Comparative Hypnotic Effects of Flurazepam

11271-11749/84/0401-0002$0:!.00/11 .Journal of Clinical PsychopharmarnloKY Copyright c HIH4 hy Williams & Wilkins Co.

Vol. ·I, No. I /)rinl1'd in l:'..'i'.A.

Comparative Hypnotic Effects of Flurazepam, Triazolam, and Placebo: A Long-Term Simultaneous Nighttime and Daytime Study MERRILL M. MITLER, PHD, 1 WESLEY F. SEIDEL, BA,c .JOHANNA VAN DEN HOED, MD/ DAVID ,J. GREENBLATT, MD, 1 WILLIAM C. DEMENT, MD, PHff

AND

1 Sleep /Jisordl'rs Cl'nler, Scripps Clinic and Research Fuundatiun, /,a .Julia, Califurnia; "S!l'l'fJ lfrsrnrch Cl'nler, /Jl'partment uf l'sychiatry, Stanford l Tnit·ersity Schou/ uf Medicine, Stanfurd, Califurnia; 1/Jit·isiu11 uf Clinirnl l'harrnarnlug\', /Jl'partmcnts uf l'sychiatry and Medicine, Tufts-Nell' f,'ng/and Mcdiml Cl'nler, Hustun, Mas.wtchusc/ls

We studied sleep and daytime function in insomniac patients who took either flurazepam, 30 mg, triazolam, 0.5 mg, or placebo 30 minutes before bedtime. Subjects were 21 patients with either a primary or a secondary diagnosis of chronic psychophysiological insomnia or insomnia associated with personality disorder. Seven subjects were randomly assigned to each condition. The study used a three group by 9 week, double-blind design with three nocturnal sleep recordings each week. During week 1, subjects took no capsules; week 2, subjects took placebo; weeks 3 to 7, flurazepam, triazolam, or placebo; weeks 8 and 9, placebo. Daytime tests for alertness and performance were administered during weeks l, 3, 5, 7, and 8. Flurazepam showed hypnotic efficacy for weeks 3 to 5. Triazolam showed hypnotic activity for weeks 3 to 7. Although not significant overall, discontinuation of flurazepam produced rebound insomnia in six of seven subjects sometime during the two withdrawal weeks. The relationship between plasma concentration of desalkylflurazepam, the principal active metabolite of flurazepam, and sleep disturbance suggested that the onset of the rebound insomnia depended on the rate of drug washout. Discontinuation of triazolam produced a significant rebound insomnia on the first and second nights of drug withdrawal. Placebo subjects showed improved sleep throughout weeks 2 to 9 of the study. Daytime testing revealed significantly decreased daytime alertness and decreased AddrPss rl'tjllests for rPprints to: llr. Mitler. Scripps Clinic and Hesearch Foundation. ]()()()() North Torrey Pinl's Hd .. La .Jolla, CA 9:W:\/. This n•st•arch was supported hy (;rant NS l!'i/81 to Dr. Mitll'r, (;rant '.VIH :\.!:!:!:\to Dr. (;reenhlatt. (;rant MH O!i804 to Dr. Dement. grants from ThP l lpjohn Compan~· to llrs. Denwnt and Mitler, and computation support from New York State Offin• of Mental Health.

performance for flurazepam subjects during weeks 3 to 7, although these effects reverted toward baseline despite continued drug administration. LEEPJN(~ pills are ubiquitous in our society. Ac-

S cording to a survey published in 1975, prescriptions

for sleeping pills are given in 8.5% of the office visits to American physicians. 1 In 1977, 25 million prescriptions for sleeping pills were filled.c The Institute of Medicine reports concluded that despite widespread use of medications to improve sleep, there are insufficient data available to make rational choices about which patients should take sleeping pills and what the long-term effects of these drugs are with respect to waking function after bedtime use. Some of the most valuable types of data for making such treatment decisions are from studies that compare drugs, or compare a drug with placebo, in well-defined samples of patients with the complaint of insomnia. Yet, there are few published reports of comparative efficacy studies for hypnotic medications in samples of patients with a common diagnosis. We are aware of no study that makes long-term comparisons between hypnotic medication and placebo. The present study of 21 insomniac patients compares flurazepam, triazolam, and placebo with respect to effects on nocturnal sleep and daytime functioning. The protocol was relatively conventional for long-term hypnotic efficacy studies in sleep laboratories: 59 nights over 9 weeks with :n nights on medication. However, several novel features deserve special introduction. First, subject inclusion criteria focused not on long objective sleep latencies or short objective sleep times during screening and diagnosis, but on whether the individual had a

