Contributions of Basal and PostprandialHyperglycemia Over a Wide
Range ofA1C Levels Before and After TreatmentIntensication in Type
2DiabetesMATTHEW RIDDLE, MD1GUILLERMO UMPIERREZ, MD2ANDRES DIGENIO,
MD, PHD3RONG ZHOU, MD4JULIO ROSENSTOCK, MD5OBJECTIVEdTo determine
the relative contributions of basal hyperglycemia (BHG)
versuspostprandial hyperglycemia (PPHG) before and after treatment
intensication in patients withglycated hemoglobin A1c (A1C) .7.0%
while on prior oral therapy.RESEARCH DESIGN
ANDMETHODSdSelf-measured, plasma-referencedglucoseproles and
A1Cvalues were evaluated fromparticipants in six studies comparing
systematicallytitrated insulin glargine with an alternative regimen
(adding basal, premixed, or prandial insulin,or increasing oral
agents). Hyperglycemic exposure (.100 mg/dL [5.6 mmol/L]) as a
result ofBHG versus PPHG was calculated.RESULTSdOn prior oral
therapy, 1,699 participants (mean age 59 years, diabetes duration9
years) had mean fasting plasma glucose (FPG) of 194 mg/dL (10.8
mmol/L), and mean A1Cwas8.7%.
BHGcontributedanaverageof7680%tohyperglycemiaovertheobservedrangeofbaseline
A1C levels. Adding basal insulin for 24 or 28 weeks lowered mean
FPG to 117 mg/dL(6.5 mmol/L), A1C to 7.0%, and BHG contribution to
3241%. Alternative regimens reducedFPG to 146 mg/dL (8.1 mmol/L),
A1C to 7.1%, and the contribution of BHG to 6471%. BHGcontributions
for patients with A1C averaging 7.67.7% were 76% at baseline and 34
and 68%after adding basal insulin or other therapies,
respectively.CONCLUSIONSdWhenA1Cis.7.0%despiteoral therapy,
BHGroutinelydominatesexposure. Intensied therapy reduces A1C and
changes this relationship, but BHG
amenabletofurtherinterventionstill accountsforone-thirdoftotal
hyperglycemiaafterbasal insulintreatment and two-thirds after
alternative methods.Diabetes Care 34:25082514, 2011The importance
of controlling hyper-glycemia as measuredby
glycatedhemoglobinA1c(A1C) levels toreducetheriskof long-term,
diabetes-relatedcomplicationsiswell understood. ThatA1Creects
contributions frombothbasal (fasting) and postprandial
hypergly-cemia(PPHG)isalsowell established.Howtreatment of basal
hyperglycemia(BHG) versus PPHGshould be prioritizedin
themanagementoftype2 diabetesisless clear.In2003, Monnier et al.
(1) pub-lishedalandmarkstudydescribingtherelative contributions of
BHG and PPHGto overall hyperglycemic exposure at dif-ferent
levelsof A1C. Thisanalysiswasbased on 1-day, four-point
daytimeglucoseprolesfrom290patientswithtype2diabeteswhoweretreatedwithdiet
therapy with or without oral antihy-perglycemicdrug(OAD)
therapyandwithoutinsulin. Thendingssuggestedthat PPHGaccounted for
;70%of overallglycemic exposure above normal levels inpatients
inthe lowest range of A1C(,7.3%), withthecontributionfromBHG
increasing with higher A1C. In thehighest A1Crange(A1C.10.2%),
thecontributions were reversed; PPHG con-tributed;30%andBHG;70%.
Thispattern has been proposed to reect a fun-damental
biologicpropertyof type2diabetes:atendencyofPPHGtoappearearly in
the natural history of the disorder,with BHGappearing later after
further de-clineof b-cell capacity(2). Apotentialclinical
implicationof
thisviewisthatwhenA1CisonlymoderatelyelevatedtherapeuticinterventionshouldtargetPPHG.
