Nonselective and Selective Benzodiazepine Receptor ...

Download PDF
3 downloads 0 Views 6MB Size Report
Comment
therapies from chloral hydrate and the barbiturates through the benzodiazepines and explores the newest selective benzodiazepine receptor agonists, including ...
NONSELECTIVE AND SELECTIVE BENZODIAZEPINE RECEPTOR AGONISTS-WHERE ARE WE TODAY?

Nonselective and Selective Benzodiazepine Receptor Agonists-Where Are We Today?J Merrill M.' Mitler, PhD

Abstract: Insomnia is problematic for many individuals, causing them to seek treatment. There is a long history of therapies aimed at restoring normal sleep patterns, each having its advantages and disadvantages. This review traces the history of insomnia drug therapies from chloral hydrate and the barbiturates through the benzodiazepines and explores the newest selective benzodiazepine receptor agonists, including zolpidem and zaleplon . The mechanisms of action of the benzodiazepine receptor agonists are compared and contrasted. A pharmacokinetic comparison is presented showing the importance that parameters such as dose, onset of action, lipophilicity, metabolites, half-life, and receptor-binding affinity have on clinical effects. The possible adverse effects of sleep aids are discussed, including residual sedation and psychomotor impairment, daytime anxiety, anterograde amnesia and cognitive impairment, rebound insomnia, and drug tolerance and dependence. Effects on sleep efficiency and staging are also discussed. Recommendations for the primary care physician on the selection of hypnotics are also provided. Benzodiazepine receptor agonists are often appropriate agents in the treatment of insomnia; however, individual drug and patient considerations are important in matching the most appropriate agent to the individual patient. Zolpidem and zaleplon , newer selective benzodiazepine receptor agonists, offer additional treatment options. INTRODUCTION Early Eras Of The Sedative/Hypnotics

INSO MN IA, THE MOST COMMON COM PLAINT ENCOUNTERED IN THE FIEL D OF SLEEP MEDI CINE, is characterized by an inabili ty to fa ll as leep or stay as leep, and/or by early awakeni ngs . The causes of thi s complaint are as numerous and varied as are the causes of the compl aint of headac he or abdo minal pain. The symptom of insomni a is age-old and affects most people at some time or another. In any one yea r, as many as 30% to 35% of Ameri can adul ts report diffic ul ty sleeping 1 and 4% take sleep medi cati ons in an attempt to reso lve their problem.2 Insomnia affects daytime performance and can cause serious medi ca l prob lems. ft im pa irs soc ial li fe, is assoc iated with dimini shed work perfo rm ance, increases the risk of automobil e acc idents and results in increased rates of hos' li ty. 2 The tota l costs of insomnia, pitali zation and morta includ ing treatment, loss of prod ucti vity, and insomni arelated acc idents, may exceed $ I 00 bi lli on per year. 3 Prescri bi ng physic ians can choose from a wide variety of med ications to treat insomni a. However, onl y sedative hypnotics that have reached the market within the past 20 years have been deve loped and eva luated under strict Food

Address correspondence to: Merrill M. Miller, PhD, Professor, 'oepartment of Neuropharmacology, The Scripps Research l nstitut~, 1 9834 Genesee Aven ue, Suite 328, La Jolla, CA 92037, Tel (858) 784.9295, Fax (858) 657.0552, E-mail: [email protected] .

'

and Drug Administrati on (FDA) guidelines, which incorporate sleep laboratory studies to accurate ly determine the effecti veness of these agents fo r appropriate durations of nightly use. 4 Modern sedati ve hypnotics therefore have obj ectively demonstrated sleep inducti on and sleep maintenance properties, and their adverse effects on attention and performance have been evaluated obj ectively. Tn contrast, research has not determined the effects on sleep and alertness of either older sedati ve hypnotics or other classes of drugs that are used "off label" as sleep aids. Today, sedati ve hypnotics can be chosen with an unprecedented degree of confi dence fo r ap propri ately evaluated and diagnosed patients on the basis of deta iled knowledge about sleep phys iol ogy and the actual benefi cial and detrimental effects of the candidate drugs. 5 This was not always true. Since antiquity, hea lers have searched for potions to produce drows iness and facilitate the onset and maintenance of a state that resembles sleep. Most of these drugs act on chl oride ionophores coupled to GABA(a) receptors. For exampl e, barbiturates and chlora l hyd rate prolong the opening ti me of the chloride channels and benzod iazepi nes increase the likelihood that GA BA will open the channels.7·8 The nonselecti ve benzodi azepines act as class ic agoni st compounds at the GABA benzodi azepine receptor complex (G RSC), binding to benzodi azepine receptors (BZ I and BZ2). Their action at the GRSC allows inhibition of neural exci tation by opening the chl oride channel in res ponse to GA BA.9 Bromide, available by the mid-I 800s, was the first age nt specifi ca ll y introduced as a sedative hypnotic. Chloral

