Diabetes mellitus: management of type 2

NICE updated its guidance on the management of type 2 diabetes mellitus (T2DM) in 2015. Key points are listed
below:

HbA1c targets have changed. They are now dependent on what antidiabetic drugs a patient is receiving and
other factors such as frailty
there is more flexibility in the second stage of treating patients (i.e. after metformin has been started) - you
now have a choice of 4 oral antidiabetic agents

It's worthwhile thinking of the average patient who is taking metformin for T2DM, you can titrate up metformin
and encourage lifestyle changes to aim for a HbA1c of 48 mmol/mol (6.5%), but should only add a second drug if
the HbA1c rises to 58 mmol/mol (7.5%)

Dietary advice

encourage high fibre, low glycaemic index sources of carbohydrates

include low-fat dairy products and oily fish
control the intake of foods containing saturated fats and trans fatty acids
limited substitution of sucrose-containing foods for other carbohydrates is allowable, but care should be
taken to avoid excess energy intake
discourage use of foods marketed specifically at people with diabetes
initial target weight loss in an overweight person is 5-10%

HbA1c targets
This is area which has changed in 2015

individual targets should be agreed with patients to encourage motivation

HbA1c should be checked every 3-6 months until stable, then 6 monthly
NICE encourage us to consider relaxing targets on 'a case-by-case basis, with particular consideration for
people who are older or frail, for adults with type 2 diabetes'
in 2015 the guidelines changed so HbA1c targets are now dependent on treatment:

Lifestyle or single drug treatment

Management of T2DM

HbA1c target

Lifestyle

48 mmol/mol

Management of T2DM

HbA1c target
(6.5%)

Lifestyle + metformin

48 mmol/mol
(6.5%)

Includes any drug which may cause hypoglycaemia 53 mmol/mol

(e.g. lifestyle + sulfonylurea)
(7.0%)

Practical examples

a patient is newly diagnosed with HbA1c and wants to try lifestyle treatment first. You agree a target of 48
mmol/mol (6.5%)
you review a patient 6 months after starting metformin. His HbA1c is 51 mmol/mol (6.8%). You increase
his metformin from 500mg bd to 500mg tds and reinforce lifestyle factors

Patient already on treatment

Management of T2DM

HbA1c target

Already on one drug, but HbA1c has risen to 58

mmol/mol (7.5%)

53 mmol/mol
(7.0%)

Drug treatment
The 2015 NICE guidelines introduced some changes into the management of type 2 diabetes. There are essentially
two pathways, one for patients who can tolerate metformin, and one for those who can't:

Tolerates metformin:

metformin is still first-line and should be offered if the HbA1c rises to 48 mmol/mol (6.5%)* on lifestyle
interventions
if the HbA1c has risen to 58 mmol/mol (7.5%) then a second drug should be added from the following list:
sulfonylurea
gliptin
pioglitazone
SGLT-2 inhibitor
if despite this the HbA1c rises to, or remains above 58 mmol/mol (7.5%) then triple therapy with one of the
following combinations should be offered:
metformin + gliptin + sulfonylurea
metformin + pioglitazone + sulfonylurea
metformin + sulfonylurea + SGLT-2 inhibitor
metformin + pioglitazone + SGLT-2 inhibitor
OR insulin therapy should be considered

Criteria for glucagon-like peptide1 (GLP1) mimetic (e.g. exenatide)

if triple therapy is not effective, not tolerated or contraindicated then NICE advise that we consider
combination therapy with metformin, a sulfonylurea and a glucagonlike peptide1 (GLP1) mimetic if:
BMI >= 35 kg/m and specific psychological or other medical problems associated with obesity or
BMI < 35 kg/m and for whom insulin therapy would have significant occupational implications or

weight loss would benefit other significant obesityrelated comorbidities

only continue if there is a reduction of at least 11 mmol/mol [1.0%] in HbA1c and a weight loss of at least
3% of initial body weight in 6 months

Practical examples

you review an established type 2 diabetic on maximum dose metformin. Her HbA1c is 55 mmol/mol
(7.2%). You do not add another drug as she has not reached the threshold of 58 mmol/mol (7.5%)
a type 2 diabetic is found to have a HbA1c of 62 mmol/mol (7.8%) at annual review. They are currently on
maximum dose metformin. You elect to add a sulfonylurea

Cannot tolerate metformin or contraindicated

if the HbA1c rises to 48 mmol/mol (6.5%)* on lifestyle interventions, consider one of the following:
sulfonylurea
gliptin
pioglitazone
if the HbA1c has risen to 58 mmol/mol (7.5%) then a one of the following combinations should be used:
gliptin + pioglitazone
gliptin + sulfonylurea
pioglitazone + sulfonylurea
if despite this the HbA1c rises to, or remains above 58 mmol/mol (7.5%) then consider insulin therapy

Starting insulin

metformin should be continued. In terms of other drugs NICE advice: 'Review the continued need for other
blood glucose lowering therapies'
NICE recommend starting with human NPH insulin (isophane, intermediate acting) taken at bed-time or
twice daily according to need

Risk factor modification

Blood pressure

target is < 140/80 mmHg (or < 130/80 mmHg if end-organ damage is present)
ACE inhibitors are first-line

Antiplatelets

should not be offered unless a patient has existing cardiovascular disease

Lipids

following the 2014 NICE lipid modification guidelines only patients with a 10-year cardiovascular risk >
10% (using QRISK2) should be offered a statin. The first-line statin of choice is atorvastatin 20mg on

Graphic showing choice of statin.

