CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2009:7:944-952

ORIGINAL ARTICLES-ALIMENTARY TRACT

Effect of Zolpidem on the Sleep Arousal Response to NocturnalEsophageal Acid Exposure

GREGG S. GAGLIARDI,' ASHISH P. SHAI,- N,4ARA GOLDSTEIN,- SUSIE DENUA_RIVERA,- KARL DOGHRAN,4JI,+

SIDNEY COHEN,- and ANTHONY J. Dll\,4AFlNO, JF-

'Divisian af Gastraenterolary and Hepatalagy, Thanas Jetferson University Haspital, Philadelphia, Pennsylvantat +Jefferson Sleep Disarders Center, ThamasJetferson Univercitv, Philadelphla, Pennsvlvania

See Editorial on page 919.

BACKGROUND & AIMS: Nocturnal acid reflux is associ-

ared t'ich complicated gastroesophageal refltu (GER) disease.

Nocturnal GER iniriates a protective arousal reflex, which in-duces a swallow to clear esophageal acid. The purpose of chis

study was to determine che effecc of zolpidem on the sleep

arousal mechanism and acid clearance in paciencs with docu-mented GER, compared with control subjects with normal acidexposure. METHODS: Eighr controls and 16 GER patiencswere enrolled in a randomized, double-blind, placebo-con-rrollrd ,rudy. Zolprdcm or placebo u:, grvcn on sepJrarrnights. Refltx events and refltrx-associated arousals or awakenings were recorded using simultaneous esophageal pH record-ing and standard polyson-rnography. RESULTS: Nocturnalacid exposure resulted in a sleep arousal 89o/o of the time inparticipants (with and wichour GER) given placebo buc only 40%

in those given zolpidem (P < .01). In controls given placebo, acidreflux events lasred 1.15 a 0.28 seconds; in conrrols givenzolpidem, they lasted 15.67 ! 12.12 seconds (P < .01). In GERpatients given placebo, the acid reflux events lasted 37.8 a 17.2

seconds compared wirh 363.3 ! 139.3 seconds witl-r zolpiderr(P < .01). \flith zolpidem reflux evencs lasted 630.6 :! 236.5

seconds when no arousal occurred and 49.2 :! 19.11 seconds

u'hen an arousal was recorded (P < .001). CONCLUSIONS:Zolpidem reduced the arousal response to nocturnal acidexposure and increased the duration of each esophagealacid reflux event in healthy individuals and patients withGER Because noctutrral acid exposure was prolonged, h1p-notic use by patients with GER could lead to increased riskfot comolicated disease.

fl,,L r.,e.ophageal reflur disease iCfRDl rs a chronrc drs-\J ord". *iih aln c.rinr.rred 20% olrhe US rdulr poprrhnonexperiencing CERD-relaced symptoms at least once a week.

Recent epidemiological srudies have srggesced that disturbedsleep occurs in as many as 75% oF patients with frequentdaycime hearrburn. Nocturnal acid reflux may be especiallydamaging because acid exposure is of longer duracion and has

been associated witl-r complications of esophagitis, includingBarrect's esophagus and adenocarcinoma.l 9

It has been shown that nocrurnal acid refltx results in a sleeparousal or awakening, which leads co a swallow reflex- Theswallowing rnechanism iniriates perisralsis which results in neu-tralization of esophageal concenrs wich saliva rich in bicarbon-ate. Studies have shown that arousals and awakenings fronsleep facilicate esophageal acid clearance bu! fragment anddisrupt sleep.ro13 Nocturnal refltx and its sleep arousal re-

sponse have been implicated in sleep disrupcion causing poordaytirne performance and productivity.'zas

lnsomnia, the complaint oF init;aring or maincaining sleep,is common, affecting 307o of che US population. Sleep aids are

commonly utilized by insonnia sufferers rvith an estimated159o of the US populacion reported using eicher a prescriptionsleep medication (87.) ar-rd/or an over tire counter (OTC) sleepaid (107o) co help them sleep.ra.rs

While hypnocic use is increasing, there is iittle informacionregarding the effect of these medications on arousals, which isan esophageal prorecrive mechanism to help clear acid in re-

sponse to noccurnal gastroesophageal reflrl (GER). Suppress-ing nocturnal arousals with hypnotic medications may lead roprolonged acid exposure and co n-rucosal injury over time.

