Perioperative fasting in adults and children

14 Pages·2011·0.21 MB·English

GUIDELINES
Perioperative fasting in adults and children: guidelines from
the European Society of Anaesthesiology
Ian Smith, Peter Kranke, Isabelle Murat, Andrew Smith, Geraldine O’Sullivan, Eldar Søreide, Claudia Spies
and Bas in’t Veld
This guidelineaims toprovide an overview ofthepresent surgery inadults andchildren, althoughpatients shouldnot
knowledgeonaspectsofperioperativefastingwithassessment havetheiroperationcancelledordelayedjustbecausetheyare
ofthe qualityofthe evidence. Asystematic search was chewing gum,sucking a boiled sweetor smoking immediately
conducted inelectronic databases toidentify trials published priortoinductionofanaesthesia.Theserecommendationsalso
between 1950andlate 2009concerned with preoperative apply topatients withobesity, gastro-oesophageal reflux and
fasting, early resumption oforal intake andthe effects oforal
diabetesandpregnantwomennotinlabour.Thereisinsufficient
carbohydrate mixtures on gastricemptying andpostoperative
evidence torecommend theroutine use ofantacids,
recovery.Onestudyonpreoperativefastingwhichhadnotbeen
metoclopramide or H -receptor antagonists before elective
included inprevious reviews and afurther 13studies 2
surgeryinnon-obstetricpatients,butanH -receptorantagonist
published since themost recentreview wereidentified. 2
should begivenbefore elective caesareansection, withan
The searches also identified 20potentially relevant studies
intravenous H -receptor antagonist given prior toemergency
oforal carbohydrates and 53onearly resumption oforal 2
caesareansection, supplemented with30ml of0.3moll(cid:1)1
intake.Publications were classifiedintermsof their
evidence level,scientific validity andclinical relevance. The sodiumcitrateifgeneralanaesthesiaisplanned.Infantsshould
ScottishIntercollegiateGuidelinesNetworkscoringsystemfor befedbeforeelectivesurgery.Breastmilkissafeupto4hand
assessinglevelofevidenceandgradeofrecommendationswas othermilksupto6h.Thereafter,clearfluidsshouldbegivenasin
used. The keyrecommendations are thatadults andchildren adults.The guidelines alsoconsiderthe safety andpossible
should beencouraged to drink clear fluidsupto 2hbefore benefitsofpreoperativecarbohydratesandofferadviceonthe
elective surgery (including caesareansection) andall butone postoperative resumption oforal intake.
memberoftheguidelinesgroupconsiderthatteaorcoffeewith EurJ Anaesthesiol2011;28:556–569
milkadded(uptoaboutonefifthofthetotalvolume)arestillclear
fluids. Solidfood should beprohibited for 6h before elective Publishedonline28June2011
Why were these guidelines produced?
WidespreadconsultationsuggestedthatguidelinesonperioperativefastingwouldbeusefultoEuropeanSocietyof
Anaesthesiology (ESA) members.
Ourguidelineaimstoprovideanoverviewofthepresentknowledgeonperioperativefastingwithassessmentofthe
qualityoftheevidenceinordertoallowanaesthesiologistsalloverEuropetointegratethisknowledgeintheirdaily
care of patients.
What is similar to previous guidelines?
The ESA guidelines endorse a 2-h fasting interval for clear fluids and a 6-h interval for solids.
What is different from previous guidelines?
The ESA guidelines:
(cid:2) are recent and include several studies published since previous guidelines;
(cid:2) increase the emphasis on encouraging patients not to avoid fluids for any longer than is necessary;
(cid:2) offer practical, pragmatic advice on chewing gum, smoking and drinks containing milk;
(cid:2) consider the safety and possible benefits of preoperative carbohydrates;
(cid:2) offer advice on the postoperative resumption of oral intake.
FromtheUniversityHospitalofNorthStaffordshire,Stoke-on-Trent,UK(IS),UniversityHospitalsofWu¨rzburg,Wu¨rzburg,Germany(PK),ArmandTrousseauHospital,Paris,
France(IM),RoyalLancasterInfirmary,SchoolofHealthandMedicine,LancasterUniversity,Lancaster(AS),Guy’sandStThomas’NHSFoundationTrust,London(GOS),
UK,DepartmentofAnaesthesiaandIntensiveCare,StavangerUniversityHospital,Stavanger,Norway(ES),DepartmentofAnaesthesiologyandIntensiveCareMedicine,
Charite´ – Universita¨tsmedizin Berlin, Berlin, Germany (CS) and Department of Anesthesiology and Pain Medicine, Haaglanden Medical Centre, The Hague, The
Netherlands(BV)
CorrespondencetoDrIanSmith,DirectorateofAnaesthesia,UniversityHospitalofNorthStaffordshire,NewcastleRoad,Stoke-on-Trent,StaffordshireST46QG,UK
Tel:[phone];e-mail:[email protected]
0265-0215(cid:1)2011CopyrightEuropeanSocietyofAnaesthesiology DOI:10.1097/EJA.0b013e3283495ba1
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Perioperativefastinginadultsandchildren 557
1. Summary of recommendations
Evidence Recommendation
Fastinginadultsandchildren
Adultsandchildrenshouldbeencouragedtodrinkclearfluids(includingwater, 1þþ A
pulp-freejuiceandteaorcoffeewithoutmilk)upto2hbeforeelectivesurgery
(includingcaesareansection)
Allbutonememberoftheguidelinesgroupconsiderthatteaorcoffeewithmilkadded
(uptoaboutonefifthofthetotalvolume)arestillclearfluids
Solidfoodshouldbeprohibitedfor6hbeforeelectivesurgeryinadultsandchildren 1þ A
Patientswithobesity,gastro-oesophagealrefluxanddiabetesandpregnantwomen 2(cid:1) D
notinlabourcansafelyfollowalloftheaboveguidelines
However,thesefactorsmayaltertheiroverallanaestheticmanagement
Patientsshouldnothavetheiroperationcancelledordelayedjustbecausetheyare 1(cid:1) B
chewinggum,suckingaboiledsweetorsmokingimmediatelypriortoinduction
ofanaesthesia