*this is a bit confusing because isn't the diagnostic criteria for T2DM HbA1c 48 mmol/mol (6.5%)? So shouldn't all
patients be offered metformin at diagnosis? Our interpretation of this is that some patients upon diagnosis will elect
to try lifestyle measures, which may reduce their HbA1c below this level. If it then rises to the diagnostic threshold
again metformin should be offered
NICE
2015 Type 2 diabetes guidelines

SIGN
2010 Management of diabetes

NICE
2014 Lipid modification guidelines

Clinical pointers: Blood glucose control in type 2

diabetes
Clinical pointers
1. An individualised HbA1c target

2. Glucose self monitoring

3. Stepwise treatment
4. Role of pioglitazone
5. Role of DPP-4 inhibitors
6. Role of SGLT-2 inhibitors
7. Role of GLP-1 receptor agonists

Continue (%7B%7Bnext.url%7D%7D)

Module Outline
Latest guidelines ({{navigation.571de1874d4b5.url}})
Clinical pointers ({{navigation.571de5eb829f0.url}})
References ({{navigation.571de655ed1fe.url}})
Further resources ({{navigation.5733478be7c3b.url}})
Final assessment ({{navigation.571de692e7c87.url}})

Clinical pointers: Blood glucose control in type 2

diabetes
Clinical pointers
1. An individualised HbA1c target

Set an HbA1c target that is personalised to the patients individual circumstances and aim lower
in newly diagnosed patients than in those with established long-term diabetes.
Type 2 diabetes is often discovered during routine blood testing, and a signicant proportion of
people already have established complications at the point of diagnosis.[1, 2, 3] This highlights
the need for early diagnosis and sustained treatment, for both control of blood glucose levels
and reduction of vascular risk factors. The UK Prospective Diabetes Study (UKPDS) and Steno-2
study have shown the benets of early intensive glucose lowering and multifactorial
interventions, including blood pressure control and lipid lowering, in reducing the microvascular
complications of retinopathy, nephropathy, and neuropathy.[4, 5, 6, 7, 8]
Intensive treatment in UKPDS targeted an HbA1c of 53 mmol/mol (7.0%) in the rst decade of
diagnosis, and showed a 25% risk reduction in microvascular complications.[4] A signicant
cardiovascular benet was also seen with the ten year UKPDS follow up, especially in the
metformin treated group, where there was a 33% reduction in myocardial infarction.[6]
The Steno-2 study showed that aggressive treatment of multiple risk factors in the early stages
of type 2 diabetes lowered the risk of microvascular complications by 50%, over a follow up
period of 7.8 years.[7] The aggressive targets included an HbA1c level of less than 48 mmol/mol
(6.5%), a fasting serum total cholesterol level of less than 4.5 mmol/l, a fasting serum triglyceride
level of less than 1.7 mmol/l, a systolic blood pressure of less than 130 mmHg, and a diastolic
blood pressure of less than 80 mmHg. In a 5.5 year extension to the Steno-2 trial,[8a] following
up patients at 13.3 years after the beginning of the trial, intensive therapy was associated with a
lower risk of death from cardiovascular causes (hazard ratio 0.43; 95% CI, 0.19 to 0.94; P=0.04),
and of cardiovascular events (hazard ratio 0.41; 95% CI, 0.25 to 0.67; P=0.001).
These studies highlight the importance of setting specic glycaemic targets, and of addressing
vascular risk factors from the time of diagnosis.[6
(https://app.elucidat.com/projects/build_view_wrapper/572b5327bfe26/571de5eb829f0#), 7
(https://app.elucidat.com/projects/build_view_wrapper/572b5327bfe26/571de5eb829f0#)]
Despite nite healthcare budgets, with the array of drugs now available, we have the opportunity
to personalise management strategies to reduce the long term complications and costs of
diabetes.[9, 10]
The 2015 guideline from NICE,[11] and a position statement from the ADA/EASD(American
Diabetes Association/European Association for Study of Diabetes) also published in 2015,[10]
both recommend taking a personalised approach to setting a blood glucose target for each
individual patient, based on consideration of both patient specic and disease related features.

The 2015 NICE guideline compared intensive glycaemic control with conventional glycaemic
control in people with type 2 diabetes, using a Cochrane review as the primary source of
evidence.[11a, 11b] Compared with conventional control, the NICE guideline found that
intensive glycaemic control did not statistically signicantly reduce death from any cause, or
death from cardiovascular causes, nor was there a statistically signicant improvement in other
macrovascular complications.
The NICE guideline found that intensive glycaemic control did reduce the risk of the composite
of microvascular complications (relative risk 0.75; 95% CI 0.61 to 0.92; 3 RCTs, n=4376), but no
statistically signicant reductions in risk were seen for the individual end points of nephropathy,
progression to end stage renal disease, or retinopathy. The guideline also found that intensive
glycaemic control increased the risk of severe hypoglycaemia (relative risk 2.23; 95% CI 1.22 to
4.08; 13 RCTs, n=5452) and mild hypoglycaemia (relative risk 1.85; 95% CI 1.53 to 2.25; 12 RCTs,
n=6320). The NICE guideline therefore supports the setting of target values, but considers it
important to ensure that a person's risk of hypoglycaemia is evaluated when setting appropriate
target levels.[11]
The 2015 NICE guideline recommendations include the following [11]:
Clinicians should involve adults with type 2 diabetes in decisions about their individual HbA1c
target. Encourage and support individuals to achieve their agreed targets unless they
experience hypoglycaemia or detrimental effects on quality of life, especially in early years of
diagnosis
An HbA1c target of 48 mmol/mol (6.5%) is recommended for adults with type 2 diabetes
managed either by lifestyle and diet, or by lifestyle and diet combined with a single drug that
is not associated with hypoglycaemia (eg metformin or DPP4 inhibitors)
An HbA1c target of 53 mmol/mol (7.0%) is recommended for adults on a drug associated
with hypoglycaemia (eg sulfonylurea or insulin)
If HbA1c levels rise to 58 mmol/mol (7.5%) or higher in a patient taking a single drug for
diabetes, lifestyle measures need to be reinforced, and drug treatment needs to be
intensied. The patient also needs to aim for an HbA1c target of 53 mmol/mol (7.0%)
HbA1c targets and management strategy should be re-evaluated at every follow up. Consider
relaxing the target HbA1c on a case by case basis, particularly for people who are older or frail.
Patients in whom relaxed targets may be appropriate include:
Those with a reduced life expectancy
Those for whom the consequences of hypoglycaemia are potentially more severe, for
example people who are at risk of falling, people who have impaired awareness of
hypoglycaemia, and people who drive or operate machinery as part of their job
Those with signicant comorbidities, especially cardiovascular disease and/or renal
dysfunction.