The purpose of the present dotble-blind placebo-controlledtrial was to detetmine che simultaneous effect of a rvldelyprescribed hypnotic on nocturnal acid exposure and sleeparousal in subjects with and withouc known GER. The sftrdydemonstraces thac a hypnoric prolongs esophageal acid expo-sure wich each ref'lu-r event by the proposed mechanism ofsuppression of the sleep arousal/acid clearance mechanism.

MethodsStttd! PoPttl4ttionTwenty-four subjects were recruited on the basis of a

chart review ofesophageal pH and notility test results from theThomas Jefferson Universiry Hospiral Division of Morility. Thereview identified 2 groups of subjects. The 6rst group hadnormal 24-hour esophageal pH tesr resulrs (normal DeMeestercriteria; nornal score <14.7) and normal esophageal m:Lnom

Abbrcviations used in this paper: ACf, acid clearance time; CNS,central nervous system; GER, gastroesophageal reflux; GERD, gastrcresophageal reflux disease.

o 2009 by the AGA Institute15/t2-3565,/09/$36.00

doi r10.1o16lj.cgh.2009.04.026

September 2009 ZOLPIDEM AND NOCTURNAL ESOPHAGEAL ACID 949

etry. The second group had abnortral 2ul hour supine esophageal pH and norll1al esophageal manometry. Lighr normal and16 GER parients w€re stu.lied in a 2:l randomizarion. Allpatients had sropped acid suppression, proron inhibicor drugs7 days prior to pH tescing. Fourteen subjects from rhe GERgroup \r,ere using proron pump inhibicor medication ar thetime oF recrlrirment_ Fifteen srrbjects from rhe GER groupfeporred regular synpronls oF heartburn. No subjecr in thenornlal group reported hearcburn.

ProtocolOn che basis of dre chart review, eligible subjcccs $,ere

contacted and con-rpleted questionnaires pertaining to inclrrsion and exclusion criter.ia_ Ifsubjecrs tere inreresred in parric_iparing in rhe srudv and met inirial criceria lor stu<ly enriy, tl-restrrdy u,as Furrher explained and rhey \yere provided a srudyconsenc forrll. A compLere hiscory and physical examination u,asthen complered_ (The clinic:rl trial is regisrered in rhe r.egiscryClinicalTrials.gov wirh identifier nLimber NCTO0462137.t Th;control and parient compensarion schedule was approved inthe clinical rrials tevieu'.

Subjecrs rvere eiigible if they rvcre 18 year.s oF age and older,and had docrmenred 2,t-hour pH testing and norrral esopha_geat manometry prior ro enrollrrent in rhe srudy. Exclusioncnteria inclrrded a hisrory of symproltls suggesti,,,e of sleepapnea syndrone, resrless legs syndrome and periodic limtr11ovenenr disorder, narcolepsy, or anv defined sleep disorder.Parients lvirh any prior histow of an esophagcal morility dis,orqer. esophagogasrnc surgery, prior use oF prescribed hypnorrcs and prior complainrs oF insomnia u,ere also excluded. Sub_Je.!s lneering diagnosric crirlrria fof sleep apnea syndrome andperiodic lilnb mov€menr disorder follorving subseqrrently con_dlLcted polysorrrnography (see below) were also excluded. Sub-Jecrs \r'irh a body trass index above J5 or who were pregnantuere nor cligible for thc study.