.I 1'1.1" l':'\1·11111•11\H\1\l'lll .. \'111. I '\11 I. FLH 1!1.'>I

chronic l'llmplaint of insllmnia. Thus, Wl' st udil'd t Ill' kind of patil'nts who an• most likPl:v to fl'('l'ivl' JHPscript ions for sll'qiing pills. Sl'cond, Wl' allowed patients t select t lwir m\·n liedt imes wit bin t lw gerwral guidelirw of maintaining t lwir usual schedules. Thus, at the ex1wnsP of sonll' JHPCision, WP introduced some "real world" varialiilit y· into this s!Pep lalmrator:v study· of hypnotic medication. Third, Wl' used Multiple Sleep LaterH'y' TPsting and othl'r pPrformarn·p tests during till' da:-.· to ass('ss alert rll'ss. Fourth. IH' third of our subjects took placebo pills on all hut t Iw first :\nights oft Iw study·. This group affords an opport unit\' t assess changl'S in sleep and insomnia s.vlllptllls O\'l'r a 9-wPPk period wit bout pharmacologic intl'l"\'l'lltion.

,)

t inw dosps: tlurazl'pam. :lo mg: t riazolam. o.:-i mg. :\II capsu]ps were ingested :m minut(•s IJpforP l)('dt inw. The stud:-· was carried out 11mll'r doulJ]p-b]ind conditions. Note that tlw d(•sign n·quin·d sulijPcts to n·adapt tot hP laboratory· l'H'ry· \\'l'l'k \'(•rt lw !l-Wl'l'k stud\'. This fpaturP may· ha\'l' introdtH·ed sonu· \·ariahility intll tlw intl'f('ndit ion comparisons. Howl'\'l'r. nightly· anal\'s('S disclosl'd only· nonsignificant first night l'ffl'ct~ throughout till' study'. \\'l' han• plottl'd all nights in tlw figun·s oft his rPport sot hat t lw first night l'ff(•l't !'an l)(• Sl'l'n in perspl'('t i\'l'. On tlw da:-·s following nights ·I. IS. :l~. Hi. and :i:l. all subjects undl'rwl'nt a hat ll•ry· of JH•rforrnan('(' t l'st s. Tlw timing and conll'nt of nwals on t lwsl' da\'s \\'('l'l' d(•tl•rmi11l'd by l'ach subj(•('! 's usual rout izws with t lw rl'st ril't ion that no !'affl'itw was ])('rm it tl'd .

Methods

.'-'uhjl'cls Tab IP 1 summarizl's t hl' n>search paradigm for the Pill irP :-i~l-night study·. All capsu]ps wen• idPnt ical in appearaJH'l'. Tlw dos(' ]pvp] for tlurazeparn and t riazolarn \\'l'rl' t lw lllanuf'act urPrs' maximum n•cornrne1HIPd bed-

\\. t't' k I

111~h 1

1111111iin

'1111d11 i< Hl ~

l\l

.1'1

~ 111:...:lit 1111r1dwr

\\·t·t·k

\\ ('('" : ;

111~h1

I >r11:...: ''''t'k l

\\"·"k I

111~h1

111111tl It'/' J

1

11•11111,..r

I )nt:...'. \\1·1·k :...: c(>/ld11 ic>ll (I>:...: l

\\ t't'h .-.

111~h1

l>rug \\i·t·k :

I ln1:...:

~ 111~111

II I.

I' I.

11 I' II

d>:'.1

1111111l1t•/'

\\1·1·k .-) (·111JC!1! H >11 !

I).-) 1

\\ t't'h ·" 111~111 1111111l1t·r \\' 11 hdr.111 "I II"""

I,

!I

I' I.

111 I' I.

11 IJ II

].-.

I'>

]'.l

IJ I.

11; IJ I.

I~

I J II

IJ I.

Ii I.

Iii I i 11

~()

~I

•)"'

~I

IJ II

I> II

Ii I.

IJ I.

I > II

11 II

::11

:1

I J II

IJ II

I> I.

IJ I.

I> I.

I i I.

Iii I i II

::1 I> II

Ii II

I> I.

I> I.