Glycaemic targets are different for pregnant women, and are covered in a separate NICE
guideline published in February 2015: Diabetes in pregnancy: management from preconception
to the postnatal period. (https://www.nice.org.uk/guidance/ng3)

The ADA/EASD 2015 position statement includes useful information about how the stringency
of HbA1c targets can be tightened or relaxed in a personalised manner, summarised in gure 1
below.

Figure 1. Approach to management of hyperglycaemia: spectrum from less to more stringent

HbA1c targets, based on patient and disease features (gure adapted from ADA/EASD 2015
position statement).[10]

2. Glucose self monitoring

3. Stepwise treatment
4. Role of pioglitazone
5. Role of DPP-4 inhibitors
6. Role of SGLT-2 inhibitors

2.Glucoseselfmonitoring
Thereiscontroversyovertheroleofselfmonitoringofbloodglucose.Evidence
suggestsitcanencouragemotivatedpatientsintheirselfmanagementeducation,and
shouldthereforebeconsideredinsomepeoplewithtype2diabetes,althoughNICE
advisesagainstroutineselfmonitoring.
Selfmonitoringofbloodglucoseeducates,motivates,andempowersthepatientindiabetes
management.[12]Selfmonitoringofbloodglucoseinpatientstakinginsulinfortype2
diabetesisanintegralpartofcare.[13]However,controversysurroundsselfmonitoringin
type2diabeteswhoarenottreatedwithinsulin,[14,15,16,17,18,19]andthe2015NICE
guidelinehasconcludedthatitshouldnotbeapartofroutinecare.[11]
Severalmetaanalyses,randomisedtrials,andreviewsshowlimitedevidencethatsupports
selfmonitoringasameansofimprovingglycaemiccontrol[16,17,18,19].NICEsreviewof
theevidenceledittoconcludethatalthoughastatisticallysignificantdifferencein
HbA1clevelswasseeninthosewhoundertookselfmonitoring,thiswasnotthoughttobe
clinicallymeaningful,androutineselfmonitoringwasthereforedeemednottobecost
effective.[11]
However,evidenceisavailablethatsupportsthevalueandutilityofselfmonitoringinnon
insulintreatedtype2diabetes,tohelpboththepatientandhealthcareprovidertounderstand
therealtimeeffectsoffoodandexerciseonbloodglucose,particularlyaspartofstructured
testingandeducation.[14,15,20,21]Inourexperiencethiscanenablethepatienttousethe
informationtoimplementamanagementplanforthemselves,includingattentiontodiet,
lifestyle,anddrugtherapy,inordertoimproveglycaemiccontrolwhileavoiding
hypoglycaemia.[14,21]
Inourview,apatientwithnewlydiagnosedtype2diabetesmaybenefitfromshorttermself
monitoringasapartofselfmanagementeducation.Thereareothersituationswhere
evaluationandpreventionoftherisksofhypoglycaemiaarecrucial,suchas:
Driving,particularlyofheavygoodsvehicles
Physicalactivityorsports
Alcoholconsumption
Shiftwork
Highriskoccupationsinvolvingmachinery
Insulinsecretagoguetherapy(sulfonylureas)
Acuteillness,fever,andsurgery
Pregnancy.
Therefore,selfmonitoringinnoninsulintreateddiabetesshouldbeconsidered.Inourview,
thedecisionshouldbebasedonarangeoffactors,includingpatientneeds,understanding
andmotivation,durationofdiabetes,andtypeoftreatment.[11,14]

The2015NICEguidelinedoesacceptthattherearecertainpatientgroupswhoarelikelyto
benefitfromselfmonitoring.Itliststhosegroupsas[11]:
Patientsoninsulin
Patientsonoralmedicationthatmayincreasetheirriskofhypoglycaemiawhiledrivingor
operatingmachinery
Patientsinwhomthereisevidenceofhypoglycaemicepisodes
Womenwhoarepregnantorplanningtobecomepregnant
NICEalsorecommendsconsideringshorttermselfmonitoringwhenstartingtreatmentwith
corticosteroids,ortoconfirmsuspectedhypoglycaemia.
TheguidelinestressesthatcliniciansshouldtaketheDriverandVehicleLicensingAgency
(DVLA)fitnesstodriveguidanceintoaccountwhenofferingselfmonitoring,asthisprovides
informationontreatmentsthatincreasesusceptibilitytohypoglycaemia,andthereforeneed
morefrequentmonitoringofplasmaglucose.Thetreatmentsmostcommonlyassociatedwith
hypoglycaemiaareinsulin,andinsulinsecretagoguessuchasgliclazideandglibenclamide.
AccordingtoNICE,cliniciansshouldalsoreviewtheongoingneedforselfmonitoring
annually.Thisshouldincludeareassessmentofthepatientsunderstandingandskillofusing
theselfmonitoringequipment,andofthecontinuedbenefittothepatient,andtheimpacton
theirqualityoflife.