Each subjecr under-rvenr 2 separare nocturnal sleep studiest'irh simulrancous polysomnography and esophage:rl pH resr_ing in the Jeffer.son Sleep Disorders Cenrer. Subjects 6rst ar-r;ved to rhe Thonas Jefferson Hospiral Mocilir,v Laboratory.rvhere an csophageal pH cathecer was placed transnasally inscandard fashion (Medrronic, Minneapolis, MN). The pH carh_erer had a buift in reference eleccrode. The cacheter $,as cali_brated at pH 7 and l, rrich a buffer solurion. The pH cathererlvas posirioned 5 cm abovc rhe krrver. esophageal sphincrer.Sul:jects thcn proceederl to rl-re Jefferson Sleep Laboritory_

In a double blind, cross-over fashion, borh normal subjecrsand GER pacienrs rvere administer.ecl zolpidem tarrrace (Am-blen@. Sanofi Avenris. Bridgewater, NewJersey) 10 mg or pla_cebo approxin-ratelv 30 m;nutes prior to bedtime. Drug andplacebo resembled ea.h other in color and size. Drugs wereadrninisrered by a coinvesrigator in a Fully blinded ranclomfashion. The pH probe posirion u,as confirnred prior ro beci_

ome. A standard polysomnograph montage \ras utilized rotecord nocrurnal physiological,:iata, including electroencephaIogram. eleccro oculogram, eleccromyogr:rrn of chin :rnd anre_rior tibralis muscles, electrocarciiogram, respirarorv florv anclexcursion, and transcucaneous oximetty. Follorving rhe over_nighr sleep study, rhc pH probe rvas rernoved_ A physical examlnaoon !r'as perlormed to evaluare subjects' levei of aiertnessancl salerl lor discharge to home. StLbjecrs rerurncd rvithin a2-rr'eek period co undergo a second sleep saudy using che aboveprotocol.

MeasarementsThe total number of reflux events *,as recorded, as lr,ell

as the number oF reflux-associaced arousals or arvakenings. Anacid reflux evenr uas defined as a pH heasurement <4.0 in rher-lisral esophagus. All polysomnographically de6ned arousaLsancl awakenings rhar had no ocher identifiable sollrce rhatoccurred during the acid reflux event and up to 5 minlrtes afretche pH retLLrn co ).1.0, were considered to be reflux,associatecl.Arousals or atr'akenings chat occurred in conjuncrion tirh anapnea or hypopnea event, an isolared hypoxernic episode (SaOzdecremenc of nore rhan 296 From baseline), or a periodic limbmovement rvere scored separately and u,ere not considered to bereflux-associared, and. thus, eliminated from the analysis. Jnaddirion, acid ciearance time (ACT) for each refltx evenr u,asmeasured and recorded as the number oF seconds rhe pH $,asbelow 4.0 All records rvere read and recorded in a blindecltashion. Blinding was noa broken uncil all stuclies \t,ere com_p1etecl.

Data Analysi.s

Dara were reviewed and v;lidated by a cerrified poly_somnographer. Polysomnograpic recororngs were scorect trran-ually by a rrained technician in 30-second epochs according rosrandard criceria (Rechrschaffen and Kales).r, Arousals werescored according ro rhe American Sleep I)isorders Associa.ioncriteria.rT Dara from rhe esophage:rl pH recording u,ere inte_grated into rhe polysomnographic recordings on a real timebasis utilizing rlte sanle software (Asrromed-Grass_TeleFactor,West Warwick, fu). A comparison of rhe mean acicl clearancecin-te between placebo and zolpidem adminisrrarion was examined fo r' ,s tacis rical significance by Studenr t cesr for p:1ired andunpaired sanples. The data are presenred as nean I srandarderror of rhe mean.

ResultsA roral of 2,1 sutrjecrs were enrolled in rhe srudy. Eighr

control subjecrc had normal 24ltollr pH resring and lr,ereclassi6ed as nornal subjects. Sixteen subjects hacl Jocurnenredabaormal supine esophageal acid exposure pr-ior to study en_rollment and nere classified as GER pacienrs. One CER pacient

Table 1. Baserine 24'Hour pH Resurtsa for Normar subjects and Gastroesophagear Refrux patients

PH results Normal subjects Gastroesophageal ref ux patients

Total (mean percent time pH <4.0)Supine (mean percent tirne pH <4.0)Upright (mean percent time pH <4.0)

1.96 I 0.360.59 :t 0.243.06 i O.45

9.23 1 1.8110.95 12.078.33 :t 2.O7

<.01<.01<.01

aData are given as mean : standard error of the mean.