Other studies support the
impor-tanceofPPHGatlowerA1ClevelsandBHGathigherlevels(28),butfeaturesof
their designs may inuence estimates oftherelativecontributionsof
BHGandPPHG. Specically, the timing of glucosemeasurements, the
level of basal glucoseconsidered above normal, inclusion or
ex-clusion of individuals not receiving antihy-perglycemictherapy,
andtheeffectsoftreatments that affect mainly BHG or PPHGall might
inuence these relationships.The purpose of this study was to
mea-sure the relative contribution of
BHGversusPPHGinacommonandspecicclinical setting. We evaluated
seven-pointglucose proles in a large group of patientswith type 2
diabetes with A1C .7.0% onOAD, for whomthe relative importance
ofBHG versus PPHG has direct relevance tofurther therapy. Because
100 mg/dL (5.6mmol/L) is considered the upper limit ofnormal
fastingplasmaglucosebytheAmericanDiabetesAssociation(9,10),thecurrentanalysisdenedbasal
valuesabove this level as elevated. We aimed
todetermine,usingourmethodofcalcula-tion, whether before
intensifying therapythese patients showed the
A1C-dependentcccccccccccccccccccccccccccccccccccccccccccccccc
cFromthe1Department of Medicine, Division of Endocrinology,
Diabetes &Clinical Nutrition, Oregon
HealthandScienceUniversity,Portland,Oregon;the2DepartmentofMedicine,DivisionofEndocrinology&Metabolism,
Emory University, Atlanta, Georgia;3sano-aventis U.S., Bridgewater,
NewJersey;4Medpace,Cincinnati, Ohio; and the5Dallas Diabetes and
Endocrine Center, Dallas, Texas.Corresponding author: Matthew
Riddle, [email protected] 1 April 2011 and accepted 5
September 2011.DOI: 10.2337/dc11-0632This article contains
Supplementary Data online at
http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dc11-0632/-/DC1.The
authors are fully responsible for all content and editorial
decisions and received nonancial support orother form of
compensation related to the development of this article. 2011 by
the American Diabetes Association. Readers may use this article as
long as the work is properlycited, the use is educational and not
for prot, and the work is not altered. See
http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.2508
DIABETES CARE,VOLUME 34, DECEMBER 2011 care.diabetesjournals.orgC l
i n i c a l C a r e / E d u c a t i o n / N u t r i t i o n / P s y
c h o s o c i a l R e s e a r c hO R I G I N A L A R T I C L
Epatternsofhyperglycemicexposurepro-posed by the earlier ndings and
to denehow adding basal insulin and other inter-ventions altered
the initially observed pat-terns of hyperglycemia.RESEARCH DESIGN
ANDMETHODSStudy and patientselectioncriteriaData from six similarly
designed random-izedtrials enrollingadult patients withtype 2
diabetes with suboptimal glycemiccontrol onoral
antihyperglycemicther-apywerepooledforanalysis(1116).Thesestudieswereselectedfromatotalof
63 supported by the sano-aventis
de-velopmentprogramforinsulinglarginebyhaving the following
characteristics: pro-spective, randomized,
active-comparatordesign;insulinglarginegivenoncedailywithout use of
prandial insulin; requireduse ofa systematic
insulin-titrationalgo-rithm; adherence to good clinical
practiceguidelines; availabilityof participant-level data;
follow-up of at least 24 weeks;and collection of at least
seven-point self-measured glucose proles.
Insulin-dosingdecisionsweremadefrequently(daily,every3days,
orweekly)seekingfastingglucose levels ,100 mg/dL (5.5 mmol/L).Entry
criteria for all studies specied thatA1Cbe.7.0%at enrollment.
Ineachstudy, theactivecomparatortreatment(human NPH insulin,
insulin lispro, pre-mixedinsulin,
orOADintensication)alsousedaspecicalgorithm.Detailsofthe studies
can be found in SupplementaryTable 1 and in the original
publications.Baseline and 24-week participantlevel data were pooled
according totreatment. One study (15) did nothave a week24visit;
week28datawere used. To be included in the presentanalysis,
participants must have nished24 to 28 weeks of treatment with
com-plete seven-point glucose prole data
atbothbaselineandweek24or28.