This materigl)gas oopied-Jonselective and Selective Benzodiazepine Receptor Agonists-Mitler att

hydrate, paraldehyde, urethan, and su lfony l fo ll owed. Chl ora l hydrate is an effective hyp notic agent and is occasiona ll y used to produce sedation prior to med ica l or dental procedures that are potentially uncomfortable. It is rap idl y absorbed from the gastro intestina l tract and metaboli zed in the li ver to trichloroethanol , whi ch is the act ive component. Ch loral hydrate is a gastri c irritant with an unpl easant taste and odor, ca using epigastri c di stress, nausea, and vom iting. Effects on the central nervous system (CNS), including lightheadedness, atax ia, ni ghtm ares, and a "hangove r" effect, are also common . Chroni c use of chl oral hyd rate may res ult in li ve r and renal failure that can be fatal. In add ition, sudden withdrawa l from chronic use may resu lt in delirium and seizures, and sometimes death. 10 The barbiturates were introduced in the early 1900s for a va ri ety of indi cations, including inso mni a. By the 1960s, they accounted for 55% of all hypnotic presc ripti ons. 1 Barbiturates (phenobarbital, secobarbi ta l, butabarbital, amoba rbital, and pentobarbital) are effective in inducing sleep. They are weak ac ids that are we ll absorbed both orally and intramusc ul arl y. Their onset of acti on correlates directl y with their durat ion o f action, and onset and duration of act ion depend on distribution more than they do on eliminatio n half- life. Di stribution is related to lipid so lubility and plasma and brain protein binding. Phenobarbital, which has the longest onset and duration of action of alI the barbiturates, has the lowest degree of lipid solubility and protein binding. The barbiturates have several disadvantages as hypnotics, including failure to induce normal slee p and frequent occ urrence o f residual sedation (hangover). Both metabolic tolerance and pharm acody namic tolerance ca n occ ur afte r repeated ad ministration, as can physica l and

~p;~~-Barbiturates t =J .

Vol. 23. S1111ple111e111 I . 2000

at the NLM and may t>e Sut>j ect US Cc$4~ight Laws Nonselective and Selective Benzodiazepine Receptor Agonists-Mitler

does triazo lam. Zo lpidem is considered a low bindingaffinity drug, as it has I 0 and 30 times less binding affinity than do flunitrazepam and triazo lam , respectively. 22 Additionally, it binds selectively to the BZ I site. Zaleplon, a pyrazo lopyrimidi ne, is a benzodiazepine receptor agonist with an elimination half-life of approx imately one hour. 25 Za leplon binds se lectively to the BZ I receptor, where it is a full agon ist. However, za lep lon's affin ity at the BZ I receptor is five times lower than zo lpidem 's, indicating that za leplon is also a low binding-affinity hypnotic. Abrupt disco ntinuation of I0 mg of za leplon after five weeks of treatment has not resulted in any rebound insomnia. 26 Effects on Sleep-efficiency and Staging

Table 2 provides a genera l overv iew of the effects on sleep of various classes of agents used on or off-lable as hypnotics. Meta-analysis of studies evaluating the sleep efficiency of the benzodiaze pines has shown that, compared with placebo, both the nonselective and selective benzodi azepine agon ists reduce sleep latency, decrease awake nin gs, increase the total amount of sleep obtained, and improve the quality of sleep.21 Quazepam and triazolam cause rapid sleep induction and significantly reduce sleep latency. In contrast, tlurazepam and temazepam are not as effective in some patients. Fluraze pam is rapidly absorbed , but its first-night effectiveness may be suboptimal, requiring two to three days' adm ini stration to produce optima l sleep efficiency.23 Zo lpidem reduces the tim e to onset of sleep, increases the

duration of sleep, and does not significantly alter slee p stages, as do th e older benzodiazepines. 9 Za lep lon improves sleep latency and slee p quality with no residual sedation or rebound inso mnia. 28-30 Zalep lon produces slightl y less sedation than zo lpidem but returns pati ents to baseline faste r (five hours vs. eight hours, respectively). 3 1.3 2 At I 0 mg and 20 mg, za leplon does not appear to alter normal sleep architecture. 33 The effects of the nonhypnotic agents are based on fewer and less well-contro ll ed studi es. Residual Sedation and Psychomotor Impairment

Strong evidence indicates that sleep deprivation compromi ses mood, alertness, and performance. 34 One of th e most common complaints about hypnotics is residual sedation (ha ngover effects), which refers to a drug's propensity to produce sleep iness, red uce alertness, and impair performance in the daytime. 35 This effect is especia ll y troublesome in elderly peopl e. Both the dose and half-life of an agent correlates with the degree and duration of res idual sedation . It appears that th e long-acting drugs produce more residual sedation than do shorter-acting drugs . Some drugs accumulate in the body, resulting in longer elimination tim es and the potential for residual sedation. However, short half-life does not ensure that no residual sedation wi ll occ ur, as dose is also an important contributor. Flurazepam is associated with significant han gove r effects, with an incidence around 24%.20 In contrast, findings on temazepam are quite variable. In some studi es, temazepam produced CNS impairment; in others, no resid-

Table 2.-Effects of Various Drug Classes on Sleep Sleep Latency

Sleep Duration

Slowwave Sleep

.j,

t

-!.

