Perioperative management of adult diabetic patients. Preoperative period

HAL Id: hal-02621402

https://hal.inrae.fr/hal-02621402

Submitted on 26 May 2020

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

Distributed under a Creative Commons Attribution| 4.0 International License

Perioperative management of adult diabetic patients.

Preoperative period

Gaelle Cheisson, Sophie Jacqueminet, Emmanuel Cosson, Carole Ichai,

Anne-Marie Leguerrier, Bogdan Nicolescu-Catargi, Alexandre Ouattara, Igor

Tauveron, Paul Valensi, Dan Benhamou

To cite this version:

Gaelle Cheisson, Sophie Jacqueminet, Emmanuel Cosson, Carole Ichai, Anne-Marie Leguerrier,

et al..

Perioperative management of adult diabetic patients.

Preoperative period.

Anaes-thesia Critical Care & Pain Medicine, Elsevier Masson, 2018, 37 (Supplement 1), pp.S9-S19.

�10.1016/j.accpm.2018.02.020�. �hal-02621402�

Guidelines

Perioperative

management

of

adult

diabetic

patients.

Preoperative

period

Gae¨lle

Cheisson

,

Sophie

Jacqueminet

,

Emmanuel

Cosson

,

Carole

Ichai

,

Anne-Marie

Leguerrier

_,

_Bogdan

_{Nicolescu-Catargi}

_,

_Alexandre

_Ouattara

_,

Igor

Tauveron

,

Paul

Valensi

,

Dan

Benhamou

*

,

working

party

approved

by

the

French

Society

of

Anaesthesia

and

Intensive

Care

Medicine

(SFAR),

the

French

Society

for

the

study

of

Diabetes

(SFD)

a_{Serviced’anesthe´sie–re´animationchirurgicale,hoˆpitauxuniversitairesParis-Sud,AP–HP,78,rueduGe´ne´ral-Leclerc,94275LeKremlin-Biceˆtre,France} b

Institutdecardio-me´tabolismeetnutrition,hoˆpitaldelaPitie´-Salpeˆtrie`re,AP–HP,75013Paris,France

c

De´partementdudiabe`teetdesmaladiesme´taboliques,hoˆpitaldelaPitie´-Salpeˆtrie`re,75013Paris,France

d

De´partementd’endocrinologie-diabe´tologie-nutrition,hoˆpitalJean-Verdier,universite´ Paris13,SorbonneParisCite´,CRNH-IdF,CINFO,AP–HP,93140Bondy, France

e

SorbonneParisCite´,UMRU1153INSERM/U1125INRA/CNAM/universite´ Paris13,93000Bobigny,France

f_{Servicedere´animationPolyvalente,hoˆpitalPasteur2,CHUdeNice,30,voieRomaine,06001Nicecedex1,France} g_{IRCAN(INSERMU1081,CNRSUMR7284),UniversityHospitalofNice,06001Nice,France}

h

Servicedediabe´tologie-endocrinologie,CHUdeRennes,CHUHoˆpitalSud,16,boulevarddeBulgarie,35056Rennes,France

i

Serviced’endocrinologie–maladiesme´taboliques,hoˆpitalSaint-Andre´,CHUdeBordeaux,1,rueJean-Burguet,33000Bordeaux,France

j

DepartmentofAnaesthesiaandCriticalCareII,MagellanMedico-SurgicalCenter,CHUdeBordeaux,33000Bordeaux,France

k

INSERM,UMR1034,BiologyofCardiovascularDiseases,universite´ Bordeaux,33600Pessac,France

l

Serviceendocrinologiediabe´tologie,CHUdeClermont-Ferrand,58,rueMontalembert,63000Clermont-Ferrand,France

m_{UFRme´decine,universite´ Clermont-Auvergne,28,placeHenri-Dunant,63000Clermont-Ferrand,France}

n_{UMRCNRS6293,INSERMU1103,ge´ne´tiquereproductionetde´veloppement,universite´ Clermont-Auvergne,63170Aubie`re,France} o

Endocrinologie-diabe´tologie,CHUG.Montpied,BP69,63003Clermont-Ferrand,France

ARTICLE INFO Articlehistory:

Availableonline17March2018 Keywords:

Diabetes Perioperative HbA1c Gastroparesis

Cardiacautonomicneuropathy Antidiabeticdrugs

ABSTRACT

Indiabeticpatientsundergoingsurgery,werecommendassessingglycaemiccontrolpreoperativelyby assessingglycatedhaemoglobin(HbA1c)levelsandrecentcapillarybloodsugar(glucose)levels,andto adjustanytreatmentsaccordinglybeforesurgery,payingparticularattentiontospecificcomplications ofdiabetes. Gastroparesiscreatesarisk ofstasis andaspiration ofgastriccontentat inductionof anaesthesiarequiringtheuseofarapidsequenceinductiontechnique.Cardiacinvolvementcanbe dividedintoseveraltypes.Coronarydiseaseischaracterisedbysilentmyocardialischaemia,presentin 30–50% of T2Dpatients. Diabetic cardiomyopathyis arealcause ofheart failure. Finally, cardiac autonomicneuropathy(CAN),althoughrarelysymptomatic,shouldbeinvestigatedbecauseitcausesan increasedriskofcardiovasculareventsandariskofsuddendeath.SeveralsignsaresuggestiveofCAN, andconfirmationcallsforcloseperioperativesurveillance.Chronicdiabetickidneydisease(diabetic nephropathy)aggravatestheriskofperioperativeacuterenalfailure,andwerecommendmeasurement of the glomerular filtration rate preoperatively. The final step of the consultation concerns the managementofantidiabetictherapy.Preoperativeglucoseinfusionisnotnecessaryifthepatientisnot receivinginsulin.Non-insulindrugsarenotadministeredonthemorningoftheinterventionexceptfor metformin,whichisnotadministeredfromtheeveningbefore.Theinsulinsareinjectedattheusual dosetheeveningbefore.Theinsulinpumpismaintaineduntilthepatientarrivesinthesurgicalunit.It shouldberememberedthatinsulindeficiencyinaT1Dpatientleadstoketoacidosiswithinafewhours.

C _{2018TheAuthor.PublishedbyElsevierMassonSASonbehalfofSocie´te´ franc¸aised’anesthe´sieetde}

re´animation(Sfar).ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/ licenses/by/4.0/).

* Correspondingauthor.

E-mailaddress:dan.benhamou@aphp.fr(D.Benhamou).

https://doi.org/10.1016/j.accpm.2018.02.020

2352-5568/ C_{2018TheAuthor.PublishedbyElsevierMassonSASonbehalfofSocie´te´ franc¸aised’anesthe´sieetdere´animation(Sfar).Thisisanopenaccessarticleunderthe}

1. Evaluationofglycaemiccontrolinadiabeticpatient

PracticalsheetBsummarisesthemainelementsofthistext.

Many studieshavebeencarriedoutindiabeticpatientsat

thetimeofsurgery,buttheyhavemainlybeeninterestedinthe

cardiovascular risk and not in the management of diabetes

itself. Glycaemic control is recommended before admission

accordingtoobjectivesadaptedtoeachpatient[1]inorderto

avoidhyperglycaemiaandhypoglycaemia.Glycaemic

variabil-ity should also be monitored during the hospital stay

[2]. Glycaemic variability is common due to fasting, stress,

developmentofinfection,useofglucocorticoids,etc.Glycaemic

controlisevaluatedduringtheanaesthesiaconsultationusing

twocriteria:(i)glycatedhaemoglobin(HbA1c);and(ii)blood

sugar (glucose) levels. Hypoglycaemic and hyperglycaemic

episodes should be treated because they may have harmful

preoperativeconsequences andmay leadto the intervention

beingpostponed.

Fig. 1 summarises the management that we recommend

depending on overall glycaemic control (HbA1c), blood sugar

levelsbeforesurgery,therecentdevelopmentofhypo-and

hyper-glycaemicepisodes,particularlythosewithassociatedketosis.

1.1. HbA1c

Inatreateddiabeticpatient,HbA1cwhichreflectsmeancontrol

over the previous 3months allows to estimate the quality of

glycaemic control before the consultation and adaptation of

treatmenttofixedobjectives[1].Somestudiesindicatethatraised

HbA1cin a diabeticpatient is associated withhigh morbidity/

mortalityandanincreasedriskofmyocardialinfarctionandearly

postoperativeinfection[3].Foreach1%increaseinHbA1c,therisk

increasesby40%[4].Theriskofsternalinfectionis5-foldhigher

(OR=5.3) when HbA1c level is greater than 7.8%. HbA1c>7%

appears to have a negative prognostic value in unrecognised

diabeticpatients[5].AcorrelationexistsbetweenHbA1candmean

bloodsugar levels(AppendixA). Itis wise topostponesurgery

(exceptinanemergencysituation) ifHbA1cisveryhigh(>9%)

becauseitdemonstratesalackofglycaemiccontrolandthepatient

is thus exposed to acute metabolic complications in the

perioperative period. Using the same reasoning, HbA1c<5%

indicatesprobablerecurrentseverehypoglycaemicepisodesina

patient treated withinsulin or hypoglycaemic sulphonamides/

glinidesandwealsorecommendpostponingsurgeryinthiscase

(Fig.1).

