeze, the bottles, (2) put two bottles by their bedside when they went to bed, (3) ingest
`one entire bottle immediately before going to bed, and (4) ingest the second entire bome
`3 hr later. Subjects were told to set an alarm to insure that they woke up 3 hr after bedtime
`to take their second placebo or GHB dose. Subjects were also instructed not to use
`alcohol, sleeping pills or other central nervous system depressants during the experiment
`and‘ to avoid drinking caffeinated beverages late in the day. Subjects were permitted to
`. use methylphenidate (up to 30 mg/day) to counter excessive daytime sleepiness, but were
`instructed not to take methylphenidate after 6:00 PM.
`Subjects were required to complete a sleep log and questionnaire for each day of the
`experiment. The sleep log consisted of»a grid on which subjects noted the time periods
`when they were awake, lying down, sleeping, and when they took methylphenidate. On
`the questionnaire, subjects recorded daily_subjective reports of (1) sleep onset latency for
`nighttime sleep, (2) number of arousals from sleep, (3) total sleep time, excluding naps,
`(4) Stanford Sleepiness Scale rating upon morning awakening, (5) number of “sleep
`attacks,” i.e., periods when they experienced an irresistible urge to sleep, (6) number of-
`cataplexy events, (7) number of naps, and (8) amount of methylphenidate taken. Patients
`reported icomplying with instructions for taking GHB. and placebo, .and most patients
`completed their daily sleep logs and questionnaires thoroughly.
`Each subject had an 8.0,-hr PSG performed onthe last night of baseline and the first
`and last nights of both treatment periods, with an MSLT performed the day following
`each PSG. (5\PSGs and 5 MSLTs); these results will be reported following statistical
`analysis of the data.
`’
`’
`,
`Patients were encouraged to call the SDC staff to report adverse reactions during the
`experiment. Patients were asked about the possible occurrence of adverse reactions in
`interviews by phone 2 weeks into each treatment period and by in-person interviews at
`the end of each treatment. 5
`
`_
`
`Data Analysis A
`
`A mixed—design Analysis of Variance (ANOVA) was used with two between-subject
`factors—gender and order of treatment—and two Within-.S11bjeCt‘faCt0I'S—-*tI'Catl'I1CIlt (GHB
`versus placebo) and week of treatment (first versus fourth week). For each daily dependent
`measure (number of cataplexy events, sleep attacks, etc.), means were computed for the
`baseline andeach week of both treatments. The units of analysis for "each subject were
`the change. scores, i.e., the differences between the baseline mean and the means f0T
`weeks 1 and 4 of GHB treatment and weeks 1 and 4 of placebo treatment. Placebo effects
`were evaluated by contrasts of the baseline mean versus the means on placebo weeks 1
`and 4. The data were also analyzed with a modification of the nonparametric'.Wilcox0Il
`test (Koch .1972), as the superiority of parametric versus nonparametric statistics has 110‘
`been demonstrated for this design. Results of the nonparametric tests will be reported
`only when they are not in agreement with the parametric statistics.
`— 1
`‘
`Washout effects, i.e., carryover or rebound effects during 5 days following C6SS3Fl°fi
`of treatment, were also tested with a mixed—design ANOVA. Separate mixed-deslgn
`ANOVAs were used to, contrast each day of the GHB washout with the corresponding
`day of placebo washout, and to contrasteach day of both washout periods with the
`baseline period. Each washout period consisted of at least 6 nights and days, but 0111)’
`the data from the first 5 days were analyzed, as subjects ‘spent the sixth day in transit t0
`the sleep laboratory and being prepared for their fourth overnight PSG.
`
`Wockhardt 1032
`
`
`
`
`
`Wockhardt 1032
`
`
`
`
`Efficacy of Gamma-Hydroxybutyrate
`
`l139Ig9l;2Ig:S3g€g[A43TRY
`'
`
`335
`
`_
`
`ii
`l
`
`
`
`-(>0I'Jl11'U(D—lZl'T1<|'T1-<><l'|'|l"'U>-l>O
`
`
`
`
`
`-fir PLACEBO TREATMENT ‘
`
`-B- GHB TREATMENT
`
`1
`
`M
`
`0 '
`
`_.i_
`BASELINE
`
`
`
`_L
`l_
`_L
`31
`2_ A
`'
`1
`TREATMENT WEE_K_
`
`i_
`4
`
`\
`
`"GHB vs Placebo: p <.
