www.revportpneumol.org
ORIGINAL
ARTICLE
Patients
with
a
high
risk
for
obstructive
sleep
apnea
syndrome:
Postoperative
respiratory
complications
H.
Pereira
a,
D.
Xará
a,
J.
Mendonc
¸a
a,
A.
Santos
a,
F.J.
Abelha
a,b,∗aDepartmentofAnesthesiology,CentroHospitalardeSãoJoão,Porto,Portugal
bAnesthesiologyandPerioperativeCareUnit,SurgicalDepartmentofMedicalSchoolofPorto,Portugal
Received3October2012;accepted30January2013 Availableonline2June2013
KEYWORDS Obstructivesleep apnea; STOP-BANGscore; Surgery; Postoperative complications Abstract
Background: STOP-BANG score (snore; tired; observed apnea; arterialpressure; body mass index;age;neckcircumferenceandgender)canpredicttheriskofapatienthaving Obstruc-tiveSyndromeApnea(OSA).TheaimofthisstudywastoevaluatetheincidenceSTOP-BANG score≥3,insurgicalpatientsadmittedtothePost-AnesthesiaCareUnit(PACU).
Methods:Observational, prospectivestudy conducted inapost-anesthesiacare unit(PACU) duringthreeweeks(2011).Thestudypopulationconsistedofadultpatientsafternoncardiac andnon-neurologicalsurgery.PatientswereclassifiedashighriskofOSA(HR-OSA)ifSTOP-BANG score≥3andLow-riskofOSA(LR-OSA)ifSTOP-BANGscore<3(LR-OSA).Patientdemographics, intraoperativeandpostoperativedatawerecollected.Patientcharacteristicswerecompared usingMann---WhitneyU-test,t-test forindependent groups,andchi-squareorFisher’s exact test.
Results:A totalof357patients wereadmittedtoPACU;340mettheinclusioncriteria.179 (52%)wereconsideredHR-OSA.Thesepatientswereolder,morelikelytobemasculine,had higherBMI,higherASAphysicalstatus,higherincidenceofischemicheartdisease,heart fail-ure,hypertension,dyslipidemiaandunderwentmorefrequentlyinsulintreatmentfordiabetes. Thesepatientshadmorefrequentlymild/moderatedhypoxiainthePACU(9%vs.3%,p=0.012) andhadahigherincidenceofresidualneuromuscularblockade(NMB)(20%vs.16%,p=0.035). PatientswithHR-OSAhadalongerhospitalstay.
Conclusions:PatientswithHR-OSAhadanimportantincidenceamongpatientsscheduledfor surgeryinourhospital.Thesepatientshadmoreco-morbiditiesandweremoreproneto post-operativecomplications.
© 2012SociedadePortuguesa dePneumologia. Publishedby Elsevier España,S.L.All rights reserved.
∗Correspondingauthor.
E-mailaddresses:fernando.abelha@gmail.com,abelha@me.com(F.J.Abelha).
0873-2159/$–seefrontmatter©2012SociedadePortuguesadePneumologia.PublishedbyElsevierEspaña,S.L.Allrightsreserved. http://dx.doi.org/10.1016/j.rppneu.2013.01.003
CORE Metadata, citation and similar papers at core.ac.uk
PALAVRAS-CHAVE Apneiaobstrutiva dosono; Pontuac¸ão STOP-BANG; Cirurgia; Complicac¸ões pós-operatórias
DoentescomaltoriscodeApneiaObstrutivadoSono:complicac¸õesrespiratórias pós-operatórias
Resumo
Introduc¸ão: Apontuac¸ãonoSTOP-BANG(Snore,Tired,Observedapnea,bloodPressure)pode preveroriscodeumdoenteterSíndromedaApneiaObstrutivaSono(SAOS).Oobjetivodeste estudofoiavaliaraincidênciadapontuac¸ãoSTOPBANG>3,empacientescirúrgicosinternados naUnidadedeCuidadosPós-Anestésica(UCPA).
Métodos: Estudo observacional e prospectivo conduzido numa UCPA, durantetrês semanas (2011).Apopulac¸ãodeestudoconsistiuemdoentesadultosapóscirurgianãocardíacaenão neurológica.OsdoentesforamconsideradoscomaltoriscodeSAOS(AR-SAOS)setinhamum scorede STOP-BANG≥3edebaixoriscodeSAOS(BR-SAOS)setinhamscoredeSTOP-BANG <3. Foramavaliadosdados demográficosdosdoentesecolhidas variáveisintraoperatóriase pós-operatórias.Ascaracterísticasdosdoentesforamcomparadasatravésdotestede Mann-Whitney,testet,qui-quadradooutesteexatodeFisher.