Therapeutic adjustments,after advicefroma general

practi-tioner(GP)oradiabetologist,shouldbediscussedforHbA1cvalues

between8%and9%(chronic,butnotthreatening,hyperglycaemia

requiring therapeutic reinforcement) or between 5% and 6%

(repeated hypoglycaemic episodes requiring to reduce the

intensityofantidiabetictherapy).

1.2. Recentbloodandcapillaryglycaemicvalues

Some studies have demonstrated a correlation between

glycaemiaatadmission(>2g/Lor11mmol/L)andpostoperative

morbidity/mortality [6–8], as well as a 10-fold higher risk of

complications when glycaemic imbalance exists before surgery

[9]. Measurement of blood sugar levels allows, for example,

stratificationoftheriskofsternalinfectionaftercardiacsurgery:a

bloodsugarlevel<1.80g/L(10mmol/L)beforetheintervention

decreasestheriskofdeath,infectionanddurationofstay[10].

Duringthepreoperativeconsultationandinthedays

immedi-ately precedingtheintervention,werecommend monitoring of

capillary blood sugar levels.A recentdisequilibrium (hyper- or

hypo-glycaemia)couldhaveaneffecton perioperative

manage-ment,evenifanadequateHbA1cvalueisobserved[11].Inthis

case, therelationHbA1c/mean blood sugar(AppendixA)isnot

applicable.

1.3. Identificationofepisodesofhypoglycaemia(sheetB)

Hypoglycaemiaisatherapeutichazardinalltreateddiabetics,

with the risk being particularly high in hospitalised diabetic

patients,duetomoresevereglycaemicvariations[12].Evenifthe

specificroleofanaesthesiaorsurgeryhasnotbeenestablished,

these situations of instability can contribute to anomalies in

glucose regulation,suchas hypoglycaemia.Hypoglycaemiais a

frequentconsequenceoftreatmentwithinsulinsecretors

(hypo-glycaemic sulphonamides/glinides) or insulin therapy (sheets E

andF).ItmainlyoccursinT1DpatientsbutalsoinT2Ddiabetics,at

anytime.

Thecut-offbloodsugarleveldefininghypoglycaemiahasbeen

thesubjectofdebate;neverthelesscurrentconsensussuggestsa

plasma bloodsugar levelof<0.7g/L (3.9mmol/L) ina diabetic

patient.Inpractice,forhealthcareteams,anyunexplainedmalaise

in a diabeticpatient shouldbe considered as a hypoglycaemic

episodeuntilprovenotherwise,evenifthebloodsugarlevelatthe

timeitismeasured(sometimesseveralminutesafterthestartof

the malaise) appears to be normal. Serious hypoglycaemia is

definedbytheneedforassistanceofanotherperson,whateverthe

bloodsugarlevel.Somehypoglycaemicepisodesarenotaseasily

apparent, particularly when hypoglycaemia occurs frequently,

diabetesisalreadyalong-lastingdiseaseorwhendysautonomiais

present.Thissituationoccursinnearly40%ofT1Dpatients,10%of

T2DpatientsoninsulinandoccasionallyinT2Dpatientsonoral

antidiabeticdrugs[13].

Episodesofhypoglycaemiaresultfromanimbalancebetween

insufficient carbohydrate supply and poorly adapted insulin or

insulin-secretor treatment. These situations are particularly

frequentintheperioperativeperiod,forexampleduetoprolonged

fastingorirregularfoodintake.Itisnecessarytobevigilantabout

theadministrationofdrugs,whichhaveadirect hypoglycaemic

effectorthosewhichincreasetheactionoforalantidiabeticdrugs

(forexamplequinolones, heparin,

b

trimethoprime-sulfamethoxazole).Deteriorationofrenalorhepaticfunctionmay

decreasetheclearanceofantidiabeticdrugsandincreasetheriskof

hypoglycaemia.

1.4. Identificationofrecentepisodesofhyperglycaemiaandketosis

(sheetB)

Ifglycaemicdisequilibriumisdetectedduringtheanaesthesia

consultation(basedonHbA1c),therapeuticadjustmentshouldbe

considered.ForexampleinT2Dpatientsonoralantidiabeticdrugs,

intensificationoftreatmentshouldbeconsidered.Therearemany

therapeutic strategies that can be considered, including those

establishedinFrancebytheHauteAutorite´ deSante´[1],andalsoby

internationalsocieties[14],validatedbytheFrenchSocietyforthe

studyof Diabetes(SFD).Intensification ofhypoglycaemic

treat-ment should be personalised and validated on by an expert

prescriber.

InT2DandT1Dpatientstreatedwithinsulin,thetimesatwhich

hyperglycaemiaoccursshouldbeassessedusingan

auto-surveil-lance booklet and the doses of insulin should be modified

accordingly.Ifnecessary,insulintherapy(transientorpermanent)

maybeproposed,particularlyifketosisisdetected.

2. Specificcomplicationsofdiabetes(sheetB)

Indiabeticpatients,theperioperativeriskmaybeincreasedby

gastroparesisand/orheartdiseaseand/orkidneydisease.

2.1. Gastroparesis

Thisdiabeticcomplicationisthemostfrequentmanifestation

ofdigestivedysautonomia,definedasdelayedgastricemptying

in the absence of mechanical obstruction. It usually affects

diabetic patients with other neuropathic diseases, affecting

30–50% of patients with T1D or T2D. Symptoms classically

includeanorexia,nausea,vomiting,abdominalpain,sensationof

bloating,earlysatietyorslowingofdigestion[15].Itmayhave

adverserepercussionsontheabsorptionofdrugsadministered

by mouth. However, there is a weak correlation between

symptomsandthe rate ofgastricemptying.Inmany diabetic

patients,gastricemptyingisevenacceleratedalthoughpatients

present with similar symptoms. Finally, some patients have

symptoms without abnormalgastric emptying. Gastroparesis

may be a factor for post-prandial glycaemic dysregulation.

Furthermore,acutehyperglycaemiaslowsdowngastric

empty-ing.Thereisthereforeareciprocalinterrelationbetweengastric

emptyingandglycaemia[16].Clinicalexaminationmayreveal

abdominal distension with classic ‘clapotement’ onan empty

stomach, a late sign of gastroparesis. Upper gastro-intestinal

endoscopy isusedasthe first-line non-clinicalassessment.It

helpstoeliminateothercausesofupperdigestivesymptomsand

torevealthe presenceoffood remaininginthestomachafter

nightfasting.Abariummealmayrevealgastricresidueoreven

phytobezoards.Thereferencediagnosticexaminationisgastric

scintigraphy carried out using a calibrated meal, preferably

solid, labelledwithtechnetium99m[15].Othertestsmaybe

proposed such as the respiratory test for carbon 13-labelled

octanoicacid.

During theanaesthesiaconsultation,gastroparesisshouldbe

investigatedbecauseitcreatesariskofstasis(fullstomach)and

aspirationatanaestheticinduction.Ataminimum,questioningof

thediabeticduringtheanaesthesiaconsultationshouldfocuson

theclassic clinicalmanifestations ofgastroparesis (sheetB and

Fig. 2). If clinicalsigns suggestive of gastroparesis are present,

measurement of the gastric antral area by ultrasound can

distinguishwhetherthestomachisfullornot[17].Echography

canalsovisualisesolidresidues.Incaseofdoubt,rapidsequence

inductionshouldbecarried out.Erythromycinand

metoclopra-mide, which can help accelerategastric motility,may be used

[18,19].

Fig.2.Preoperativeassessmentofthespecificcomplicationsofdiabetes.PH:previoushistory;UB:urinestrip;ECG:electrocardiogram;HR:heartrate;HRV:heartrate variability;MI:myocardialinfarction;SMI:silentmyocardialinfarction;MET:metabolicequivalent;ABP:arterialpressure.

2.2. Effectsofdiabetesontheheart

2.2.1. Cardiovascularriskofdiabetes

Approximately75%ofdiabeticpatientsdiefromcomplications

ofatherosclerosis.However, thedegreeof cardiovascularriskis

heterogeneous within the diabetic population, linked to

co-existingriskfactorssuchasarterialhypertension,lipid

abnormali-ties,smoking,familialhistory of earlycardiovascularproblems,

andalsotospecificfactorssuchasglycaemiccontrol,theduration

of diabetes and presence of nephropathy in particular.

Micro-albuminuriaisassociatedwithanincreasedcardiovascularriskin

bothT1DandT2D,andtheriskisevengreaterinthepresenceof

macroproteinuriaor renal failure [20]. The presence ofarterial

damageinoneareaisclearlyassociatedwithanincreasedriskthat

arteriopathicdamagealsoexistselsewhere.