Theaboveisbasedsolelyoneffectsongastricemptyingandnicotineintake(includingsmoking,
nicotinegumandpatches)shouldbediscouragedbeforeelectivesurgery
Fastingininfants
Infantsshouldbefedbeforeelectivesurgery.Breastmilkissafeupto4hand 1þþ A
othermilksupto6h.Thereafter,clearfluidsshouldbegivenasinadults
Prokineticandotherpharmacologicalinterventions
Thereisinsufficientevidenceofclinicalbenefittorecommendtheroutineuseof 1þþ A
antacids,metoclopramideorH-receptorantagonistsbeforeelectivesurgeryin
2
non-obstetricpatients
AnH2-receptorantagonistshouldbegiventhenightbefore,andonthemorningof, 1þþ A
electivecaesareansection
Theguidelinesgrouprecognisesthatmostoftheevidencerelatestosurrogatemeasures,
suchaschangesingastricvolumeandpH,ratherthanaclearimpactonmortality
AnintravenousH2-receptorantagonistshouldbegivenpriortoemergencycaesarean 1þþ A
section;thisshouldbesupplementedwith30mlof0.3mollS1sodiumcitrate
ifgeneralanaesthesiaisplanned
Theguidelinesgrouprecognisesthatmostoftheevidencerelatestosurrogatemeasures,
suchaschangesingastricvolumeandpH,ratherthanaclearimpactonmortality
Oralcarbohydrates
Itissafeforpatients(includingdiabetics)todrinkcarbohydrate-richdrinksupto 1þþ A
2hbeforeelectivesurgery
Theevidenceforsafetyisderivedfromstudiesofproductsspecificallydevelopedforperioperative
use(predominantlymaltodextrins);notallcarbohydratesarenecessarilysafe
Drinkingcarbohydrate-richfluidsbeforeelectivesurgeryimprovessubjective 1þþ A
wellbeing,reducesthirstandhungerandreducespostoperativeinsulinresistance
Todate,thereislittleclearevidencetoshowreductionsinlengthofpostoperativestayandmortality
Fastinginobstetricpatients
Womenshouldbeallowedclearfluids(asdefinedabove)astheydesireinlabour 1þþ A
Solidfoodshouldbediscouragedduringactivelabour 1þ A
Theguidelinesgrouprecognisethatitmaybeimpracticaltostopallwomenfromeatingduringlabour,
especiallylow-riskwomen.Considerationshouldbegiventoeasilydigestible,low-residuefoods
Postoperativeresumptionoffluids
Adultsandchildrenshouldbeallowedtoresumedrinkingassoonastheywishafter 1þþ A
electivesurgery.However,fluidintakeshouldnotbeinsisteduponbeforeallowing
dischargefromadayorambulatorysurgeryfacility
,recommendedbestpracticebasedontheclinicalexperienceoftheguidelinesdevelopmentgroup.
2. Purpose and development of the guideline withexpertiseintheirfieldtojointhetaskforce.Further
The European Society of Anaesthesiology (ESA) is experts were co-opted onto the task force as required.
committed to the production of high-quality, evidence- SeveralEuropeannationalanaesthesiologysocietieshave
based clinical guidelines. After the formation of the alreadyproducedrecommendations for aspectsof perio-
Guidelines Committee in 2008, a prioritisation exercise perative fasting. Our guideline aims to provide an over-
suggested that guidelines on perioperative fasting view of the present knowledge on the subject with
would be useful to ESA members and a task force was assessment of the quality of the evidence in order to
established in June 2009 to produce this guideline. The allow anaesthesiologists all over Europe to integrate –
chairpersons of the relevant subcommittees (Evidence- wherever possible – this knowledge in their daily care
based Practice and Quality Improvement, Ambulatory of patients.
Anaesthesia,ObstetricAnaesthesia,PaediatricAnaesthe-
siaandAnaesthesiafortheElderly)oftheESAScientific Evidencetosupporttherecommendationswasobtained
Committee were asked to nominate an ESA member as follows. A systematic search was conducted by
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
558 Guidelines
members of Cochrane Anaesthesia Review Group of European nation. Although national societies and
the electronic databases Ovid, MEDLINE and Embase individuals are free to use the guidelines, modified as
to identify trials published between 1950 and late 2009 necessaryforlocalandnationalpracticecontexts,theyare
concernedwithpreoperativefasting,earlyresumptionof undernoobligationtodoso.Further,thepotentiallegal
oralintakeandtheeffectsoforalcarbohydratemixtures implications may be a point of concern.6 It cannot be
on gastric emptying and postoperative recovery. A total emphasised enough that guidelines may not be appro-
of3714abstractsfromMEDLINEand3660fromEmbase priate forallclinicalsituations.Thedecision whether or
were identified from the search. After elimination nottofollowarecommendationfromaguidelinemustbe
of duplicates, irrelevant studies, non-clinical trials and madebytheresponsiblephysicianonanindividualbasis,
studies with a non-clinical outcome, one study on takingintoaccountthespecificconditionsofthepatient
preoperative fasting which had not been included in and the available resources. Therefore, deviations from
previous reviews1–4 and a further 13 studies published guidelinesforspecificreasonsshouldremainpossibleand
since the most recent review4 were identified. The cancertainlynotbe interpretedasabasefor negligence
searches also identified 20 potentially relevant studies claims. However, we hope that these guidelines will
of oral carbohydrates and 53 on early resumption of oral both assist anaesthesiologists throughout Europe to
intake. bringresearchevidencetobearontheirclinicalpractice
and also provide support to colleagues and healthcare
These publications were classified in terms of their
fundersinmakingchangesandimprovementsnecessary
evidence level, scientific validity and clinical relevance.
to enhance patient care.