6/2/2016

BMJLearningModule

3.Stepwisetreatment
Takeastepwiseapproachtobloodglucoselowering:intensificationoftreatment
andinsulininitiationareoftendelayedfortoolong.
Theconceptofclinicalinertia,thefailuretointensifytreatment,hasbeendescribedinthe
managementoftype2diabetes.[22,23]Thisinertiaisnotedateverystageofdisease
management,includingmovingfromsingledrugtherapytocombinationsoforal
medications,andinmakingadecisiontoinitiateinsulintherapy.Factorsrelatedto
clinicians,patients,andhealthcaresystemsareallbelievedtocontribute.[22]Clinicians
facetimeconstraints,complexandvariedguidelines,knowledgegaps,andconcerns
aroundincreasedriskofhypoglycaemia.Patientbarriersareprimarilytheavoidanceof
injectabletreatment,poorcompliance,andconcernsaboutweightgainand
hypoglycaemia.Treatmentcostsalsoplaysaroleindelaysininitiatinganduptitrating
therapy.[22]
Delayedinsulininitiationisaparticularconcern.Studieshaveshownthatthereisadelay
ofalmost12to18monthsininitiationofinsulin,leavingpatientswithpoorglycaemic
controlforasustainedperiod.[24,25,26]TheINSTIGATEstudyshowedthattheaverage
HbA1creading12monthspriortoinsulininitiationwas73mmol/mol(8.8%),andthree
monthspriortoinsulininitiationwas88mmol/mol(10.2%),muchhigherthanthe
recommendedguidelinethresholdsforinitiatinginsulin.[24]Inspiteofdiabetesrelated
complications,almost50%ofpatientsdidnotstartinsulintherapyforapproximatelyfive
yearsafterfailureofglycaemiccontrolwithmultipleoralagentshadbeenidentified.[26]
Evenafterinitiationofinsulin,approximately75%ofpatientshadnochangeintheir
insulinregimenoversixmonthsoffollowup.[25]Thishighlightstheneedforachangein
practice,andimprovementsinourroleinearlyinitiationandintensificationofinsulin
treatmentfortype2diabetes.
The2015NICEguidelinedescribesastepwisealgorithmforthetreatmentofblood
glucoseloweringtherapy,withthreestagesofintensificationfordrugtreatment.[11]
Inpatientswhocantakemetformin,thestepsareasfollows:
IfthepatientsHbA1crisesto48mmol/mol(6.5%)onlifestylemeasuresalonethen
youshouldoffermetformin,withatargetHbA1cof48mmol/mol(6.5%)
IftheirHbA1crisesto58mmol/mol(7.5%)whileonmetforminandlifestylemeasures,
thenfirstintensificationisrecommended,withdualoraltherapyandarevised
HbA1ctargetof53mmol/mol(7.0%)
IftheirHbA1crisesto58mmol/mol(7.5%)whileondualoraltherapy,thensecond
intensificationisrecommended,consistingoftripleoraltherapyorinsulinthe
HbA1ctargetis53mmol/mol(7.0%).
Thisapproach,togetherwiththemedicationoptionsrecommendedateachstageof
intensification,issetoutinfigure2below.
http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

1/5

6/2/2016

BMJLearningModule

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

2/5

6/2/2016

BMJLearningModule

Figure2.Treatmentalgorithmforpatientswithtype2diabeteswhocantake
metformin.
Inthoseforwhommetforminiscontraindicatedornottolerated,NICErecommendsthe
followingsteps:
IfthepatientsHbA1crisesto48mmol/mol(6.5%)onlifestylemeasuresalone,then
youshouldconsideroneof:aDDP4inhibitor,pioglitazone,orasulfonylurea.Ifthe
patienttakesaDDP4inhibitororpioglitazone,theirtargetHbA1cshouldbe48
mmol/mol(6.5%).Iftheytakeasulfonylurea,theirtargetHbA1cshouldbe53
mmol/mol(7.0%)
IftheirHbA1crisesto58mmol/mol(7.5%)whileononedrug,thenfirstintensification
isrecommended,withdualoraltherapyandarevisedHbA1ctargetof53mmol/mol
(7.0%).Fordrugcombinations,seefigure3
IftheirHbA1crisesto58mmol/mol(7.5%)whileondualoraltherapy,thensecond
intensificationisrecommended,consistingofinsulinbasedtreatmentwithan
HbA1ctargetof53mmol/mol(7.0%).
Thisissummarisedinfigure3below.

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

3/5

6/2/2016

BMJLearningModule

Figure3.Treatmentalgorithmforpatientswithtype2diabeteswhocannottake
metformin.
Whenstartinginsulintreatment,therecommendedinsulinishumaninsulin(NPH)over
analogueinsulin,asthereisamuchlowercostoftreatmentwithNPH,andnodifference
inHbA1cloweringisnotedbetweenthetwo.[11]
NICEalsorecommendsthatifanadultwithtype2diabetesissymptomatically
hyperglycaemicatanystageoftreatment,youshouldconsiderinsulinorasulfonylurea,
andreviewtheirtreatmentwhentheyhaveachievedcontroloftheirbloodglucose.[11]
Longactinginsulinanalogues(eginsulindetemirorinsulinglargine)canbeconsideredin
specialcircumstances,suchasifthereisaneedforlessfrequentdosingbecausea
healthcareassistantneedstovisitandadministerinsulin,orifsymptomatic
hypoglycaemiaisoccurring.NPHcanalsobechangedtoalongactinginsulinanalogueif
[11]:
http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