2.2.2. Heartdisease

Heart diseasehasseveral characteristicsindiabeticpatients.

Myocardial infarction is usually silent and silent myocardial

ischaemia (SMI)can befoundin 30–50%of asymptomaticT2D

patients who have no cardiac history but who present with

cardiovascularriskfactors[20,21].Itsnegativeprognosticvalue

hasbeendemonstrated[22–24].InvestigationsforSMIarebased

onprovocationtests(exercisetolerancetest,myocardial

scintig-raphywithexerciseand/oradministrationofdipyridamole,stress

echographyoreven stressmagneticresonance imaging).IfSMI

detectionisfollowedbycoronarography,thisexaminationreveals

significantcoronarystenosisin30–70%ofpatients[20].Detection

ofheartdiseaseappearslogicaltopreventcardiacevents,butit

shouldbereservedforpatientswithaveryhighcardiovascularrisk

[20]. Measurement of the coronary calcium score by CT scan

(withoutinjectionof iodinatedproduct) contributestoevaluate

thecardiovascularrisk;ascoreof>400Agastonunitsisassociated

witha worse prognosisanda highprevalence ofSMI [25]and

shouldleadtoinvestigationsforSMI[20].However,itshouldbe

noted that many rigorousstudies have shown that systematic

coronaryrevascularisationbeforenon-cardiacsurgeryisnotuseful

toreducetherateofpostoperativemyocardialischemicepisodes

[26].Conversely,repeatedperioperativemeasurementoftroponin,

associatedwith ECG,allows to detectperioperativemyocardial

damageandhelpstoinitiatecardiovasculartreatment[27].

2.2.3. Heartfailure

Theriskofcongestiveheartfailure(CHF)is2–3-foldhigherin

diabeticpatients.Age,durationofdiabetes,heartdiseaseandthe

presenceofalbuminuriaareallassociatedwithanincreasedriskof

heartfailure[28].Althoughheartdiseaseandarterialhypertension

arethemainfactorsresponsibleforCHF,diabeticcardiomyopathy

is a well-established clinical entityand a provencause of CHF

[29].MortalityafterafirstepisodeofCHFis10-timeshigherinT2D

patients than in non-diabeticpatients. Atthe preclinicalstage,

structural and functional alterations of the left ventricle are

frequentlysuspectedontheECGinrelationtodiabetic

cardiomy-opathyandmightbeobservedevenintheabsenceofheartdisease

orhypertension.Leftventricularhypertrophyisoftenpresent,but

diastolicorsystolicdysfunctionmightalsobeobserved[30].

Mea-surementofbrainnatriureticpeptide(BNP)orpro-BNPlevelshas

goodsensitivitytodetectdiastolicorsystolicdysfunctionatthe

preclinicalstage[31].

2.2.4. Cardiacautonomyneuropathy(CAN)(AppendixB)

The dysautonomic cardiaccomplicationof diabetes israrely

expressed by clinical symptoms: permanent tachycardia and

orthostatichypotensioninparticular(whichisofteniatrogenicin

origin),post-prandialhypotensionorseverehypoglycaemiaswith

noeffects.ThesearesymptomsofsevereCAN.Attheinfra-clinical

stage,themostfrequentcomplicationsofCANincludeanomalies

of heart rate(HR) at baseline and abnormal variability during

standardisedtests.Theprevalenceoftheseabnormalitiesincreases

withthedurationofdiabetesand ifglycaemicdisequilibriumis

long-lasting [32]. In a French multicentre study, confirmed or

severeCANdefinedbytwoorthreetestsassessingvariationsinHR,

was present in 20% of patients and CAN was significantly

associatedwiththepresenceofmicroangiopathiccomplications

[33].TheriskslinkedtoCANincludemyocardialinfarction(MI),

which is painless or discovered only on a systematic ECG, an

increaseincardiovasculareventsandanincreaseinmortality.The

riskofsuddendeathsecondarytoseriouscardiacrhythmdisorders

shouldbenoted[32,34].Amongmechanismsimplicatedarethe

absenceofanocturnalfallinbloodpressure(non-dippingtypeor

evenreversedipping)duringambulatorybloodpressure

measu-rements,alterationsinventricularrepolarisationwithlengthening

oftheQTcinterval(>440ms)andtheco-existenceofSMI[32,35].

ThedetectionofCANattheinfra-clinicalstageisbasedonan

analysis of variationsin HR during standardisedtests for deep

respiration, active orthostatism and the Valsalva manoeuvre

[32].Thesetestsareusually carriedout witha simple ECGbut

shouldbeinterpretedasafunctionofage(duetothephysiological

reductioninHRvariabilitywithage).Althoughthesetestsarequite

easy to perform, they take time and manual calculation is

necessary, making their use limited outside a specialised

consultation,unlessasoftwareprogrammeisused.Variationsof

HRessentiallybutnon-specificallyrevealparasympatheticheart

damage. Current recommendations suggest carrying out these

testsinT1Dpatientswithknowndiseaseforatleast5yearsandin

allT2Dpatients,particularlyifmicroangiopathiccomplicationsof

diabetesexist[36].CANisgradedaccordingtotheresultofthese

tests[32].

If CAN is detected, drugs that may induce orthostatic

hypotension should be avoided, ambulatory measurement of

bloodpressureshouldbecarriedouttolookfornocturnal

non-dippingpatterns andthe QTinterval shouldbemeasured on a

standard ECG (at minimum). If it is prolonged, problems with

paroxysmalventricularrhythmshouldbedetectedby24-hECG

continuousmonitoring[32].

Anaesthesia, whether general (GA) or regional (RA), has

pronouncedeffectson perioperative sympatheticnervoustone

[37].Manyclinicalinvestigationshaveevaluatedtheinfluenceof

IV anaestheticagents onperipheralautonomicnervousinflux.

Microneurographyallowstoassesssympatheticnervousactivity

inlowerlimbmusclesbyplacingelectrodesintheperonealnerve

[37].StudieshavealsoassessedtheeffectsofRAonautonomic

nervoustone.Whateverthetypeofneuraxialanaesthesiaused,

spinalblockorepidural,theauthorsreportasignificantdecrease

insympatheticnervousinflux[38,39].Ithasalsobeenshownthat

postoperativeepiduraladministrationofmorphineafter

abdom-inal aortic surgery decreases postoperative hypertension by

reducingsympathetichyperactivity[40].Inthediabeticpatient,

andinthecaseofmetabolicsyndrome,peripheralsympathetic

hyperactivityoccurs[41].Interactionsbetweenanaesthesiaand

CAN leadtoan increasedriskof perioperative haemodynamic

instabilityalthoughmechanismsareunclear.Indiabeticpatients,

apreoperativedecreaseinrespiratoryHRvariability(respiratory

sinus arrhythmia) is associated with a risk of perioperative

haemodynamicinstability[42,43].Someauthorshavereported

that perioperative requirement for vasopressor support is

correlatedwiththedegreeofdysautonomia[44,45].Withregard

to its influence on perioperative haemodynamics, it is not

surprising to observe that perioperative dysautonomia has a

postoperative prognosticimpactonthediabeticpatientin the

long-term[44].Preoperativeevaluationofdiabeticpatientsusing

appears useful to identify patients at risk of perioperative

haemodynamicinstabilityandof cardiovascularcomplications

despitetheabsenceofclinicalsymptomsofCAN[32,44].Diabetic

patients suffering from dysautonomia also have a decreased

ventilatoryresponsetohypoxaemiaandhypercapnia[46]andto

perioperativehypothermia[47].Inthesepatients,more

sophisti-catedhaemodynamicmonitoringincluding a continuous

mea-surement of arterial pressure and cardiac index can be

recommended[48].

2.2.5. Cardiovascularassessmentattheanaesthesiaconsultation

Specificmanagementofapatientwithcoronaryarterydisease

undergoingnon-cardiacsurgeryhavebeendescribedindepthin

the2011SFAR/SFCGuidelines[49].Thepresentreportsummarises

theminimalcardiovascularassessmentandthepossibleadditional

assessmentsthatshouldbedoneinthediabeticpatient.

Questioningthepatientwilldetermine:

 ifthepatientishypertensive;

 ifthereisanycardiovascularhistory,inparticulartheexistence

ofheartdisease,cardiacischaemia(CI),arrhythmia,

cerebrovas-cularaccidentorlowerlimbsarteriopathy;

 ifthepatientisunderthecareofacardiologist;

 ifthepatienthasalready undergonecoronary,supra-aorticor

peripheralvascularrevascularisation;

 if there are recent symptoms of angina, these may appear

atypicalormildsuchasdyspnoeaorexerciseepigastricpain,i.e.

symptoms suggesting cardiac ischaemia or of peripheral

arteriopathy,symptomssuggestingorthostaticorpost-prandial

hypotension,episodesofseriousunfelthypoglycaemia;

 currenttreatments.