WeusedtheScottishIntercollegiateGuidelinesNetwork
(SIGN)scoringsystemforassessinglevelofevidenceand Differences from existing guidelines
grade of recommendations (Fig. 1).5 Highest priority
Although there is little new evidence relating to fasting
was given to meta-analyses of randomised, controlled
for fluids and solids, the current guidelines review more
clinicaltrials.Inreachingconsensus,particularemphasis
recent literature than any of the existing guidelines.
was placed on the level of evidence, ethical aspects,
In addition, the American Society of Anesthesiology
patientpreferences,clinicalrelevance,risk/benefitratios
(ASA) guidelines on the subject7 were published in
and degree of applicability. For example, a pragmatic
1999 and contain little on preoperative carbohydrate,
solution to an acceptable amount of milk in tea or
whereas the UK Royal College of Nursing guidelines4
coffee was agreed based on the unpublished experience
dealwiththesafetyaspectofpreoperativecarbohydrate,
accumulated by several members of the group over
but not possible benefits. In these current guidelines,
many years.
we have also tried to address practical problems such as
chewing gum.
These guidelines have undergone the following review
process. The final draft was reviewed by members of
the relevant Subcommittees of the ESA’s Scientific 3. Fasting
Committee who were not involved in the initial pre-
3.1. Fluids
paration of the guideline. It was posted on the on ESA
Recommendation
websitefor4weeksandallESAmembers,individualand
national,werecontactedbyelectronicmailtoinvitethem Adultsandchildrenshouldbeencouragedtodrinkclear
to comment on the draft. It was also sent to the Inter- fluids (including water, pulp-free juice and tea or coffee
national Association for Ambulatory Surgery (IAAS) for withoutmilk)upto2hbeforeelectivesurgery(including
information and comment. All those who commented caesarean section) (evidence level 1þþ, recommen-
are listed in the ‘Acknowledgements’ section below. dation grade A).
Comments were collated by the chair of the guideline
Allbutonememberoftheguidelinesgroupconsiderthatteaor
task force and the guideline amended as appropriate.
coffeewithmilkadded(uptoaboutonefifthofthetotalvolume)
The final manuscript was approved by the Guidelines
are still clear fluids.
CommitteeandBoardoftheESAbeforesubmissionfor
publication in the European Journal of Anaesthesiology. Rationale
SincethelandmarkworkofMaltbyetal.8in1986,alarge
These guidelines are produced as a service to ESA
body of evidence has been accumulated to show that
members and other anaesthesiologists and healthcare
theoralintakeofclearfluidsupto2hbeforeanelective
staff in Europe. The ESA recognises that practice and
operation is safe.3,9,10 Many countries have, therefore,
opinion varies in different European countries. Despite
changed their fasting guidelines, allowing most patients
theavailabilityofthesamescientificinformation,theway
take clear fluids (water, clear juices and coffee or tea
in which healthcare services are organised may result in
without milk) up to 2h before elective surgery.11
different practices in the various European countries.
Thus, it is not always possible to produce guidelines In addition to the liberalising of fasting guidelines, the
which will be both appropriate and relevant for every emphasis is now changing, with the realisation that
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Perioperativefastinginadultsandchildren 559
Fig.1
Key to evidence statements and grades of recommendations
1++ High-quality meta-analyses, systematic reviews of RCTs, or RCTs with a very low risk of bias
1+ Well conducted meta-analyses, systematic reviews, or RCTs with a low risk of bias
1 - Meta-analyses, systematic reviews, or RCTs with a high risk of bias
2++ High-quality systematic reviews of case–control or cohort studies
Well conducted case–control or cohort studies with a low risk of confounding or bias and a moderate
2+ probability that the relationship is causal
2 - Case–control or cohort studies with a high risk of confounding or bias and a significant risk that the
relationship is not causal
3 Non-analytic studies, e.g. case reports, case series
4 Expert opinion
Grades of recommendation
Note: the grade of recommendation relates to the strength of the evidence on which the recommendation is
based. It does not refect the clinical importance of the recommendation.
At least one meta-analysis, systematic review, or RCT rated as 1++,
A and directly applicable to the target population; or
a body of evidence consisting principally of studies rated as 1+,
directly applicable to the target population, and demonstrating overall consistency of results
A body of evidence including studies rated as 2++,
B directly applicable to the target population, and demonstrating overall consistency of results; or
extrapolated evidence from studies rated as 1++ or 1+
A body of evidence including studies rated as 2+,
C directly applicable to the target population, and demonstrating overall consistency of results; or
extrapolated evidence from studies rated as 2++
Evidence level 3 or 4; or
D extrapolated evidence from studies rated as 2+
Good practice points
√
Recommended best practice based on the clinical experience of the guideline development group
ScottishIntercollegiateGuidelinesNetwork(SIGN)gradingsystem.RCT,randomisedcontrolledtrial.5
prolonged fasting is an inappropriate way to prepare drinking up until 2h before surgery in order to reduce
for the stress of surgery. Abstaining from fluids for their discomfort and improve their well being.
a prolonged period prior to surgery is detrimental for
patients,especiallytheelderlyandsmallchildren.Rather 3.1.1. Milkintea or coffee
than ensuring a minimal fasting interval has been Milk in large quantities curdles in the stomach and acts
achieved, it is important to encourage patients to keep likeasolid,butsmallerquantitiesarehandledlikeother
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
560 Guidelines
liquids and are safe. There is anecdotal evidence boiled sweet or smoking immediately prior to induction
(includingfromsomemembersofthisgroup)thatwhen of anaesthesia
milk is allowed to be added to tea or coffee consumed
The above is based solely on effects on gastric emptying and
before elective surgery, regurgitation and aspiration are
nicotine intake (including smoking, nicotine gum and patches)
no more likely to occur, but no randomised studies
should be discouraged before elective surgery.
have looked specifically at the safety of this practice.