Resultados: Umtotalde357doentesforamadmitidosdeUCPAe340preencheramoscritérios deinclusão.Centoesetentaenove(52%)tinhamAR-SAOS.Estesdoenteserammaisvelhos, tinham maiorprobabilidadede serem dosexo masculino, tinhamum Índice MassaCorporal superior,tiveramumaclassificac¸ãomaiornoestadofísicoAmericanSocietyAnesthesiologists, umamaiorincidênciadedoenc¸acardíacaisquémica,insuficiênciacardíaca,hipertensão, dis-lipidemia eerammais frequentementedoentescomdiabetes em tratamentocominsulina. Esses doentestiverammaisfrequentementehipóxialeve/moderado naUCPA(9%versus3%, P=0,012)etiveram maiorincidênciadebloqueio neuromuscularresidual (NMB)(20%versus 16%,P=0,035).OsdoentescomAR-SAOStiverammaiortempodeinternamentohospitalar. Conclusões: OsdoentescomAR-SAOStemumaaltaincidênciaentreosdoentessubmetidosa intervenc¸õescirúrgicasprogramadasnonossohospital.Essesdoentestinhammais comorbili-dadeseforammaispropensosatercomplicac¸õespós-operatórias.
© 2012SociedadePortuguesa dePneumologia. PublicadoporElsevier España,S.L.Todosos direitosreservados.
Introduction
Obstructivesleepapnea(OSA)isacommonmedical condi-tion affecting 2---26% of the general population and it is the mostprevalentbreathing disturbancein sleep.1,2 OSA
is characterized byrepeated complete or partialcollapse ofthepharyngealairwayduringsleep,causingcessationof airflow(apnoea)orshallowbreathing(hypopnea).This pat-ternofsleeparousal,coupledwithintermittenthypoxemia, isassociatedwithseriousadversecardiovascularoutcomes, including strokes, and leads to daytime somnolence and compromisedneurocognitivefunction.3
TheprevalenceofOSAinthesurgicalpopulationishigher thaninthegeneralpopulationanditcanvarywidely accord-ingto theunderlying medicalcondition.3 Inparticular, as
manyas70%ofpatientsundergoingbariatricsurgerywere foundtohaveOSA.4OSAhasbeenrecognizedasapotential
independentriskfactorforadverseperioperativeoutcome.5
Compared to non-OSA patients, OSA patients undergoing surgicalproceduresarevulnerabletopostoperativeairway obstruction,6myocardialischemiacongestiveheartfailure,
strokeandoxygendesaturation.5,7,8
The goldstandardfor thediagnosis ofOSA, polysomno-graphy (PSG), is impracticable as a routine preoperative assessmenttoolforOSAbecauseitisanexpensiveandlabor intensivetest.5Manytoolshavebeenproposedforscreening
patientsforOSAsuchastheBerlinquestionnaire,theSTOP questionnaire and theAmerican Society of Anaesthesiolo-gists(ASA)checklist---andtheuseofthesetoolsimproves
thelikelihoodofidentifyingOSApreoperatively.3TheSTOP
questionnaire,showeda highsensitivityfordetecting OSA (whenscoreis≥3):93%and100%formoderateandsevere OSA,respectively;however,thespecificityatthesame cut-off is low: 47% and 37% for moderate and severe OSA, respectively,resultinginhighfalse-positiverates.TheSTOP questionnaireisascoringmodelconsistingof8easily admin-istrated questions starting with the acronym STOP-BANG
(Appendix)andis scoredbasedonYes/Noanswers(scores
1/0).1
The aimofthisstudy wastoevaluate theincidenceof patients witha STOP-BANG score≥3, in surgical patients admittedinthePost-AnesthesiaCareUnit(PACU).
Methods
Subjectsandsettings
Ethicalapproval for thisstudy wasprovidedby theEthics CommitteeofCentroHospitalarSãoJoão,Porto,Portugal. Written informed consent was obtained from all partici-pants.
Centro Hospitalar São João,Porto, is an 1124-bed ter-tiary hospital in a major metropolitan area that serves 3,000,000 people. This prospective study was conducted ina 12-bed PACUover a 3-week period (fromMay 9th to May27th,2011).EverypatientadmittedtothePACUwho wasabletoprovidewritteninformedconsentwasincluded
inthestudy.Exclusioncriteriawerepatientrefusal, inca-pacity to provide informed consent, a score of <25 in themini-mentalstateexamination(MMSE),under18years ofage, foreignnationality, knownneuromusculardisease, urgent/emergentsurgeryandcardiacsurgery,neurosurgery orotherproceduresthatrequiredtherapeutichypothermia. AllthepatientswereaskedtocompletetheSTOP ques-tionnaire. Information concerning body mass index (BMI) age,neckcircumference,andgender(Bang)werecollected byaresearchassistant.