Theresultsofearlierinvestigationsareexamined:mostrecent

ECGavailable,echocardiogram,arterialecho-Doppler,

investiga-tions made to detect SMI. A meticulousclinical cardiovascular

examinationiscarriedout.Orthostatichypotensionis

investigat-ed.Criticalischaemiaofthelowerlimbsisruledout.ECGisdone

againifthelastonehadbeencarriedoutseveralmonthsagoand

searches for signs of ischaemia or even SMI, tachycardia,

arrhythmia, prolonged QTc interval. Measurement of BNP or

pro-BNPlevels,oranimmediateECGmaybeprescribedinthecase

ofpossiblecardiacischaemia.

InvestigationsforSMIarethereforerecommendedinpatients

scheduledformajorsurgeryiftheLee(revisedcardiacriskindex)

scoreis2andiffunctionalcapacityis<4metabolicequivalents

(METs).Thepatientwillbereferredtoacardiologistforspecific

managementandfortestsforischaemia(sheetBandFig.2).Ifthe

patienthasno history nor anycoronarysymptoms, and in the

absenceofaspecificanomalyonECG,asilentheartdiseaseshould

be suspected in patients with a high cardiovascular risk, in

particular in the presence of other arterial damage, macro

proteinuria, renal failureand when thecoronary calciumscore

is>400Agastonunits[20].

WhenCANis suspectedduetothepresenceofsymptomsor

complications(permanenttachycardia,QTc>440ms,MIorSMI,

orthostatic or post-prandial hypotension, serious unfelt

hypo-glycaemia, absence of nocturnal decrease in blood pressure,

unexplained), it should be confirmed by tests analysing HR

variations. These tests should also be used to investigate CAN

when microangiopathic complications exist. The procedure

proposed includes theanalysis of HR variations during a deep

respiration test and a test for orthostatism (Appendix B). The

presenceofCAN,whenconfirmedbytwoabnormaltestsorwhen

symptomatic or complicated, should lead to intra- and

post-operativemonitoring (ina highdependency unit)(sheet Band

Fig.2).

2.3. Diabeticnephropathy

2.3.1. Epidemiology

Diabeticnephropathy(DN)isoneofthemostfrequentmicro

vascular complicationsof diabetes.Itoccursin30% ofT1D and

approximately 20% of T2D patients [50–52]. However, the

frequencyof thiscomplicationhasplateauedorevendecreased

overthepastfewyearsinsomecountriesduetoearlierandmore

appropriatemanagement[53,54].IntheUSA,therelativeriskof

DNhasdecreasedbyone-half(13.7to6.1%)in20years[55].The

riskfactorsforDNarewellknown:malesex,SouthAsianor

Afro-Caribbeanethnicityandprolongedevolutionofdiabetes.DNisthe

mostfrequentcause ofend-stage renalfailure,affecting45% of

individualswithrenalfailureintheUSA.Twenty-fourto50%of

patients receiving dialysis are diabetics with end-stage renal

failure [52].Theincidenceof end-stagerenal failureindiabetic

patients hasgreatly increased compared to non-diabetics with

renal failure [51,56]. This progression classically occurs more

frequentlyinapatientwithadvancedT1D;itisoftenassociated

with neuropathy and diabetic retinopathy (revealing micro

vascular complications) [50,51]. However, recent studies show

thatprogressionofDNtowardsend-stagerenalfailureappearsto

becomparableforthetwotypesofdiabetesandthatithasslowed

down overthepastfewyears:4to15%in20yearsand16% in

30years.TheriskfactorsforprogressionofDNtoend-stagerenal

failurearethefollowing[57]:raisedHbA1c,raisedbloodpressure,

presence ofalbuminuria, earlydecreasein glomerularfiltration

rate(GFR),age,durationofdiabetesandraisedlevelofuricacid.

DNincreasestheriskofmortalitywhateverthetypeofdiabetes

(relativerisk40–100-timeshigherthaninnon-diabetics)[51].In

theUK,themortalityrateofdiabeticpatientsbetween18and

44-yearsofage,ondialysis,reaches30%in5yearscomparedto11%in

dialysednon-diabeticpatients.Itisalsoamajorindependentrisk

factorforcardiovascularcomplications,atherosclerosisandinsulin

resistance,fullyjustifyingitsprevention[52,58].

2.3.2. Physiopathologyanddiagnosis

The physiopathological concept of DN and its management

have evolved greatly over the past few years, explaining the

reduction in its incidence and its consequences in terms of

morbidity/mortality. Untilthe start of thelast decade,DN was

considered tobecharacterisedbylesionsofnodular

glomerulo-sclerosisand progressingtowardsrenalfailure classifiedin five

stages of increasing severity based on GFR values (Fig. 3)

[50,51,59].Theclinicaldiagnosisdependsontheclassicaltrioof

microalbuminuria,arterialhypertensionandrenaldysfunction(of

variable severity). Histopathological data have advanced this

conceptandthetermDNhasbeenreplacedbydiabeticchronic

kidneydisease(DCKD).Alongsidetheclassicglomerulosclerosis,

therenallesionsobservedduringdiabetesmayaffectthetubules,

therenal interstitialtissueandthevessels[60].Thus, although

microalbuminuriaisthemostfrequentbiological anomaly,it is

notconstant(becauseitonlyrevealsglomerulardamage)anddoes

notnecessarilyreflecttheseverityofrenaldysfunction.There is

indeednocorrelationbetweenGFRandmicroalbuminuria,now

namedmoderatelyincreasedalbuminuria(valuesbetween30and

300mg/g).

The pathogenic mechanisms of diabetic kidney lesions are

complexandarelinkedtoanomaliesotherthanhyperglycaemia.

Glomerulosclerosis ismainly induced by hyperglycaemia,which

triggersmesangialexpansionandtubularhypertrophywithcellular

oedemaresponsible forinitialglomerular hyperfiltration.These

modifications resultinlocalactivation ofthe

renin-angiotensin-aldosterone system with glomerular efferent arteriolar

vasocon-striction [51,57,58]. This reaction triggers many pathways of

production, oxidative stress, growth factors, etc. The structural

modifications which result include thinning of the basement

membrane,alterationofpodocytesandmesangialhyperplasia[58].

Measurementoftherenalexcretionrateofalbumin

(albumin-uria)ina24-hurinesamplehasbeenthe‘‘goldstandard’’forthe

earlydiagnosisofDNformanyyears.However,thisparameteris

unreliable and is affected by sampling conditions with an

appreciable intra-individual variability. Furthermore, this simple

measure,revealingtubularlesions,doesnotreflecttheseverityof

renaldysfunctionmeasuredbyGFR.Thereisnorelationbetweenthe

progressionofalbuminuriaandthedecreaseinGFR.Thus,apositive

diagnosisandtheseverityofDCKDisbasedonnew

recommenda-tions[61].Theratioofalbuminuriatourinarycreatinine(ACR)and

GFRallow the classification of DCKD. There are threestages of

increasingseverityofACR:stageA2(ACRbetween30and300mg/g)

corresponds to moderately high albuminuria previously termed

microalbuminuria.ThedecreaseinGFRisdefinedbyfivestagesof

increasingseverity(G1toG5)(Fig.3).Itisadvisedtomeasurethe

ACRandtheestimatedGFRononeurinesampleratherthanon24-h

urinesandtoperformatleast2or3measurementsin6monthsto

confirmthediagnosis[62].AlthoughtheestimatedGFRisusually

based on the MDRD formula, other formulae are also reliable

(CockroftandGault,CKD-EPI).Nevertheless,theyallunderestimate

theGFRforsubnormalrenalfunction(GFR90mL/min)[62].The

stagesA1/G1andA1/G2areconsideredstableandshouldleadto

annual measurements. For all the other stages, the risk of

progressiontowardchronicrenalfailureincreases,aswellasthe

cardiovascular risk. Follow-up with biological control should be

carriedout2–4timesa year asafunction ofseverity.Theearly

diagnosis and follow-up of DCKD should be facilitated by the

measurementofurinarybiomarkers.Theseareproteinsotherthan

albuminsuchascystatinC,

b

a

immunoglobulin G, transferrin, nephrin and metalloproteins,all

indicators of tubular or glomerular lesions. Even though some

studieshaveshownthatthesebiomarkersaremoresensitiveand

specific for the diagnosis of DCKD, their usefulness in clinical

practiceremainstobedemonstrated.