`
`.01
`
`Figure 1. Cataplexy events per day during baseline, placebo, and GHB treatment. For each patient, " ‘<4
`the mean number of cataplexy events per day was calculated for the baseline and for each week
`of both treatments. The ‘figure shows the means and standard errors of patients’ baseline and weekly
`means for cataplexy.
`0
`
`Cataplexy Frequency A
`Figure 1 shows the means and standard errors for cataplexy events during baseline,
`placebo, and GHB treatment periods. GHB treatment caused a significantly" greater decline
`in cataplexy frequency .than did placebo treatment (F = 6.58; df '=. 1,15; p = 0.022).
`Compared to cataplexy events per day during baseline (mean : standard error 3.1 :
`
`,
`
`0.5), cataplexy events per day declined during GHB weeks 1!(l.5 : 0.2) and 4'(0.9 i
`
`0.2) by 52% and 69%, respectively, and declined during placebo weeks 1 (1.7 i 0.3)
`and 4 (2.0 i 0.3) by 43% and 33%, respectively. Placebo effects versus baseline were
`significant during placebo week 1 (F =0 10.36; df = 1,15; p = 0.006) and were nearly
`significant for placebo week 4 (F = 4.01; df = 1,15; p = 0.064).. Sex of the patients
`did not have a significant interaction with the effect of GHB on cataplexy frequency.
`Of 19 patients with log entries for cataplexy during baseline, placebo, and GHB
`treatments, 16 (84%) had fewer cataplexy events per day during the fourth week of GHB
`versus the fourth week of placebo. Nine (47%) had at least 1 less cataplexy attack per
`day, and 7 (37%) had 0. l—0.99 fewer cataplexy attacks per day with GHB. Four patients
`(21%) had no cataplexy attacks during the fourth week of GHB treatment, whereas 1
`patient (5%) had no cataplexy attacks during the fourth week of placebo treatment. Seven
`patients (37%) continued to have at least 1 cataplexy attack per day during the fourth
`week of GHB treatment. The daily logs of one patient (“p”) were collected but subse-
`
`quently lost, and therefore, thispatient could not be includediin the statistics. Like all
`
`3,
`
`:3
`
`‘
`
`Wockhardt 1032 I
`
`
`
`....r....._.,~
`
`1-»_......
`
`j
`
`i i 1
`
`l
`1‘
`
`1
`
`.
`
`Results
`
`V
`
`,
`
`’
`
`Wockhardt 1032
`
`€
`
`
`
`
`<>Ummnm4zm<m<xmrv>4>O
`
`'%'P|acebo Treatment
`
`'E'GHB Treatment
`
`I
`
`(.0
`
`l
`
`._L J
`'N
`
`I
`‘I’
`BASELINETREATMENT
`.
`WEEK4
`
`.
`
`I
`1
`
`I’ 0
`‘2
`
`T
`4
`
`'
`
`»
`
`0'
`
`A
`
`,
`
`;
`1‘;
`if
`
`i
`
`1
`
`;
`
`i
`
`‘av
`
`
`
`r‘
`3 .
`V
`.
`‘
`WASHOUTA DAY -
`
`‘-
`* GHB vs Placebo: p < 0.02‘
`“GHB vs Placebo: p < 0.01
`Figure 2. Cataplexy events per day during’ baseline and washout periods. Means and standard
`errors of the patients’ baseline means and cataplexy events per day during washout periods are
`shown.
`'
`
`/
`
`£6
`99
`' participants, patient p was interviewed at the cndof the patient’s participation in the
`study, with experimenters and patient still blind to treatment order. The patient indicated
`that cataplexyi frequency was 10/day during baseline, 2.5/day during treatment 1 (GHB),
`and 10/day during treatment‘ 2 (placebo).