950 GAGLIARDI ET AL

lvas eliminated frorn the stucly after he was diagnosed u'irhobsrmcrive sleep apnea on the 6rsr sleep sftldv nighr. A total oF23 subjecrs completed the study. The mean age for rhe conrrolsubjects was 41.3 years compared lr'irh 51.1 years in the CERpatienrs (P : not sr:rtisticallv significant). Among normalcontrol sllbjects. mean supine acid exposure was 0.5996 a0.24% conrpared wirh 10.95% ! 2.07a/o tn CER parients (P <.01j Table 1).

ln rl-re normal subjects there tere a total of22 acid refluxevents. or 2.8 acid reflux events per sleep study (95% confidenceinten'al, 1.7 4.2). l.-ifteen occurred during placebo adninisrration. Scvcn acid reflux evenrs occurred du|ing zolpidem admin-istrarion. ln rhe GER patienrs there lvere a cotal of 125 refluxevents or 8.3 acid reflux events per sleep study (95"/" confidenceinterval. 6.8 9.8). Seventy'one occurred with placebo administration rl4rile 54 occurred g,ith zolpidem. C-hanges in number ofacid reflrx events in ea.h group \r,as not significanr (P > .05)rvidr zolpidem.

ReJIoox-As sociate d AroasalsArnong all 23 suLrjecrs (controls and GER), a nocturnal

acid reflux evenr led to an arousal or atakening 8996 ofthe rimewhen pla.ebo was adminiscered. ln comparison. as shown inIjigure l, reflux ever-rts \r,ere associated \r'ith an arousal or alr'akening only 40Yo of the time on nights when zolpidem was

administered (P < .01). Absence of rhe arousal response was

noted in the first part of the evening (6rst 3 hours post zolpidem) with an intact arousal response prior to morning arvak-ening. Thus, che arousal response was presenc in borh groupsregardless of rhe dtration of acid exposure suppressed by zol,pidem.

Esophageal Acid. Clearance Times

Mean esopl-rageal acid cl€arance cine for an acid GERevent was measured in the GER patiencs and norlral slLbjecrs

!r'hen given placebo versus zolpiden-r. ln rhe 8 norrnal subjeccs,rhe mean ACT for each nocrurnal acid reflux event rvas 1.15 l:0.28 seconds. On the nights when zolpiderTr was adninisreredas sho$n in Figure 2, the mean ACT For an acid CER event rvas

15.67 1 13.42 seconds t'frich u,as significanrly higher whenconpared wich placebo (P < .01).

BO

Placebo ZolpidemFigure 1. Percent of GER events assocated with an arousa or awakening as a functon of placebo vs zopder| use. n the presence olzolpidem, there was a rnarked reduction in reflux associated seeparousas n the normal subjects and GEF patents (P < .0T).

cLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol.7. No.9

Placebo ZolpidemFigure 2. Mean ACT (in seconds) for a nocturna GER event in thecontro sublects with normal nocturna acd exposure when laking placebo vszo pidem. ACTwas sgnificanUy proionged wth zo pidem com-pared wth pacebo (P < .05). These physio ogica Teflux events in norma sublects, athough proonged with zolpdem, st lreman relatvelybrief when compared wth those of the GER pat ents.

As shorvn in Figure 3A. in the GER patients rhe lrrean ACTlor.r grverr r'flrLX r\, r'r rvlr,.n .iren placcbo u.r. \-.-9 -: l- 25

seconds compared wirh 363.3 | 139.29 seconds when zolpidernwas given (P { .01). Thlls, noccurnalacid exposure tas dramat-ically increasecl per each event over rhe enrire sleep period.

To Further exanine the inportance oF the arousal responsein acid clearance, we examined the dac:r for acid clearance rimesin GER pacienrs in che presence or abscnce of an arousal rvichzolpiden as seen in Figure 38.