Ii 11

I i 11

II

I~

., I ,,

11

,.-)

I> II

I> II

I > I.

Ii I.

I> I.

.-,]

.-):...:

\\ I.

\\ I.

.-1 ~

.-i'.1

...,

·;,'°'I

"

1.-.

l!I

I > II

Ii 11

\\' I.

\\ II

\\ II

,,, \\ I.

IJ I.

I

I' II

•1,

111

I~

Iii I i 11 •I

\\ I.

l't1JJd1l t11JI I \ \ ' j I

\\ t'l'h !I 111~111 lllJlltf tt•J' \\'i1hdr:111"I 11 .. ,.1; ~

I ii I' II

I' I.

1J11111I1t·r 1'J111d1t1(111

\\'t·

H I.

>.1!

I' II

I' II

1'Lll'1·fi11 1 11ndl! H>n 11'!.1

I

''"'"

Tl1>1r-

.,

T111·....,

~llll

Ba...,t·l1 rlt·

Subjl't'ts \\'l'l'l' mal(• and f'l>malP adults: t hl' f(•lllal(' subjl'cts Wl'n' post ml•nopausal or ot lwrwisl' inl'apalil(• of lll'aring childn•n. Inclusion criteria ('1111sistl'd of 111 a long-standing complaint of insomnia. and (~I a diag1wsis

Iii \\ 11

\\ I.

\\ I.

er n1d1t 11111 1 \\·:.._:) u

l.t•!..,:t'!l!i 1~1r (' and sleep duration as represPntativP slPPp paranH·U·rs. \\'p also disl'uss suhjel'tiH• sleep laIPn'P mm·pnwnt (HEM) sleep time. and per waking hours of laboratory da>·s. all suhjl'l'I s l'omplPt l'd t lw ProfilP of Mood St at PS and 111 lwr paper and pencil assessnwnts of their !'motional status. Da>·tinw ll•sting for 1wrformam·rformancp testing was donl' in fi\'l' ! -hour sPssions. with t hl' first sPssion sclwclukd Ii() to ~J() minutes aftl'r t lw subjl'ct awakerwd in t lw morning. Suhst>quPnt te-;ts were donl' at :.!-hour intl'n·ak Till' halll'r>· consisted of a Digit S>·mbol Suhst itut ion Tl•st. a Wilkinson Addition Test. a Four-( 'ho in• Hem·t ion Time Test. and a l'omhirwd Target Pursuit and l>i1·idPd Alll'ntion Tl'sl. Tlw MultiplP Sll'l'Jl Lall'IH'>' TPsl was gin•n sl'1·en tinws throughout the da>· at :.!-hour intl·n·als. with t lw first test sclwduled Ii() minutes at'tt>r t lw first pprformam·p I Psi ing sPssion. SlePp latPnl'>' on t lw Mull iplt> Sll•l•p Latl'n· was carril'd ot11. Hecl'nt I>·· Mitler and a,;,;ociatps' hm·e suggestt>d a Mainll'natH'P of \\'akl'fulnl',;,; Tt>sl im'!il\'ing instrul'tion to sla\ awakt:>

I )u/u unu/_,·sis Tlw gPrwral stat ist il'al approal'h consist Pd of a :l groups hy ~l \\'l'l' h l»· :i nights (within week) anal>·sis of variarH'l' with I suhjl•11diti"" l"r till' t111r;i11·J""" 1.\ I. tri;imla111 1/i1. and plan·l1ho group "Spp TahlP I for c•xplanat ion of abhrPviat ions.

HL

PL

DI

D:!

l'i.1 14.1 :!0.4

11.8 I :u JIU

fi.G IJ.:l :!:!.!

l'i.O I:!. I

:l'i4 ;l(j/

:141 :!G.!J :l4.8 41.0 :1;,o :l44 :l:lG

401 :l8:! :lfi8

4:!'i 418

:lr>7

l'i.O :l:!.8 !i0.8

408 404 :l'i9

D4

10.8 l'i.9 19.fi

IG.'i 11).(i

4:!1 401i :l'il

8.9 :!2.Il

9.:1 :!:L:l

10.:! :l8.:!

4:!.:!

[j;l,]

-1:!.0

4:l0 :l91i :l48

:l/9 :l!iO :l:!li

7.7

D:l

40:! :l'iO :lli4

411 :HiG :lS7

fj_()

4:!8 ;199

:rno 8.4 :!·U :!1.8 4:!0 :lf>7 :l9:l

J)[j

WI

W:!