ThereisnoimprovementinHbA1cduetosignificanthypoglycaemia,OR

4/5

6/2/2016

BMJLearningModule

ThereissignificanthypoglycaemiairrespectiveofHbA1c,OR
ThepatientisunabletousetheNPHdeliverydevice.
Whenstartinginsulintherapy,NICErecommendsthatyoushouldcontinuetoofferthe
patientmetforminiftheycantoleratethedrug,andreviewthecontinuedneedforother
antihyperglycaemicdrugs.TherecommendedstartingregimenisNPHinsulin,injected
onceortwiceperdayaccordingtoneed.[11]
IftheHbA1cremainssuboptimal,especiallyiftheHbA1cis75mmol/mol(9.0%)or
higher,youcanconsiderescalationtoNPHplusashortactinginsulin.Thesecanbe
givenseparatelyorasapremixed(biphasic)preparation.[11]
NICErecommendsthatyouconsiderpremixedpreparationsthatincludeshortacting
insulinanalogues,ratherthanpremixedpreparationsthatincludeshortactinghuman
insulinpreparations,ifapersonprefersinjectinginsulinimmediatelybeforeameal,if
hypoglycaemiaisaproblem,orifbloodglucoselevelsrisemarkedlyaftermeals.[11]The
analogueshaveaquickeronsetofaction,whichavoidstheneedtotaketheinjection
severalminutesbeforethemeal,andalsotargetspostprandialhyperglycaemia.

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

5/5

6/2/2016

BMJLearningModule

4.Roleofpioglitazone
Despiterecentsafetyfearsrelatingtothiazolidinediones,NICEnowrecommends
pioglitazoneasaninitialandsecondlineoptioninselectedpatients.
ThethiazolidinedionerosiglitazonewaswithdrawnfromtheEuropeanmarketin2010due
toreportedincreasesinischaemicheartdisease,andthishasledtosafetyconcernsover
otherdrugsinthesameclass.[27,28]However,theUSFoodandDrugAdministration
(FDA)hasremovedsomeofitsprescribingrestrictionsforrosiglitazonefollowinga2013
reviewofthesafetydata,andareevaluationofdatafromtheRECORDtrial.
[29,30,31,32]ReexaminationofthelargeVADTstudyshowedthatrosiglitazonewas
associatedwithreducedriskoftheprimarycardiovascularcompositeoutcomeand
cardiovasculardeath.[33,34]
Pioglitazonecontinuestoplayanimportantroleinthemanagementoftype2diabetes,as
wellashavingacardiovascularbenefitinthesepatients.[35]ThePROactivestudy
suggestedthat48peopleneededtobetreatedtopreventonecardiovasculareventover
threeyears.[35]
Thiazolidinedionesregulategenesinvolvedinlipidandcarbohydratemetabolism,and
beneficialeffectsincludeimprovedinsulinsensitivityinmuscles,andreducedhepatic
glucoseproduction.[10]Thismeansthattheypotentiallyhaveabeneficialroleinfatty
liverdiseaseandsevereinsulinresistance.
Themajorandwellknownsideeffectsofthiazolidinedionesareweightgain,fluid
retention,precipitationofheartfailure,andbonefractures.[10,11]In2011,theMHRA
issuedanalertontheriskofcardiacfailureassociatedwithpioglitazoneincombination
withinsulin.[36]Worseningofmacularoedemahasalsobeenreportedinpatientstaking
pioglitazone,andifpatientsdevelopvisualdisturbancesonstartingtherapy,adetailed
eyeexaminationiswarranted.[37,38]Theriskofbladdercancerhasalsobeena
concern,andtheMHRAreleasedanalertonthematterin2011.[39]However,recent
reviewsandamultivariateanalysishavenotfoundadirectcausativeassociation.
[40,41,42]Whenprescribingthisclassofdrug,youshouldmaintainpharmacovigilance,
anddiscusstherisksandbenefitsofthedrugwithyourpatientsbeforestartingtreatment.
[10,11]
The2015NICEguidelinerecommendsachoicebetweenaDPP4inhibitor,pioglitazone,
orasulfonylureaasfirstlineglucoseloweringmedicationinpatientsinwhommetformin
iscontraindicatedornottolerated.[11]Pioglitazonemaybeusedincombinationwith
metformin,asulfonylurea,aDPP4inhibitor,and/orinsulinforfirstorsecondstage
intensificationofmedicationinpatientsnotreachingtheirHbA1ctarget,assetoutin
figure2above.[11]
However,pioglitazoneshouldnotbeinitiatedorcontinuedinadultswithtype2diabetesif
theyhaveanyofthefollowing[11]:
http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

Currentorpasthistoryofheartfailure

1/2

6/2/2016

BMJLearningModule

Hepaticimpairment
Diabeticketoacidosis
Currentorpasthistoryofbladdercancer
Uninvestigatedmacroscopichaematuria.
TheMHRArecommendsreviewingthesafetyandefficacyofpioglitazoneinpatientsafter
threetosixmonthsoftreatment,toensurethatonlypatientswhoarederivingbenefit
continuetobetreated.Youshouldstopthedruginpatientswhodonotrespond
adequatelytotreatment,egifthereisnoreductioninHbA1c.TheMHRAalso
recommendsthatelderlypatientsshouldstartonthelowestpossibledoseandbe
regularlymonitored,becauseoftherisksofbladdercancerandheartfailureassociated
withthedrug.[39]Ifpioglitazoneisusedincombinationwithinsulin,patientsshouldbe
monitoredforsignsofheartfailure,weightgain,andoedema.[36]