Patientswereclassifiedasbeingathighriskfor obstruc-tivesleepapnea (HR-OSA)iftheirSTOP-BANG scorewas3 orgreaterandasbeingatlowriskofOSA(LR-OSA)iftheir scorewaslessthan3.
The anesthesiologist in charge was blinded to patient involvement in the study. Anesthesia was provided and monitoredaccording totheanesthesiologistin charge cri-teria, but this conduct followed minimum departmental standards.Neuromuscularblockingdrugs(NMBD)wereused for tracheal intubation, and additional boluses were pro-vided, if needed. No written policy exists concerning the use of neuromuscular monitoring so this was per-formed at the discretion of the anesthesiologist. To ensure that the anesthesiologist remained blinded to the patients’ participation in the study, we did not attempt to observe the use or interpretation of TOF intraoperatively. The anesthesiologist was free to decide whether to reverse the neuromuscular blockade (NMB) withneostigmine at the conclusionof the surgical proce-dure.
Usually,thepatientwasextubatedintheoperatingroom andtransferredtothePACU.Criteriaforextubationincluded sustainedheadliftorhandgripformorethan5s,the abil-itytofollowsimplecommands,astableventilatorypattern withacceptablearterialoxygensaturation(SpO2)>95%,and aTOF ratiogreater than 0.80. Allsubjects were adminis-tered100%oxygenbyafacemaskaftertrachealextubation. Theanesthesiologistwasfreetodecidewhetherto admin-isteroxygenduringthe timebetween transfertothecart andadmissiontothePACU.
UponarrivalatthePACUallsubjectswereprovidedwith oxygeneitherbyanasalcannulaorfacemask.
Astandardizeddatacollectionsheetwascompletedfor each patient. The recorded patient characteristics were: age,weight,height,bodymassindex(BMI),benzodiazepine administrationbeforesurgery,chronicbenzodiazepineuse, site of surgery (intra-abdominal, musculoskeletal or head andneck), American Society of Anesthesiologists physical status (ASA-PS), Revised Cardiac Risk Index (RCRI), drug or alcohol abuse, duration of preoperative fluid fasting, type of anesthesia, duration of surgery, use of nitrous oxide,adverserespiratoryeventsinthePACU,postoperative pain level (VAS score), postoperative nausea and vomit-ing6hoursafter surgeryand 24h after surgery, length of stayinthePACU,andpostoperativelengthofstayin hospi-tal.
The surgical procedure wasclassified interms of mag-nitude,asmajor(surgeryinwhich bodycavities or major vesselsareexposedtoambienttemperaturesuchasmajor abdominal, thoracic, or major vascular, thoracic spine surgerywithinstrumentation,orhiparthroplasty),medium (surgery in which body cavities are exposed to a lesser
degree suchasappendectomy), andminorsurgery (super-ficial surgery). Major surgery was defined as a surgery requiringahospitalstayof2ormoredays.
Clinical risk factors (a history of ischemic heart dis-ease,historyofcompensatedorpriorheartfailure,history of cerebrovascular disease, diabetes mellitus, and renal insufficiency)andsurgicalrisk(high-riskdefinedas intratho-racic,intraperitoneal,orsuprainguinalvascularsurgery,or surgeryinvolvinglargebloodlossorfluidshifts)weredefined accordingtotheCardiac RiskStratificationfor Noncardiac SurgicalProceduresofthe2007guidelinesonPerioperative CardiovascularEvaluationandCareforNoncardiacSurgery oftheAmericanCollegeofCardiology/AmericanHeart Asso-ciationTaskForceonPracticeGuidelines.9
Data for other preoperative clinical information regarding chronic obstructive pulmonary disease (COPD), hypertension and dyslipidemia were collected from rou-tine clinical documentation entered into the institution’s perioperativeclinicalinformationsystem.
Residual neuromuscular blocking was defined as a TOF<0.9 and it wasquantified at admission to the PACU using acceleromyography of the adductor pollicis muscle (TOF-Watch®).10,11
The patients were screened for delirium at PACU dis-charge and in the ward the next day after surgery using the nursingdelirium screening scale (Nu-DESC).12 Patients
witha Nu-DESC score of 2or morepoints during at least at1evaluationwereconsidereddeliriumpositive.Patients weretestedfordeliriumbytheresearchteamatthetime theywereformallydeclaredtobe‘readyfordischarge’to theregularwardbythephysicianinchargeoftherecovery room.Inaddition,thepatientswereseenonthemorningof thefirstpostoperativeday.