-!.

t

-!.

Short-acting benzodiazepines

-1-H

-1-H

Selective benzodiazepine receptor agonists

-1-

Hypnotic Agents Barbiturates

Long-acting benzodiazepines

Nonhypnotic agents used for sleep Antihistamines



-1-H

t

-1-H

TC As

-1-H

tH

-!.

SSRls

t

.j,

-1-H

TCAs

REM Sleep

=tricyclic antidepressants; SSRls =selective serotonin reuptake inhibitors.

SLEE!~

Vol. 23. S11pple111ent I. 2000

I h1.s mate ri al w a.s cop1ed siia may De Nonselective and Sei

at the NLM sut>ject

usCeWHW Laws

=t' B . . ec ive enzod1azepine Receptor Aqonists-M1tler

of slee p, cause fewer cog111t1ve and psychomotor side effects, and may have fewe r withdrawal effects. 19 These se lecti ve BZ I-binding agents have littl e or no clinical effi cacy as anti co nvul sants, myorelaxants, or anxiolytics.'l

lipop hili c, but there are differences in lipophilicity among the vari ous drugs. Quaze pam is the most lipop hilic, followed by diaze pam , flurazepam , temazepam, and triazolam. 23 Lipid solubility also influences the extent of drug di stribution into the periph eral ti ss ues, whi ch takes th e active Importance of Pharmacokinetics drug away from the prefe rred site of action in th e brain. Individual pharmacokinetic characteristics determine Reaardless of their elimination half-l ife, hi ghly lipophili c the type and severity of adverse effects of the different bencoi~pound s can have a short duration of act ion due to th e zod iazepines. The characteri sti cs are especial ly relevant to shuntin g actio n to periph era l tiss ues. 2-1 Add iti onall y, the use of hypnoti cs in elderl y and/or debilitated patients lipop hilicity influences the equi librium di stribution. ratio who may be more susce ptible than their respective counbetween brain tiss ue and the unbound compound 1n th e plasma. . terparts to CNS depressants.20 Dose and duration of adminMetabolites: Triazo lam, temaze pam , estazo lam, zolp1ist ration are com mon to all th e drugs. Factors such as onset of action, lipophili city, presence of metabolites, eliminat ion dem, and zaleplon do not have c lini ca ll y act ive meta bolites. rate, receptor-binding affinity, and effects on the sy mpaFlurazepam and quaze pam are transform ed into a sequence thetic nervo us system and hypothalami c- pituitary-adrenal of act ive metabolites (N -d esa lk y ltlu razepa m an d N(HPA) axi s are characteri stic of individual drugs. (Table I) de sa lkyl-2-o xo quazepam ,respective ly) . Desa lky ltluDose: S ide effects are dose related, hence the recomrazepam is hi ghl y protein bound and lipophilic and has a mendation to use the lowest poss ible dose. The pote ncies of hi gh intrinsic potency or low rece ptor affinity constant. the different benzodiazepine ago ni sts va ry widely, so that During ni ghtl y adm inistration, this metabo lite accumu lates 21 equiva lent doses may vary as much as 20-fo ld . Dose must substa ntially and has bee n assoc iated with next-day CNS be se lected prudently when switching age nts, and must be depress ion (drows iness, dizziness, li ght-h eadedness, ~ nd reduced in elderl y patients. incoord inat ion). Converse ly, the presence of m etabolit~s The difficulty of estab lishing dose eq ui va lency amon g provides a slow, metered "se lf-taperin g" when th e drug is benzodiazepines g ives ri se to methodological problems withdrawn, reducing rebound inso mni a or w ithdrawa l sy nwhen conducting comparative studies. Thus, reports that a drome. particular drug has stronger effects than anoth er may be Elimination Half-l(fe: The elimination rate of the parent due onl y to the use of nonequival ent doses . Research ers co mpound and act ive metabolites appears to be a criti ca_I have not establi shed a standard measure of eq uipotency. factor in the deve lopment of w ithd rawa l sy mptoms. HalfSome researchers use data from animal studies or in vitro life can be used to predict th e extent of drug accumulation binding affin ity studi es to determine potency, whereas othduring multiple dosing. Long half-life hypnotics acc umuers use a mi II igra m-based or performance-based system. late slowly and extensively and d isa ppear slowly afte r t.he Based on recall performance tests, tri azo lam, on a mil last dose. Short ha lf-li fe hy pnotics reac h steady-state rap 1dligram-pe r-milli gra m bas is, wou ld be 150 times more ly and have a low degree of accum ul ation ; they are rapid ly eliminated after the last dose . On the one hand, drug acc upotent than flu razepa m. Onset of Actio11: Onset of act ion is particularly impormul at ion may lead to such side e ffects as day time sedation . 23 tant to patients having difficulty fa lling as leep. The rate of On the other hand , it may atten uate effects such as rebound drug absorpt ion is the principal determinant of the onset of insomnia due to the se lf-tapering e ffect. Recep;or-bimliug A.ffini~y . A high receptor-bi1~di.n g action. Lag time, during which the drug di ssolves, becomes absorbed by the gastroi ntestinal mucosa into portal circ ul aaffinity may also play a role in the deve lopment of·. side tion, and passes through th e hepatic circulation and into the effects occ urrin g wit h ab rupt withdrawal. Drugs with .a systemic blood, can take anywhere from I 0 to 20 minutes hi gh receptor-binding affinity inc lude lorazepam, tlun1depending on the drug and the individual's metaboli sm. traze pam, triazo lam, and mid azo lam; those with a low Ti me to max im al plasma concentrati on ranges from 30 to receptor-binding affi nity include te maze pam , flurazepam, 90 minutes. Abso rpti on rate is based on the drug's intrinsic and quazepam. Drugs such as tri azo lam, whi ch have both a so lubili ty, the character of the pharmaceutical preparation hi gh binding affi nity and a short half- li fe, have a greater (pa rti cle size, capsu le co mpos iti on), and the pati ent's fa stpropensity for causing rebound in somnia. 22 F lu ra~e p~m 1 ing status, as food wi ll slow the rate of a bso rption. ~ and quazepam have a moderately low receptor-b1nd1ng Upopliilicity: Lipid so lubil ity, as measured in vitro, affi nity and longer half-lives and are assoc iated with less determ ines how quickly th e drug mol ec ul e can cross the and milder rebound inso mni a. Relativ e to tri azo lam, blood-brain barrier. Highl y lipid-so lubl e drugs have a rapid temazepam, whic h is more slow ly eliminated and has a low onset of action when co mpared with l.ess lipoph ili_c drugs, receptor-binding affin ity, produces less freque nt and milder beca use they reac h the brain quickly .(1f the rate ot ab~o rpsy mptoms afte r ab rupt di scontinuation of ni ghtly use than tion is equall y rap id ). All benzod 1aze p1nes ar0i id1"tghlbla lwasrnp1e-cl at th.e NLM a nd may be