`There was a significant interaction between the effects of treatment and the week of
`treatment on cataplexy frequency (F = 7.45; df = ‘1,15;p = 0.016). Therefore, cataplexy
`frequency was analyzed with week held constant and with treatment held constant, re-
`spectively. At week 1 of each treatment, the number of cataplexy attacks per day Was
`not significantly different with GHB than with placebo treatment. At week 4, cataplexy
`frequency was significantly less with GHB than with placebo (F = 10.50; df =_ 1,153
`p = 0.006) .' Post hoc analysis of the second and third weeks of treatment indicated that A
`cataplexy frequency was significantly less with GHB than placebo during the third treat-
`ment week (F = 10.02; df = 1,15; p = 0.006), but not during the second W691‘-***e
`Analyses with treatment held constant indicated that cataplexy attacks declined sigr1ifi-
`cantly from week 1 to week 4 of GHB treatment (F = 15.10; df = 1,15; p = 0.002)-
`g
`Post hoc trend analysis of all 4 weeks of GHB treatment indicated that there W3S;a,
`significant linear component to the decline in cataplexy frequency during GHB treatment W 7
`(F = 18.81; df =_ 1,15; p = 0.0006). There was noisignificant change in cataplexy
`frequency from week 1 to week 4 of placebo treatment.
`Figure 2 shows themeans and standard errors for cataplexy events during each day
`of the washout periods following placebo and GHB treatment, respectively. Fewer 03‘
`taplexy events tended to occur during GHB washout than placebo washout (p' = 0-091);
`Analysis of each day .of the washouts separately indicated that although the number OI
`
`-
`—
`
`1
`
`{
`‘
`is
`
`*‘
`g
`'
`
`,~
`
`Wockhardt 1032
`
`
`
`Wockhardt 1032
`
`
`
`i
`
`‘gfficacy of Gamma-Hydroxybutyrate
`
`BIOL PSYCHIATRY
`1989;2_6:331——343
`
`337
`
`
`
`
`
`<>Uwmvmxo>44>vmmrm
`
`-Av PLACEBO‘ TREATMENT
`
`-E-.-GHB TREATMENT
`
`1.
`
`3T
`
`2r
`
`1?‘
`
`BASELINE
`
`1
`
`.
`
`T2
`
`3
`
`.
`
`4
`
`TREATMENT WEEK
`
`_
`
`.
`
`i“i
`
`3
`
`errors shown were derived as in Figure 1.
`
`Figure3.Sleepattacksperdayduringbaseline,placebo,andGHBtreatniept.Meansandstandard
`
`cataplexy events was lower on each day of the GHB washoutversus the corresponding
`day of the placebo washout, this difference was significant on only the first day of the
`washout periods (p = 0.010). "Compared to the baseline period, cataplexy frequency was
`significantly lower during GHB washout days 1-5 (p < 0.05 for‘ each comparison) and .
`during placebo washout days 1, 2, and 4 (p < 0.05 for each comparison), and tended
`to be lower during placebo washout days 3 and 5 (0.05 < p < 0.10).. Although treatment
`effects extended into the washout week, the main treatment effect of
`versus placebo
`on cataplexy was not influenced by carryover effects, i.e. , the interaction between treat— .
`ment and order of treatment was not significant (p = 0679).
`
`V
`
`,
`
`Daytime Sleepiness
`Figure 3 shows the means and standard errors for sleep attacks during baseline, placebo,
`and GHB treatment periods. Compared to sleep attacks per day during baseline (mean
`: SE 3.3 i 04), sleep attacks «per day declined during GHB weeks 1 (1.8 i 0.2) and
`4 (2.1 t 0.3) by 47% and 38%, respectively, and declined during placebo weeks 1 (2.1
`i 0.3) and 4 (2.0 i 0.2) by 37% and 41%, respectively. There were no significant
`differences between GHB and placebo effects on sleep attacks. Contrasts between each
`condition and baseline indicated that the number of sleep attacks declined significantly
`during placebo week 1 (F = 23.37; df = 1,13; p = 0.0003), placebo week 4 (F =
`21.37; df = 1,l3;p = 0.0005), GHB week 1 (F = 24.93;,df = 1,l3;p = 0.0002),
`and GHB week 4 (F = 7.71; df = 1,13; p .=~0.0l6). The frequency of sleep attacks
`declined to <1/day in only 1 patient (6%) during placebo week 4 and in 3 patients (16%)
`during GHB week 4.