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

2/2

6/2/2016

BMJLearningModule

5.RoleofDPP4inhibitors
DespiterecentsafetyconcernsaboutDPP4inhibitors,NICEnowrecommends
themasinitialandsecondlineoptionsinselectedpatientswithtype2diabetes.
DPP4inhibitors(gliptins)workbyincreasingincretinlevels,whichinturnreduces
glucagonsecretionandincreasesinsulinrelease.The2015NICEguidelinerecommends
thatyouconsiderinitialdrugtreatmentwithaDPP4inhibitor,pioglitazone,ora
sulfonylureainpatientsinwhommetforminiscontraindicatedornottolerated.NICEalso
recommendsaDPP4inhibitorasanoptionforfirstintensificationoftreatment,for
patientswhoarenotreachingtheirHbA1ctargetonmetforminalone(seefigure2above).
[11]
AreductioninHbA1cof0.5%to1.0%isnotedwithDPP4inhibitors,withbenefitsoflow
ratesofhypoglycaemiaandnogaininweight.[10,11,43]ThesafetyconcernsofDPP4
inhibitorsaretherisksofpancreatitisandpancreaticcancer,aswellascardiovascular
safety.
Apopulationbasedcohortstudyfrom2014foundnoincreaseinthecrudeincidencerate
ofpancreatitiswithincretinbasedtherapies(GLP1receptoragonistsandDPP4
inhibitors)comparedtosulfonylureas.[44]TheEMAandFDAconfirmedin2014that
currentdatadonotestablishacausalrelationshipbetweenincretinbasedtherapyand
pancreatitisorpancreaticcancer.However,asmallincreaseintheriskofpancreatitis
cannotbecompletelyexcludeduntilmoredataareavailable,andallDPP4inhibitors
includeawarningabouttheriskofacutepancreatitisintheirsummaryofproduct
characteristics.[45]A2016international,multicentrecohortstudyfoundnoevidenceofan
increasedriskofpancreaticcancerassociatedwiththeuseofincretinbaseddrugs
comparedwithsulfonylureas.However,theauthorsnotedthatthispotentialadversedrug
reactionwillneedtobemonitoredinthelongtermduetothelatencyofthecancer.[45a]
ThecardiovascularsafetydataintheSAVORTIMIandEXAMINEstudiesshowedno
impactoncardiovasculareventsorallcausemortalitywithsaxagliptinandalogliptin,
respectively.[46,47]Saxagliptinwasassociatedwithanincreaseinhospitalisationfor
heartfailure.[46]InApril2015,theFDArecommendedaheartfailurewarninglabelfor
saxagliptin,butconfirmednoseriouscardiovascularoutcomesorallcausemortality
concernsforDPP4inhibitors.[46,47,48]InApril2016,theFDAaddedwarningsabout
theriskofheartfailuretothelabelsforsaxagliptinandalogliptin,particularlyforpatients
withheartorkidneydisease.[48a]TheTECOStrialresultspresentedattheAmerican
DiabetesAssociationmeetinginJune2015suggestedthatsitagliptinwasnoninferiorto
placeboforthecompositecardiovascularendpoint(cardiovasculardeath,nonfatal
myocardialinfarction,nonfatalstroke,orhospitalizationforunstableangina),andshowed
noincreaseinheartfailure.[49]
A2016systematicreviewandmetaanalysisofrandomisedandobservationalstudies
examinedtheuseofDPP4inhibitorsandriskofheartfailureintype2diabetes.It
concludedthattherelativeeffectofDPP4inhibitorsontheriskofheartfailureinpatients1/2
http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

6/2/2016

BMJLearningModule

withtype2diabetesisuncertain,giventherelativelyshortfollowupandlowqualityof
evidence.Bothrandomisedcontrolledtrialsandobservationalstudies,however,suggest
thatthesedrugsmayincreasetheriskofhospitaladmissionforheartfailureinthose
patientswithexistingcardiovasculardiseases,ormultipleriskfactorsforvascular
diseases,comparedwithnouse.[49a]
TherearenoprescribingrestrictionsforDPP4inhibitors,butinourviewcliniciansshould
avoidtheiruseinpatientswithahistoryofheartfailure,pancreatitis,orpancreaticcancer.
Rhinitis,nasopharyngitis,hypersensitivityreaction,andangioedemahavealsobeen
reported.Clinicalvigilanceisrequiredforsignsoftheaboveinallpatients.