Prevention and management of DCKD are currently well

codified. Many studies have shown that they depend on

multimodalmanagement,whichslowsdown theprogressionof

DCKDandreducesthecardiovascularcomplicationsandmortality

of these patients [51,58]. The two main objectives of this

management are the administration of antihypertensive

treat-ments and closer control of glycaemia. All antihypertensive

treatments havedemonstrated efficacy(

b

a

diuretics,hydralazine).However,blockersofthe

renin-angioten-sin-aldosteronesystemhavethegreatestefficacy.Thus,

angioten-sinconvertingenzymeinhibitors(ACEinhibitors)andangiotensin

antagonists(sartans)decreasemortality,slowdownthe

progres-sionof DCKD toend-stage renal failure and reduceglomerular

hyper filtration and its consequences. It is not currently

recommendedtocombineACEinhibitorswithsartansandthere

isnoproofofsuperiorefficacyofthiscombination[50,51,62].The

initiationoftreatment withACEinhibitorsor sartansis

recom-mended from stage A2 moderate albuminuria and strongly

recommendedincaseofseverealbuminuria(A3)and/orestimate

GFR<60mL/min/1.73m2 _{[63,64].These agentsare also}

recom-mendedindiabeticpatientssufferingfromarterialhypertension.

No randomised study has been carried out to determine the

optimum arterial pressure cut-off value in these patients.

However, experts recommend to maintain arterial pressure at

lessthan140/85–90mmHg[20,64–66].Thesecond-linetherapy

forDCKDiscontrolofglycaemiaanditsefficacyonmicrovascular

complications has been widely established. Thus, the current

recommendationsadvisemaintainingHbA1c7%[64,67].Control

ofhyperlipidaemia isalsoimportant,reducing albuminuriaand

slowing down the decrease in GFR. This is carried out by

preferentialadministrationofstatins[51].Earlyrenal

transplan-tation before carrying out dialysis may also contribute to an

improvementinmorbiditylinkedtoDCKD.

2.3.3. ImpactofDCKDonanaesthesia

Diabetesisanindependentriskfactorforthedevelopmentof

acuterenalfailureintheperioperativeperiod.Thismaydevelopin

theabsenceofpreviousrenaldysfunctionorinpatientswithDCKD.

Postoperatively,thepresenceofDCKDincreasesthisrisk[68].The

perioperativeevaluationofACRandGFRisessentialduringmajor

surgery, as a matterof urgencyor if the patient presents with

diabeticinstabilityorpoorlycontrolledglycaemia.TheGFRcanbe

estimated usingclassicformulae (MDRD, CKD-EPI,Cockroft and

Fig.3.Evaluationofdiabeticchronickidneydisease(DCKD)anditsseverityusingtwoparameters:albumin/creatinineclearanceratio(ACR)andglomerularfiltrationrate (GFR)[61,62].InstagesA1/G1andA1/G2,thediseaseisstable.Theotherstages,1–4,indicateanincreaseinriskofprogressiontoend-stagerenalfailure.

Gault)ifthepatientisstablebutitshouldalsobemeasuredinother

situations.TheperioperativemanagementofDCKDisnotspecific.It

isnecessarytoavoidtheadministrationofnephrotoxicagentsor

drugs in the perioperative period. Haemodynamic optimisation

aims for a mean arterial pressure between 60 and 70mmHg

and>70mmHgifthe patientis hypertensive,tomaintain renal

perfusionpressure.Toachievethisobjective,itisrecommendedto

carryouthaemodynamicmonitoringinordertoevaluatethestroke

volume to guide vascular filling and the administration of

vasopressorsduringsurgicalproceduresassociatedwithariskof

haemodynamicinstability(haemorrhagicsurgery,majorsurgeryor

emergencysurgery)[68].Allotherstrategiesarenon-specificand

shouldfollowtherecommendationsforthemanagementofacute

renalfailure[68].Administrationofanaestheticagentsshouldtake

intoaccountthepharmacokineticandpharmacodynamic

modifi-cations resulting from chronic renal failure without particular

specificityinrelationtodiabetes.

3. Treatmentmanagement(sheetB)

3.1. Metformin(sheetsBandE)

Theefficacyofmetforminonglycaemiccontrolandreductionof

mortality and complications was demonstrated in the UKPDS

study [69]and confirmed in the meta-analysis of Selvin et al.

[70].ThisbeneficialeffectwasalsoobservedintheREACHregistry

in patients with moderate renal failure (creatinine clearance

between 30 and 60mL/min) or a history of heart failure (HF),

whichareclassiccontraindicationstoitsuse[71].Inthestudyof

Duncanetal.,postoperativecomplicationswereidentical

what-everthetreatmentused(metforminornot)[72].These

observa-tions have led to the restriction of contraindications and the

prescriptionofmetforminasfirst-linetherapyinT2D.

Howevertheriskoflacticacidosis,whosemaincauseisrenal

failure,shouldnotbeforgotten.Theincidenceofthiscomplication

is2–9/100,000patients/yearanditsmortalityraterangesfrom30–

50%.In France, the numberof casesof lactic acidosis linked to

metforminincreasedfrom10to72 peryearbetween2005and

2010:inmostcases,patientswereelderly(68%were>65years)

andtookahighdoseofmetformin(meanof2600mg/day).Analysis

of cases alsoshowed acute renal failure in almost all patients.

Progressiontowarddeathwasobservedin20%ofcases[73].

However,publishedstudiesarecontradictory:somehavefound

nocase[74,75]whileothershavefounditonlyinthepresenceof

risk factors [76]. Some authors have even reported a better

cardiovascularprognosisinpatientswithHF[77,78].Incontrast,

other studies have reported very severecases, but these were

alwaysduetotheuntimelyprescriptionofmetformin(severerenal

or heart failure)[79].It is thereforeimportant to lookfor risk

factorsbeforecarryingoutsurgery:

 renalfailure(creatinineclearance<60mL/min);

 administrationofiodinatedcontrastagents;

 situationsthatcouldalterrenalfunction:dehydration,fastingor

medicaltreatments(ACEinhibitorsandsartans,diuretics,

non-steroidalanti-inflammatorydrugs(NSAIDs);

 severeHF(leftventricularejectionfraction<30%);

Thepresenceoftheseriskfactorsalsomeansthatmetformin

shouldnotberestartedtooquicklyinthepostoperativeperiod.

Inpractice,werecommend:

 stopmetforminthenightbefore;

 donotrestartbefore48hformajorsurgeryandafterassuring

adequaterenalfunction;

 donotstopincase ofminororambulatorysurgeryexceptif

thereissevererenalfailure.

3.2. Othernon-insulindrugs(sheetsBandE)

Theothernon-insulintreatmentsfordiabetesarenottakenon

themorningofminorormajorsurgeryandarecontinuedinthe

case ofambulatorysurgery. Hypoglycaemicsulphonamides and

glinides may cause hypoglycaemia. Taking these medications

beforeemergencysurgerymeansthataglucoseinfusionshouldbe

setupifthepatientremainswithanemptystomach.Thisisnotthe

casewiththeothernon-insulindrugs.

3.3. Insulins(sheetsBandF)

3.3.1. Subcutaneous(SC)injections

In T1D, basal insulin should never be stopped, whether

administered as one or two injections, due to the risk of

ketoacidosis.

3.3.2. Particularcaseofinsulinpump(sheetH)

Themainriskduringtheperioperativeperiodisketoacidosisin

T1D patients if continuation of insulin is not immediate after

stopping thepump (SCinjection by thebasal-bolus scheme or

continuousintravenousinsulininfusion).

Inthepreoperativeperiod,andwithpatientquestioning,itis

necessary to gain an understanding of ‘‘total basal delivery’’,

allowing the prescription of an analogue of long-acting (slow)

insulinifthepumpisstopped;ortonotethereplacementscheme,

whichshouldbeknownbythepatient(doseoflong-acting(slow)

insulinifthepumpisstopped).

Inthecaseofambulatorysurgeryorsurgeryofshortduration,

it is necessary to retain the pump which will continue to

administer the basal delivery (for example, patients on

long-acting(slow)insulinwhodonotstoptheirinsulinduringthese

types of surgery). Correction of hyperglycaemia during the

procedurewillbecarriedoutusingacorrectivebolus

adminis-teredbySCinjectionofanultra-rapidanaloguefollowinga

basal-bolusprotocol(sheetsLandQ).

4. Fasting

Ifthepatientneedstobeleftwithanemptystomachwhilehe/

sheistreatedwithinsulin,werecommendsettingupaglucose

infusion (from 7.00 a.m.) which is stopped if blood glucose

is>16.5mmol/L (sheetsG, H, K, L). Taking sulphonamides or

glinides before emergency surgery also requires a glucose

infusionif thepatient remainswith anempty stomach.If the

patientdoesnotuseinsulin,itisnotnecessarytoadministera

glucosesolution.

5. Choiceofagentsandanaesthesiatechniques

5.1. Agents

To date, there is no proof that any anaesthetic agent is

associatedabetteroutcomeindiabeticpatients.

5.2. Techniques

Inparticular,thereisnoproofthatGAprovidesbetterresults

thanRAindiabeticpatients,evenifRAisassociatedwithaslight

increase in glycaemia preoperatively. Spinal and epidural

toahaemodynamicrisk.ThechoicebetweenGAandRAwillbe

madeasfor anyotherpatient.Peripheralnerveblocksare not

contraindicatedindiabeticpatients[80–82]buttheyshouldbe

carriedout after takingand recordingin the patient’s records

severalprecautions,whichare moreimportantthanin a

non-diabeticpatient:preoperativeclinicalexaminationwith

investi-gationsforsignsofdysautonomiaandpre-existing

polyneuropa-thy.