Postoperativerespiratorycomplications
EachpostoperativeAREwasdefinedonthedatacollection sheetusingthefollowing criteriaaccordingtothe classifi-cationdescribedbyMurphyetal13:
1. Upperairwayobstructionrequiringanintervention(jaw thrust,ororalornasalairway);
2. Mild-moderatehypoxia(SpO2of93---90%)on3Lnasal can-nulaO2thatwasnotimprovedafteractiveinterventions (increasingO2flowsto>3L/min,applicationofhigh-flow face mask O2, verbal requests to breathe deeply and tactilestimulation);
3. Severehypoxia(SpO2<90%)on3LnasalcannulaO2that wasnotimprovedafteractiveinterventions(increasing O2flowsto>3L/min,applicationofhigh-flowfacemask O2,verbalrequeststobreathedeeply,andtactile stim-ulation);
4. Signsofrespiratorydistressorimpendingventilatory fail-ure(respiratoryrate>20breathsperminute,accessory muscleuse,andtrachealtug);
5. Inability to breathe deeply when requested to by the PACUnurse;
6. Patientcomplainingofsymptomsofrespiratoryorupper airwaymuscleweakness(difficultybreathing, swallow-ing,orspeaking);
8. Clinical evidence or suspicion of pulmonary aspiration aftertrachealextubation(gastric contentsobservedin theoropharynxandhypoxemia).
The PACU nurses observed the patients continuously during the PACU stay and contacted a study investigator immediatelyifanAREwasobserved.Theinabilitytobreathe deeply and theassessment of symptomsof respiratory or upperairwaymuscleweaknesswereassessedat10-minute intervals.Astudyinvestigatorexaminedthepatientto con-firmthatthepatientmetatleastoneofthecriteriaforan ARE.
Statisticalmethod
Descriptive analysis of variables was used to summarize data. Ordinal and continuous data found not to follow a normaldistribution,basedontheKolmogorov---Smirnovtest fornormalityoftheunderlyingpopulation,arepresentedas medianandinterquartilerange.Normallydistributed data arepresentedasmeanandstandarddeviation(SD).
A univariate analysiswas performed toidentify differ-ences between HR-OSA and LR-OSA patients by using the Mann---WhitneyU-testtocomparecontinuousvariablesand the Chi-square or Fisher’s exact test to compare propor-tions between the2 groups of subjects.Differences were consideredstatisticallysignificantwhenPwas<0.05.
Aunivariateanalysiswasperformedtoidentifypredictors forARE.
Multipleregressionbinarylogisticwasusedwithforward conditionalmethodinthemodelinordertoidentify inde-pendent predictors for ARE. In this model, all covariates withp<0.05intheunivariateanalyseswereenteredandan oddsratio(OR)and95%ConfidenceInterval(95%CI)were calculated.
DatawereanalyzedusingSPSSsoftwareforWindows Ver-sion19.0(SPSSInc.,Chicago,IL,USA).
Results
Fromthe357patientsconsecutivelyadmittedtothePACU duringthestudyperiod,atotalof340patientswerestudied. Seventeenpatientswereexcluded:7patientswere admit-tedtoasurgicalintensivecareunit,3patientswereunable toprovideinformedconsentorhadaMMSE<25,3patients didnotundergosurgery,1patientunderwentneurosurgery, 1patientwaslessthan18yearsold,1patientdidnotspeak Portugueseand1patientrefusedtoparticipate.
Ofthe340patientsincludedintheanalysis(Table1),179 (52%)wereclassifiedasHR-OSA.Thesepatientswereolder (meanageof61.7±13.4vs.48.8±16.0yearsandmedian age63vs.47years,p<0.001),weremorelikelytobemen (63%vs.21%,p<0.001),hadhigherBMI(mean29.1±6.0vs. 25.1±4.4kg/m2,median28vs.24kg/m2,p<0.001)higher ASAphysicalstatus(ASAIII,IVorV26%vs.10%,p<0.001) andhigherRCRIscores(RCRI>2in7%vs.2%).Patientswith HR-OSA had a higher incidence of ischemic heart disease (10%vs.2%,p=0.001),heartfailure(8%vs.3%,p=0.032), hypertension(62%vs.19%,p<0.001),dyslipidemia(41%vs. 12%,p<0.001)andunderwentmorefrequentlyinsulin treat-mentfordiabetes(24%vs.4%,p<0.001).Theprevalenceof
cerebrovasculardisease(4%vs.1%,p=0.116)and preoper-ativeserumcreatinine>2mg/dl(6%vs.4%,p=0.394)were notsignificantlydifferentbetweenthe2groups.
