PAIN(cid:3) 154 (2013) 677–689 www.elsevier.com/locate/pain Intravenous acetaminophen reduces postoperative nausea and vomiting: A systematic review and meta-analysis Christian C. Apfela,b,⇑, Alparslan Turanc, Kimberly Souzaa, Joseph Pergolizzid,e, Cyrill Hornussf aDepartmentofAnesthesiaandPerioperativeCare,UCSFMedicalCenteratMtZion,SanFrancisco,CA,USA bDepartmentofEpidemiologyandBiostatistics,UCSFMedicalCenteratMtZion,SanFrancisco,CA,USA cInstituteofAnesthesiologyandOutcomesResearchInstitute,ClevelandClinicFoundation,Cleveland,OH,USA dDepartmentofMedicine,JohnsHopkinsUniversitySchoolofMedicine,Baltimore,MD,USA eDepartmentofPharmacology,TempleUniversitySchoolofMedicine,Philadelphia,PA,USA fDepartmentofAnaesthesiology,Ludwig-Maximilians-University,Munich,Germany Sponsorshipsorcompetingintereststhatmayberelevanttocontentaredisclosedattheendofthisarticle. a r t i c l e i n f o a b s t r a c t Articlehistory: Opioidsareakeyriskfactorforpostoperativenauseaandvomiting(PONV).Asintravenous(i.v.)acetami- Received25August2012 nophenreducespostoperativepainandopioidrequirements,onewouldexpecti.v.acetaminophentobe Receivedinrevisedform21December2012 associatedwithalowerincidenceofopioid-inducedsideeffects,includingPONV.Weconductedasys- Accepted31December2012 tematicsearchusingMedlineandCochranedatabasessupplementedwithhandsearchofabstractpro- ceedings to identify randomized-controlled trials of i.v. acetaminophen. Inclusion criteria were (a) randomized for i.v. acetaminophen vs a placebo control, (b) general anesthesia, and (c) reported or Keywords: obtainable PONV outcomes. Primary outcome was postoperative nausea and secondary outcome was PONV postoperativevomiting.Weincluded30studieswith2364patients(1223intheacetaminophengroup, Acetaminophen 1141intheplacebogroup).Therelativerisk(95%confidenceinterval)was0.73(0.60–0.88)fornausea Systematicreview Meta-analysis and 0.63 (0.45–0.88) for vomiting. Data showed significant heterogeneity for both nausea (P=0.02, I2=38%)andvomiting(P=0.006,I2=47%),butwerehomogeneouswhenstudiesweregroupedaccording totimingoffirstadministration:i.v.acetaminophenreducednauseawhengivenprophylacticallyeither beforesurgery,0.54(0.40–0.74),orbeforearrivalinthepostanesthesiacareunit,0.67(0.55–0.83);but notwhengivenaftertheonsetofpain,1.12(0.85–1.48).Wheni.v.acetaminophenwasgivenprophylactically,thereductionofnauseacorrelatedwiththereductionofpain(oddsratio0.66,0.47–0.93),butnot withreductioninpostoperativeopioids(oddsratio0.89,0.64–1.22).Prophylacticallyadministeredi.v. acetaminophenreducedPONV,mainlymediatedthroughsuperiorpaincontrol. (cid:2)2013InternationalAssociationfortheStudyofPain.PublishedbyElsevierB.V.Allrightsreserved. 1.Introduction drugs(NSAIDs)andpregabalinaspartofmultimodalanalgesiahave demonstratedareductioninPONVbyabout30%,whichiscompara- Postoperativenauseaandvomiting(PONV)occursinabout30% bletotheefficacyknownfromconventionalantiemetics[29,54]. ofallpatientsundergoinggeneralanesthesia,andthemaintriggers While oral and rectal acetaminophen has been on the market are known to be inhalational anesthetics and opioids [2,3,27,51]. formanydecades,i.v.acetaminophenwas introduced onlyabout Whiletheemetogenicinhalationalanestheticsmaybesubstituted 10years ago in Europe and 2years ago in the United States. The withintravenous(i.v.)propofolinfusions,itismoredifficulttore- key advantage of i.v. acetaminophen seems to be that 1g of i.v. placeoratleastminimizetheuseofopioids[2,50].However,even acetaminophenisassociatedwithabouttwicetheplasmaandef- ifitmaynotbegenerallypossibletoeliminateopioids,thereduction fect site concentrations as 1g of its oral or rectal applications, ofopioidconsumptionisgenerallythoughttobeassociatedwith resultingingreatercentralnervoussystempenetration[44],which feweropioid-relatedsideeffects.Forexample,systematicreviews corroborates the superior analgesic efficacy seen with i.v. com- and meta-analyses of the use of nonsteroidal anti-inflammatory paredtooralacetaminopheninthesurgicalsetting[38]. Accordingtosystematicreviewsandmeta-analyses,acetamino- phenhasalsobeenshowntoreduceopioidrequirements[32,47]. ⇑ Correspondingauthor.Address:DepartmentofEpidemiologyandBiostatistics, However,tooursurprise,thisdidnottranslatetoareducedinci- UCSFMedicalCenteratMtZion,1600Divisadero,C-447,SanFrancisco,CA94115, denceofPONV.Acloserlookrevealedthatthosereviewsfocused USA.Tel.:+14158857842;fax:+14158857770. E-mailaddress:apfelc@anesthesia.ucsf.edu(C.C.Apfel). on studies with pain or opioid consumption as the primary end 0304-3959/$36.00 (cid:2)2013InternationalAssociationfortheStudyofPain.PublishedbyElsevierB.V.Allrightsreserved. http://dx.doi.org/10.1016/j.pain.2012.12.025 678 C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 point,andveryfewstudiesdidactuallyreportPONV.Furthermore, comparable to otherfactors.Reportingofpostoperative painwas inthelast2years,anumberofnewstudieshavebeenpublished ratherheterogeneous,and conversions wereperformedas neces- reviewingPONVandacetaminophen[15,30]. sary. First, we converted pain values to an 11-point pain scale Thus,ourobjectivewastosystematicallyreviewtheeffectofi.v. (0–10). Second, we calculated the area under the pain curve for acetaminophenonnauseaandvomitinginrandomizedcontrolled each group and divided the between-group differences by the trialsinpatientswhoreceivedgeneralanesthesia. duration of observation (usually 24 hours), which resulted in an averagepainintensitydifference. 