Some studies of preoperative tea and coffee12–14 did Rationale
allow milk to be added if desired (R. Maltby, personal
There is ongoing debate on how to deal with patients
communication),butthisisnotrecordedinthepublished
chewing gum in the immediate preoperative period and
textandthenumberofsuchpatientswassmall.Unpub-
what constitutes a safe fasting interval. There are only
lished work has shown that adding small quantities of
three (partly) randomised controlled studies concerning
milk(froma12mlsingle-portionpot)toamodelstomach
the intake of chewing gum during the perioperative
caused no restriction inemptying, butthat addingthree
fasting period.
ormoremeasurescausedclumping(R.Maltby,personal
communication).However,the model compriseda glass In one comparison of 77 patients,16 16 did not chew
vesselwithafixedburettetapasanoutletand,therefore, any kind of gum, 15 patients were allowed to chew
did not mimic either a sphincter which can relax or the gum until transfer to the operating room and 46 were
effects of churning from muscle contractions. allowed to chew gum as long as they wished, even until
thetimeofanaesthesiainduction.Thelastgroupwasnot
The lack of evidence from human studies and the
allocated by means of randomisation. Both gastric fluid
uncertainty in defining and controlling a safe amount
volume as well as pH did not differ significantly among
ofmilk,meansthatmostguidelinesonlyadvocateblack
the three groups.
tea orcoffee. Althoughthis may appear asafe approach,
somepatientswouldratherhavenothingatalliftheyare Another study compared 46 children between 5 and
denied milk in their morning cup of tea or coffee. With 17yearsoldwhowereallowedtocheweithersugar-free
one exception, the guidelines group considered that tea or sugared gum up to 30min before transfer to the
or coffee with a modest amount of milk added (up to operating room. Both the sugar-free and sugared gum
about one fifth of the total volume) should still be chewershadsignificantlyhighergastricfluidvolumeand
considered as clear fluids and, therefore, safe up to pHthanthecontrolgroupwhodidnotchewanygum.17
2h before the induction of anaesthesia. Drinks made
Søreideetal.18compared 106femalepatientsscheduled
predominantly from milk, however, should be treated
for elective gynaecologic surgery. They were either
as solids.
smokers or non-smokers and were allowed to chew
nicotine-containing chewing gum or nothing (smokers)
3.2. Solid food orsugar-freegumornothing(non-smokers),respectively.
Recommendation Up to one chewing gum per hour was given until
transportation to the operating room. The non-smoking
Solid food should be prohibited for 6h before elective
chewers as well as the smokers (chewing or not) had
surgeryinadultsandchildren(evidencelevel1þ,recom-
significantly higher gastric fluid volume than the non-
mendation grade A).
smokers who did not chew gum. As far as gastric pH
Rationale values are concerned, the levels were higher in both
non-smoking groups than in both smoking groups.
Norecentstudieshaveattemptedtodefineaminimalsafe
Nocaseofaspirationorothercomplicationduringanaes-
periodforpreoperativefastingforsolidfood.Oneprevious
thesia induction was reported. Although the differences
study found no increase in gastric volume after a light
in pH and gastric volumes were statistically significant,
breakfastofteaandbutteredtoastconsumed2–4hbefore
electivesurgery,15butthepresenceofresidualsolidsinthe the authors did not believe the difference (30 versus
20ml) was clinically significant (E. Søreide, personal
stomachatinductionofanaesthesiacouldnotberuledout
communication).
bythemethodologyused.Itremainscommonpracticeto
avoid solid food for at least 6h before elective surgery.
Most patients will accept this if they are permitted to 3.4. Patients with delayed gastric emptying
drinkuntilclosertotheirsurgery.Thereisnoclearbenefit Recommendation
to reducing the fasting time for solids below 6h.
Patients with obesity, gastro-oesophageal reflux and
diabetes and pregnant women not in labour can safely
3.3. Chewing gum, sweets and smoking follow all of the above guidelines (evidence level 2(cid:1),
Recommendation recommendation grade D).
Patients should not have their operation cancelled or These factors may, however, alter their overall anaesthetic
delayed just because they are chewing gum, sucking a management.
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Perioperativefastinginadultsandchildren 561
Rationale metoclopramide10mgorally60–90minbeforeinduction
of anaesthesia. There was no significant difference in
A great number of factors can potentially delay gastric
either gastric pH or fluid volume.21
emptying. These include obesity, gastro-oesophageal
refluxanddiabetes.Studiesofpreoperativefastinghave So far, valid studies that investigate the effect of
notevaluatedthesegroupsofpatientsadequatelyenough preoperatively administered metoclopramide alone on
to provide definitive evidence. However, the evidence gastricpHandgastricfluidvolumearelacking,although
which does exist suggests that limitation of gastric metoclopramide significantly improves gastric emptying
emptying is, at most, mild and that these patients can incardiacsurgerypatients18hpostoperativelycompared
followthesameguidelinesashealthyadults.Thisadvice with placebo.22
also applies to pregnant women who are not in labour.
So far, there is insufficient evidence that preoperatively
Opioid analgesia can also delay gastric emptying, but
administered metoclopramide alone improves clinical
again there is insufficient evidence to make any
outcome, reduces gastric fluid volume or increases
recommendation. However, patients who have recently
gastric pH.
taken sufficient opioids to have a significant effect on
gastric emptying are unlikely to be undergoing elective
4.2. Histamine H -antagonists and proton pump
surgery often. 2
inhibitors
The mechanisms of action of H -antagonists and
4. Medications 2
proton pump inhibitors (PPIs) differ. The former block
Recommendation
H -receptors on the stomach’s parietal cells, thereby
2
There is insufficient evidence of clinical benefit to inhibitingthestimulatoryeffectsofhistamineongastric
recommendtheroutineuseofantacids,metoclopramide acid secretion; the latter block the enzyme system
orH -receptorantagonistsbeforeelectivesurgeryinnon- of hydrogen/potassium ATPase (Hþ/Kþ ATPase), the
2
obstetricpatients(evidencelevel1þþ,recommendation ‘proton pump’ of the gastric parietal cell, such that the
grade A). stimulatory actions of histamine, gastrin and acetyl-
choline are inhibited. Both have been applied with the
Rationale
aimofdecreasingtheriskofdeleteriouseffectsresulting
from a potential acid aspiration syndrome.