5.3. Investigationsfordifficultintubation

Preoperativeassessmentofrisksatintubationshouldbeclearly

specified because tracheal intubation may be difficult due to

densification of the periarticular collagen structures of the

temporomandibularand atlanto-occipital joints.Asthese

meta-boliccollagendisorders(non-enzymaticglycosylationand

anom-alies of metabolism) simultaneously affect the interphalangeal

joints,itisusuallyadvisedtoevaluatethedifficultiesintracheal

intubationusingthepalmprinttest inpatients withlong-term

diabetes.

Disclosureofinterest

Theauthorsdeclarethattheyhavenocompetinginterest.

The members of the working party have received support

from the French Society of Anaesthesia and Intensive Care

Medicine(SFAR)andtheFrenchSocietyfortheStudyofDiabetes

(SFD)fortransportandaccommodationwhennecessary.Thetwo

nationalbodieshavealsoprovidedhonorariafortranslationof

thetexts.

AppendixA

EquivalencebetweenbloodglucoselevelsandHbA1c

ThereisadirectcorrelationbetweenHbA1clevelin%andmean

glycaemia over the previous 3months [11], according to the

formula:

Meanglycaemiag/L=[(28.7T_{HbA1c%)–46.7]/100,or}

Meanglycaemiammol/L=(1.5944T_{HbA1c%)–2.5944}

The table below shows the corresponding values for the

standardvaluesofHbA1c.

HbA1c Meanbloodglucoselevel

% g/L mmol/L 5 0.97[0.76–1.20] 5.4[4.2–6.7] 6 1.26[1.00–1.52] 7.0[5.5–8.5] 7 1.54[1.23–1.85] 8.6[6.8–10.3] 8 1.83[1.47–2.17] 10.2[8.1–12.1] 9 2.12[1.70–2.49] 11.8[9.4–13.9] 10 2.40[1.93–2.82] 13.4[10.7–15.7] 11 2.69[2.17–3.14] 14.9[12.0–17.5] 12 2.98[2.40–3.47] 16.5[13.3–19.3]

Thevaluesinbrackets[]arethecorresponding95%confidence

intervals.

AppendixB

Investigationsandgradingofcardiacautonomyneuropathyin

theanaesthesiaconsultation.

1.Investigationsfororthostatichypotension:

Measurebloodpressureafter10minofdecubitusandthen1,

2and3minafterchangingtoorthostatism.Orthostatic

hypoten-sionisdefinedbyadecreaseinsystolicbloodpressureofatleast

20mmHg(30mmHginhypertensivepatients)and/orofdiastolic

bloodpressureofatleast10mmHginorthostatism.Intheabsence

ofanyiatrogenicfactor,hypovolaemiaandanaemia,itspresenceis

indicativeofserioussympatheticdysautonomicdamage.

2.Testsexploringthecardio-vagalchangesinheartrate(HR):

Thesetestsshouldbeperformedatrestandlongaftercoffeeor

tobaccoconsumption.

Deep respiration tests: the patient is placed in decubitus

positionandshouldtrainbeforehandtothistypeofrespiration.

Thepatientisaskedtoperformsixcyclesofdeeprespirationin

1min:duringeachcycle,thepatientisstimulatedtoinspire(for

5 sec) and then expire (for 5 sec) deeply. ECG is recorded

continuouslyduringthetestandthebeginningofbothinspiration

(I)andexpiration(E)aremarkedonthetracing.

Physiologically,HRincreasesoninspiration(i.e.RRdecreases)

anddecreasesonexpiration(i.e.RRincreases).TheRR-Emaximum

andRR-Iminimumaremeasuredforeachcycle.TheratioRR-E/

RR-Iiscalculatedforeachcycle.

Forexample:2.3/2.1forthefirstcycle(1.0952=1.10).

Replace2.1or2.3squaresby21mmor23mm

Themeanoftheratiosfor6respiratorycyclescanbecalculated.

Forexample:(1.10+1.21+1.11+1.09+1.26+1.15)/6=1.15

Correlations (discontinuous line) between the RR ratio

duringthedeeprespirationtestandageforhealthysubjects.

Resultsforthepatientareanalysedbypositioningthemonthe

graph.Theyareabnormalifbelowthelineforthe5thpercentile

(continuousline)[83].Theresultisnormalifitisonorbelowthe

thickline.Forexample,thetestisnormalforapatient40yearof

agebutnotforayoungerpatient.

Orthostatictest:

After10minofdecubitus,thepatientstandsupquickly.The

ECGisrecordedcontinuouslybeforerisingandduringtheminute

following change to the standing position. The HR normally

increasesinthefirstsecondsfollowingpassagetoorthostatismto

reachitsmaximumtowardsthe15th_{second,thenslowsdownand}

reachesit minimalvaluetowardsthe30th_{second.Theresultis}

expressedbytheratioRR30/RR15.

Correlations (discontinuous line) between the RR ratio

duringtheorthostatictestandageinhealthysubjects.

Theresultsforthepatientareanalysedbypositioningthemon

thegraph.Theyareabnormalifbelowthelineforthe5th_percentile

(continuousbold line)[83].Inourexample,RR30/RR15=3.9/

3.3=1.18.Thisresultisthenprojectedonthecurveasafunctionof

age,forinterpretation.

3.Practicalprocedureproposed

Thepatientplaceshim/herselfindecubitusposition

AstandardECGisrecorded

After10minofrest,adeeprespirationtestiscarriedout

Measurearterialpressureandleavethebloodpressuremonitor

inplace

RestarttheECG

Thepatientstandsquickly

TheECGisrecordedfor1min

Arterialpressureismeasuredagainat1,2and3min.

4.GradingofCAN

Fromthesetests,itispossibletogradetheCAN:

-analteredcardio-vagaltestidentifiespossibleorearlyCAN

-definiteorconfirmedCANcorrespondstotwoaltered

cardio-vagaltests

-complicatedCANisidentifiedbythepresenceoforthostatic

hypotensionoranomaliesintheheartratetests

References

[1]HauteAutorite´ deSante´.Recommandationpopurlapratiqueclinique. Stra-te´gieme´dicamenteuseducontroˆleglyce´miquedudiabe`tedutype2;2013 [http://www.has-sante.fr/portail/upload/docs/application/pdf/2013-02/ 10irp04_reco_diabete_type_2.pdf].

[2]MeynaarIA,EslamiS,Abu-HannaA,vanderVoortP,deLangeDW,deKeizerN. Bloodglucoseamplitudevariabilityaspredictorformortalityinsurgicaland medicalintensivecareunitpatients:amulticentercohortstudy.JCritCare 2012;27:119–24.

[3]SatoH,CarvalhoG,SatoT,LattermannR,MatsukawaT,SchrickerT.The associationofpreoperativeglycemiccontrol,intraoperativeinsulinsensitivity, andoutcomesaftercardiacsurgery.JClinEndocrinolMetab2010;95:4338– 44.

[4]HalkosME,LattoufOM,PuskasJD,Kilgo P,CooperWA,MorrisCD, etal. ElevatedpreoperativehemoglobinA1clevelisassociatedwithreduced long-term survival after coronary artery bypass surgery. Ann Thorac Surg 2008;86:1431–7.

[5]O’SullivanCJ,HynesN,MahendranB,AndrewsEJ,AvalosG,TawfikS,etal. HaemoglobinA1c(HbA1C)innon-diabeticanddiabeticvascularpatients.Is HbA1Canindependentriskfactorandpredictorofadverseoutcome?EurJ VascEndovascSurg2006;32:188–97.

[6]LazarHL,ChipkinSR,FitzgeraldCA,BaoY,CabralH,ApsteinCS.Tightglycemic controlindiabeticcoronaryarterybypassgraftpatientsimproves periopera-tive outcomes and decreases recurrent ischemic events. Circulation 2004;109:1497–502.

[7]BhamidipatiCM,LaParDJ,StukenborgGJ,MorrisonCC,KernJA,KronIL,etal. Superiorityofmoderatecontrolofhyperglycemiatotightcontrolinpatients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 2011;141:543–51.

[8]DesaiSP,HenryLL,HolmesSD,HuntSL,MartinCT,HebsurS.Strictversus liberaltargetrangeforperioperativeglucoseinpatientsundergoingcoronary arterybypassgrafting:aprospectiverandomizedcontrolledtrial.JThorac cardiovascSurg2012;143:318–25.

[9]FishLH,WeaverTW,MooreAL,SteelLG.Valueofpostoperativebloodglucose inpredictingcomplicationsandlengthofstayaftercoronaryarterybypass grafting.AmJCardiol2003;92:74–6.