`,
`There was a statistical tendency for sleep attacks to occur less frequently during placebo
`’ ”washout than during GHB washout (p = 0.0996), but differences between GHB and
`
`Wockhardt 1032
`
`35
`
`
`,
`"ll
`
`3
`
`
`
`l ji
`
`I
`
`1
`;
`
`
`
`Wockhardt 1032
`
`
`
`
`
`.-Av-PLACEBO TREATMENT
`
`-E!-GHBTREATMENT
`
`4;’
`
`.
`
`3L
`
`So
`
`U f
`
`L
`-S
`
`2‘
`
`ER
`
`5‘
`:2 4
`T
`
`(I
`_ 4_ '
`J
`'1
`2
`3'
`TREATMENT‘ WE:EK
`Figure 4. Subjective arousals per night during the baseline, placebo, and GHB treatment. Means
`and standard errors shown were derived as in Figure 1. The overall GHB versus placebo_ treatment
`effect was significant (p = - 0.035). The treatrnent-by-week interaction was not significant, so GHB
`versus placebo comparisons were not made for each week separately.
`
`4
`
`8
`
`u
`4 -
`
`0;‘:
`
`1_
`BASELINE
`
`
`
`
`
`
`
`placebo washout conditions were slight, ranging from 0.0 to 0.4 sleep attacks per washout
`.day. Analysiszof each day separately indicated that no placebo washout day was signif-
`icantly different from the corresponding GHB washout day for sleep attacks; Compared
`to sleep attacks during the baseline period (mean : SE 3.3 i 0.4), the" number of sleep
`attacks was significantly lower (p < 0.05) during placebo washout days 3, 4, and 5 (2.0
`i 0.2, 1.8 : 0.3, and 2.0 1- 0.2, respectively) and during GHB washout days -2-5
`(2.3 : 0.4 and 2.51-0.3, respectively). The main comparison of GHB versus placebo
`effects on sleep’ attacks was not affected by carryovereffects, i.e., the interaction between
`treatment and order of treatment was not significant for sleep attacks (p = 0.599).
`Stanford Sleepiness Scale (SSS) ratings at morning wake-up time, methylphenidate
`usage, and the number of daytime naps were not significantly different during GHB versus
`placebo treatment or placebo versus baseline. SSS ratings (means : SE) at baselinewere 5
`3.0 i 0.3; at placebo weeks 1 and 4 were 2.7 i 0.2 and 2.7 i 0.2, respectively; afld '""fl’" ’
`at GHB weeks 1 and 4 were 2.7 : 0.2 and 2.5 i 0.2, respectively. The number of__5:,,
`mg methylphenidate tablets taken per day-(means : SE) at baseline was 2.7 i 0.6; at
`placebo weeks 1 and 4 was 3.18 1- 0.5 and 2.9 : 0.5, respectively, and at GHB weeks
`l and 4 was 3.0 i 0.6 and 2.8 : 0.6, respectively. The number of daytime naps (meaI1S..._ .
`: SE) at baseline. was 1.3 : 0.3; at placebo weeks 1 and 4, 1.2 1 0.2 and 1.1 : 0.2’;”’:*‘*'
`and at GHB weeks land 4, 1.1 1- 0.2 and 1.1 -J: 0.3.
`8
`"
`
`T
`
`_
`
`i\
`
`Subjective Noctiufnal Sleep
`
`Figure 4 shows the means and standard errors for subjective nocturnal arousals during
`baseline, placebo, and GHB treatment periods. Compared to subjective nocturnal arousals
`at baseline (3.6 . i 0.8), subjective arousals from sleep declined during GHB weeks 1-.__
`Wockhardt 1032
`
`Wockhardt 1032
`
`
`
`
`
`Efficacy of Gamma-Hydroxybutyrate
`
`BIOL PSYCHIATRY
`1'989;26:331—343
`
`.
`
`339
`
`(2.5 i 0.4) and 4 (2.4 i 0.3) by 32% and 33%, respectively, and increased during
`both placebo weeks 1 (3.8 :2 0.8) and 4 (3.8 t 1.1) by 4%. The difference between
`GHB versus placebo treatment effects on arousals from sleep wasjsignificant (F = 5.43;
`df = 1,14; p = 0.035). Neither sex nor order of treatment interacted significantly with
`the effect of treatment on subjective arousals from sleep. Of 18. patients with log entries
`for nocturnal arousals during baseline, placebo, and GHB treatments, 11 (61%) had fewer
`subjective arousals on GHB_ than on placebo. Placebo. effects were not significant for
`subjective arousals.