2.Materialandmethods 3.Results According to a prospectively drafted protocol, we searched Medlineusingthesearchstring‘‘(acetaminophenORparacetamol) 3.1.Search AND(PONV)’’usingthelimits‘‘(humanANDclinicaltrialANDran- domizedcontrolledtrial)’’onMarch4,2012.Eligiblestudieswere SearchusingMedlineandCochranedatabasesrevealed178and abstractsorfullpaperswithnolanguagerestrictions,butallstud- 62hits,respectively,with33duplicates(Fig.1).Handsearchofref- ieshadtoberandomized,placebo-controlled,andreporttheinci- erences and contacting Cadence Pharmaceuticals, Inc (San Diego, dencesornumberofpatientswithnausea,vomiting,orboth. CA)revealed10additionalreferences,leadingtoatotalof217ref- Nauseaandvomitingare2distinctphenomenaand,aswehave erences. After reviewing the title and abstracts, 143 references pointed out before, the study of PONV should assess and report were excluded because they did not meet the inclusion criteria thesevariablesseparately[4].Butbecausefewpatientsexperience (eg,treatmentofacetaminophenoverdose)and74full-textarticles vomiting without nausea, the incidence of postoperative nausea were assessed for final eligibility. Of those, a further 44 articles (PON)andPONVisgenerallyfairlysimilarandthus,originalpapers wereexcludedforthefollowingreasons(notethatonestudycould oftendonotdistinguishbetweenthesevariables[50].Thus,when have more than one reason): not randomized for i.v. acetamino- bothPONandPONVwerereported,weusedthenauseavalues,and phen(n=15),noinactivecontrol(n=20),patientsdidnotreceive ifPONVbutnotPONwasreported,weconsideredthePONVvari- surgery under general anesthesia (n=12), and PONV data were ables as a very close surrogate for PON. The most commonly re- missingandnot obtainable (n=4).Thus,weidentified30 papers ported time interval to measure the efficacy of antiemetics is 24 (with32comparisons)with2364patients(1223intheacetamino- hours[4].Ifanothershorterorlongertimeintervalwasreported, phengroup,1141intheplacebogroup;Table1[5–7,9–11,13,17– weusedthetimeintervalthatwasclosestrelativetothe24-hour 23,25,26,28,33,34,36,37,39,42,43,45,46,48,49,52]). interval.Ifdatawereunclearormissing,wecontactedtheauthors toaskforadditionalinformation. 3.2.Efficacydata Included studies had to report PON, PONV, or postoperative vomiting, either as a primary end point, secondary end point, or Intravenousacetaminophenwasassociatedwitharelativerisk asadverseevents.Ourprimaryhypothesiswasthati.v.acetamino- (95%confidenceinterval)of0.73(0.60–0.88)fornausea(Fig.2)and phenwillreducePON,possiblythroughreductioninpostoperative 0.63(0.45–0.88)forvomiting(Fig.3),butwithsignificanthetero- opioid requirements. In addition, we expected the effect size for geneityforbothnausea(P=0.02)andvomiting(P=0.006;Table2). vomitingtobesimilar;acknowledgingthatduetothemuchlower Numberneededtotreatfori.v.acetaminophenwas12.3(7.6–32.3) incidenceofvomitingcomparedtonausea,therecouldbeinsuffi- fornauseaand14.2(8.3–50.8)forvomiting.Sensitivityanalysisre- cientpowerforastatisticallysignificanteffecteventhoughtheef- vealed that i.v. acetaminophen reduced nausea (0.63, 0.54–0.75) fect size could be similar to that for nausea. Furthermore, we andvomiting(0.42,0.31–0.56)ininvestigator-initiatedtrials,but planned sensitivity analyses for the source of funding, dosage, didnotreducenausea(1.12,0.85–1.48)andevenincreasedvomit- andmeta-regressionstostudytheeffectsofpostoperativeopioid ing(1.41,1.02–1.96)inindustry-sponsoredclinicaltrials.Acloser and pain reduction. Because we found that i.v. acetaminophen look showed that i.v. acetaminophen was generally started pro- did not reduce PONV when it was dosed many hours and often phylacticallyininvestigator-initiatedtrials,whileitwasgenerally the following days after surgery as a rescue medication, we conducted sensitivity analyses on patients in whom i.v. acetaminophen was given prophylactically. We analyzed all reported outcomesforpotentialpublicationbiasbyexaminingfunnelplots 178 records identified through PubMed, 10 additional records identified andperformingEgger’sregressiontests[16]. 