Allglucoseprolesincludedmeasurementbeforeand2haftereachmeal
andatbedtime.Statistical analysisCalculation of the relative
contribu-tions of glucose. Agraphical depiction ofthe area under
the curve (AUC) for nor-mal, basal, and postprandial glycemic
ex-posure is shown in Supplementary Fig. 1.The daily blood glucose
(BG) response tomeals was estimated by calculating the
in-crementalAUCofdaytimeBGfromtheoverall glucose prole. The total
durationfor the seven-point prole was 24 h, withthe glucose level
at the 24th h imputed bythe value at fasting before breakfast.
Fourareas were calculated geometrically fromthe seven-point curve
as follows: 1) nor-mal
glycemicexposure(AUCN)d100mg/dL324h=2,400mg/dLperhofexposure;2) BHG
(AUCB)dthe area be-tween100mg/dLandalineprojectedrightward for 24 h
from the fasting (beforebreakfast) glucose value in the prole
(thearea is taken to represent the daily abnor-mal
glycemicexposureresultingfromBHG); 3) PPHG (AUCP)dthe area abovethe
line projected rightward from the fast-ing sample before breakfast
and below theline connecting the six remaining
points,minusanyareabelowthelineprojectedfromthebasal value, if
applicable(thisarea is considered a reection of the post-prandial
glycemic responses to breakfast,lunch, anddinner); and4)total
glucose(AUCG)dthe total area under the glucosecurve is the sum of
the other three areas[AUCG=AUCN+AUCB+AUCP].
Asaresult,therelativecontributionsofpost-prandial
andfastingBGtothetotal BGincrementwerecalculated, respectively,by
using the following equations:[AUCP/(AUCB + AUCP)] 3 100% for
thepostprandial
contributionand[AUCB/(AUCB+AUCP)]3100%forthebasalcontribution.
Negativevaluesweresetto zero.Outcomes. The outcomes of interest
werecorrelations betweenA1CandAUCG,AUCB, andAUCP, andtheoverall
basalcontribution(percentage)tohyperglyce-mia at baseline and end
point; seven-pointglucose proles at baseline and after 24
or28weeksoftreatmentanalyzedbyA1Ccategory;
andtherelativecontributionsof postprandial and fasting glucose to
totalhyperglycemia before and after treatmentbyA1Ccategory,
andbydifferenttreat-ment groups (basal insulin vs. other,
andinsulinglarginevs. NPHinsulin). TheA1C categories were ,8.0, 8.0
to,8.5, 8.5to,9.0, 9.0to,9.5, and$9.5%. Hypoglycemic events were
alsoassessed.Statistical methodologyPearson correlation analysis
was per-formed between outcomes and predictors.Table 1dPatient
demographics and baseline characteristicsCharacteristic Overall
Insulin glargine NPH insulinBasal insulin(total glargine and
NPH)Other (lispro,premix, or OAD)n 1,699 1,026 235 1,261 438Age
(years) 59.4 (9.4) 59.9 (9.5) 56.8 (8.7) 59.3 (9.4) 59.5 (9.3)Male
(%) 57.7 58.0 53.6 57.2 59.4White (%) 94.6 95.5 85.4 93.8
96.6Weight (kg) 87.6 (15.9) 86.7 (15.9) 94.2 (16.1) 88.1 (16.2)
86.4 (15.1)Duration of diabetes (years) 9.0 (6.0) 9.1 (6.3) 9.0
(5.2) 9.1 (6.1) 8.8 (5.9)Baseline fasting plasmaglucose (mg/dL)
193.5 (47.6) 194.4 (47.1) 200.8 (47.3) 195.6 (47.2) 187.6
(48.3)Baseline A1C (%) 8.69 (0.94) 8.73 (0.95) 8.62 (0.92) 8.71
(0.95) 8.62 (0.93)Baseline A1C category, n (%),8.0 422 (24.8) 236
(23.0) 63 (26.8) 299 (23.7) 123 (28.1)8.0 to ,8.5 348 (20.5) 220
(21.4) 44 (18.7) 264 (20.9) 84 (19.2)8.5 to ,9.0 298 (17.5) 173
(16.9) 47 (20.0) 220 (17.4) 78 (17.8)9.0 to ,9.5 245 (14.4) 147
(14.3) 38 (16.2) 185 (14.7) 60 (13.7)$9.5 386 (22.7) 250 (24.4) 43
(18.3) 293 (23.2) 93 (21.2)Values are mean (SD) unless otherwise
noted.care.diabetesjournals.org DIABETES CARE,VOLUME 34, DECEMBER
2011 2509Riddle and AssociatesA regression analysis adjusted to
study orother exploratory factors was performed totest the
robustness of the correlationanalysis. In addition, A1C
correlations atweek 24 or 28 in patients on basal
insulin(insulinglargine or NPHinsulin) werecompared with those on
another therapy(insulin lispro, premixed insulin, or
OADintensication) andinthesubgroupsofpatients taking insulin
glargine versusthose taking NPH insulin.Week24or
28outcomescompari-sons(basalinsulin vs.other,insulinglar-gine vs.