In the GER patienrs rhe nean ACT in response to an acidCER event lr.hen an arousal occurred \r'as ,19.18 I 19.11 sec-onds compared with 630.58 :! 236.52 se.onds $,hen no arousaloccurred (P < .001). The more malked acid exposure times \r'ithzolpidem were seen in rhe first 3 hours of sleep t'hen rhearousal response rvith acid exposure was infrequenr.

DiscussionThe presenc scudy indicares that a nedication com-

rnonly used to induce sleep prolongs nocrurnal esophageal acidexposure in conjuncrion *'irh inhibicion of rhe centrally mediated, sleep arorLsal acid clearance nechanism. It has been pre-viously demonscrated that nocturnal esophageal acid reflux iscieared by an esophageal iniriared central nervous sysrern (CNS)reflex. Hydrogen ions initiare the CNS response which rhencauses a sleep arousal and a su,allorving response. This sleeparousal mechanism was inicially demonstrated in normal sub-jects using 15 mL acicl perfusions of different pH.r1 The sleeparousal mechanism in our srudv was not conrrolled becalrsevolume oF acid refluxare and pH were clinical events and norexperimentally regul:rred. However, prior studies have collectedreflux volunres oF 2 7 mL per horr in che fasrillg state.l3Despire these srudy differences, pathologic: reflrrr evenrs ofvarying volumes

'nitiate acid clearan.e through rhe CNS-medi-

ated alousal rcsponse simiiar ro the earlier reports-rlThis srudvalso indicated thar rhe vely brief "p hysiological" reflux event innormal subjects may also iniriace an arousal response.

The duration oF esophageal acid exposure during sleep hassignificant implicarions on rhc developnenr of complicated

30

25

2A

15

10

70

60

30

20

10

Placebo

89o/"

September 2009

Figure 3. F) N,/ean ACT (n seconds) for a nocturnareflux event in the GER patents when tak ng pacebo vszo p dem. Zo p dem was associated wth proongationof ACT n patents with abnorma screening pH studies(P < .01). (B) Mean ACT (in seconds) for a nocturnalrefux event n the GER patents n the presence or ab-cor e o .r clro, d

^ rtr opden. lh- doc6^ e or d

sleep arousal with zolpidern ed to a marked proongation of ACT with each reflux event (P < .001).

GIJRD including elosive esophagiris. srricture Formarion, Bar-rett's €sophagus and esophageai adenocarcinorna.l 6.1e.20 lr isuell escablisl-red that esophageal acid exposure during sleep isprolonged in both normal subjects and in parients wich CERD.and that arorLsals and ar.vakenings in response to acid reflux arecritical ro enable acid clearance during sleep-7!)12.21 Orr er alhave previously clemonscrared thar slcep inpairs esophagealacid clearance cime and rhar acicl clearance occurs ln assoctanoltrvith arousals From sleep.T Thus. facrors that suppress arousalsan,-l arvakenings rnal fllrther prolong esophageal acid conra.rtime. Insonrnia, or self-described r-lisrurbed sleep, is a commonconplaint in rhe US aclult popularion.

In rhis stlrdy, rve Found rhar zolpiden. a conrnonly pre-sclibed sleep-inducing hypnotic, suppressed nocmrnal arousalsand arvakenings in lcsponse co acid reflru events which resuftedin plolonged esophageal acicl clearance rime. Thus. zolpidemhaci che effecr of enabling suLrjecrs to "sleep chrough" reflurevenrs. thereby increasing nocturnal acid exposure.