14..t lfi.fj J:UJ

1:!.f) 44_;, lli.:l

:!f1.S 1:u :! I.G

:l98 .lflfi :19fi

7.7

:l9:! :l:!8 :188

:!4.:-i

1:u ;, 1.1

:io.o

:i:u1

:rn4 :l'i4

:l:).r)

:rn:1 :!!)] :l!JI

:Hi4 :I/Ii :l9:! :! 1.8 19.8 4:U1 :~;)-;

:Hi! :l'i:!

:!9.'i :!!J.1 :n.G

:14_;,

:rn.o

18.!l :!Ii.I :l:!.!J

:!'i.-1 :!9.!J

4"-·'.'>:L;> :lG.I

:l/.8 41i.9 :!:u;

9.S.f1 9:!.ll 84.:l

!J;J_;, 8!Ui 81i.8

!J:!.I 88.:! HG.I

!l 1.'1 87.8 88.li

81.:l 8!U 88.1

8fi.1 1:i.11 81i.fi

81.:l 8:L:l 88.:-i

:!II.I :!IU :! I.II

11.4 I I.I :! I.I

I fi.G 11.8 18.'i

lli.fi 19.fi :!IJ.!i

Iii.II l!J.IJ :!1.:l

JG.I 18.1 I !Ui

:!O.ll :!:!.:l :! J.(i

:!O.:l :!4.'1 HUi

;,;J.]

ii:l.ii

fil.1

1)/.1)

r:. ....

lifi.:l li:U 49.1

fi-1.'.! '14.:l ·l!l.4

;,4,]

G].()

li!i.:l li0.8 ;, 1.4

li4.fi

;, I.I

li4.8 41i.8

li!i.!i li:!.fj ;, 1.9

70.4

;,.1,0

4:u1 4;i,;,

:l8.ll 4fi.li 40.:l

11.'i :!ll.-1 :H>.:l

8:-i.:! H:u; 8:U

88.:l 8-Ul 8fi.4

:! I.:l :z 1.'1 :!ll.:!

;,o_;,

7

~

1:u; 18.!l

;"°1:2.:1

ii/.4 ;'""1:2.:2

.J

('1.1:-> l',-,c·110J>IL\H\L\C' wen• gn•atPr than haselirw valuPs. hut onl:-.· drug WPPk 1 was grPater than valtrp:-; during t lw placebo week. Subjt•(·ts in t lw pla\'Pho group showt·d a t rPncl tow a rd grPat Pr sk•Pp Pffi('il'lH'\,' as t lw stud:-.· progrp:-;sPd. hut t lwrP wPrP no signifi('ant differPlH'l':-; among WPPk:-;.

Our data di:-;clo:-;pd no signifi('ant changps in HE:V1 sleep throughout the stud:-." Tlwrf::' appean•cl to llf::' a trf::'ml toward redu('ecl HEM (ll'rcl'nt a:-; the drug ingestion pPriod progrPsst•d for hot h tlurazPpam and t riazolam subject:-;, hut \'ariahilit:-.· pn·cludPd :-;tat ist icall:-.· significant liet Wl'l'll or within sulijPcl effects. Hesults for 1wrcent stage ~ sleep indicated that both llurazppam and t riazolam signifi('ant I:·• in('n•asPd stage ~ sk•ep throughout t lw drug ingf::'st ion period. Th est, increases were mirrored by· ('ommensurate statistically signifi\'ant dPcreases in 1wn·pnt st age :l and ~ :-;leep and wakt• time aft pr slPep onsPt.

/'/asma

dru.~

c'l11u·entratiuns

Table ~ prPsent s mean plasma ll'vl'ls of I he principal nwtaholitt• of llurazepam. IV- I-de:-;alk:-.·lf1urazeparn. Thi:-; wa:-; the only· henzodiazepirw or henzodiazepine nwtaholite detectl'd in an:• suhje(·t for

].-).-)

:!:!,....;)