NPHinsulin, or other
comparisons)wereperformedwithtwosamplettestswithSatterthwaiteunequal
varianceap-proximation. To consider the A1C effect,the comparisons
were also performed us-ing ANOVA modelswith week 24 or 28A1C
category, study, and comparison cat-egory as factors. The
comparisons were alsorepeatedusinganANCOVAmodel withthe change in
A1C as a covariate and studyand comparison category as
factors.Hypoglycemia incidence (n/N for eachgroup) was calculated
for symptomatic hy-poglycemia (all reported events),
glucose-conrmedhypoglycemia(symptomaticevents with reported glucose
values ,50mg/dL[2.8mmol/L]), andseveresymp-tomatic hypoglycemic
events (any symp-tomatichypoglycemicevent
requiringassistanceandaBGof ,36mg/dL[2.0mmol/L], if available, or
withpromptrecovery after oral carbohydrate, intrave-nousglucose,
orglucagonadministra-tion). Odds ratios andPvalues
werecalculatedusinglogisticregressionwithstudy and comparison
category as factors.RESULTSDemographics and
baselinecharacteristicsThe demographic and clinical
character-istics of the 1,699 study participants areshown in Table
1. Oral therapies used
atbaselineweremostlylimitedtometfor-minalone(5%),
asulfonylurea(45%),orthetwotogether(46%). Fewerthan5% of patients
were taking another mon-otherapy or combination regimen or
diettherapyonly. Of
the1,261patientsas-signedtobasalinsulin,1,026(81%)re-ceivedinsulinglargineand235(19%)receivedhumanNPHinsulin.
Of the438assignedtoother regimens, 32%were
treatedwithpremixedinsulin,36%withprandial insulin,
and32%withadditional oral therapy. Baselinecharacteristicsof
thesesubgroupsweresimilar.Glycemic patterns andhyperglycemic
exposureby A1C ranges at baselineThe overall mean 6SD BG
concentrationfrom baseline glucose proles was 189 649 mg/dL (10.5
62.7 mmol/L), and totalglycemicexposure(AUCG) correlatedvery
signicantly withbaseline A1C(r2= 0.545; P ,0.0001). The mean
rela-tive basal contribution to total hypergly-cemicexposureat
baselinewas
78%.Meanseven-pointglucoseconcentrationtimeprolesatbaselineforeachof
theA1Cranges are shown in Fig. 1A. The over-night values imputed by
a line between thebedtimevalueand anext-morning
valueprojectedtobethesameasthefastingvalueof thatdayarenot
showninthegure. Figure1Bshowsthebasal andpostprandial contributions
computed foreachA1Crange. TheBHGcontributionFigure 1dA: Baseline
seven-point glucose proles by A1C category. B: Relative
contributions of BHG and PPHG to overall hyperglycemia by
A1Ccategory at baseline. C: The seven-point glucose proles by A1C
category at week 24 or 28. D: Relative contributions of BHG and
PPHG to overallhyperglycemia by A1C category at week 24 or 28.2510
DIABETES CARE,VOLUME 34, DECEMBER 2011
care.diabetesjournals.orgBasal versus postprandial
hyperglycemiaranged from 76 to 80% of hyperglycemicexposureandthat
fromPPHGrangedfrom 24 to 20% from the lowest (,8.0%,meanA1C7.6%)
tothehighest A1C($9.5%, mean A1C10.0%) ranges.A tendency toward a
greater contributionfromPPHGat lowerandfromBHGathigher ranges of
baseline A1C approachedbutdidnotreachstatistical signicance(P =
0.074).Effects of intensication oftreatment on A1C and
fastingplasma glucoseAfter 24 or 28 weeks of intensied treat-ment,
mean 6 SD basal glucose from theprolesfor
thewholepopulationwas121635mg/dL(6.761.9mmol/L),andmeanweek24A1Cwas
7.0460.91%. Meanseven-point glucosecon-centration time proles at
week 24 or 28by end point A1C category are shown inFig. 1C. After
intensication of treatmenttotal glycemic exposure correlated
signif-icantly with achieved A1C (r2= 0.471; P, 0.0001). The mean
relative basal con-tribution to total hyperglycemic
exposureaftertreatmentintensicationwas43%.Figure 1Dshows the BHGand
PPHGcon-tributions for each A1C category for week24 or 28 for all
participants. Intensica-tion of therapy reduced the
contributionfrom BHG from 76 to 80% at baseline to4148%.