Zolp;dem is an imidazopyridine agent rhar is an agonisr arthe beozodiazepine recepror component of rhe 7 arninobutyricacid d-receptor colrrple-\. Zolp;dem enhances the inhibirorveTccr olt y .rurr,.'buLr rlc .r.,ci "', n, rrr ral excrr.rr orr rrr, di;LirSics hypnotic cfFects. Ir h:ls been shown ro redlLce rime to onserand co prolong the duration of sleep withouc affecring orhL'raspects of sleep archirecrure. The drug is rapicllv absorbed,reaching n&\il11unl concentration in the serum at 1,6 hoursafrer administration. Ir is cxrensivclv nerabolized by rhe cyco-chrome P450 isoenzyrnes, resulting in a short elimination half-Iile of 2.5 hours. Therefore, psvchomotor and nemory impair-n1enr is limited ro rhe first few hours afcer adminisrracion andthese effecrs are generallv noc obsen'ed ar 6 hours afrer adniniscration.lz This shorc durarion of act;on of zolpidem mayac.ounc For rhe no|e marked suppfession oF che arousal re-sponses eariy in the evening, rvhen CI-R evenrs are most com-trton, wich rerurn ro nomral in che period beFore morningarvakening.zl This efFect of zolpitlem rvas also seen in normalsubjects u4rere prolongation in acrd exposure was recorded,albeir stiLl brief in duration compared wirh the reflux parientsrvho were selected on rhe b:rsis of prior abnorrn:rl pH studies.

Singh ec al have previously shown thar noccurnal adrniniscration of rhe benzodiazepine, alp|azolam, prolongs esophagealacid .learance ti111e versus placebo.2r Hou'ever, polysomnogr:rpfry *as nor utilized and therefore it is unclear from dris studyrvlrat eFFecc alprazolun had on arousals and arvakenings andoverall sleep:uchitecrure. Fass er ai, in a longirudinal srudy oFover 15,000 sllbjects in rhe Narional Instirrrtes of Health SleepHeart Hcalch srudy, found benzodiazepine use, along withinson-rnia, bocly mass index, sofr drink consunption,:rnd hy-perrension to be scrong preclictors oF heartburn drLring sleep.l5

ZOLPIDEITI AND NOCTURNAL ESOPHAGEAL ACID 951

Placebo Zolpidem

C)ur scudv objectively' denonstraced. rrsing polyson-rnographv.that zolpidem impairs arousals. This is likely a resulr ol :rblunting of the CNS response co acid in rhe distal esophagus.Failure to induce an arousal prevents a prorecdve swalLo\r'ingreflex. As a resLllt. rhere is increased esophageal acid exposure.uhich may lead ro addirional cornplications of CERD. Sinilarly. other CNS depressants or ps1'chotropic nredicarions mayalso have rhis effect.

This mechaniscic srudy may have clinical relevance. Therr crd. rcc of esoplrrg, :l adcnoc.,r'crn,,rna c^n r il| , s r, \ 'nr r..lscin mosr of the developed world.r, The cause of rhis rise is srillnoc cLear. Ho$,ever, CERD. and especiallv Barrett's esopha!!s.is a fecognized precursor lesion. Increasingly. Americans areusing sleep aids on a Frequenr or rcgular basis.lrThis srudydemonstrates the CNS bltrnring eFfecas of such rnedicarions onthe arousal response to acid in rhc distal esophagts drar prolongs acid exposure. This increased acid exposLlre afrer hvpnot,ics occurs boch in normal subjects trrt cven nrore so in refluxpatienrs, !r'hich would be expected to increase mucosal damag-ing efFects. The inc|eased acid exposure in pacienrs with GERDmay contribute. along v,ith other faccors, ro complicaced gasrroesophageal refltrx disease.

To prove a long term delererious cffecc of hypnoric use ongas.roesophageal reflux clisease will require an epiderniologicalcohort study in a large popuLation base over time. However, asmany as 15o/o 3096 ofpadenrs rvith disturbed sleep may h:rfboruncliagnosed GERD.2r.16 27 If rhis effect of blrrnred arousals orawakenings by frypnorics noted in chis sturly is subsranriared.d-ris t,ordd suggesr caution in the use ofsleep aids uithout firsrconsidering GERD as an eriologic fxctor in patienrs rvirh complainrs of disturbed sleep.']s.'?s.l'l

One possible limirarion of rhis study is a function of cherelatively small san,ple size. Howevet. ue ninimized stLch el.-

fects by analyzing dara by evenrs and not by subjecr. In additionchere are linitations inrrinsic in any study using polysornnog-rapl-rv. Slecp pacrerns lrrav vary from nighr to night in a givensLLbject. In addition, previous sruclies which rrtilizecl polysomnography. have demonsrrated thac an adaprarion effecr is notedafter the initial sleep srudy. characterized by improvemenr insleep conrinuity measures in srLbsequenr sleep studies.rrr rr Bvrandonizing che patients ro placebo or zolpidem on differenrnigbcs we arrenpted co limir rhis effect.