4.1. Prokinetic medications
In contrast to the prevalence of the perioperative use A recent meta-analysis23 comparing these medications
of prokinetics, there is limited evidence to support to therapeutic targets suggests that pre-medication
the prophylactic use of these agents to reduce the risk with ranitidine is more effective than PPIs in reducing
of perioperative aspiration of gastric contents. the volume of gastric secretions (by an average of
0.22mlkg(cid:1)1, 95% confidence interval 0.04–0.41) and
There are single studies that investigate the effect of
increasing gastric pH (by an average of 0.85 pH units,
prokineticsongastricpHandgastricfluidvolumeduring
95% confidence interval 1.14–0.28). These conclusions
anaesthesiainduction.Iqbaletal.19compared75women
could be drawn based on nine randomised controlled
undergoingcaesareansectionundergeneralanaesthesia.
trials,ofwhichsevenweresuitableformeta-analysis.In
Twenty-five women were administered both an
these trials a total of 223 patients received ranitidine,
H -antagonist (ranitidine) with a prokinetic drug (meto-
2 whichwasthesoleH -blockerusedintheincludedtrials,
clopramide), whereas 25 women served as a placebo 2
and 222 patients received different PPIs (omeprazole,
control group (another 25 patients received only raniti-
lansoprazole,pantoprazoleandrabeprazole).Overall,the
dine).Thecombinationofthetwodrugswassignificantly
size of the trials is rather small. Further, heterogeneity
more effective in increasing the pH and reducing the
couldalsobedetectedwithrespecttopreoperativefast-
gastric fluid volume than placebo.19
ing time, route of administration, repeat administration
Hong20 investigated the effect of ranitidine and meto- and the specific PPIs used. It is interesting to note
clopramide versus placebo. Forty patients scheduled for that patients in the trials received ranitidine at doses
laparoscopic gynaecological surgery were administered equivalent to,orlessthan, thedailyrecommendeddose
either 50mg ranitidine with 10mg metoclopramide for the maintenance of peptic ulcer disease, whereas
intravenously (n¼20) or the same volume of isotonic patientsreceivedPPIsatdoseshigherthanthoserecom-
saline in the control group. Gastric fluid volume was mended for this purpose.
significantly higher in the placebo group, as was the
It is not clear how long the potential protective effect
gastric pH in the treatment group.20
on gastric volume or pH lasts. It is also unclear whether
Bala et al.21 compared the combination of ranitidine– these observed effects can be extrapolated to patient
erythromycin with ranitidine–metoclopramide. Forty populations with a higher risk of aspiration, as all the
ASA I or II patients were given either erythromycin included trials appeared to be in patients at very low
250mg and ranitidine 150mg or ranitidine 150mg and riskofaspirationandtheobservedparameterswereused
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
562 Guidelines
asasurrogateforthe‘trueoutcome’,thatis,mortalityor 5.1. Carbohydrates versus clear liquids or intravenous
adult respiratory distress syndrome following gastric infusion
aspiration, which could not be evaluated. Taniguchi et al.28 investigated the safety and effective-
ness of oral rehydration as compared with intravenous
5. Preoperative carbohydrates: gastric rehydration prior to general anaesthesia. Fifty patients
emptying and potential benefits were randomised to either 1000ml of oral rehydration
Recommendation solutionor1000mlofanintravenouselectrolytesolution.
Volume of gastric contents, as measured directly after
Itissafeforpatients(includingdiabetics)todrinkcarbo-
induction,wassignificantlylowerintheoralrehydration
hydrate-rich drinks up to 2h before elective surgery
group.
(evidence level 1þþ, recommendation grade A).
Kaska et al.29 performed a randomised controlled trial
The evidence for safety is derived from studies of
comparing preoperative fasting with preoperative
products specifically developed for perioperative use
preparation with either oral or intravenous intake of
(predominantly maltodextrins); not all carbohydrates
carbohydrates, minerals and water.29 Oral intake shortly
are necessarily safe.
before surgery did not increase gastric residual volume
Rationale and was not associated with any risk.
Studiesinanimalmodelsofseverestress,suchashaemor- In the study by Nygren et al.,30 gastric emptying of a
rhageandendotoxaemia,showedthatseveralkeysystems carbohydrate-richdrinkwasinvestigatedbeforeelective
involvedinthestressresponsesweremarkedlyimpaired surgery and in a control situation. Patients served as
eveniftheanimalhadbeenfastedforabriefperiodbefore their own control pre and postoperatively. Despite the
the onset of a given stress. These key systems included increasedanxietyexperiencedbypatientsbeforesurgery,
fluid homeostasis, stress hormone release, aspects of gastricemptyingdidnotdifferbetweentheexperimental
metabolism,musclefunctionandgutintegrity.24Ifthese and control situations.
models were fasted for as long as 24h, there was also a
Jarvela et al.31 investigated the effect of a preoperative
difference in survival. This indicates that the metabolic
oralcarbohydratedrinkversusovernightfastingonperi-
change caused by a recent meal (as opposed to fasting)
operative insulin requirements in 101 non-diabetic
andthelossofglycogenoccurringevenafterabrieffastis
patientsundergoingelectivecoronaryarterybypassgraft-
sufficientto alter the stress response.
ing.Accordingtotheirfindings,itissafetoallowcardiac
Allowing patients to drink clear fluids up to 2h prior to surgery patients to drink clear fluids up to 2h before
surgery is not likely to produce any major change in inductionofanaesthesia,becausegastricemptyingofthe
metabolism,asthesedrinksusuallydonotcontainsuffi- drink was almost total and no aspiration occurred.
cient energy. The best known method for changing
Breuer et al.32 studied the effects of preoperative oral
metabolism from the overnight fasted state to that of a
carbohydrate administration on gastric fluid volume.
fed state is the use of carbohydrates. The key change
Before surgery, 188 ASA physical status III–IV patients
required to be achieved is a prompt insulin response,
undergoing elective cardiac surgery were randomised to
preferably to an extent similar to that observed after
receiveaclear12.5%carbohydratedrink,flavouredwater
intake of a meal.