62 records identified through Cochrane through other sources This study was conducted in accordance with the Preferred ReportingItems for Systematicreviews and Meta-Analysis(PRIS- MA)statement[35].AnalyseswereprimarilyconductedwithRev- 217 records after 33 duplicates removed Man5.1(CochraneCollaboration,Copenhagen,Denmark),andwe calculatedtherelativeriskoftheeventsfortheinterventionversus 217 abstracts screened 143 records excluded theplacebocontrolgroupusingthemoreconservativerandom-ef- fectsmodel.Wealsocalculatedthenumberneededtotreat;however, this measure needs to be cautiously interpreted since 74 full-text articles 44 full-text articles excluded numbersneededtotreatforprophylacticinterventionsarehighly (multiple reasons possible): assessed for eligibility dependentonbaselineincidencesandthusshouldnotbeusedto Not randomized for IV compare interventions among trials with different baseline inci- acetaminophen, n=15 No inactive control, n=20 dence [8]. For meta-regressions and Egger’s regression tests, we 30 studies included in used Comprehensive Meta-Analysis 2.2 (Biostat; Englewood, NJ, qualitative and No general anesthesia, n=12 quantitative synthesis No PONV data, n=4 USA)andconfirmedtheresultswithSTATA9.0(StataCorpLP,Col- (meta-analysis) legeStation,TX,USA).Postoperativeopioidconsumptionwascon- vertedtomorphineequivalentsandtheregressionwasconducted Fig.1. Flowchartofsystematicidentification,screening,andeligibilityassessment on 10-mg morphine equivalents to make the coefficient more ofstudies.IV,intravenous;PONV,postoperativenauseaandvomiting. C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 679 Table1 Characteristicsofincludedstudies. Study(first Numberof Number Riskofbias Patients/typeof Intervention Outcomes Comments author)Year patients of assessmenta surgery acetaminophen patients placebo Arici[5] 55 27 U/U/H/H/U ASAI-II,elective 1gi.v.paracetamolbefore Morphineconsumption1,2, 2009 abdominal surgery1gi.v.paracetamol 4,8,12,24h hysterectomy (priortoskinclosure) PainVAS(0–10)score1,2,4, 8,12,24h Nausea,vomiting,itching, SpO20–24h Hospitalstay Arslan[6] 20 20 L/H/L/L/L ASAI-II,elective 1gi.v.acetaminophen PainVAS0.25,1,2,4,6,8, 2011 thyroidectomy intraoperatively8mg 12,18,24 lornoxicamintraoperatively Timetofirstanalgesic (tramadol) Patientsatisfaction Nausea,vomiting0–24h Allergy,hypotension0–24h Atef[7] 38 38 L/L/L/L/L Electivetonsillectomy 1gi.v.paracetamol PainVAS1,2,3,4,24h 2008 immediatelyafterendof Numberofpethidinedoses surgeryandforevery6hfor Nauseaandvomiting0–24h 18h Headache0–24h Brodner[9] 49 49 L/L/L/L/L ASAI-III,minor-to- 1gi.v.acetaminophenevery PainVAS1,6,18,30,42h 2010 intermediatesurgery 6hintraoperatively1g Timetofirstpiritramide dipyroneevery6h Piritramideconsumption intraoperatively40mg Patientsatisfaction parecoxibevery12h Respiratorydepression intraoperatively Nauseaandvomiting Sedation,sweating Cakan[10] 20 20 L/L/H/L/L ASAI-III,elective 1gi.v.acetaminophenat PainVAS1,2,4,6,12,18, 2008 lumbarlaminectomy endofsurgeryandforevery 24h anddiskectomy 6hfor24h Patientsatisfaction Nauseaandvomiting0–24h Sedation0–24h Dizziness,headache, drowsiness,hypotension0– 24h Candiotti 166 165 L/L/L/L/U ASAI-III,lower 1gi.v.acetaminophenafter Sumofpainintensityat24 [11]2008 abdominal,non- surgeryandforevery6hfor and48h laparoscopic, 48h PainreliefTimetofirst gynecologicalsurgery rescuemorphine Totalmorphine consumptionafter24and 48h Patients’satisfaction Cattabriga 56 57 L/L/L/L/L Electivecardiac 1gi.v.acetaminophenat PainVAS12,18,24,30,36, [13]2007 surgery endofsurgery 42,48,54,60,66,72h Andforevery6hfor72h Morphineconsumption0– 72h Nauseaandvomiting Cardiorespiratory parameters Cok[15] 41 45 L/L/L/L/L ASAI-II,strabismus 15mgkg(cid:2)1i.v.paracetamol Facespainscale0–24h Allpatients 2011 surgeryinchildren2– Numberofanalgesic received 14years interventions dexamethasone Nauseaandvomiting0–24, 0.1mgkg(cid:2)1 24-48h 10mgkg(cid:2)1 paracetamolas analgesicrescue medication Emir[17] 30 30 L/H/H/H/U Electivespinal 1gi.v. 0,0.25,0.5,1,2,4,6,12,18, Averagewas 2010 vertebralsurgery paracetamol+0.75mgkg(cid:2)1 24h takenGroupT: tramadolimmediatelyafter Numericratingscaleforpain 17/7=2.43 surgery assessment4-pointnausea GroupP:4/ 1.5mgkg(cid:2)1tramadol scale 3=1.33 immediatelyaftersurgery Cardiorespiratory parameters Fadly[18] 10 10 U/U/H/H/U ASAI-II,surgical 1g Painscore0,0.25,0.5,0.751, Additional 2006 releaseofpostburn paracetamol+0.05mgkg(cid:2)1 2,4,6h morphine neckcontractures morphine1gparacetamol Timetofirstmorphine 0.05mgkg(cid:2)1 Studydruggiven rescuedose immediatelyafter Totalmorphine extubation consumption Patientsatisfaction Nauseaorvomiting0–6h (continuedonnextpage) 680 C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 Table1(continued) Study(first Numberof Number Riskofbias Patients/typeof Intervention Outcomes Comments author)Year patients of assessmenta surgery acetaminophen patients placebo SedationSpO2 Gimbel[19] 30 31 L/L/L/L/L Electivehip 1gi.v.acetaminophenfirst Firstdose:Painintensity 2008 arthroplasty morningaftersurgeryand difference0.25,0.5,0.75,1, repeatedafter4,10,16h 2,3,4hpostdose Painintensityandpainrelief derivedscores PAR,SPID,TOTPAR,MAXPID, MAXPAR Rescue,frequency, administration,amount Repeateddoses:PIDat1,5, 11,and17h Nausea Gokten[20] 20 20 U/H/H/L/L Percutaneous 1gi.v.paracetamol PainVASat6and24h Additional0.25% 2011 nephrolithotomy postoperativelyand Percentagemeperidine levobupivacaine repeated rescuemedication infiltration 4/day20mL0.25% Sideeffects:Nauseaor levobupivacaine+1gi.v. vomiting