`'
`A
`1
`The difference between nocturnal arousals during placebo washout versus GI-[B wash-
`out was not significant overall or for any single washout day. There were no significant
`differences in noctumal arousals during any placebo or GHB washout day compared to
`baseline, but there was a trend (p = 0.091) for fewer nocturnal arousals to occur on
`GHB washout day 1 compared to baseline.
`‘
`Subjective sleep onset latency and total ‘sleep time were not significantly different for
`GHB versus placebo treatment or placebo versus baseline. Subjective sleep onset latency
`(means t SE) at baseline was 13.9 :' 4.8 min; at placebo weeks 1 and 4, 14.1 i 4.2
`min and 13.1 i 4.1 min, respectively; and at GHB weeks 1 and 4, 9.1 : 1.5 min and
`7.8 i 1.2 min. Total sleep time (means t SE) at baseline was 6.7 i 0.4 hr; at placebo
`weeks .1 and 4, 6.9 i 0.3 hr and 7.1 1 0.3 hr, respectively; and at GHB weeks 1 and
`\
`(4, 7.1‘ : 0.2hrand 7.1 : 0.2 hl‘.
`'
`
`Hypnagogic Hallucinations and Sleep Paralysis
`Ten narcolepsy patients (5 women and 5 men) ratedlthe severity of hypnagogic hallu-
`cinations and sleep paralysis (1-10 rating scale with 10 being most severe) at the end of
`eachltreatment period. Of seven patients who reported hypnagogic hallucinations during .
`placebo treatment (medianseverity rating 4.5, range 3-8), only one also reported hyp-
`nagogic hallucinations during9GHB treatment. This patient rated hypnagogic hallucina-
`tions to be slightlyless severe during GHB treatment than during placebo treatment. No
`other patients reported hypnagogic hallucinations during GHB treatment. The effect of _
`GHB versus placebo treatment on hypnagogic hallucinations was significant (sign test,
`1 p = 0.008).
`Only 3 of 10 patients reported having sleep paralysis during either treatment. Two
`patients had sleep paralysis during placebo but not GHB treatment, and the third had
`sleep paralysis duringGHB treatment only.
`_
`
`Side Eflects
`Physical symptoms and problems reported by patients during GHB treatment, but not
`during placebo treatment or baseline, were as follows (number of patients in parentheses):
`upset stomach (3); upset stomach with vomiting during the first week of GHB treatment
`(1); sluggishness and stiffness upon morning awakening (2); dizziness after the second
`‘GHB dose of the night (1); urinary urgency (1); occurrence of cataplexy during physical
`exertion, which had not occurred prior to GHB treatment (1); general weakness and
`fatigue (1); one episode of dizziness during the day (1). The total number of adverse‘
`reactions reported during GHB treatment was less than during placebo treatment.
`Two episodes that occurred in the sleep laboratory were noteworthy. One woman
`fainted for a few seconds before taking the GHB doseonthe evening of her finalnight
`of GHB treatment; she did not faint at any other time in the study. Secondly, a technician .
`
`Wockhardt 1032
`
`/,vvlt
`
`_
`
`._...__._a—~___..__.__._...._.A\__._.._....__..H..._._..s-_M_....
`-. __—_m_~__gg.____
`
`
`
`Wockhardt 1032
`
`
`
`mistakenly allowed one man to get out of bed and smoke a cigarette immediately following
`the second GHB dose on the first GHB treatment night. The patient then experienced
`dizziness and nausea followed by 2 cataplexy attacks. The rule that patients must not get
`up immediately__ after taking GHB or placebo was reinforced.
`No patient discontinued participation in the study because of side ‘effects. The blood
`test results during baseline and at the end of each treatment periodindicated that GHB
`_ did not cause hypokalemia or other marked changes in blood chemistry.
`
`~;
`
`Discussion
`
`The main result was that 50 mg GHB/kg body weight‘ (25 mg/kg h.s. ‘and 25 mg/kg 3 '
`hr later) caused a 52% reduction in cataplexy on the first week of tre