(placebo), or to fast overnight (control). Carbohydrates
In the first instance, intravenous glucose has been and placebo were treated in double-blind format and
proposedandusedforthispurpose.Theinsulinresponse patients received 800ml of the corresponding beverage
toglucoseinfusionsisdeterminedbytherateofdelivery in the evening and 400ml 2h before surgery. Ingested
of glucose in a dose-dependent manner. Infusion of liquids did not cause increased gastric fluid volume or
glucose (and insulin) has been shown to induce an other adverse events.
insulin response to levels of about 60mUml(cid:1)1,25 reduce
Inthese fiverandomisedstudies,therewasno evidence
postoperative insulin resistance26 and retain substrate
of an increased gastric volume after ingesting carbo-
oxidation. This is important because postoperative
hydrates. Care should be taken in extrapolating this
insulinresistanceandhyperglycaemiaareassociatedwith
evidence beyond those specific carbohydrates which
an impaired outcome after surgery.27 Preoperative oral
havebeenstudied;notalloralcarbohydrateswillnecess-
carbohydrate loading in humans also reduces postopera-
arily behave similarly.
tive insulin resistance. Dietary interventions, therefore,
representapromisingandattractivetherapeuticstrategy
to optimise postprandial glycaemia. Thus far, inter- 5.2. Diabetic patients versus healthy individuals
ventions with respect to the preoperative addition of Investigators have been reluctant to give diabetic
carbohydrateshavefocusedonsafety,metaboliceffects, patients oral carbohydrates because of the unknown
personal perioperative well being and postoperative effects on preoperative glycaemia and gastric emptying.
length of stay. Gustafssonetal.33investigatedtheeffectofpreoperative
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Perioperativefastinginadultsandchildren 563
oral carbohydrate loading in type 2 diabetic patients. study of 14 patients displayed less reduced insulin
Twenty-fivepatientswithtype2diabetesand10healthy sensitivityaftercolorectalsurgeryfollowingpreoperative
controlswerestudied.Acarbohydrate-richdrink(400ml, oral carbohydrate administration as compared with
12.5%) was given with paracetamol 1.5g for deter- patients who were operated on after an overnight fast.39
mination of gastric emptying. Patients with type A recently published study in patients undergoing open
2diabetesshowednosignsofdelayedgastric emptying, colorectal surgery also showed reduced postoperative
suggesting that a carbohydrate-rich drink may be safely insulin resistance after preoperative oral carbohydrates,
administrated180minbeforeanaesthesiawithoutriskof as well as reduced thirst and hunger.40 However, a pre-
hyperglycaemia or aspiration preoperatively. operative oral carbohydrate drink did not reduce post-
operative insulin resistance or postoperative nausea and
On the basis of this limited evidence, diabetes (of
vomitinginaninvestigationof101non-diabeticpatients
either type) should not be seen as a contraindication to
undergoing elective coronary artery bypass grafting.31
preoperative oral carbohydrates.
In a randomised study in 65 patients undergoing
major abdominal surgery, carbohydrates contributed to
5.3. New formulas for preoperative drinks
the maintenance of muscle mass.41 In two randomised
Beverages containing either amino acids (glutamine)
trials in 8642 and 172 patients43 undergoing laparoscopic
orpeptides(soypeptides)havebeenstudiedwithregard
to their safety.34,35 Glutamine (15g) with carbohydrate cholecystectomy, there was either no effect42 or only a
reduction in postoperative nausea and vomiting.43 Faria
in 300–400ml of water seems to be safe to give 3h
et al.44 showed improved glucose metabolism and
preoperatively in healthy volunteers based on stomach
organic response in 21 female patients participating in
emptying time. A drink containing soy peptide given to
a randomised controlled trial and undergoing laparo-
patientsadmittedforelectivebowelresectionshasbeen
scopic cholecystectomy.
shown to be safe. There was no difference in gastric
emptying time between the carbohydrate group (12.5g Helminen et al.45 studied 210 patients, undergoing
per100mlcarbohydratedrink)andcarbohydrate/peptide gastrointestinal surgery, randomly assigned to fasting,
group(12.5gper100mlcarbohydrateand3.5gper100ml intravenous or oral carbohydrates. Intravenous glucose
of hydrolysed soy protein).35 More research is necessary infusiondidnotdecreasethesenseofthirstandhungeras
todeterminetheeffectsofclearliquidswithaminoacid effectivelyasintheoralintakegroup,butitdidalleviate
or hydrolysed protein in metabolic response and insulin the feelings of weakness and tiredness.
sensitivity after surgery.
Taniguchietal.28investigated50patientsrandomisedto
either 1000ml of oral rehydration solution or 1000ml of
5.4. Carbohydrates, metabolic response and
anintravenouselectrolytesolution.Patients’satisfaction
postoperative discomfort
favoured oral rehydration as they experienced less
Recommendation
feelingsofhunger,lessoccurrenceofdrymouthandless
Drinkingcarbohydrate-richfluidsbeforeelectivesurgery restrictionofmovement.Similarsubjectivebenefitswere
improvessubjectivewellbeing,reducesthirstandhunger observed in a recent small study of gynaecological
and reduces postoperative insulin resistance (evidence patients.46
level 1þþ, recommendation grade A).