paracetamol Sedation,respiratory postoperativelyrepeated4/ depression,hypotension, day20mL0.25% pruritus,dizziness,sweating, levobupivacaine drymouth Grundmann 20 20 L/L/L/L/L ASAI-II,unilateral 1gi.v.paracetamol Alloutcomesforfirst2h [21]2006 microsurgicallumbar 40mgparecoxib postsurgery diskectomy 1gmetamizol PainVAS Beforetheendofsurgery Patientsrequiringadditional paintherapy Timetorescuepiritramide Cumulativepiritramidedose Nauseaandvomiting Shivering,pruritus,mean arterialpressure,heartrate Hong1[23] 31 32 L/L/L/L/L Children6-24months, 15mgkg(cid:2)1iv Postoperativefentanyldose Additional 2010 elective acetaminophen+0.5 at1,6,12,24,36,48,60,72h fentanyl ureteroneocystostomy lgkg(cid:2)1intraoperatively Children’sHospitalof 0.5mgkg(cid:2)1 0.5lgkg(cid:2)1intraoperatively EasternOntarioPainScaleat Parentnurse-controlled 1,6,12,24,36,48,60,72h analgesiawithi.v. Parentsatisfaction acetaminophenin Sideeffects(0–72h): treatmentgroup Vomiting,sedation,pruritus, poororalfeeding, desaturation Hong2[22] 63 61 L/U/L/L/L ASAI-II,women, 1gi.v.paracetamolbefore PainVASat1,3,6,12,24,48, 2010 robot-assisted inductionofanesthesiaand 72h endoscopictotal forevery6hfor24h Timetorescuefentanyl thyroidectomy Sideeffects:Nauseaand vomiting Drainoutput Jokela[25] 40 40 L/L/L/L/L ASAI-III,laparoscopic 1gacetaminophenat NRSpainscaleat1,2,4,6,8, 2010 hysterectomy inductionofanesthesiaand 18,24h forevery6hfor24h Timetofirstrescue 1gacetaminophenat oxycodone inductionofanesthesiaand Totaloxycodonedose0–24h forevery6hfor24h+4mg Patientsatisfaction ondansetron Sideeffects(0–24h):PONV, vomiting,dizziness, headache,lackof concentration,pruritus Kiliçaslan 25 25 L/L/L/L/L ASAI-II,cesarean 1gparacetamolbeforeend Pain(VAS)1,3,6,12,24h [26]2010 section ofsurgeryandforevery6h postoperative for24h Sedationscore(5-point)1,3, Patient-controlledtramadol 6,12,24hpostoperative analgesia PCAdemandanddelivery first24h Totaltramadolconsumption first24h Sideeffects(0–24h): Nausea,vomiting,useof antiemetics,drymouth, dizziness-drowsiness C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 681 Table1(continued) Study(first Numberof Number Riskofbias Patients/typeof Intervention Outcomes Comments author)Year patients of assessmenta surgery acetaminophen patients placebo Lee[28] 20 20 U/U/H/L/L ASAI-II, 1gi.v.paracetamol VASpainscore0.5,1,2,4, 2010 thyroidectomy 700mgi.v. 6hpostoperative paracetamol+3mgi.v. Incidenceofrescue morphine pethidine 30mgi.v.Ketorolac Patientsatisfaction 30minbeforeendof Sideeffects(0–6h):Nausea, surgery vomiting,headache, sedation,dizziness, respiratorydepression Memis[33] 20 20 L/L/H/H/L ASAII-III,major 1gi.v.paracetamol+i.v. Behavioralpainscale(10- 2010 surgery,ICUpatients meperidinei.v.meperidine pointscale)untilextubation every6hfor24hafter VAS10-pointpainscaleafter arrivalinICU extubationat24h Sedationscoreatextubation Sedationscoreat24h Adverseeffects:respiratory depression,reintubation, nauseaandvomiting, treatmentfornauseaand vomiting,pruritus Minkowitz 23 21 L/L/L/L/U ASAI-III,vaginal 1gacetaminophenon Painreliefandpainintensity Earlystudy [34]2008 hysterectomy morningofpostsurgeryday difference(4-point)at terminationdue 1andafter4,10,16h 0,1,2,3,4h toparticulatesin Timetomaximumpain placebovials relief,maximumpainrelief SPID4,TOTPAR4 Nauseaandvomiting,liver enzymes Moon[36] 36 35 L/L/L/L/H ASAI-II,abdominal 2gacetaminophen30min Totalpatientcontrolled 2011 hysterectomy beforeinductionof hydromorphone anesthesia consumption0-24h VASpainscoreat1,2,6,12, 24haftersurgery Sideeffects(0-24h):Nausea andvomiting,sedation, respiratorydepression, pruritus,useofantiemetics Ohnesorge 27 26 L/U/L/LU ASAI-III,elective 1gi.v.paracetamol NRS(11-point)painscale, [37]2009 breastsurgery 1gi.v.metamizol vigilance,nausea0.5,1,2,4, 20minbeforeendof 6,10,24haftersurgery surgeryand4,10,16hafter Needforrescuemorphine surgery Totalmorphine consumption24h Patientsatisfaction, cognitivefunction Timetofirstambulation Platzer[39] 40 39 L/L/H/L/L ASAI-II,childrenage 15mgkg(cid:2)1paracetamol 5-pointsmileyscaleat0.5,1 Additional 2011 3–13years, 2mgkg(cid:2)1ketoprofen 2,3,4hafterPACUarrival 2mgkg(cid:2)1 tonsillectomy, Morphineconsumption ketoprofen adenoidectomyor Timetorescuemorphine additional adenotonsillectomy Vomiting propofol Salihoglu 20 20 L/L/L/L/L ASAI-II,laparoscopic 1gi.v.paracetamolafter Verbal(4-point)andVAS [42]2009 cholecystectomy intubation (11-point)painscalesat5, 10,15,30,60minafter recovery Timeoffirstrescue morphineinPACU Totalmorphine consumption DurationofstayinPACU Awakeningtime,extubation time Nausea,vomiting Sinatra[43] 49 52 L/L/L/L/L ASAI-III,orthopedic 1gi.v.acetaminophenon VAS(0-100)painscale, Patient- 2005 surgery morningofpostsurgeryday verbalpainscale(4-point), controlled 1andforevery6hfor24h painrelief(5-point)at0.25, morphine 2gi.v.propacetamol 0.5,0.75,1,2,3,4,5,6,18, analgesia 20,24hafterfirstdose Timetorescuemedication Totalmorphine consumption Patientssatisfactionover (continuedonnextpage) 682 