SLEEP Vol. 23. Supplement I. 2000

Subj en USCSl$2right Lawf!ilonselective

.

.

and Selective Benzodiazepine Receptor Agonists-M1tler

days or as long as two wee ks or ca n lead to a longe r-term persistent phenomenon . Because withdrawal sy mptoms are o ften the sa me as presenting sy mptoms, it is difficult to deter111ine whether they are a side effect of drug use or are mere ly a return of previous sy mptoms. Personality factors innuence the type and degree of withdrawa l sy ndrom e expe ri e1:ced, as do pri or alcohol and CNS depressa nt drug use, whi ch may also be factors in psychological dependence. Physiologic dependence does not appear to lead to dose esca lati on or psyc hologi cal dependence.45 Selection of a Hypnotic Agent in the Primary Care Setting Primary insomni a is sleeplessness not attributable to a 111edi ca l, psychiatric, or enviro n111 ental ca use (refe r to the Diagnostic and Statistical Manual of Mental Disorders Fo urth Editi on, for di agnosti c criteri a) . Before suogestin~ . b b treat111ent, primary care phys icians 111ust correctl y diannose pri111ary inso mnia by sc ree ning for 111edi ca l disorder~ th at 111ay contribute to inso mni a, such as hype rtension, ca rdi ovasc ular insufli ciency, and hypothyro idi s111. The patient's 111ental status should be eva luated because 33% to 67% of pat ients with insomnia have an underlying psyc hi atri c diso rd e ~-.~ Anxiety, panic ~ttacks, and depress ion 111ay req uire spec1he treat111ent. Pattents presenting with inso 111 nia are rarely found to be substance abusers. Use of all sedative hyp notics, alcohol , and illicit drugs shou ld be determined . Th.e pr.i111ary car~ phy sician should consider the possibl e contn button .of pn~11ary s.leep di so rders, such as slee p apnea,. to . pn111 ~ ry .111so111n1 a. Deter111ination 111ay req uire exte nsive 111terv1ew 1ng of both the patient and the patient's sleeping part ner. Although not necessary in a routin e eva luat ion of primary insom ni a, polyso111nograp hy and refe rra l to a sleep di so rd er clini c 111ay prove helpful when circad ian rhythm d i.s~ rd e rs are suspected .. ~n add iti on, the prilllary ca re physicia n should rule out sh1tt wo rk , j et lag, and restless legs sy ndrome as contributing factors. A consultation for inso mnia should always include a di sc uss ion of sleep hyg iene and recommendations for behavioral change. Pri111ary ca re phys icians who choose pharmacologic treat111ent to manage inso111ni a must clea rl y und erstand the properties of the va ri ous benzodiaze pines that may influTable 3.-Common Nonprescription Agents Used to Indu ce Sleep