HR-OSA patients were more likely to undergo intra-abdominalsurgery (43%vs. 29%, p=0.011) andsurgery of majormagnitude (32% vs. 22%, p=0.033), had more fre-quentlylocoregionalanesthesiaasauniquetechnique(26% vs.14%,p=0.009)andlessfrequentlyhadgeneralor com-bined general and loco regional anesthesia (74% vs. 86%, p=0.009).PatientsclassifiedasHR-OSAhadahigher inci-denceofresidualNMB(RNMB)(20%vs.16%,p=0.035).
Inrelationtorespiratoryevents(Table2)thatoccurredin thePACU,HR-OSApatientshadmorefrequentlyexperienced mild/moderatedhypoxia (9%vs.3%,p=0.012).Thelength ofthehospitalstay(median5daysvs.3days,p=0.01),but notthelengthofthePACUstay(96vs.91minutes,p=0.064), washigherinthesepatients.
InmultivariateanalysesforAREoccurrence(Table3)and after adjustment for univariate predictors (gender, intra-abdominalsurgery,major surgery,cardiovascular high-risk surgery, general anesthesia and/or combined anesthesia, use of neuromuscular relaxants and postoperative resid-ual neuromuscular blockade) the occurrence of residual neuromuscularblockadewasidentified asan independent predictorofARE.
IntheseanalysesHR-OSAwasnotadeterminantforARE (p=0.142)andwasnotenteredinthemultivariatemodel. HR-OSA is shown in Table 3 but it wasnot introduced in themodel becauseitwasnot considereddeterminant for ARE.
Discussion
TheincidenceofpatientswithHR-OSAwas52%inourstudy, whichindicatesthatthisisanimportantmatterforconcern amongpatientsscheduledforsurgeryinourhospital.
WereliedonaSTOP-BANGscore≥3forthisstudy,since thiswasdemonstratedtobeahighlysensitivetestfor mod-erate/severeOSA.Indeed,Chungetal.14suggestedthatthis
cut-offisappropriateforthesurgicalpopulation14and
fur-ther stated thatthe STOP-BANG questionnaire is concise, easy to use, and its importance lies in the identification ofpatientsat highrisk ofOSA whichmayhelp toprevent postoperativecomplications.
The HR-OSA group of patients had a significantly higher percentage of men and as expected the BMI was significantly different between the LR-OSA and HR-OSA groups (p<0.001). Similar results have been reported previously.15,16Theimportanceofthesefindingsisrevealed
bythe presence ofthese variablesin different toolssuch asthesleepapneaclinicalscore(SACS)andeventhe STOP-BANG,usedtoassessOSAseverity.
The association between increased morbidity and untreatedOSAis wellestablished, whichmayresultin an increasedmortality rate.17,18 Furthermore,these patients
are at a greater risk of developing cardiovascular dis-ease.Indeed,HR-OSApatientshadmorefrequentlycardiac comorbidities(ischemicheartdiseaseandcongestiveheart failure),hypertension,dyslipidaemiaandpulmonarychronic disease in our study. These comorbidities may, however, have reflected higher ASA and RCRI scores, which would
Table1 Patientbaselinecharacteristics(n=340).
Variable HR-OSA(n=179) LR-OSAn=161 p
Age,median(IQR) 63(55---71) 47(36---61) <0.001c
Gender,n(%) <0.001a Male 113(63) 33(21) Female 66(37) 128(80) SiteofSurgery,n(%) Intraabdominal 76(43) 47(29) 0.011a Musculoskeletal 88(49) 91(57) 0.175a
HeadandNeck 15(8) 23(14) 0.084a
Bodymassindex(kg/m2),median(IQR) 28(25---32) 24(22---28) <0.001c
ASAphysicalstatus,n(%) <0.001a
I/II 132(74) 145(90)
III/IV/V,n(%) 47(26) 16(10)
Highrisksurgery,n(%) 57(32) 36(22) 0.050a
Ischemicheartdisease,n(%) 18(10) 3(2) 0.001b
Congestiveheartdisease,n(%) 15(8) 5(3) 0.032b
Cerebrovasculardisease,n(%) 7(4) 2(1) 0.116b
Insulintherapyfordiabetes,n(%) 42(24) 7(4) <0.001a
Renalinsufficiency,n(%) 10(6) 7(4) 0.394a RCRI,n(%) 0.030b RCRI≤2 167(93) 158(98) RCRI>2 12(7) 3(2) Hypertension,n(%) 111(62) 30(19) <0.001a Dyslipidemia,n(%) 73(41) 19(12) <0.001a COPD,n(%) 15(8) 4(3) 0.015b Typeofanesthesia,n(%) 0.009a Locoregional 46(26) 23(14) General/combined 133(74) 138(86) Magnitudeofsurgery,n(%) Minor 14(8) 21(13) 0.114a Medium 108(63) 104(65) 0.481a Major 57(32) 36(22) 0.033a Bariatricsurgery,n(%) 12(7) 2(1) 0.016b
Durationsurgery,median(IQR) 87(50---135) 75(50---120) 0.173c
Post-operativedeliriumhospital,n(%) 23(13) 16(10) 0.400a
Temperatureonadmission,median(IQR) 35.3(34.9---35.7) 35.4(34.9---35.7) 0.158c
<35◦C,n(%) 6(25) 104(33) 0.424a
LengthofPACUstayminutes,median(IQR) 96(67---145) 91(65---120) 0.064c
LengthofHospitalstay(days),median(IQR) 5(2---8) 3(2---6) 0.010c
Perioperativemusclerelaxantsuse,n(%) 99(55) 106(66) 0.048a
Neuromuscularresidualblocking,n(%) 35(20) 25(16) 0.035a
aPearson’schi-squaredtest. b Fisher’sexacttest.