(:!q

l:.!:!J

t inuat ion on night ~8 for a:-; long a:-; ;> day·s aft pr dis('ont inuat ion. Thus. it is oh\·imrs that our tlurazepam suhjel·ts eliminated de:-;alk:-.·lllurazepam at differPnt ratP:-;. This mean:-; that it ma.\' not ht• valid to stud:-.· dis('()ntinua I ion e ffPl't s in t ]w:-;e :-;u hje('t s h.v grouping t lwm in term:-; of nights aft pr dis('ontinuat ion. SuhjPct:-; with high blood le\·els ma~· not :-;]rnw the sanw :-;lel'Jl effl·1·t:-; a:-; :-;uh_ie\'t :-; with intl'rnwdiatl' IP\·Pls or as suhjP('t:-; with low IPwk Did tlw :-;jppp disturharn·es wl' saw aftpr dis('ontinuation of tlurazppam correlate with \'hangp:-; in blood IPvPls of ,V-1-dt•:-;alk.vlflurazt•pam'! \\'p attemptPd to furtlwr anal~·zl' tlw p]a:-;ma data to asses:-; an~· rPlat ion:-;hip bet Wl'l'n plasma ll'\'Pl of N-1dPsalk:-.·ltlurazppam and slPep disturhan\'e:-; in :-;onw llurazPpam suhjP\'I s. Our approach dPpemk·d on the fa('\ that t lw oh:-;pn·pd change:-; in blood pla:-;ma ]l'\'t•ls of N-1dt•salk.vlllurazq>am parallPI \'hangp:-; I hrmrghout t lw hod:-.· and brain. Thl'rl'fore. we ('omparl'd. within individual suhjl'l'b, tlw plasma ('OlH'Pntration:-; of N-1-desalk:-.·ltlurazepam on nights ~8 and :1r> with ,;]pep paranwlPrs for Pa('h night of t lw withdrawal ('()ndit ion:-;. Blood !t'VPls wPre l'Xpn•:-;:-;pd as pPrcl'nt of maximum blood ll'\'l'l n•cordl'd for each suhje('t. SIPl'p parametl'rs wPre expres:-;l'd a:-; (ll'fl'f::'nt of hast'line nwan (m·pragp of night:-; I to:\). FigurP :1 prp:-;pnt :-; t lw rl':-;ults oft his approa\'h for slPL•p duration in a rPJ)fPsPntati\'P :-;uhjP\'I who l'XpPriPnn•d markl'd :-;IPPJl disturliatH'P after discontinuation of llurazepam. \\'p also altt•mpted to a:-;:-;p:-;:-; t lw relation:-;hip hPtWl'en decrenwnt:-; in p]a:-;ma lt•\'Pls of IV-1-dPsalkvlllurazepam and lo:-;:-; of nocturnal sll'l'Jl a\'ro:-;:-; :-;uhje\'b. For all :-;p\'l'll llurazppam suhjl'\'t s. WP ('omparPd I lw blood lt•vPI data on night r1:i and slPep paranwtpr:-; on tlw wor:-;t night aftl'r di:-;('ontinuation of drug. To do this WP ~l di\'idt•d b:-.· shortp:-;t slPep du rat ion among nights 1 to :n. with t lw final rPstilt exprpsspd in pprn•nt. For examplP. t lw subject in Figurp :l PxperieIH'Pd about an 80'; slt:'ep Jo:-;s on :-;t ud:-.· night f'1/. Figurp ·I summarizPs t lw:-;p \'aluPs for t lw :-;p\·en lluraZPJ>am subjects. As ('an hP sl'Pn. a marked rPduct ion below hasPlirw IP\'Pis of :-;]ppp duration appParpd to O('('Ur in thosP two suh_je('ts who had Pliminated GO to (j()'; of tlw IV- l-dp:-;alky·lllurazPparn from their blood plasma h',' night :i:-1. Tlw ow rail ('ur\'l' is suggest iw of an inverted l J with maximum slt•Pp di:-;t urharn·p O\'\'t1rring wlwn GO to (j()';, of N-1-de:-;alk:-.·lflurazPpam has hPPn Pliminated.