C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 Table1(continued) Study(first Numberof Number Riskofbias Patients/typeof Intervention Outcomes Comments author)Year patients of assessmenta surgery acetaminophen patients placebo 24h TOTPAR6,SPID6,SPRID6 Sideeffects:nausea, vomiting,constipation, abdominalenlargement, injectionsitepain/reaction, fever,anemia,pruritus, coughing Topal[45] 20 20 U/U/H/U/L ASAI-II,inguinal 1gparacetamolbeforeend VAS(11-point)painscale Additional 2009 herniarepair ofsurgeryandforevery6h andsedationat0.25,0.5,1, 0.05mgkg(cid:2)1 for24h 2,4,6,8,12,16,24h morphineatend postsurgery ofsurgeryand Totalmorphine patient consumption controlled Nausea,vomiting, morphine antiemeticuse,respiratory analgesia depression,pruritus Toygar[46] 60 30 U/H/H/H/U Lumbardiskectomy 1gparacetamol15min VASpainscale0,1,2,3,6, 2008 beforeinductionof 12,24h anesthesia Timetorescuemorphine 1gparacetamol15min Totalmorphine beforeendofsurgery consumption Sideeffects:Nausea, vomiting,urinaryretention Uvarov[48] 25 24 L/U/H/H/L Thoracotomy 1gi.v.paracetamolbefore VASpainscale3,6,12,18,24h Additional 2008 endofsurgeryandevery6h totalconsumption epiduralpatient for24h ropivacaine/fentanyl controlled Rectalparacetamol Sideeffects:Nauseaor analgesiawith vomiting,skinitching, ropivacaineand urinaryretention fentanyland 30mgi.m. ketorolac Viscusi[49] 35 34 L/L/L/L/L Totalhiparthroplasty Onpostsurgeryday1:1g Nausea:i.v.acetaminophen BMSstudydata 2008 paracetamolvs.placebo. 2/35 perstudyreport Placebo0/34 receivedfrom Vomiting:i.v. Cadence acetaminophen13/35 Pharmaceutical Placebo7/34 Wininger 134 110 L/L/L/L/L Abdominal Startofmedicationat Painintensity(4-point),VAS [52]2010 laparoscopicsurgery postsurgeryday1:650mg (100mm)painscale,Pain acetaminophenevery4hfor relief(5-point)at0,1,2,3,4, 24h 5,6,7,8,9,10,11,12,14,16, 1000mgacetaminophen 18,20,22,24hon every6hfor24h postsurgeryday1 Amountofandtimetofirst morphineor hydromorphone Patientsatisfaction(4-point) after24h TOTPARat4,6,24h,SPIDat 4,6,24h Sideeffects:Nausea, vomiting,constipation, diarrhea,flatulence, infusion-sitepain,pyrexia, incision-sitepain,backpain, headache,psychiatric disorders,insomnia, dyspnea U,unclear;L,low;H,high;ASA,AmericanSocietyofAnesthesiology;i.v.,intravenous;VAS,visualanaloguescale;PAR,painrelief;SPID,summedpainintensitydifference; TOTPAR,totalpainrelief;MAXPID,maximumpainintensitydifference;MAXPAR,maximumpainrelief;PID,painintensitydifference;NRS,numericalratingscale;PONV, postoperativenauseaandvomiting;PCA,patient-controlledanalgesia;PACU,postanesthesiacareunit;SPRID,sumofpainreliefandpainintensitydifference;BMS,Bristol- MyersSquibb. a The categories in Risk ofBias Assessment are: random sequence generation, allocation concealment, blinding ofparticipants and personnel, blinding of outcome assessment,andincompleteoutcomedata. giventhedayfollowingsurgeryinindustry-sponsoredregistration irrespective ofwhetherit wasstartedbeforesurgery,intraopera- trials. tively, or immediately after surgery (Fig. 4, Table 2). However, a Further sensitivity analyses revealed that prophylactically singleprophylacticdoseofi.v.acetaminophenwasassociatedwith administered i.v. acetaminophen reduced nausea and vomiting lessnausea(0.50,0.38–0.66)thanifthedosewasrepeated(0.72, C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 683 Acetaminophen Control Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Arici 2009 BS 3 28 9 27 2.2% 0.32 [0.10, 1.06] Arici 2009 IS 3 27 9 27 2.2% 0.33 [0.10, 1.10] Arslan 2011 IS 9 20 18 20 6.6% 0.50 [0.30, 0.83] Atef 2008 IR 7 38 10 38 3.6% 0.70 [0.30, 1.65] Brodner 2011 IR 15 49 21 49 6.3% 0.71 [0.42, 1.22] Cakan 2008 IR 12 20 17 20 7.9% 0.71 [0.47, 1.06] Candiotti 2008 RR 96 166 88 165 10.6% 1.08 [0.89, 1.32] Cattabriga 2007 IR 3 56 1 57 0.7% 3.05 [0.33, 28.48] Cok 2011 IS 6 41 15 45 3.6% 0.44 [0.19, 1.02] Fadly 2006 IS 0 10 3 10 0.4% 0.14 [0.01, 2.45] Gimbel 2008 RR 4 30 7 31 2.4% 0.59 [0.19, 1.81] Gokten 2011 PR 1 20 6 20 0.9% 0.17 [0.02, 1.26] Grundmann 2006 IS 1 20 1 20 0.5% 1.00 [0.07, 14.90] Hong2 2010 BR 14 63 27 61 6.2% 0.50 [0.29, 0.86] Jokela 2010 BR 8 40 12 40 4.1% 0.67 [0.31, 1.45] Kilicaslan 2010 IR 11 25 12 25 5.5% 0.92 [0.50, 1.67] Lee 2010 IS 2 20 1 20 0.7% 2.00 [0.20, 20.33] Memis 2010 PR 1 20 8 20 0.9% 0.13 [0.02, 0.91] Minkowitz 2008 RR 9 23 9 21 4.6% 0.91 [0.45, 1.86] Moon 2011 BS 13 36 21 35 6.5% 0.60 [0.36, 1.00] Ohnesorge 2009 IR 9 27 7 26 3.7% 1.24 [0.54, 2.83] Salihoglu 2009 BS 2 20 5 20 1.4% 0.40 [0.09, 1.83] Sinatra 2011 RR 13 49 7 52 3.7% 1.97 [0.86, 4.53] Topal 2009 IR 9 20 10 20 5.1% 0.90 [0.47, 1.73] Toygar 2008 BS 4 30 7 30 2.4% 0.57 [0.19, 1.75] Toygar 2008 IS 3 30 7 30 2.0% 0.43 [0.12, 1.50] Uvarov 2008 IR 1 25 1 24 0.5% 0.96 [0.06, 14.50] Viscusi 2008 RR 2 35 0 34 0.4% 4.86 [0.24, 97.69] Wininger a 2010 RR 16 91 3 43 2.2% 2.52 [0.78, 8.19] Wininger b 2010 RR 4 43 9 67 2.4% 0.69 [0.23, 2.11] Total (95% CI) 1122 1097 100.0% 0.73 [0.60, 0.88] Total events 281 351 Heterogeneity: Tau² = 0.08; Chi² = 46.93, df = 29 (P = 0.02); I² = 38% 0.01 0.1 1 10 100 Test for overall effect: Z = 3.21 (P = 0.001) Favors acetaminophen Favors placebo Fig.2. Effectofintravenousacetaminophenonpostoperativenausea.CI,confidenceinterval;BS,beforesurgery,singledose;IS,duringorimmediatelyaftersurgery,single dose;IR,duringorimmediatelyaftersurgery,repeateddose;RR,rescuemedication,repeateddose;PR,postoperative,repeateddose;BR,beforesurgery,repeateddose. 