c Mann---WhitneyU-testIQR,interquartilerangeHR-OSA,high-riskofOSA;ASA,AmericanSocietyofAnesthesiologists;RCRI,revised cardiacriskindex;COPD,ChronicPulmonaryObstructiveDisease;PACU,PostAnesthesiaCareUnit.
indirectlyindicatethepresenceofmoreimportant associ-atedcomorbiditiesinpatientsatahigherriskofhavingOSA. OurresultsshowthatHR-OSApatientsmorefrequently underwentbariatricsurgery,abdominalsurgery,andmajor surgery.Thus, it is possible thatthe risksassociated with major surgery and bariatric surgery may account for the highercomplicationrates ratherthanthediagnosis of HR-OSAinitself.
Wefoundthatthelengthofthehospitalstaywashigher inHR-OSApatients,whichisinagreementwiththestudyof Liuetal.19 This suggestspostoperativehypoxemiais
asso-ciatedwithadverseoutcomesincludingtheneed formore frequentrespiratoryinterventionsandmoreintensive mon-itoringwhichmayledtoalongerhospitalstay
TheincidenceofRNMBwashigherinHR-OSApatients.In addition,ahigherincidenceofRNMBwasfoundinpatients
Table2 Respiratorycomplications(n=340).
Variable All(n=340) HR-OSAn=179 LR-OSAn=161 p
Respiratoryevents 64(19) 39(22) 25(16) 0.140
Obstructionairway 3(1) 1(1) 2(1) 0.475
Hypoxia(mild/moderate) 19(6) 15(9) 4(3) 0.012
HypoxiaSevere 5(2) 3(2) 2(1) 0.529
Respiratoryfailure 7(2) 5(3) 2(1) 0.253
Inspiratorycapacitydecreased 49(15) 28(17) 21(13) 0.407
Muscleweakness 11(3) 8(5) 3(2) 0.132
Table3 Multivariateregressionanalysisforpredictorsofadverserespiratoryevents.
Variable SimpleOR(95%CI) p bOR(95%CI) pa
Gender 1.9(1.0---3.3) 0.038
---Intra-abdominalsurgery 1.8(1.0---3.0) 0.049
---General/combinedanesthesia 2.9(1.3--- 6.6) 0.013 ---
---NMBDuse 2.8(1.5---5.3) 0.002 ---
---RNMB 6.1(3.1---12.2) <0.001 5.9(2.8---12.5) <0.001
Highrisksurgery 2.3(1.3---4.1) 0.004 ---
---Majorsurgery 1.8(1.0---3.2) 0.045 ---
---HR-OSA 1.5(0.9---2.6) 0.145 ---
---a Logisticregressionanalysiswithforwardconditionaleliminationmethodwasusedwithanentrycriterionofp<0.05. OR;oddsratio;CI,ConfidenceInterval;NMBD,neuromuscularblockingdrugs;RNMB,residualneuromuscularblockade.
b AdjustedORforgender,intra-abdominalsurgery,high-risksurgery,majorsurgery,NMBDuseandRNMB(variablesconsideredtobe determinantsforAREinaunivariateanalyses).
withahighBMI,20,21whichmaybeduetotheaccumulation
ofneuromuscularblockerintheadiposetissue.Thesedata suggestthatenhancedmonitoringisimportantwhenHR-OSA patientsareexposedtoadditionalriskfactorsforrespiratory complicationslikeRNMB, orwhenthesepatientsaregiven drugsthatenhanceneuromuscularblockade.