In conclusion, rhis srtLdy. uriltzing objeccive paralneters ao

measure acid leflux and sleep nechanics demonstrared drathypnoric medication use in pacienrs s,ith and l,ithour estab-lished noccurnal GER prolongs esophageal acid conract time byimpairing reflui associated arousals and alakenings. H1'pnoticmedicarions should be used wich caution in parienrs withknouer supine cERD. GERD should be suspecred and invesri,

600

500

300

200

100

0

t-952 GAGLIARDI FT AL

gated as a cause in patients wirh insomnia or disrurbed sleep.Furthermore, use of hypnoric medicarions shorld be restrictedduring 2,1 hour pH testing to avoid false posirive results.

References

1. Locke G lli, Talley NJ, Ferl S, et al. Prevalence and clinical spectrumof gastroesophageal reflux: a populatjon-based study in OlmstedCounty, tulinnesota. Gastroentero oey 1997 t!121t448-1456.

2. Fatup C, Kleinman L, Sloan S, et al. The impact of nocturnalsymptoms associated with gastroesophageal reflux disease onhealth-re ated quality of life. Arch Intern Med 2001;161:45 52.

3. Johnson DA, Orr WC, Crawley JA, et al. Effect of esomeprazole onnighttime heartburn and sleep quality in patients with GERD: arandomized, placebo{ontrolled trial. Am J Gastroenterol 2005;704'.7974 7922.

4. Shaker R, Castell D0, Schoenfeld PS, et al. Nighttime heartburnis an underappreciated clinical problem that jmpacts sleep anddaltime functionr the resu ts of a GaLlup SuNey conducted onbehaf of the American Gastroenterological Association. Am J

Gastroenterol 2OO3:98'. J-487 -t493.5. Orr WC. S eep and gastroesophageal reflux disease: a wake-up

cal . Rev Gastroenterol Disord 2OO4;4 Suppl 4t525-532.6. Lagergren J, Bergstron R, Lindgren A, et al. Symptomatic gastro-

esophageal reflux as a risk factor for esophageal adenocarci-noma. N Engl J l,4ed 1999;340:825-831.

7. Ofi WC, Johnson LF, Robinson I\,4G. Effect of s eep on swal owing,esophageal perista sis, and acid c earance. Gastroenterology1984;86:814-819.

8. Orr WC, Allen l\y'1, Robinson N4. The pattern of nocturnal anddiurnal esophageal acid exposure in the pathogenesis of erosivernucosal damage. Am J Gastroenterol 1994;89:509-512.

9. Orr WC. Sleep and gastroesophageal reflux: what are the risks?Am J Med 2003;115 Supp 34:1095 1135.

10. OrrWC, Robinson MG, Johnson LF. Acid clearance during sleep inthe pathogenesis of reflux esophagitis. Dig Dis Sci 1981;261423 427.

11. Orr WC, Johnson LF. Response to diflerent leve s of esophagealacidifcation during waking and sleep. Dig Dis Sci 1998;43:241-245.

12. ofi WC, Elsenbruch S, Harnish MJ, et al. Proximal migration ofesophageal acid perfusion durjng waking and sleep. Am J Gastroentero 2000;95:37 42.

13. Dickman R, Green C, Fass S, et al. Relationships betlveen sleepquality and pH monitoring frndings in persons with gastroesophageal reflux disease. J Clin Sleep Med 2007;3:505,513.

14- National Institltes of Health. National Institutes of Health Stateof the Science Conference statement on Manifestations andManagement of Chronic Insomnia in Adults,.June 13-15, 2005.Sleep 2005;28r1049-1057.

15. Sleep in America Poll, National Sleep Foundation, 2002. Avail-able at www.sleepfoundation.org. Accessed April 23, 2009.