[10]FurnaryAP,ZerrKJ,GrunkemeierGL,StarrA.Continuousintravenousinsulin infusionreducestheincidenceofdeepsternalwoundinfectionindiabetic patientsaftercardiacsurgicalprocedures.AnnThoracSurg1999;67:352–60 [discussion360-362].

[11]NathanDM,KuenenJ,BorgR,ZhengH,SchoenfeldD,HeineRJ,etal. Translat-ing theA1C assayinto estimatedaverage glucose values.DiabetesCare 2008;31:1473–8.

[12]BrutsaertE,CareyM,ZonszeinJ.Theclinicalimpactofinpatienthypoglycemia. JDiabetesComplications2014;28:565–72.

[13]SchopmanJE,GeddesJ,FrierBM.Prevalenceofimpairedawarenessof hypo-glycaemiaandfrequencyofhypoglycaemiaininsulin-treatedtype2diabetes. DiabetesResClinPract2010;87:64–8.

[14]InzucchiSE,BergenstalRM,BuseJB,DiamantM,FerranniniE,NauckM,etal. Managementofhyperglycemiaintype2diabetes:apatient-centered ap-proach:positionstatementoftheAmericanDiabetesAssociation(ADA)and theEuropean AssociationfortheStudyofDiabetes(EASD).DiabetesCare 2012;35:1364–79.

[15]KemplerP,AmarencoG,FreemanR,FrontoniS,HorowitzM,StevensM,etal. Managementstrategiesforgastrointestinal,erectile,bladder,andsudomotor dysfunction in patients with diabetes. Diabetes Metab Res Rev 2011;27:665–77.

[16]PhillipsLK,RaynerCK,JonesKL,HorowitzM.Measurementofgastric empty-ingindiabetes.JDiabetesComplications2014;28:894–903.

[17]BouvetL,MazoitJX,ChassardD, AllaouchicheB,BoselliE,BenhamouD. Clinicalassessmentoftheultrasonographicmeasurementofantralareafor

estimating preoperative gastric content and volume. Anesthesiology 2011;114:1086–92.

[18]JanssensJ, PeetersTL,VantrappenG,TackJ, UrbainJL,De RooM,etal. Improvementofgastricemptyingindiabeticgastroparesisbyerythromycin. Preliminarystudies.NEnglJMed1990;322:1028–31.

[19]EllishWS,KarthaV,FluderE,SlogoffS.Effectofmetoclopramideongastric fluidvolumesindiabeticpatientswhohavefastedbeforeelectivesurgery. Anesthesiology2005;102:904–9.

[20]Ryde´nL,GrantPJ,AnkerSD,BerneC,CosentinoF,DanchinN,etal.ESC Guidelinesondiabetes,pre-diabetes,andcardiovasculardiseasesdeveloped incollaborationwiththeEASD:theTaskForceondiabetes,pre-diabetes,and cardiovascular diseasesof theEuropean Societyof Cardiology(ESC) and developedincollaborationwiththeEuropeanAssociationfortheStudyof Diabetes(EASD).EuropHeartJ2013;34:3035–87.

[21]ValensiP,CossonE.Itisnotyetthetimetostopscreeningdiabeticpatientsfor silentmyocardialischaemia.DiabetesMetab2010;36:91–6.

[22]ValensiP,Parie`sJ,Brulport-CerisierV,TorremochaF,SachsRN,VanzettoG, etal.Predictivevalueofsilentmyocardialischemiaforcardiaceventsin diabeticpatients:influenceofageinaFrenchmulticenterstudy.DiabetesCare 2005;28:2722–7.

[23]CossonE,NguyenMT,ChanuB,BanuI,ChihebS,BaltaC,etal.Cardiovascular risk prediction is improved by adding asymptomatic coronary statusto routine risk assessment in type 2 diabetic patients. Diabetes Care 2011;34:2101–7.

[24]YoungLH,WackersFJ,ChyunDA,DaveyJA,BarrettEJ,TailleferR,etal.Cardiac outcomesafterscreeningforasymptomaticcoronaryarterydiseaseinpatients withtype2diabetes:theDIADstudy:arandomizedcontrolledtrial.JAMA 2009;301:1547–55.

[25]HackerM,BeckerC.Theincrementalvalueofcoronaryarterycalciumscoresto myocardialsinglephotonemissioncomputertomographyinriskassessment. JNuclearCardiol2011;18:700–11[quiz712-6].

[26]McFallsEO,WardHB,MoritzTE,GoldmanS,KrupskiWC,LittooyF,etal. Coronary-arteryrevascularizationbeforeelectivemajorvascularsurgery.N EnglJMed2004;351:2795–804.

[27]FoucrierA,RodsethR,AissaouiM,IbanesC,GoarinJP,LandaisP,etal.The long-termimpactofearlycardiovasculartherapyintensificationforpostoperative troponin elevation after major vascular surgery. Anesth Analg 2014;119:1053–63.

[28]NicholsGA,HillierTA,ErbeyJR,BrownJB.Congestiveheartfailureintype 2 diabetes: prevalence, incidence, and risk factors. Diabetes care 2001;24:1614–9.

[29]Amour J, Kersten JR.Diabetic cardiomyopathyand anesthesia:bench to bedside.Anesthesiology2008;108:524–30.

[30]PhamI,CossonE,NguyenMT,BanuI,GenevoisI,PoignardP,etal.Evidencefora SpecificDiabeticCardiomyopathy:anObservationalRetrospective Echocardio-graphicStudyin656AsymptomaticType2DiabeticPatients.InternJEndocrinol 2015;2015:743503.http://dx.doi.org/10.1155/2015/743503[EpubMay13]. [31]lbertiniJP, CohenR,Valensi P,SachsRN,CharniotJC. B-typenatriuretic

peptide,amarkerofasymptomaticleftventriculardysfunctionintype2 dia-beticpatients.DiabetesMetab2008;34(4Pt1):355–62.

[32]SpalloneV,ZieglerD,FreemanR,BernardiL,FrontoniS,Pop-BusuiR,etal. Cardiovascularautonomicneuropathyindiabetes:clinicalimpact, assess-ment,diagnosis,andmanagement.DiabetesMetabResRev2011;27:639–53.

[33]Valensi P, Paries J, Attali JR.Cardiac autonomic neuropathy in diabetic patients: influence of diabetes duration, obesity, and microangiopathic complications – the Frenchmulticenter study. Metab Clin Experimental 2003;52:815–20.

[34]Pop-BusuiR,EvansGW,GersteinHC,FonsecaV,FlegJL,HoogwerfBJ,etal. EffectsofcardiacautonomicdysfunctiononmortalityriskintheActionto Control Cardiovascular Risk in Diabetes (ACCORD) trial. Diabetes Care 2010;33:1578–84.

[35]ValensiP,SachsRN,HarfoucheB,LormeauB,PariesJ,CossonE,etal.Predictive valueofcardiacautonomicneuropathyindiabeticpatientswithorwithout silentmyocardialischemia.DiabetesCare2001;24:339–43.

[36]TesfayeS,BoultonAJ,DyckPJ,FreemanR,HorowitzM,KemplerP,etal. Diabeticneuropathies:updateondefinitions,diagnosticcriteria,estimationof severityandtreatments.DiabetesCare2010;33:2285–93.

[37]Neukirchen M,KienbaumP.Sympathetic nervous system:evaluationand importanceforclinicalgeneralanesthesia.Anesthesiology2008;109:1113–31.

[38]KirnoK,LundinS,ElamM.Effectsofintrathecalmorphineandspinal anaes-thesiaonsympatheticnerveactivityinhumans.ActaanaesthesiolScand 1993;37:54–9.

[39]LundinS,WallinBG,ElamM.Intraneuralrecordingofmusclesympathetic activityduringepiduralanesthesiainhumans.AnesthAnalg1989;69:788–93.

[40]BreslowMJ,JordanDA,ChristophersonR,RosenfeldB,MillerCF,HanleyDF, etal.Epiduralmorphinedecreasespostoperativehypertensionbyattenuating sympatheticnervoussystemhyperactivity.JAMA1989;261:3577–81.

[41]GrassiG,Dell’OroR,Quarti-TrevanoF,ScopellitiF,SeravalleG,PaleariF,etal. Neuroadrenergicandreflexabnormalitiesinpatientswithmetabolic syn-drome.Diabetologia2005;48:1359–65.

[42]Knu¨ttgenD,TrojanS,WeberM,WolfM,WapplerF.[Pre-operative measure-mentofheartratevariabilityindiabetics:amethodtoestimatebloodpressure stabilityduringanaesthesiainduction].DerAnaesthesist2005;54:442–9.

[43]HuangCJ,KuokCH,KuoTB,HsuYW,TsaiPS.Pre-operativemeasurementof heartratevariabilitypredictshypotensionduringgeneralanesthesia.Acta AnaesthesiolScand2006;50:542–8.