Diphenhydramine (Nytol®, Sleep-eze® , Sominex® , others) Diphenhydramine in co mbination (Anacin PM ®, Doan's PM ®, Extra Strength Exced rin PM ®, Tylenol PM ®, Unisom with pain relief® ) Doxylamine (Uniso m Nighttime® ) Tryptophan

ence their se lecti on, because 111 atc hing patient needs to the age nts' properties is the goa l of drug treat111ent. The following recom111 end ations are suggested : I) use the lowest e1Tective dose, 2) use on an "as needed " basis (e.g., 2 to 4 ti111es week ly), 3) use fo r a short ti111 e (four-week maxi111um), 4) di sco ntinue the medication grad uall y, and (5) be alert fo r rebound inso mnia. Physicians must also consider patient type when se lecting a phar111acologic age nt. Elderl y pat ients and patients wi th renal or hepatic insuflici ency may be at greater ri sk for sedative side e ffects. On occasion, it may be desirable to have daytime sedation and anxiolyt ic effects, and a longacti ng agent wo uld be preferable. In ge neral, most patients will benefit from a short-acting agent with a rapid onset of ac ti on. Za leplon ca n be taken four to five hours before wak ing and has no res idual sedation. Thi s drug may therefo re be desirabl e for patients who require a hypnotic age nt after attempting, but failing, to foll as lee p without phar111aco logic manage111ent. All other hypnotics must be taken at least eight hours before wak ing to avo id dayti111 e sedation. They are taken before diffi culty falling as leep or staying asleep act uall y occ urs. Pati ents who co111pl a in of ea rly- 111orning awakening 111ay require an interm ed iateacting age nt, such as temazepam. Drugs should be se lected and used with ca utio n, but the pri111ary ca re physician presc ribing hyp notic age nts for short-term use need not be overly conce rned about drugseek ing behavior or patients' tea r of dependence. Epidemi ologic and laboratory ev idence suggests that benzod iazep ines have a relati ve ly low ab use li ab ility. 41 Of the benzod iazep ines, diazepa111 see111 s to ha ve the greatest abuse potential.' ~ No npresc ripti on agents (Tab le 3) are used by I 0% of patients wi th inso111ni a to induce sleep; anot her 13% use alco hol to improve sleep.3 These practices shoul d be di sco uraged, as nonpresc ription drugs are onl y 111ini111all y elfoctive, have side effects, and reduce sleep quality. In additi on, so111e over-th e-co unter preparations adverse ly affect next-day perfor111ance. In conc lusio n, since the introduction of flura zepa m in 197 1, benzodiazepine receptor ago ni sts ha ve pro ved to be effective hypnotic age nts. During the subsequent two decades, introd uction of age nts with shorter du rati ons of act ion and improved se lectivity ha ve prov ided clini cians with 111ore treat111 ent options fo r better 111anage111ent of pat ients with inso111nia. Increased experi ence with newer se lective benzodiazepine receptor ago ni sts, such as zo lpide111 and za lep lon, will better defin e the role of th ese agents vs. traditional benzod iazep ines. It is hoped that future research will deline the act ion of th e va ri ous benzodiazep ine rece ptors. In thi s way, drugs ca n be deve loped to interact with recepto rs that produce the desired eflect, whil e having few adve rse effects.

Melatonin wascopie.d mayt>e . . . . sut>ject us'Cb"pyright Lal'l~nselect1ve and Selective Benzod1azepine Receptor Agonists-Mitler

---------------------..i~-~ ial

SLEEP Vol. 23. Supplement I. WOO

attheNU~

ual sedation occurred. CNS depress ion occurs in abo ut 14% of patients taking 0.25 mg of triazo lam and in 20% of th ose tak ing 0.5 mg, demonstrating the dose-dependent effect.20 However, eve n in the elderl y, a 0.125 mg dose of tria zolam does not impa ir dayt ime wakefulness . 18 Quazepa m is assoc iated with an overa ll frequ ency of sedati on of 12%. In placebo-controlled trial s, zo lpidem and placebo produce psychomotor impairment in small percentages of study parti cipants, whereas those given tri azolam showed significant dayt i111e impair111ent co111pa red with placebo.'l In a placebo-controlled trial , za lep lon had no adverse effect on perfo r111 ance in four psyc hom oto r tests at 8.25, I 0.25 , and 12.25 hours after ad mini strati on.JGWh en given in doses up to 20 mg, zalepl on produced no res idual sedative effocts. 33-' 7 In addition, no residual sedative e ffects were observed after zalepl on ( I 0 mg) was given in the middle of the ni ght, whereas flura zepa111 30 111g produced significant sedation . Even if za leplon is taken three hours prior to ri sing, no res idual sedation is observed, an adva ntage to patients not wishing to use 111edication si111pl y in anticipation of ex peri encin g slee p probl e111s .37 -3'>

chan ce of rebound ; the shorter the half- li fe the greater the chance of rebound. The probab iIity or intensity of rebound inso111nia does not increase with increas ing duratio n of benzodiaze pine use. Giving the lowes t effec ti ve close and withdrawing the dru g slow ly ca n reduce the potenti al for rebo und inso mni a. Co un se ling the pati ent to expect a few ni ghts of poor slee p will help reduce anxiety over lost sleep. Te111azepa111 ca uses s ignifi ca nt rebound insomni a after 14 ni ghts of a 30 mg close, but not at lower doses. Rebound inso mnia with tri azo lam is dose re lated, as it has been observed at a 0.5 111g dose but not at a 0.25 mg dose. Several studi es suggest that abrupt di sco ntinuat ion of zo lpidem does not ca use rebound inso111nia.'l Howeve r, in a placebo-co ntroll ed co mparati ve study o l-' za lep lon and zolpicle111 , no s ignifi ca nt di ITerences we re detec ted in the incidence of rebound in so mni a or withdra wa l effects between za leplon (5, I 0, and 20 mg doses) and placebo, but rebound insomni a and withdrawal effects we re significantly greater a fter one ni ght in th e zo lpide111 group compared with pl acebo.42 Tolerance