0.58–0.89, P-value for sub-group differences 0.04, Fig. 5). Results 4.1.Critiqueagainstprevioussystematicreviews forvomitingweresimilar,withP-valueforsub-groupdifferences of 0.13 (Table 2). Reduction of postoperative opioids (average Atthetimethismanuscriptwasbeingwritten,severalsystematic reductions were about 9mg of morphine equivalents) did not and narrative reviews have been published, all noting that i.v. contributetotheantiemeticeffectofprophylactici.v.acetamino- acetaminophendoesnotreducePONV[24,30,31,40,41,47,53].This phen,withanoddsratioof0.89(0.64–1.22)for10mgofmorphine includes a Cochrane review that identified only 4 papers, even equivalents(Fig.6,P=0.45).However,reductioninpostoperative thoughatthetimeofthelastsearch,May2010,19ofthe30papers pain (average reduction was about 0.9 points on a 0–10 scale) thatweidentifiedhadalreadybeenpublished[47].Itisdisconcert- wasassociatedwithasignificantreductioninpostoperativenau- ingthatsomanystudieshadnotbeenidentifiedintheCochrane sea, with an odds ratio of 0.66 (0.47–0.93) per 1 point (Fig. 7, review,despitetheexplicitreportingandapplicationoffairlycom- P=0.02). Egger’s regression tests did not reveal any evidence of plexsearchstrategies.Furtherresearchintowhythishappensiscer- publicationbias for any of the outcomesstudied (P > 0.05 for all tainlyindicatedifwewanttoimprovethereliabilityofsystematic outcomes,Table2). reviews.Therefore,weinvestigatedtheeffectofi.v.acetaminophen asourmainoutcomeandincludedthenewevidenceinouranalysis. 4.Discussion 4.2.Timingofi.v.administration Wefoundthatprophylactici.v.acetaminophenreducedPONV. Interestingly, timing of treatment markedly influenced the treat- Cumulative results showed a high degree of heterogeneity, menteffect:i.v.acetaminophenwasmosteffectiveagainstPONV which prompted further sensitivity analyses. We analyzed the when given before surgery and intraoperatively, but not when results according to the source of funding. To our surprise, treatmentwasinitiatedafteronsetofpain.Further,wheni.v.acet- industry-sponsoredtrialsdidnotshowareductionofPONV,while aminophenwasgivenprophylactically,thereductionofPONVcor- independent,investigator-initiatedtrialsshowedaclearreduction relatedwiththereportedreductionofpain,butnotpostoperative inPONV.Wethusreviewedthestudydesignsandfoundthati.v. opioidconsumption. acetaminophen was given for rescue treatment only in industry- 684 C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 Acetaminophen Control Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Arici 2009 BS 2 28 9 27 3.7% 0.21 [0.05, 0.90] Arici 2009 IS 3 27 9 27 4.7% 0.33 [0.10, 1.10] Arslan 2011 IS 5 20 14 20 6.8% 0.36 [0.16, 0.80] Brodner 2011 IR 9 49 12 49 7.1% 0.75 [0.35, 1.62] Cakan 2008 IR 7 20 14 20 7.8% 0.50 [0.26, 0.97] Candiotti 2008 RR 37 166 28 165 9.3% 1.31 [0.84, 2.04] Cok 2011 IS 3 41 11 45 4.7% 0.30 [0.09, 1.00] Emir 2010 IR 0 30 2 30 1.2% 0.20 [0.01, 4.00] Gimbel 2008 RR 1 30 2 31 1.8% 0.52 [0.05, 5.40] Hong1 2010 IR 5 31 18 32 6.5% 0.29 [0.12, 0.68] Hong2 2010 BR 4 63 13 61 5.3% 0.30 [0.10, 0.86] Jokela 2010 BR 2 40 4 40 3.1% 0.50 [0.10, 2.58] Kilicaslan 2010 IR 3 25 4 25 3.9% 0.75 [0.19, 3.01] Minkowitz 2008 RR 7 23 7 21 6.5% 0.91 [0.38, 2.17] Ohnesorge 2009 IR 1 27 1 26 1.4% 0.96 [0.06, 14.60] Platzer 2011 IS 1 40 1 39 1.4% 0.97 [0.06, 15.05] Salihoglu 2009 BS 1 20 3 20 2.0% 0.33 [0.04, 2.94] Sinatra 2011 RR 6 49 3 52 4.1% 2.12 [0.56, 8.02] Topal 2009 IR 3 20 4 20 4.0% 0.75 [0.19, 2.93] Toygar 2008 BS 0 30 5 30 1.3% 0.09 [0.01, 1.57] Toygar 2008 IS 1 30 5 30 2.2% 0.20 [0.02, 1.61] Viscusi 2008 RR 13 35 7 34 6.9% 1.80 [0.82, 3.97] Wininger a 2010 RR 7 91 0 43 1.3% 7.17 [0.42, 122.80] Wininger b 2010 RR 4 42 2 67 3.1% 3.19 [0.61, 16.66] Total (95% CI) 977 954 100.0% 0.63 [0.45, 0.88] Total events 125 178 Heterogeneity: Tau² = 0.28; Chi² = 43.35, df = 23 (P = 0.006); I² = 47% 0.05 0.2 1 5 20 Test for overall effect: Z = 2.67 (P = 0.008) Favors acetaminiphen Favors placebo Fig.3. Effectofintravenousacetaminophenonpostoperativevomiting.CI,confidenceinterval;BS,beforesurgery,singledose;IS,duringorimmediatelyaftersurgery,single dose;IR,duringorimmediatelyaftersurgery,repeateddose;RR,rescuemedication,repeateddose;PR,postoperative,repeateddose;BR,beforesurgery,repeateddose. Table2 Efficacyofi.v.acetaminophentoreducenauseaandvomiting. Comparison Acetaminophen Control Riskratio 95%CI P-valueofeffect P-valueheterogeneity P-valueEgger’stest Nausea 281/1122 351/1097 0.73 0.60–0.88 0.001 0.02 0.07 Industrysponsoredtrials/asrescue 144/437 123/413 1.12 0.85–1.48 0.42 0.32 0.60 Investigatorinitiatedtrials/prophylactic 137/685 