[44]LankhorstS,KeetSW,BulteCS,BoerC.Theimpactofautonomicdysfunctionon peri-operativecardiovascularcomplications.Anaesthesia2015;70:336–43.

[45]BurgosLG,EbertTJ,AsiddaoC,TurnerLA,PattisonCZ,Wang-ChengR,etal. Increasedintraoperativecardiovascularmorbidityindiabeticswith autonom-icneuropathy.Anesthesiology1989;70:591–7.

[46]SobotkaPA,LissHP,VinikAI.Impairedhypoxicventilatorydriveindiabetic patientswithautonomicneuropathy.JClinEndocrinolMetab1986;62:658– 63.

[47]KitamuraA,HoshinoT,KonT,OgawaR.Patientswithdiabeticneuropathyare atriskofagreaterintraoperativereductionincoretemperature. Anesthesiol-ogy2000;92:1311–8.

[48]KadoiY.Anestheticconsiderationsindiabeticpatients.PartI:preoperative considerationsofpatientswithdiabetesmellitus.JAnesth2010;24:739–47.

[49]Socie´te´ franc¸aised’anestyhe´sieetdere´animation(Sfar),Socie´te´ franc¸aisede cardiologie(SFC).[Perioperativeassessmentofcardiacriskpatientin non-cardiac,surgery].AnnFrAnesthReanim2011;30:e5–29.

[50]Martı´nez-CastelaoA,Navarro-Gonza´lezJF,Go´rrizJL,deAlvaroF.TheConcept andtheepidemiologyofdiabeticnephropathyhavechangedinrecentyears.J ClinMed2015;4:1207–16.

[51]EbohC,ChowdhuryTA.Managementofdiabeticrenaldisease.AnnTranslMed 2015;3:154.

[52]GhaderianSB,HayatiF,ShayanpourS,BeladiMousaviSS.Diabetesand end-stagerenal disease;a review article onnewconcepts. JRenal InjPrev 2015;4:28–33.

[53]BurrowsNR,LiY,GeissLS.Incidenceoftreatmentforend-stagerenaldisease amongindividualswithdiabetesintheU.S.continuestodecline.DiabetesCare 2010;33:73–7.

[54]DounousiE,Duni A,LeivaditisK,VaiosV,EleftheriadisT,LiakopoulosV. Improvementsinthemanagementofdiabeticnephropathy.RevDiabetStud 2015;12:119–33.

[55]GreggEW,LiY,WangJ,BurrowsNR,AliMK,RolkaD,etal.Changesin diabetes-related complications in the United States, 1990–2010. N Engl J Med 2014;370:1514–23.

[56]NarresM,ClaessenH,DrosteS,KvitkinaT,KochM,KussO,etal.Theincidence ofend-stagerenaldiseaseinthediabetic(comparedtothenon-diabetic) population:asystematicreview.PloSOne2016;11:e0147329.

[57]RadcliffeNJ,SeahJM,ClarkeM,MacIsaacRJ,JerumsG,EkinciEI.Clinical predictivefactorsindiabetickidneydiseaseprogression.JDiabetesInvestig 2017;8:6–18.

[58]MonteroRM,CovicA,GnudiL,GoldsmithD.Diabeticnephropathy:Whatdoes thefuturehold?InternUrolNephrol2016;48:99–113.

[59]Toth-ManikowskiS,AttaMG.Diabetickidneydisease:pathophysiologyand therapeutictargets.JDiabetesRes2015;2015:697010.

[60]TervaertTW,MooyaartAL,AmannK,CohenAH,CookHT,DrachenbergCB, etal.Pathologicclassification ofdiabeticnephropathy.JAmSocNephrol 2010;21:556–63.

[61]LeveyAS,deJongPE,CoreshJ,ElNahasM,AstorBC,MatsushitaK,etal.The definition,classification,andprognosisofchronickidneydisease:aKDIGO ControversiesConferencereport.KidneyIntern2011;80:17–28.

[62]KimSS,KimJH,KimIJ.Currentchallengesindiabeticnephropathy:early diagnosisandwaystoimproveoutcomes.EndocrinolMetab2016;31:245–53.

[63]TheRenalAssociation.UKRenalregisttry.Thesixtheenthannualreport;2013 [http://www.renalreg.org/reports/2013-the sixteenth-annual-report/. Last accessApril24,2017].

[64]AmericanDiabetesAssociation.9.Microvascularcomplicationsandfootcare. Diabetescare2016;39(Suppl.1):S72–80.

[65]BlacherJ, HalimiJM,Hanon O, MouradJJ, Pathak A,SchnebertB, etal. Managementofhypertensioninadults:the2013FrenchSocietyof Hyperten-sionguidelines.FundamClinPharmacol2014;28:1–9.

[66]EmdinCA,RahimiK,NealB,CallenderT,PerkovicV,PatelA.Bloodpressure loweringintype2diabetes:asystematicreviewandmeta-analysis.JAMA 2015;313:603–15.

[67]NationalKidney,Foundation.KDOQIclinicalpracticeguidelinefordiabetes andCKD:2012update.AmJKidneyDis2012;60:850–86.

[68]IchaiC,VinsonneauC,SouweineB,ArmandoF,CanetE,Clec’hC,etal.Acute kidneyinjuryintheperioperativeperiodandinintensivecareunits(excluding renalreplacementtherapies).AnaesthCritCarePainMed2016;35:151–65.

[69]Noauthors. Effectofintensiveblood-glucosecontrolwith metforminon complicationsinoverweightpatientswithtype2diabetes(UKPDS34).UK ProspectiveDiabetesStudy(UKPDS)Group.Lancet1998;352:854–65.

[70]SelvinE,Bolen S,Yeh HC,Wiley C, WilsonLM,Marinopoulos SS,etal. Cardiovascularoutcomesintrialsoforaldiabetesmedications:asystematic review.ArchInternMed2008;168:2070–80.

[71]RousselR,TravertF,PasquetB,WilsonPW,SmithJrSC,GotoS,etal.Reduction ofAtherothrombosisforContinuedHealth(REACH)RegistryInvestigators. Metforminuseandmortalityamongpatientswithdiabetesand atherothrom-bosis.ArchInternMed2010;170:1892–9.

[72]DuncanAI,KochCG,XuM,ManlapazM,BatdorfB,PitasG,etal.Recent metforminingestiondoesnotincreasein-hospitalmorbidityormortalityafter cardiacsurgery.AnesthAnalg2007;104:42–50.

[73]TrinkleyKE,MaloneDC,NelsonJA,SaseenJJ.Prescribingattitudes,behaviors andopinionsregardingmetforminforpatientswithdiabetes:afocusgroup study.TherapAdvChronicDis2016;7:220–8.

[74]SalpeterSR,GreyberE,PasternakGA,SalpeterEE.Riskoffatalandnonfatal lacticacidosis withmetforminuseintype2diabetesmellitus.Cochrane databasesystrev2010;4:CD002967.

[75]BandoK,OchiaiS,KunimatsuT,Deguchi J,KimuraJ,FunabashiH,etal. Comparisonofpotentialrisksoflacticacidosisinductionbybiguanidesin rats.RegulToxicolPharmacol2010;58:155–60.

[76]RendaF,MuraP,FincoG,FerrazinF,PaniL,LandoniG.Metformin-associatedlactic acidosisrequiringhospitalization.Anational10yearsurveyandasystematic literaturereview.EurRevMedPharmacolSci2013;17(Suppl.1):45–9.

[77]ScheenAJ,PaquotN.[Useofmetforminindiabeticpatientswithcardiac disease:benefit-riskbalance].RevMedSuisse2013;9:1527–33.

[78]ScheenAJ,PaquotN.Metforminrevisited:acriticalreviewofthebenefit-risk balanceinat-riskpatientswithtype2diabetes.Diabetesmetab2013;39:179–90.

[79]MerckerSK, MaierC, Neumann G, Wulf H.Lactic acidosis as a serious perioperativecomplicationofantidiabeticbiguanidemedicationwith met-formin.Anesthesiology1997;87:1003–5.

[80]GebhardRE, NielsenKC,PietrobonR,Missair A,Williams BA.Diabetes mellitus,independentofbodymassindex,isassociatedwitha‘‘higher success’’rateforsupraclavicularbrachialplexusblocks.Reganesthpain med2009;34:404–7.

[81]WelchMB,BrummettCM,WelchTD,TremperKK,ShanksAM,GuglaniP,etal. Perioperativeperipheralnerveinjuries:aretrospectivestudyof380,680 cases during a 10-year period at a single institution. Anesthesiology 2009;111:490–7.

[82]WilliamsBA,MurinsonBB.Diabetesmellitusandsubclinicalneuropathy:a call for new paths in peripheral nerve block research. Anesthesiology 2008;109(3):361–2.

[83]ValensiP,FeuvreayD,SachsRN.CoeuretDiabe`te.Priseenchargeetsuivides patientsdiabe´tiques.Paris,France:E´ditionsFrisonRoche;1999.