Anterograde Amnesia and Cognitive Impairment The a111nes ia assoc iated with benzod iazep ine use cou ld be considered a retrograde effec t of sleep that di srupts long-term 111e111ory consolidation. A Ithough inso111ni a itse lf 111 ay be associated with memory problem s, anterograde amnesia is a pote ntial e ffect of a ll benzodi azep ines. Thi s drug class is associated with a temporary impairm ent in information acqui sition and subseq uent reca !l.1 7 The degree of impairment depe nds on the route of administrati on and dose . Intrave nous admini stration of diazepam often res ults in anterograde amnesia, intramusc ular ad ministration ca uses fewer epi sodes, and oral admini stration is infrequently associated with a111nesia. 22 The 111ore pote nt the drug the more 111 emory impairment observed. Measurement tim e afte r drug admini strati on is also a factor. Res ul ts fro111 reca ll tests g iven at peak plas111a co ncentrations will ohen be poorer than res ults fro111 tests give n se veral hours after ad mini strati on. 17 Even at max imal se rum co nce ntrations, I 0 111g o f za leplon has been mini111 all y assoc iated with either anterograde amnesia or cogniti ve i111 pai r111 ent. JX.40 Anterograde amnesia appears to be signifi ca ntly less with zo lpidem than with fluraze pa111.9 Rebound Insomnia Rebound inso111nia is a worsening of s lee p beyond baseline leve ls of di sturbance that lasts one or two ni buhts after ab rupt di scontinuation of so111 e sedat ive hyp not ics.4 1 Rebound inso111ni a can occ ur afte r one ni ght of benzodiazep ine use .22 Drug dose and half-1 i fe are i111portant factors in rebound insomnia. The hi gher th e close th e greate r the

Tol erance is a reducti on in pharmacolog ic eflects with repeated ad111inistrat ion of a drug. Compared with the older sedat ive hypnotics, such as the barbiturates, tol erance to the hypnotic effects of benzodi aze pines is minimal, especially if th ese age nts are used for no longer th an four wee ks, as reco mm end ed. Tol erance is Iinked to th e changes in rece ptor binding that occ ur with chronic admini stration. It is not known whether thi s degrading rece ptor-effe ctor co upling, whi ch occ urs over tim e, interle res w ith eflicacy.'13 No ev idence of tolerance has been observed with zo lpide111 or za leplon used eve n afte r as long as 179 and 365 days, rcs pectively44 Dependence Nightl y use o f benzodia ze pincs is reco111m end ed for no lon "'uer than four weeks , beca use 111 ore chronic aclmini stration may lead to phys io logic and, poss ibl y, psyc hologica l dependence. Physiol ogic dependence is assoc iated with withdrawal sy ndrome on di scontinuation . Shorter-ac ting drugs produce an earli er and more intense wi thdrawal sy ndrome relat ive to longe r-acting drugs .'1" Rapidly eliminated drugs ca use 111ore withdra wa l proble111s than do s low ly eliminated ones. Fluraze pam and quazepam are not typica ll y associated with withdrawa l sy ndrom e. All the age nts have similar depe nd ence profiles. It is not known whether the ri sk of dependence is based on a minimum dosage or ex posure to the drugs, nor is it known what proportion of pati ents rece iving normal doses are likely to ex peri ence withdrawal sy ndrome.4< Withdrawal sy ndrom e ca n last as littl e as one to four 1