228/684 0.63 0.54–0.75 <0.001 0.60 0.36 Beforesurgery 44/217 81/213 0.54 0.40–0.74 <0.001 0.92 0.31 Duringorimmediatelyaftersurgery 93/468 147/471 0.67 0.55–0.83 <0.001 0.39 0.50 Prophylacticsingledose 46/282 96/284 0.50 0.38–0.66 <0.001 0.94 0.64 Prophylacticrepeateddoses 91/403 132/400 0.72 0.58–0.89 0.002 0.41 0.74 Vomiting 125/977 178/954 0.63 0.45–0.88 0.008 0.006 0.20 Industry-sponsoredtrials/asrescue 75/436 49/413 1.41 1.02–1.96 0.04 0.57 0.40 Investigator-initiatedtrials/prophylactic 50/541 129/541 0.42 0.31–0.56 <0.001 0.90 0.40 Beforesurgery 9/181 34/178 0.29 0.14–0.57 <0.001 0.87 0.55 Duringorimmediatelyaftersurgery 41/360 95/363 0.46 0.33–0.63 <0.001 0.84 0.93 Prophylacticsingledose 16/236 57/238 0.31 0.19–0.51 <0.001 0.96 0.52 Prophylacticrepeateddoses 34/305 72/303 0.49 0.35–0.70 <0.001 0.77 0.96 i.v.,intravenous;CI,confidenceinterval. sponsoredtrials,andi.v.acetaminophenwasgivenprophylactically 4.3.Prophylacticsingleversusrepeateddosage in investigator-initiated trials. Furthermore, heterogeneity van- ished when the studies were stratified for timing. We classified As a secondarysensitivity analysis,we separated studies with thestudiesinto3mainsubgroups:studiesinwhichi.v.acetamino- prophylacticallyappliedi.v.acetaminophenintothosewhereasin- phenwasadministereda)beforesurgery,b)duringorshortlyafter gledosewasgiven,comparedtorepeateddosage.Whilethereduc- theendofsurgery,andc)afterrecoveryfromanesthesiaatonsetof tion of vomiting was not statistically significantly different pain. Intravenous acetaminophen was shown to be effective for betweensingleandrepeateddoses,nauseawasmorereducedby both nausea and vomiting only when administered prophylacti- thesingledose(riskratio0.31)comparedtorepeateddoses(risk cally.Incontrast,inindustry-sponsoredtrialsinwhichi.v.acetami- ratio0.49,P-valueforsinglevs.repeatedcomparison=0.04).These nophenwasgivenaftertheonsetofpain(oftenonlyinitiatedonthe datasuggestthati.v.acetaminophenreducesPONVatleastaswell firstpostoperativeday),itdidnotreducePONV.Hence,thediffer- asestablishedantiemetics,whichgenerallyhaveariskratioof0.60 encesbetweeninvestigator-initiatedandindustry-sponsoredtrials to0.80[1,12].However,itwouldbesurprisingifasingledoseof canbeattributedtothedifferencesintimingoftreatment. i.v.acetaminophenwouldbemorethantwiceaseffectiveascon- C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 685 Acetaminophen Control Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI 1.13.1 Before surgery Salihoglu 2009 BS 2 20 5 20 1.2% 0.40 [0.09, 1.83] Arici 2009 BS 3 28 9 27 1.9% 0.32 [0.10, 1.06] Toygar 2008 BS 4 30 7 30 2.2% 0.57 [0.19, 1.75] Jokela 2010 BR 8 40 12 40 4.5% 0.67 [0.31, 1.45] Hong2 2010 BR 14 63 27 61 9.4% 0.50 [0.29, 0.86] Moon 2011 BS 13 36 21 35 10.5% 0.60 [0.36, 1.00] Subtotal (95% CI) 217 213 29.7% 0.54 [0.40, 0.74] Total events 44 81 Heterogeneity: Tau² = 0.00; Chi² = 1.42, df = 5 (P = 0.92); I² = 0% Test for overall effect: Z = 3.93 (P < 0.0001) 1.13.2 Intraoperatively Uvarov 2008 IR 1 25 1 24 0.4% 0.96 [0.06, 14.50] Grundmann 2006 IS 1 20 1 20 0.4% 1.00 [0.07, 14.90] Lee 2010 IS 2 20 1 20 0.5% 2.00 [0.20, 20.33] Cattabriga 2007 IR 3 56 1 57 0.6% 3.05 [0.33, 28.48] Toygar 2008 IS 3 30 7 30 1.7% 0.43 [0.12, 1.50] Arici 2009 IS 3 27 9 27 1.9% 0.33 [0.10, 1.10] Cok 2011 IS 6 41 15 45 3.8% 0.44 [0.19, 1.02] Ohnesorge 2009 IR 9 27 7 26 4.0% 1.24 [0.54, 2.83] Topal 2009 IR 9 20 10 20 6.4% 0.90 [0.47, 1.73] Kilicaslan 2010 IR 11 25 12 25 7.6% 0.92 [0.50, 1.67] Brodner 2011 IR 15 49 21 49 9.7% 0.71 [0.42, 1.22] Arslan 2011 IS 9 20 18 20 10.7% 0.50 [0.30, 0.83] Cakan 2008 IR 12 20 17 20 17.0% 0.71 [0.47, 1.06] Subtotal (95% CI) 380 383 64.8% 0.70 [0.57, 0.86] Total events 84 120 Heterogeneity: Tau² = 0.00; Chi² = 10.68, df = 12 (P = 0.56); I² = 0% Test for overall effect: Z = 3.36 (P = 0.0008) 1.13.3 Immediately after surgery Fadly 2006 IS 0 10 3 10 0.3% 0.14 [0.01, 2.45] Gokten 2011 PR 1 20 6 20 0.7% 0.17 [0.02, 1.26] Memis 2010 PR 1 20 8 20 0.7% 0.13 [0.02, 0.91] Atef 2008 IR 7 38 10 38 3.8% 0.70 [0.30, 1.65] Subtotal (95% CI) 88 88 5.5% 0.31 [0.11, 0.89] Total events 9 27 Heterogeneity: Tau² = 0.40; Chi² = 4.50, df = 3 (P = 0.21); I² = 33% Test for overall effect: Z = 2.18 (P = 0.03) Total (95% CI) 685 684 100.0% 0.63 [0.54, 0.75] Total events 137 228 Heterogeneity: Tau² = 0.00; Chi² = 19.80, df = 22 (P = 0.60); I² = 0% 0.05 0.2 1 5 20 Test for overall effect: Z = 5.40 (P < 0.00001) Favors acetaminophen Favors placebo Test for subgroup differences: Chi² = 3.70, df = 2 (P = 0.16), I² = 46.0% Fig.4. Effectofprophylacticintravenousacetaminophenonpostoperativenauseaaccordingtotimingofadministration.CI,confidenceinterval;BS,beforesurgery,single dose;BR,beforesurgery,repeateddose;IR,duringorimmediatelyaftersurgery,repeateddose;IS,duringorimmediatelyaftersurgery,singledose;PR,postoperative, repeateddose. ventionalantiemetics.Inanyevent,fromtheperspectiveofPONV, expectations, reduction of PONV was not significantly associated asingledoseofprophylactici.v.acetaminophenisatleastaseffec- with