Kaska et al.29 performed a randomised controlled trial
Rationale comparing preoperative fasting with preoperative pre-
paration with either oral or intravenous intake of carbo-
In postoperative patients in need of intensive care,
hydrates,mineralsandwater.Consumptionofthemixof
studies have shown that, when glucose is controlled by
water, minerals and carbohydrates offered some protec-
intensiveinsulintherapy,mortalityandmorbiditycanbe
reduced.27 In addition, data suggest that postoperative tionagainstsurgicaltraumaintermsofmetabolicstatus,
cardiac function and psychosomatic status.
discomfort can be reduced when patients are given a
carbohydrate-rich beverage preoperatively.
Breuer et al.32 studied the effects of preoperative oral
In a placebo-controlled randomised trial of 252 patients carbohydrate administration on postoperative insulin
undergoingelectivegastrointestinalsurgery,itwasshown resistance,preoperativediscomfortandvariablesoforgan
that the intake of carbohydrate-rich clear fluid until dysfunction in 188 ASA physical status III–IV patients
2h before the operation led to less thirst, restlessness, undergoing elective cardiac surgery, including those
weakness and concentration problems as compared with non-insulin-dependent type-2 diabetes mellitus.
toplacebo.36Twosmallplacebo-controlleddouble-blind Carbohydratesandplacebowereadministeredindouble-
studies in 1537 and 1438 patients, respectively, under- blind format and patients received 800ml of the corres-
going hip surgery, showed that the intake of a carbo- ponding beverage in the evening and 400ml 2h before
hydrate-rich clear fluid until 2h before the operation surgery. Blood glucose levels and insulin requirements
reduced insulin resistance on days 1 and 3.37,38 Another did not differ between the groups. Patients receiving
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
564 Guidelines
carbohydrateandplacebowerelessthirstycomparedwith preoperative period does not appear to impact on the
controls. The carbohydrate group, however, required intragastricvolumeorpHofchildren.2Thisalsoapplies
less intraoperative inotropic support after initiation of to overweight and obese children.56
cardiopulmonary bypass weaning(P<0.05).32
6.1. Breast milk and infant formula
One study of 36 patients undergoing colorectal surgery
Fastingtimeforbreastmilkandinfantformulaisslightly
has demonstrated a reduction in median length of stay
more controversial. It was demonstrated more than
associatedwithoralcarbohydratetherapy.47Aretrospec-
25 years ago that the gastric emptying of 110–200ml
tive analysis of three small prospective randomised
of human milk was 82(cid:3)11% after 2h in neonates
trials (one matched-control study), primarily investi-
and infants of less than 1 year of age, 84(cid:3)21% after
gating postoperative insulin resistance,24 showed that
whey-hydrosylatedformula,74(cid:3)19%afterwhey-predo-
althoughthestudiesweretoosmalltoshowasignificant
minant formula, 61(cid:3)17% after casein-predominant
reduction in length of stay individually, the combined
formula and 45(cid:3)19% after cow’s milk.57 Thus, human
effect was a significant reduction of about 20%.24 This
milkandwhey-predominantformulaemptiedfasterthan
was confirmed in the randomised trial of Yuill et al.41 in
casein-predominant formula and cow’s milk. Two other
2005 in 72 patients undergoing elective abdominal
studies performed before anaesthesia also demonstrated
surgery. However, the recently published randomised
that breast milk empties from the stomach faster than
trial of Mathur et al.48 in 142 patients undergoing
most formulas in infants and both require more than 2h
colorectalsurgeryorliverresectiondidnotconfirmthese
to ensure complete gastric emptying.54,58 According
results.
to these data, the American guidelines recommended
4hfastingtimeforbreastmilkand6hforinfantformula
6. Perioperative fasting in children and and non-human milk.7 These recommendations were
infants also endorsed by the Royal College of Nursing that
Recommendations considered there was insufficient evidence to change
contemporary best practice (i.e. breast milk up to 4h
Children should be encouraged to drink clear fluids
and formula and cows’ milk up to 6h).4 Scandinavian
(includingwater,pulp-freejuiceandteaorcoffeewithout
guidelines recommended 4h fasting for breast milk but
milk) up to 2h before elective surgery (evidence level
also for formula milk in infants of less than 6 months
1þþ, recommendation grade A).
ofage.3Thus,itisrecommendedtofinishbreastfeeding
Allbutonememberoftheguidelinesgroupconsiderthatteaor 4h before anaesthesia and to stop infant formula 4–6h
coffeewithmilkadded(uptoaboutonefifthofthetotalvolume) prior to anaesthesia depending on the age and on local
are still clear fluids. considerations. Both cow’s milk and powdered milk are
considered as solid food.
Infantsshouldbefedbeforeelectivesurgery.Breastmilk
is safe up to 4h and other milks up to 6h. Thereafter,
6.2. Solid food
clear fluids should be given as in adults (evidence level
Recommendationsforfastingofsolidfoodinchildrendo
1þþ, recommendation grade A).
notdifferfromthoseproposedforhealthyadults.There
Rationale is no evidence against these recommendations.