16- Rechtschaffen A, Kales A. A manua of standardized terminolo€l/,techniques, and scoring system for sleep stages of human sub-jects. Los Angees: UCLA Brain Information Service/Brain Re,

search Institute, 1968.17. EEG arousa s: scoring rules and examplest a preliminary reporl

from the Sleep Disorders Atlas Task Force of the American SleepDisorders Association. S eep 1992;15:173-184.

18. SifrimD,Mitta R, Fass R, et al. Review articlet acidity and volume

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7. No. 9

of the refuxate in the genesis of gastro-oesophageal refluxdisease symptoms. Aliment Pharmacol Ther 2007;25:1003-to77 .

19. Adachi K, Fujishiro H, Katsube T, et a. Predominant nocturnalacid refux in patients wjth Los Angeles grade C and D refluxesophagitis. J Gastroenterol Hepatol 2001;16:1191 1196.

20. Orr WC, Lackey C, Robinson N4G, et ai. Esophageal acid cearance during s eep in patients with Barrett's esophagus. Dig DisSci 19aa;33:654 659.

21. Drnarlo fu Jr. Bdnwdit KS. Eschtrge. L. e' al. Tl_e effect ofgastro-oesophageal ref ux and omeprazole on key s eep param-

eters. Aliment Pharmacol lhet 2005:221325-329.22. Holm V,J, Goa KL. Zolpidem: an update of jts pharmacology,

therapeutic efficacy and tolerability in the treatment of insomnia.Drugs 2000;4:865 489.

23. Hila A, Castell D0. Nighttime refux is primarily an early event.I Clin Gastroentero 2005139:579 583.

24. Singh S, Bailey RT, Stein HJ, et al. Effect of alprazo am (Xanax)esophageal motility and acid reflux. Am J Gastroenterol 1992;B7:483-488.

25. Fass R, Qlan SF, o'Connor cT, et al. Predictors of heartburnduring seep in a large prospective cohort study. Chest 2005;127r1658-1666.

26. 0rrWC, Goodrich S, Robertl. The effect of acid suppressjon onsleep patterns and sleetrrelated gastro oe so ph ageal reflux. Aliment Pharmaco Ther 2005;21:103-108.

27. Guda N, Partington S, Shaw M, et al. Unrecognized GERD symp-toms are associated with excessive daltime sleepiness in pa,tients undergoing sleep studies. Dig Dis Sci 2OO7;522a73247 6.

28- Johnson DA. Gastroesophageal reflux disease and sleep djsorders: a wake'up call for physicians and their patients. Rev Gastroenterol Disord 2005;5 Supp 2:S3 S11.

29. Bruley des Varannes S, Errieau G, Tessier C. Two thirds ofpatients with gastroesophageal reflux have nocturna symptoms:survey by 562 general practitioners of 36,663 patients [inFrenchl. Presse fMed 2O07;36t591-597.

30. Agnew HW.Jr, Webb WB, Wi iams RL. The first night effect: anEEG study of sleep. Psychophysiolog)i 19662:263-266.

31. Browman CP, Cat|wright RD. The first-night effect on s eep anddreams. Biol Psychiatry 1980;15:aO9-812.

32. Curio G, Ferrara M, Piergianni A, et al. Paradoxes of the irst nighteffect: a quantitative anaiysis of antero-posterior EEG topogra,phy. Clin Neurophysiol 2A04:775:717 B -7t88.

Repfint requestsAddress requests for reprints to: Anthony J. Dilvlarino, Jr, lvlD, Divi-

sion of Gastroenterology and Hepatology, Thomas Jefferson UniversityHospital, 132 South 10th Street, Suite 480 Main, Philadelphia, Penn-sylvania 19107. e-mail: anthony.dimarino@jefferson.edu; fax: (215)955-O472.

Conflicb of intercstKarl Doghramji, MD, serues as a consultant for Sandolfi-Aventis. The

remaining authors disclose no conflicts.

FundingThis research was supported mainly through a grant from the Com-

monwealth of Pennsylvania with approximately 2070 support from thelnvestigatols Sponsored Study Program of Astrazeneca.