/)aytime testing Anal:-.·:-;e:-; of da:-.·t ime t L':-;t i ng fo('u,.; on the Multiple SlPP]l Latl'IH'\,' TPst and tlw TargPt l'ur:-;uit TPst. Tabular

,J CI.IN

Flurazepam, Triazolam, and l'laceho 140

20




SLEEP "?. BASELINE

120 17

DFLZ BLD

100

w z w BO w w 60 u

..."' "' 8'

DFLZ BLD

40

I

20 0

47

51

49

53

57

55

(

59

BL

> (

DI

> (

03 )

(

05

> (

WI

>

( Wl

>

FLURAZEPAM GROUP

STUDY NIGHT

Fu;. :l. The relationship between change in sleep and change in blood level of desalkylflurazepam for a representative subject. The blood level of desalkylflurazepam (expressed as percent of highest blood level recorded during the study) has been co-plotted with sleep duration (expressed as percent of shortest sleep duration recorded during the baseline condition). 100

11

90

BO

El

~ 70

::; ~ 60 .,..,

(

BL

>

< 05 >

TRIAZOLAM GROUP

;i: 50 0 ~ 40

.,

El

20

" 30

El

20

El

El

10 0

20

40

60

c

El

80

100

DFLZ "Y. MAXIMUM

F1 (

01

> (

03

> (

05

>

(WI

>

PLACEBO GROUP Fu;. f>. Daytime alertness, based on the multiple sleep latency test, as a function of study condition for the flurazepam (A), triazolam (/Jl, and placebo((') groups, respectively. Mean sleep latencies are plotted for each group from each of seven sleep latency tests given on one day during the BL, DI, D:l, Dfi, and WI conditions (see Table I for explanation of abbreviations).

creased alertness during the drug ingestion period for the triazolam subjects, but none of the Dl-Wl condition means differed to a statistically significant degree from baseline. There were no significant changes in alertness throughout the study for placebo subjects. The Target Pursuit and Divided Attention Tests are given together and designed to assess skills related to driving an automobile. Figure 6, A to C, presents mean distance off-target for the last 2 minutes of the Target Pursuit Test. Analyses of variance revealed no signifi-

10

,J CuN 125

A

placebo subjects showed no significant effect of weeks, probably because of large intersubject variability and particularly poor performance of one placebo subject on the WI test day. Table 5 summarizes condition means for the Multiple Sleep Latency Test and the Target Pursuit Test as well as the other performance tests. Inspection of Table 5

120 115

a:

1t0

~ 105

a:

"" 100

"'~

Mit/f'r and As.,ucintes

Psvnt0PllAHMACOL, YoL 4/No I, FEH 1984

95

~ 90

CJ)

0

TAHU~ '"

Summarv of performance testing. week of testing da~·

\'t'rsus drug group








FLURAZEPAM GROUP 125

B

Flurazepan1 group Triazolan1 group

115

~

a:

l'laceho group Target Pursuit TPsl I nwan distance error in units of tell'vision dot diameters)

110

105

""100

tj z

95

Flurazt'parn group

~ 90 CJ) 0








TRIAZOLAM GROUP 125

115 110 105

UJ

100

""~

95

I·!.:!

111.:.! lli.:l !:!.Ii

J:l.:! 16.!J 12.9

14.1 J:i.11 1:u;

l!i.11 J:l.H 14.1

Triazolam group l'lacPho group Wilkinson Addition Test lnumlwr attempted I FlurazPpam group

HI. I HO.I H4.9

1(1.1 14.I "i"i.(I

90.1 II"i.!1 IX.Ii

H1.H

/().!i 14.0 H9. I

Triazolan1 group Pla('eho group

li:l.:! HI.Ii lil.H

lil.8 l(J.li fiH.:.!

liH.4 IH.1 li4.H

~ 90

lill ..I



' studiPs of h>'lllH>til' drugs: eftical'y and withdrawal effel'ts. .J C'lin l's>Thopharmacol 198:l;:l:l.t0 ;,(), 21. Mitler MM, Carskadon MA, Phillips l{L, Stnling W, Zarrnm· \'I'. Spiegel H. Cuilleminault (', Dement WC. Hypnotil' Pllil'acv of temazepam: a long-term sleep laborator>· Pvaluat ion. Br .J ('!in Pharmarnl 19'i9:8:ti:lri-8fi. 22. Carskadon MA, Seidel WF. C:reenblatt D.J, Denwnt WC. Davtinw carryover oft riazolam and llurazepam in elderly insomnial's. Sll'PP 1982:fl::Hi 1-'i I. 2:l. Morgan K, Oswald I. AnxiPt>· l'aused I»· a short half-life h>·pnt ic. Br Med .J 1982:284:942 .t. 24. Kales A, Soldatos (', Bixler E. Kales .JD. Earl>· morning insomnia with rapidly Pliminated henzodiazepirll's. Scil'm'P l!J8:l:~20:9r1 7.