Perioperative complications, including respiratory events, were more frequent in patients with HR-OSA. Compared with patients without this diagnosis, we found that the HR-OSA patients more frequently experienced mild/moderated hypoxia in the PACU and there was a greaterriskofpostoperativehypoxemiainHR-OSApatients in comparison with those without the diagnosis. This is consistent with the results of the study of Gali et al.,15
whichfoundthatrespiratoryeventsweremorefrequentin patientswithahighriskofOSA.However,inourstudy,asin thestudyofChungetal.22analysisofperioperativeadverse
events did not show significant respiratory morbidity in HR-OSApatientscomparedtotheLR-OSApatients. Infact inourstudywedidnotfindthatHR-OSAwasadeterminant forARE.
This study has several limitations. The first and most importantlimitationisthatitwasnotpossibletocompare theresultsoftheSTOP-BANGquestionnairewithadefinitive polysomnographicdiagnosis.
Second, our patients may have been at high risk for postoperative events without havingOSA. The use of the STOP-BANGscoreisasaninstrumenttopredicttheriskof apatienttohavingOSAmayhaveledtoahighincidenceof
falsepositiveresults.Itispossiblethatahigherscoreshould havebeenusedbecauserecentstudieshaveshownthatthe STOP-BANGscorehasahigheraccuracyindetecting moder-atetosevereOSApatientsinasurgicalpopulationwhenthe scoreisgreaterthanorequalto5.1
Third, the respiratory events were only registered in thePACUandcomplicationsthatcouldhaveoccurredafter PACUdischargewerenotconsidered.Thismightbeviewed asamajorlimitationsincerespiratorycomplicationsafter surgerymaybeassociatedwithadverseoutcomesthatoccur duringthehospitalstayafterPACUdischarge.
Fourth, thefact that HR-OSApatients morefrequently underwentmajorsurgerymaybeviewedasaconfounding factorandalimitationofthestudybecauseitindicatesthat thepatientgroupswerenon-homogeneous.
Inconclusion,theprincipalfindingsofthisstudyareas follows:
- HR-OSApatientshaveahigherincidenceofpostoperative respiratorycomplications;
- mild/moderatehypoxiawasthemostfrequentpulmonary adverseeventintheimmediatepostoperativeperiodthat occurredmorefrequentlyinHR-OSApatients;
- HR-OSA patients had more comorbidities including ischemicheartdisease,congestiveheartfailure,diabetes treated withinsulin, chronic pulmonary disease, hyper-tensionanddyslipidemia;
- HR-OSApatientsrequiredalongerstayinthehospital. - HR-OSAwasnotadeterminantforARE.
Ethical
disclosures
Protection of human and animal subjects.The authors declarethatnoexperimentswereperformedonhumansor animalsforthisstudy.
Confidentialityofdata.Theauthorsdeclarethattheyhave followedtheprotocolsoftheirworkcenteronthe publica-tionofpatientdataandthatallthepatientsincludedinthe studyreceivedsufficientinformationandgavetheirwritten informedconsenttoparticipateinthestudy.
Righttoprivacyandinformedconsent.Theauthorshave obtainedthe written informed consentof the patients or subjectsmentionedinthearticle.Thecorrespondingauthor isinpossessionofthisdocument.
Conflicts
of
interest
Theauthorshavenoconflictsofinteresttodeclare.
Appendix
A.
STOP-Bang
Scoring
Model
References
1.ChungF,SubramanyamR, LiaoP,SasakiE,ShapiroC,SunY. HighSTOP-Bangscoreindicatesahighprobabilityofobstructive sleepapnoea.BrJAnaesth.2012;108:768---75.
2.Chung F, Yegneswaran B. STOP questionnaire----a tool to screen patients for obstructive sleep apnea. Anesthesiology. 2008;V:812---21.
3.Shafazand S. Perioperativemanagementof obstructivesleep apnea: ready for prime time? Cleve Clin J Med. 2009;76 Suppl.:S98---103[Review].
4.Romero-Corral A, Caples SM, Lopez-Jimenez F, Somers VK. Interactions between obesity and obstructive sleep apnea: implicationsfor treatment.Chest. 2010;137:711---9[Research Support,NIH, ExtramuralResearchSupport,Non-USGov’t Review].
5.TraceyL, WrightC,George A.Risk assesmentof obstructive sleepapneainapopulationofpatientsundergoingambulatory surgery.JClinSleepMed.2010;6.
6.Stierer TL, Wright C, George A, Thompson RE, Wu CL, Collop N. Risk assessment of obstructive sleep apnea in a population of patients undergoing ambulatory surgery. J Clin Sleep Med. 2010;6:467---72 [Research Support, Non-USGov’t].
7.Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea---hypopnoeawith or without treatmentwith continu-ouspositiveairwaypressure:anobservationalstudy.Lancet. 2005;365:1046---53[ComparativeStudyResearchSupport,Non-U SGov’t].