r his mate·r i al w as copied SLEU~

Vo l. 23. S11p11le111e111 I. 2000

atthe N L~4~d m ay be Nonselective Subject US"'CDpyr ight Law s

and Selective Benzodiazepine Receptor Agonists-Mitler

Nakazawa Y. Effic acy of L-846 in patie nts with insomni a: eva luation by polyso mnograph y. Psychi atry Cl in Ne urosc i 1998;52: 156-1 57. 27. Nowe ll PD, Maz umdar S, Buysse DJ , Dew MA, Rey nolds CF Ill, Kupfe r DJ . Benzodi aze pines and zolpiclem for chroni c insomnia: a meta, I. Walsh J K, Engelhardt CL. Trends in th e pharm aco logic treatment ol' analys is o f treatment e ffic acy. JAMA I997;278 :2 170-2 177. insomni a. J Clin Psychiatry 1992;53 : I0-1 8. 28 . Vogel G, Erwin CW, Roth T, Zammit GK, Whitake r T, Fujimori M, 2. Meyer T J. Eva luation and manage ment of insomnia. Hos p Pract Dec 15, 1998 :75-86. ct al. Enects o f a single close o f' za leplon on transient insomnia. 2 1St Congress Co ll egium Intern ationa le neuropsychologicum (CINP); July 3. Eddy M, Walbroehl GS . Insomnia. Am Fam Phys ician 1999;59: 19 1119 16, I 9 I 8. 12-16, 1998; Glasgow, Scotland. Abstract. 29. Hedn er J, Emili en G, Sa linas E, and th e Za lep lon Clinica l 4. Kales A, Kales JD. Shortcomings in th e eva luati on and promoti on o f hypnotic dru gs . N Eng J Med I975;23 9: 826-827. Investi gator Study Group. Improvement in sleep latency and slee p qual, ity with za leplon in elderl y patients with primary insomnia. 14tl1 5. Mitl er MM , Mil ler JC. Meth ods fo r testing slee plessness . Behav Med 1996 ;2I: 17 1- 183 . Congress o f th e European Slee p Research Society; September 8-1 2, 6. Paul SM. GA B/\ and glyc ine. In: Bloom FE, Kupfer DJ , eds. 1998; Madrid, Spa in. Abstract. Psychopharmacology : th e fourth ge neration ol'progress. New York, NY: 30. Ancoli -l srae l S, Wa lsh JK , Mangano RM , f'u jimori M, Zaleplo11 Rave n Press; 1995:89. C li nica l Study Group . Zaleplon, a nove l nonbenzodia ze pine hypnotic, 7. Lovinger DM , Zimmerman SA, Levitin M, Jones MY, Harri son NL. efle cti ve ly treats insomn ia in elderly patients without causing rebouncl Trichloroethan ol potentiates synaptic transmi ss ion medi ated by geffe cts. Primary Care Companion to th e J or CJin Psychiatry 1999; I: J J4, aminobuty ri c acidA receptors in hippocampal neurons. J Pharmacol Ex p 120. Ther 1993;264: 1097-1 103 . 3 1. Drove r DR , Lemmens HJM , Na idu S, et al. Doub le-b lind, placebo8. Costa E, Guidotti A. Neuropeptides as cotransmi!ters: modu latory controll ed, randomi zed, crossover, pharmacodynamic study of I0 and 20 effects at GA BAerg ic sy napses . In: Me ltzer HY , ed. mg doses or zaleplon and zolpiclem. Clin Phann Ther 1998;63: 199. Psychopharm acology : the third generation. New York , NY: Ra ven Press, 32 . Fry J, Scharf MB, Berkowitz DY, et al. A phase Ill , 28 day, multi1987:425-435 . center, randomi zed, doub le-blind comparator- and pl ace bo-controlled, 9. Langtry 1-1 , Benfi eld P. Zolpiclem: revie w of its pharmacoclynami c and para ll el-group sa lety, tol erabil ity, and effi cacy study of5 , 10, and 20 111g pharmacokin etic properti es and therapeutic potenti a l. Drugs or zaleplon, compared with IO mg of zolpiclem or placebo, in adu lt outI 990 ;40:291-3 13. pati ents with insomnia. Sleep 1998;2 I(s upp1):262. I0. Hobbs WR, Rall TW, Yercloorn TA . Hypnoti cs and sedati ves; 33. Ware JC, All en R, Scharf' MB, el al. An evaluati on or res idual sedaethanol. In: Gi lman AG, consulting eel ; Hardman JG, Limbircl LE, eclstion follo wing ni ghttime adm in istration or I0 or 20 mg of za leplon, 30 in-chief; Molino ff PB, Roddon RW, eds. Gilman's the pharmacological mg o f tlura ze pam , or placebo 111 hea lthy subj ects. Sl eep bas is of th erapeutics. 9th eel. New York, NY: McGraw-Hi ll Book Co. , 1998;2 I(suppl):263 . 1996. 34. Bonnet Ml-I , Arand DL. We are chronically slee p deprived. Sleep I I. Meltzer EO. Antihi stamines and decon gestant- induced perform ance 1995; 18:908-91 I. decrements. J Occup Med I990;32 :32 7-334 . 35 . Roth T, Roehrs TA. Iss ues in the use of benzocliaze pine therapy. J 12. Nicholson AN, Stone BM . Antih istam ines : impaired perform ance Clin Psychi atry 1992;53: 14-1 8. and the tendency to sleep. Eur J Clin Pharmaco l I986;3 0:27-32. 36. Di etri ch B, Imi li en G, Salinas E. Za leplon does not produce res idua l 13. Roeh rs TA , Ti etz EL, Zori ck FJ, Roth T. Day time sleepiness and sedation in a phase-advance 111oclel o f' tran sient insomnia. 