The recommendations are based on reviews and
6.3. Trauma
guidelines published in the late 1990s and more
Data on fasting in injured children are minimal. One
recently.2–4,7,11,49–52 Fasting is aimed at decreasing the
studysuggestedthatthevolumeofgastriccontentsmay
risk of pulmonary aspiration, but the incidence of this
depend on the nature of the trauma, but gastric content
complicationisverylowinrecentseriesand,althoughthe
wasnotrelatedtothelengthoffasting.59Gastricvolume
riskofaspirationappearstobeslightlygreaterinchildren
was better linked to the interval between the last
thaninadults,53thedifferenceislessthanthatpreviously
meal and the trauma. Thus, the injured child should
reported. All recent surveys indicate the relatively
beconsideredasapatientwithafullstomach.However,
good outcome of this event in the paediatric population
an increasing number of minor surgical procedures are
compared with previous series.
doneundersedationintheemergencydepartment.The
There is a lot of evidence that clear fluids can be given available literature does not provide sufficient evidence
up to 2h prior to surgery in neonates, infants and to conclude that pre-procedure fasting results in a
children. In neonates and infants, gastric emptying of decreased incidence of adverse outcomes in children
clearfluidsfollowsfirst-orderkineticsasinolderchildren undergoing either moderate or deep sedation.60,61
and adults.54 Allowing clear fluids prior to surgery
improvescomfortofthechildandtheparents,decreases 6.4. Postoperative fluids
thirstanddecreasestheriskofpreoperativedehydration Oral fluid intake is usually allowed within the first 3
inyounginfants.55Thevolumeoffluidspermittedinthe postoperative hours in most paediatric patients. Early
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Perioperativefastinginadultsandchildren 565
oral fluid intake was previously required in most Low-risk nulliparous women in labour (n¼2443) were
institutionsbeforedischargingthepatientfromhospital. randomised to either an ‘eating’ or a ‘water only’ group.
This view was challenged, as it has been reported that The results were analysed by intention to treat. No
withholding oral fluids postoperatively from children significant difference was found in the normal vaginal
undergoing day surgery reduces the incidence of vomit- delivery rate; the instrumental vaginal delivery rate;
ing.62,63However,themostrecentstudydidnotfindthat the caesarean section rate; the duration of labour; or
postoperative fasting reduces the incidence of vomiting the incidence of vomiting.67
after general anaesthesia in children when compared
Maternal death from aspiration of regurgitated gastric
with a liberal regimen.64 Thus, it seems reasonable to
content is now extremely rare, and its decline probably
letchildreneatanddrinkaccordingtotheirowndesires,
owesmoretothewidespreaduseofregionalanaesthesia
but not to insist on oral intake before discharge.
foroperativeobstetricsthantofastingpolicies.Inviewof
the predominant use of regional techniques on most
7. Fasting in obstetric patients
deliveryunits,rigidfastingpoliciesarearguablynolonger
Recommendations
appropriateduringlabourandmothersshould,therefore,
Women in labour should be allowed clear fluids (as beallowedtoalleviatethirstduringlabourbyconsuming
defined above) as they desire. (evidence level 1þþ, ice chips and clear fluids (isotonic sports drinks, fruit
recommendation grade A). juices, tea and coffee, etc).
Solid food should be discouraged during active labour Aseatingconfersnobenefittoobstetricoutcome,women
(evidence level 1þ, recommendation grade A). should be discouraged from eating solid food during
labour. However, in view of the almost negligible inci-
Pregnant women, including obese individuals, can
dence of deaths from aspiration, low-risk women could
consume clear liquids until up to 2h prior to surgery
consume low-residue foods (such as biscuits, toast or
(under regional or general anaesthesia) (evidence level
cereals) during labour. In addition, when deciding
2(cid:1), recommendation grade D).
whether or not women should eat during labour, the
An H -receptor antagonist should be given the night use of parenteral opioids should also be considered
2
before,andonthemorningof,electivecaesareansection because of their profound delay on the rate of gastric
(evidence level 1þþ, recommendation grade A). emptying. Units who perform a significant volume of
their emergency obstetric surgery under general anaes-
An intravenous H -receptor antagonist should be given
2 thesiashouldprobablynotallowwomeninlabourtoeat.
prior to emergency caesarean section; this should be
supplemented with 30ml of 0.3moll(cid:1)1 sodium citrate Inhigh-riskpregnancies,itremainsappropriatetonoteat
if general anaesthesia is planned (evidence level 1þþ, during labour and to achieve hydration by limited
recommendation grade A). volumes of oral clear fluids or by the intravenous route.
Theguidelinesgrouprecognisesthatmostoftheevidence
7.2. Preparation for caesarean section
relates to surrogate measures, such as changes in gastric
volume and pH,ratherthan a clear impact on mortality. 7.2.1. Preoperative fastinginelective obstetric surgery
Evidencesuggeststhatpregnantwomen,includingobese
Rationale
individuals,canconsumeclearliquidsuntilupto2hprior
to surgery (under regional or general anaesthesia)68,69
7.1. Oral intake during labour
(evidence level 1þ, recommendation A).
Surgery during labour is usually unplanned, and when it
occursthedegreeofemergencycanrangefromminimalto
7.2.2. Recommended drugregimens indetail
surgerythatislifesavingforeithermotherorbaby.Against
thisbackground,logicdictatesthatallmothersshouldbe 7.2.2.1. Elective obstetric surgery All mothers should
starved during labour. However, it is often argued that be actively encouraged to have regional anaesthesia for
allowing mothers to eat and drink during labour will an elective caesarean section.
prevent ketosis and dehydration and, thereby, improve
An H -receptor antagonist (e.g. 150mg ranitidine) or a
2
obstetric outcome. There is currently wide variation in
PPI(e.g.omeprazole40mg)shouldbegivenatbedtime
practice with respect to eating during labour in Europe.
andagain60–90minbeforetheinductionofanaesthesia.
However, it has now been shown that although eating a
Theadministrationoforalmetoclopramide10mgatthe
light diet during labour will prevent ketosis, it will also
same time as the H -receptor antagonist or PPI should
increase gastric volume,65 whereas when isotonic ‘sport 2
also be considered.
drinks’ are consumed during labour,66 ketosis can be
eliminated without an increase inintragastric volume.
7.2.2.2. Emergency obstetric surgery under regional
A recent randomised controlled study evaluated the anaesthesia Intravenous H -antagonist (e.g. ranitidine
2
effectoffoodintakeduringlabouronobstetricoutcome. 50mg)attimeofdecisionforsurgery.Inhigh-riskwomen
EuropeanJournalofAnaesthesiology 2011,Vol28No8
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.