8.Isono S. Obstructive sleep apnea of obese adults: patho-physiology and perioperative airway management. Anesthe-siology. 2009;110:908---21 [Research Support, Non-U S Gov’t Review].
9.Fleisher LA, Beckman JA, Brown KA, Calkins H, Chaikof E, FleischmannKE,etal.ACC/AHA2007guidelineson periopera-tivecardiovascularevaluationandcarefornoncardiacsurgery: a report of the American College of Cardiology/American HeartAssociation TaskForce onPracticeGuidelines (Writing CommitteetoRevisethe2002GuidelinesonPerioperative Car-diovascularEvaluation forNoncardiacSurgery):developed in collaborationwiththeAmericanSocietyofEchocardiography, AmericanSocietyofNuclearCardiology,HeartRhythmSociety, SocietyofCardiovascularAnesthesiologists,Societyfor Cardio-vascularAngiography and Interventions, Societyfor Vascular MedicineandBiology,andSocietyforVascularSurgery. Circu-lation.2007;116:e418---99[ConsensusDevelopmentConference PracticeGuideline].
10.Murphy GS, Brull SJ. Residual neuromuscular block: lessons unlearned. PartI: Definitions,incidence, and adverse physi-ologiceffectsofresidualneuromuscularblock.AnesthAnalg. 2010;111:120---8[Review].
11.Murphy GS. Residual neuromuscular blockade: incidence, assessment,andrelevanceinthepostoperativeperiod.Minerva Anestesiol.2006;72:97---109[ResearchSupport,Non-USGov’t Review].
12.GaudreauJD, GagnonP, HarelF,Tremblay A, Roy MA.Fast, systematic,andcontinuousdeliriumassessmentinhospitalized patients:thenursingdeliriumscreeningscale.JPainSymptom Manage.2005;29:368---75[ClinicalTrialResearchSupport,Non-U SGov’t].
13.MurphyGS,SzokolJW,MarymontJH,GreenbergSB,AvramMJ, VenderJS.Residualneuromuscularblockadeandcritical respi-ratoryeventsin thepostanesthesiacare unit.Anesth Analg. 2008;107:130---7.
14.Chung F, Subramanyam R, Liao P, Sasaki E, Shapiro C, Sun Y. High STOP-Bang score indicates a high prob-ability of obstructive sleep apnoea. Br J Anaesth. 2012;108:768---75 [Evaluation Studies Research Support, Non-USGov’t].
15.Gali B, Whalen FX, Schroeder DR, Gay PC, Plevak DJ. Identification of patients at risk for postoperative respi-ratory complications using a preoperative obstructive sleep apnea screening tool and postanesthesia care assess-ment. Anesthesiology. 2009;110:869---77 [Research Support, Non-USGov’t].
16.Ramachandran SK, Kheterpal S, Consens F, Shanks A, DohertyTM,MorrisM,etal.Derivationandvalidationofa sim-pleperioperativesleepapneapredictionscore.AnesthAnalg.
2010;110:1007---15[ResearchSupport,Non-USGov’tValidation Studies].
17.ChungSA,YuanH,ChungF.Asystemicreviewofobstructive sleepapneaanditsimplicationsforanesthesiologists.Anesth Analg. 2008;107:1543---63 [Research Support, Non-U S Gov’t Review].
18.Hung J, Whitford EG, Parsons RW, Hillman DR. Association ofsleep apnoea withmyocardial infarction in men. Lancet. 1990;336:261---4[ResearchSupport,Non-USGov’t].
19.LiuSS,ChisholmMF,NgeowJ,JohnRS,ShawP,MaY,etal. Post-operativehypoxemiain orthopedicpatientswithobstructive sleepapnea.HssJ.2011;7:2---8.
20.GaszynskiT,SzlachcinskiL,JakubiakJ,GaszynskiW.Reversal fromnon-depolarisingneuromuscularblockadeinthe postop-erativeperiod.AnestezjolIntensTer.2009;41:11---5.
21.GaszynskiT,SzewczykT,GaszynskiW.Randomizedcomparison ofsugammadex and neostigminefor reversalof rocuronium-inducedmusclerelaxationinmorbidlyobeseundergoinggeneral anaesthesia.BrJAnaesth.2012;108:236---9[ComparativeStudy RandomizedControlledTrialResearchSupport,Non-USGov’t]. 22.Chung F, Ward B, Ho J, Yuan H, Kayumov L, Shapiro C. Preoperativeidentificationofsleepapneariskinelective sur-gical patients,using theBerlin questionnaire.J ClinAnesth. 2007;19:130---4[ClinicalTrialResearchSupport,Non-USGov’t].