14th Congress antihi stamines. Slee p 1984 ;7: 137-141. of the European Slee p Research Soci ety; September 8-12, 1998; Madrid, 14. Kupfer DJ , Spiker DG, Co bel PA , et al. Slee p and treatm ent predi cSpain. Abstract. ti on in endogenous depress ion. Am J Psychiatry 198 1; 138 :429-434. 37. Wa lsh JK , Scharf MB, Pollak C, et al. Lack of res idual sedation fol15. Kupfer DJ , Spiker DG, Rossi A, et al. Nortripty line and EEG slee p lowing middle-of-the-night za leplon administration in sleep maintein depressed pat ients. Biol Psychiatry 1982; 17:535 -546. nance insomnia. Slee p 1998;2 1(supp1 ):247. 16. Shipl ey .IE, Kupfer DJ , Dea ly RS , et al. Differential effects o r 38 . Paty I, Danjou P, Frunc ill o R, Selmani A. lack of res idual e ffoc ts with amitriptyline and of zimelicline on th e sleep electroencephalogram of zalepl on com1;arecl with zolpi dem a fter adm ini stration 5 to 2 hours depressed patients. Clin Pharm acol Ther I984;3 6: 25 1-259. before awakening .. 2 1st Congress Co ll egium Internationale neuropsy17. Greenblatt DJ. Pharmacology of' benzodiazepine hypnotics. J Cl in chologicum (CI NP); July 12-16, 1998; Glasgow, Scotland. Abstract. Psychi atry I992;53 (supp l 6):7- 13. 39. Vermeeren A, O'H anl on JF, Danjou PE, Fournie P. Lack of residual 18. Mendelson WB. Clin ical distinctions betwee n long-acting and shorteffe cts or evening and midcl l e-of~th e~ni ght admin istration or zaleplon I0 act ing benzod iaze pines. J Clin Psychiatry I992 ;53(suppl):4-9. and 20 mg on actua l driving perform ance. 2 1st Congress Coll egiulll 19. Kupfer DJ , Reyno lds CF 111. Management of insomnia. N c ng l J Internati onale neuropsychologicum (CINP); Jul y 12-16, 1998; Glasgow, Med I997;33 6:3 41-346. Scotland . Abstract. 20. Roth T, Roehrs TA . A review o f the safety profil es o rbenzodiazepine 40. Darwish M, Danjou P, Paty I, Fournie P. Overa ll peak psychomotor hypnotics. J Clin Psychiatry 199 1;52(suppl 9) :38-4 1. and memory effocts o f zaleplon in clinical pharmacology studi es . 21 st 2 1. Ashton 1-1 Guide lines for th e rationa l use of' benzodiazepines . Drugs Congress Collegium In ternational e Neuropsychologicum (C INP); July I 994 ;48:25-40. 12- 16, 1998; Glasgow, Scot land . Abstract. 22 . Ygontzas AN , Ka les A, Bi xler EO . Benzodiazepine side e ffects. Rol e 4 1. Roeh rs T, Vogel G, Roth T. Rebound insomnia: its determinants and of pharmaco kineti cs and pharm acocly nami cs . Pharm aco logy signi licance. Am J Med I990 ;88(suppl 3A) :3 9-42. 1995;5 I :205-223 . 42 . Elie R, D'A vignon M, Emilien G, Salinas E. Zaleplon decreases slee p 23. Mendels J. Criteri a for se lection of appropri ate benzodi aze pine hyplatency in outpati ents without producing rebound insomn ia after 4 weeks noti c therapy. J Clin Psychiatry 199 1;52(suppl 9):42-46 . of treatment. 14th Congress of th e European Sleep Research Society; 24. Greenbl att DJ. Benzod iazepi ne hypnotics: sortin g th e pha rmacokiSeptember 8-12 , 1998; Madrid, Spain. Abstract. netic facts. J Cl in Psychiatry 1991 ;52( suppl 9):4-10. 43 . Lader M. Biol ogical processes in benzodiaze pine dependence. 25 . Greenbla!t DJ , Harmatz JS, von Moltke LL, et al. Comparati ve kin etAddicti on 1994;89: 141 3-141 8. ics and dynami cs of zaleplon , zo lpiclem, and placebo. Clin Pharmaco l 44. Mendelson WB . Comparative results or long-term hypnotic stu di es. Ther I 998;64: 553 -56 1. Presented at th e Ameri can Sleep Di sorders Associat ion poster session, 26. Sakamoto T, Uchimura N, Mukai M, Mizuma 1-1 , Shirakawa SI, "Awakenin g to iss ues about insomn ia" ; May 15, 1999 ; Washington , DC. Thi;; mat eri al was. cop.ie-d at th e NLM ~':\~ m aybe 1\1 . I . c1· · R t A · t M. I SLEEP Vol. 23. Supplement I, 2000 Subj ect usd>)'Yright Law~onselect1ve and Se ect1ve Benzo 1azepme ecep or gorns s- 1t er

REFERENCES

45. Woods JI-I , Winge r G. C urren t be nzodi azep ine iss ues. Psychophannaco l 1995; 11 8: I 07- 15. 46. l'etu rsso n 1-1 . T he benzodi aze pine wi th drawa l sy ndrome . Addi ctio n 1994;89 : 1455- 1459 .

This m at erial w asropie{j SL!~'EI~

Vol. 23. Supplement I. 2000

at th e N U~jfi d m ay l>Ef-Jonselective and Subj ect US'COpyright Laws

Selective Benzodiazepine Receptor Agonists-Mitler