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

2/2

6/2/2016

BMJLearningModule

6.RoleofSGLT2inhibitors
SGLT2inhibitorsareeffectiveinimprovingglycaemiccontrolandweight
reduction,andlatestevidencesuggestsabenefitintermsofcardiovascular
outcomes,butsafetyconcernsremaintobeaddressed.
SGLT2inhibitorssuchascanagliflozin,dapagliflozin,andempagliflozinareeffectiveat
improvingglycaemiccontrolintype2diabetes.[10,50,51]Theyreversiblyinhibitthe
sodiumglucosecotransporter2(SGLT2)intherenalproximalconvolutedtubule.This
reducesglucosereabsorptionandincreasesurinaryglucoseexcretion.Astheireffects
areindependentofinsulintheycanbeusedatanystageoftype2diabetes,andhavea
lowriskofhypoglycaemia.[10]SGLT2inhibitorsreduceHbA1cby0.5%to1.0%when
usedasmonotherapy,orincombinationwithotherglucoseloweringdrugs.
[51,52,53,54]Additionalbenefitsincludeweightandbloodpressurereduction.
[10,50,55]
EmergingevidencesuggestsapossibleroleforSGLT2inhibitorsinreducing
cardiovascularriskintype2diabetes.Thecardiovascularsafetytrialinempagliflozin
(EMPAREGOUTCOME)waspresentedattheEASDconferenceinSeptember2015
andpublishedatthesametime,andshowedthebeneficialeffectsintermsofalowerrate
oftheprimarycompositecardiovascularoutcome,andofdeathfromanycause,in
comparisontoplacebowhenthestudydrug,empagliflozin,wasaddedtostandardcare.
[60]Thereasonsforthisareunclear,althoughthespeedoftheeffectsuggestsittobe
haemodynamic(reducedcirculatingvolume)inorigin,asopposedtoaneffectonthe
developmentofatheromaandvasculardisease.FurtherresultsfromEMPAREGshowed
thatinpatientswithtype2diabetesandhighcardiovascularrisk,empagliflozinreduced
hospitalisationduetoheartfailure,andcardiovasculardeath,withconsistentresultsin
patientswithandwithoutbaselineheartfailure.[55a]
Asystematicreviewof57trialsandsixregulatorysubmissionspublishedinMarch2016
concludedthatSGLT2inhibitorsappeartohaveanetprotectiveeffectagainst
cardiovasculareventsanddeathsaswellasallcausemortality.Theauthorssaidtheir
findingssuggesteda"strongrationaletoexpectbenefit"fromusingthisclassofdrugsin
patientswithtype2diabeteswhoareathighriskofcardiovascularevents.However,they
alsohighlightedthatempagliflozinistheonlySGLT2inhibitorforwhichlongterm
cardiovascularoutcomesdataisavailable,andthatresultsfromongoingstudieswillbe
requiredtosubstantiatethebenefitsacrossthewholeclassofSGLT2inhibitors.[55b]
Itwillbeinterestingtoseeifthesefindingsarereplicatedinfuturetrialsofotherdrugsin
thisclass.Thecardiovasculartrialindapagliflozin(DECLARETIMI58)isexpectedto
reportin2019,andcanagliflozin(CANVAS)in2017.
CommonsideeffectsofSGLT2inhibitorsincludeanincreasedrateofgenitalfungal
infection(egforcanagliflozinversusplacebo:11%ofwomenand4%ofmen,versus3%
and1%intheplacebogroups),andasmallincreaseinurinarytractinfections.
[10,50,56,57]ThediureticeffectofSGLT2inhibitorscanprecipitatevolumedepletion, 1/2
http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

6/2/2016

BMJLearningModule

andyoushouldusethesedrugswithcautioninelderlypatients.[10]Anincreasein
urinarycalciumandincreasedfracturesareseenwithcanagliflozin,andthesearebeing
monitoredbytheFDA.[10]Asmall,reversibleincreaseincreatinineandincreaseinlow
densitylipoproteincholesterol(5%)isalsonoted.[10]
In2015,boththeFDAandtheMHRAissuedawarningoftheriskofpotentiallylife
threateningdiabeticketoacidosis(DKA)withSGLT2inhibitorsintype2diabetes.[58,59]
SGLT2inhibitorsshouldthereforebeusedwithcautionininsulintreatedtype2diabetes.
TheMHRAhighlightstheimportanceofinformingpatientsofthesymptomsandsignsof
ketoacidosis,andadvisingthemtogetimmediatemedicalhelpiftheseoccur.TheMHRA
alsoadvisesdoctorstotestforraisedketonesinpatientswithacidosissymptoms,evenif
plasmaglucoselevelsarenearnormal,asomittingthistestcoulddelaythediagnosisof
DKA.[59]
The2015NICEguidelinementionsSGLT2inhibitorsasapossibleoptionforcombination
treatmentoftype2diabetes,althoughthedetailsofpatienteligibilityforthesedrugsare
dealtwithinseparateNICEtechnologyappraisals.[11,61,62,63]Anyoneof
dapagliflozin,canagliflozinorempagliflozinisrecommendedasanoptionfordualtherapy
incombinationwithmetformin,onlyifasulfonylureaiscontraindicatedornottolerated,or
thepatientisatsignificantriskofhypoglycaemia.Allthreedrugsarealsorecommended
asoptionsincombinationwithinsulin.[11,61,62,63]Canagliflozinandempagliflozincan
becombinedwithmetforminandasulfonylurea,ormetforminandathiazolidinedione,as
tripletherapy.[62,63]Dapagliflozinisnot,however,recommendedforcombinedusewith
metforminandsulfonylureasintheformoftripletherapy,althoughthisiscurrentlyunder
reviewbyNICE.