reduction of morphine equivalents; however, it was associ- tiveasrepeateddosesorasantiemetics. atedwithareductioninpainintensity.Painitselfiscommonlybe- lievedtobeariskfactorforPONVamongclinicians,andthismay 4.4.Potentialmechanisms bethefirstreporttosupportthisnotion.Anotherpotentialmech- anismcouldbeadirectantiemeticeffectofacetaminophen.Infact, ToelucidatethemechanismsofPONVreductionbyi.v.acetami- acetaminophenismetabolizedinthebrainintoAM404,ametabo- nophen, we performed meta-regressions of the effects of opioid lite that is able to inhibit the reuptake of anandamide, a known consumptionandpainintensityonpostoperativenausea.Itisgen- cannabinoidCB1andCB2receptoragonist.Ithasbeenshownthat erallyacceptedthattheantiemeticeffectofNSAIDsismediatedby decreasedanandamidelevelsareassociatedwithanincreasedrate adose-dependentreductioninopioidconsumption;thus,itseems of nausea and vomiting in humans [14]. Therefore, it is possible plausible that i.v. acetaminophen might also reduce PONV by thatacetaminophensimplyhasadirecteffectonPONVbyincreas- reducing postoperative opioid requirements. In contrast to our inganandamidelevels. 686 C.C.Apfeletal./PAIN(cid:3)154(2013)677–689 Acetaminophen Control Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI 1.15.1 Single Dosage Arici 2009 BS 3 28 9 27 1.9% 0.32 [0.10, 1.06] Arici 2009 IS 3 27 9 27 1.9% 0.33 [0.10, 1.10] Arslan 2011 IS 9 20 18 20 10.7% 0.50 [0.30, 0.83] Cok 2011 IS 6 41 15 45 3.8% 0.44 [0.19, 1.02] Fadly 2006 IS 0 10 3 10 0.3% 0.14 [0.01, 2.45] Grundmann 2006 IS 1 20 1 20 0.4% 1.00 [0.07, 14.90] Lee 2010 IS 2 20 1 20 0.5% 2.00 [0.20, 20.33] Moon 2011 BS 13 36 21 35 10.5% 0.60 [0.36, 1.00] Salihoglu 2009 BS 2 20 5 20 1.2% 0.40 [0.09, 1.83] Toygar 2008 BS 4 30 7 30 2.2% 0.57 [0.19, 1.75] Toygar 2008 IS 3 30 7 30 1.7% 0.43 [0.12, 1.50] Subtotal (95% CI) 282 284 35.3% 0.50 [0.38, 0.66] Total events 46 96 Heterogeneity: Tau² = 0.00; Chi² = 4.20, df = 10 (P = 0.94); I² = 0% Test for overall effect: Z = 4.84 (P < 0.00001) 1.15.2 Repeated Dosage Atef 2008 IR 7 38 10 38 3.8% 0.70 [0.30, 1.65] Brodner 2011 IR 15 49 21 49 9.7% 0.71 [0.42, 1.22] Cakan 2008 IR 12 20 17 20 17.0% 0.71 [0.47, 1.06] Cattabriga 2007 IR 3 56 1 57 0.6% 3.05 [0.33, 28.48] Gokten 2011 PR 1 20 6 20 0.7% 0.17 [0.02, 1.26] Hong2 2010 BR 14 63 27 61 9.4% 0.50 [0.29, 0.86] Jokela 2010 BR 8 40 12 40 4.5% 0.67 [0.31, 1.45] Kilicaslan 2010 IR 11 25 12 25 7.6% 0.92 [0.50, 1.67] Memis 2010 PR 1 20 8 20 0.7% 0.13 [0.02, 0.91] Ohnesorge 2009 IR 9 27 7 26 4.0% 1.24 [0.54, 2.83] Topal 2009 IR 9 20 10 20 6.4% 0.90 [0.47, 1.73] Uvarov 2008 IR 1 25 1 24 0.4% 0.96 [0.06, 14.50] Subtotal (95% CI) 403 400 64.7% 0.72 [0.58, 0.89] Total events 91 132 Heterogeneity: Tau² = 0.00; Chi² = 11.35, df = 11 (P = 0.41); I² = 3% Test for overall effect: Z = 3.04 (P = 0.002) Total (95% CI) 685 684 100.0% 0.63 [0.54, 0.75] Total events 137 228 Heterogeneity: Tau² = 0.00; Chi² = 19.80, df = 22 (P = 0.60); I² = 0% 0.05 0.2 1 5 20 Test for overall effect: Z = 5.40 (P < 0.00001) Favors acetaminophen Favors placebo Test for subgroup differences: Chi² = 4.06, df = 1 (P = 0.04), I² = 75.4% Fig.5. Effectofsinglevsrepeateddosingofintravenousacetaminophenonpostoperativenausea.CI,confidenceinterval;BS,beforesurgery,singledose;IS,duringor immediatelyaftersurgery,singledose;IR,duringorimmediatelyaftersurgery,repeateddose;RR,rescuemedication,repeateddose;PR,postoperative,repeateddose;BR, beforesurgery,repeateddose. 4.5.Limitations 4.6.Conclusion The validity of any systematic review and meta-analysis de- Insummary,thissystematicreviewandmeta-analysisdemon- pends on the validity and quality of the published material, the stratedthatprophylactici.v.acetaminophenreducespostoperative methodological approach, and appropriate critical appraisal of nauseaandvomitingwithaneffectsizethatcompareswellwith the study results. We have thus analyzed the data regarding the data known from other antiemetics. The results from our meta- riskofbiasofvariouscomponentsanddidnotfindsignificantdif- regressionsuggestthattheantiemeticeffectofi.v.acetaminophen ferences.Inparticular,neitherfunnelplotsnorEgger’sregression isnotmediatedthroughthereductionofpostoperativeopioidcon- tests suggested any evidence for a publication bias. However, sumption, but through direct mechanisms or through the reduc- investigator-initiated studies were generally small and of similar tionofpostsurgicalpain. size, which limits the chance to detect publication bias. Thus, although the effectiveness of prophylactically administered Conflictofintereststatement i.v. acetaminophen was consistent (without any significant in- between-study heterogeneity), a large, well-designed random- Dr.ApfelwasontheSpeaker’sBureauofCadencePharmaceuti- ized-controlled trial would be warranted to confirm these cals.Noneoftheotherauthorshaveanyotherpotentialconflictof findings. interesttoreport.

Intravenous acetaminophen reduces postoperative nausea and

Similar Intravenous acetaminophen reduces postoperative nausea and