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

2/2

6/2/2016

BMJLearningModule

7.RoleofGLP1receptoragonists
GLP1receptoragonistshavearoleinasmallgroupofpatientsforwhomtriple
therapyisineffective,orasdualtherapyincombinationwithbasalinsulin(for
whichspecialistinputisrequired).
TreatmentwithacombinationofinsulinandaGLP1receptoragonisthasshown
promisingresults,and2015EuropeanandUSguidelinesfromtheAACE/ACEandthe
ADA/EASDbothencouragetheuseofabasalinsulin/GLP1receptoragonist
combinationinpatientswithpoorlycontrolledtype2diabetes,especiallyiftheirHbA1cis
above75mmol/molto86mmol/mol(9%to10%).[10,64,65]
ThecombinationofbasalinsulinandGLP1receptoragonistformsaneffectivetreatment
forpeoplewithtype2diabetes.[66,67]Itcanleadtogoodimprovementsinglycaemic
control,withlessweightgainandlesshypoglycaemiacomparedwithinsulinalone,and
fewergastrointestinalsideeffectscomparedtoGLP1receptoragonisttherapyalone.
[66,67,68]
Inapooledanalysisof15trials(n=4348participants),thecombinationofbasalinsulin
andGLP1receptoragonistshowedagreatermeanreductionofHbA1c(0.44%(95%CI,
0.6%to0.29%))thananyotherstrategy.[67]Therewasalsonoincreaseinrelativerisk
ofhypoglycaemia,andameanreductioninweightof3.22kg(4.9kgto1.54kg).[67]
StudieshaveshownthatGLP1receptoragonistscanbeaddedtobasalinsulintherapy,
orbasalinsulincanbeaddedtoGLP1receptoragonisttherapy,withtheorderof
treatmentbasedontheindividualpatient.[66,67,68]
DifferentGLP1receptoragonistshavedifferentpharmacokineticandpharmacodynamic
propertiesthisresultsinarangeofefficacies,dosefrequencies,andsideeffectsforthe
differentdrugsintheclass,withsomebettersuitedtoactasanalternativetoprandial
insulin,andothersleadingtoanoverallimprovementinglycaemiawithadominanteffect
onfastingglucose.[69,70,71]
The2015NICEguidelineisclearthatspecialistadviceisrequiredbeforeofferingaGLP
1receptoragonistincombinationwithinsulin,andthereshouldbeongoingsupportfrom
aconsultantledmultidisciplinaryteam.
LatestevidencealsosuggestsapossibleroleforGLP1receptoragonistsinweight
managementinpatientswithtype2diabetes.Asystematicreviewthatlookedat25trials
withGLP1receptoragonistsshowedthatpeoplewithdiabetesachievedreductionsin
bodyweight(2.8kg).[72]
The2015NICEguidelinegivesalimitedroletoGLP1receptoragonistsinfurther
intensificationoftreatmentinthesmallgroupofpatientsforwhomstandardtripletherapy
hasprovedineffectiveinreachingtheirHbA1ctarget,orforthoseinwhomstandardtriple
therapyiseithercontraindicatedorpoorlytolerated.[11]Insuchpatients,aGLP1
receptoragonistcanbeconsideredincombinationwithmetforminandasulfonylurea,in
http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi 1/2
patientswho:

6/2/2016

BMJLearningModule

EitherhaveaBMIof35kg/m2orhigher(alowerBMIthresholdisrecommendedfor
peopleofblack,Asian,orotherminorityethnicbackground),and
Havespecificpsychologicalorothermedicalproblemsassociatedwithobesity,OR
HaveaBMIlowerthan35kg/m2,and
Forwhominsulintherapywouldhavesignificantoccupationalimplications,or
Forwhomweightlosswouldbenefitothersignificantobesityrelatedcomorbidities.

YoushouldcontinuetreatmentwithaGLP1receptoragonistonlyifabeneficial
metabolicresponseisnoted,intheformofareductionofatleast11mmol/mol(1.0%)in
HbA1c,andaweightlossofatleast3%ofinitialbodyweightinsixmonths.[11]

http://learning.bmj.com/learning/modules/elucidat/572b542da215e.html?moduleId=10056517&status=LIVE&action=start&_flowId=ELU&sessionTi

2/2

Managing Type 2 diabetes in adults

If patient is hyperglycaemic,
consider:

Adult with type 2 diabetes

Lifestyle interventions

SU

MET

MET

Consider
modied
release
if not
tolerated

Oer
standard
release
initially

DPP

or

MET

or

MET

SU

or

MET

MET

Metformin contraindicated
If HbA1c rises to
48 mmol/mol

48

Consider
monotherapy

DPP

If HbA1c rises to
58 mmol/mol

DPP

PIO

53

or

SGLT

SU
or

MET

MET

SGLT

PIO

53

SU

53

SGLT

MET

Adjust BMI for black

and asian people

If any of:

or

Psychological/medical issues
associated with obesity
therapy would have
+ Insulin
signicant occupational
implications
or Weight loss would benet other
obesity-related comorbidities

MET = Metformin

= Sulfonylurea

53

Review need for other

HbA1c lowering drugs

Consider insulin
programme

Consider:

Person needs help

injecting
Lifestyle restricted
by hypoglycaemic
episodes

NPH insulin
No

MET

Yes

2 Inhibitors

DPP

Read the full

article online

Dipeptidyl
= peptidase-4
Inhibitor

SU

PIO

48

Short acting
insulin

Insulin detemir
or

Glucagon-like
peptide-1 (GLP-1)
mimetic

Sodium-glucose

SGLT = cotransporter

(once/twice daily)

Consider:

BMI under 35

SU

Insulin Programme

Would otherwise need

twice-daily NPH insulin

BMI over 35

SU

DPP

or

PIO

or

Continue

If triple
therapy fails:

Key

53

PIO

Consider
triple therapy

or

SU

SU

If HbA1c rises to
58 mmol/mol

PIO

or

or

Consider insulin programme

MET

DPP

PIO

or

48

Consider
dual therapy

or

SU

review treatment when blood

glucose has been controlled.

Dietary control

Metformin tolerated

Short acting
insulin

or

Person prefers
injecting before meals
Blood glucose rises
markedly before meals
Hypoglycaemia
is a problem

= Pioglitazone

= Ideal HbA1c target

http://bmj.co/diaT2A

Yes

Glargine

Short acting
insulin analogues

Monitor people on
insulin for the need to
change the regimen

An individualised target may be

needed, depending on persons:
Preferences
Comorbidities
Risks from polypharmacy
Risks from tight blood glucose control
Ability to achieve long term benets

Designed by: Will Stahl-Timmins

2016 BMJ Publishing group Ltd.