Astatisticallysignicant in-crease of the contribution from BHG
rel-ative to PPHGfromlower to higher rangesof A1C was evident (P =
0.0155).Effects of basal insulin versus otherforms of intensied
treatmentThe1,261participantswhosetreatmentwas
intensiedwithinsulinglargineorNPHinsulin and the 438 who
usedpremixed insulin, prandial insulin, or addi-tional oral therapy
achieved similar meanlevels of A1C (7.02 vs. 7.09%) (Table 2).Mean
total glucose levels were also similarforthetwomeansof
intensicationoftreatment (3,596mg z h z dL21[199.6mmol z h z L21]
vs. 3,437mmol z h z L21[190.8mmol z h z L21]). However,
themeanbasalglucosefromtheprolesforeachgroupdiffered(115mg/dL[6.4mmol/L]
vs. 137 mg/dL [7.6 mmol/L] forbasal insulinvs. other treatment;
P,0.0001), andtheseven-point prolesshowed different patterns (Fig.
2A). Treat-ment with basal insulin reduced the
meanrelativecontributionfromBHGto34%.Theothertreatment
methodsresultedinameanrelativeBHGcontributionof68%.
ThecontributionsfromBHGandPPHGafter treatment are shown by rangesof
achievedA1Cfor thebasal insulingroup in Fig. 2B and the group
treated inotherwaysinFig. 2C. Afteradditionofbasal insulin, the
BHGcontribution rangedbetween 31 and 41%, with a greater
con-tribution from BHG at the higher ranges ofachievedA1C.
Inasubanalysis, partici-pants treatedwithinsulinglargineorNPH
insulin were compared directly; pat-terns of hyperglycemia were
generally alikewith the two insulins (Table 2).Intensicationof
therapy withpre-mixedor prandial insulinor additionaloral agents
resulted in a range of contribu-tion fromBHGfrom63.5 to 71%of
overallhyperglycemic exposure. Despite the sim-ilarityof
achievedA1Clevelsaftertreat-ment intensication with basal
insulinTable 2dBaseline, week 24 or 28, and change in A1C and
glucose variables and correlations with change in A1C by treatment
categoryCharacteristic and visit Overall Basal insulin Other
treatmentSubanalysisInsulin glargine NPH insulinn 1,699 1,261 438
224 235A1C (%)Baseline 8.69 (0.94) 8.71 (0.95) 8.62 (0.93) 8.66
(0.91) 8.62 (0.92)Week 24/28 7.04 (0.91) 7.02 (0.88) 7.09 (0.98)
7.00 (0.85) 6.90 (0.70)Change 21.65 (1.03) 21.69 (1.00) 21.53
(1.12) 21.66 (1.00) 21.73 (0.89)AUCB (mg z h z dL21)Baseline 2,155
(1,176) 2,271 (1,183) 1,821 (1,091) 2,598 (1,220) 2,449 (1,256)Week
24/28 590 (739) 477 (654) 917 (864) 596 (822) 587 (734)Change
21,565 (1,243) 21,794 (1,204) 2904 (1,112) 22,003 (1,250) 21,863
(1,341)Correlation 0.425* 0.449* 0.371* 0.480* 0.460*AUCP (mg z h z
dL21)Baseline 560 (592) 573 (611) 521 (533) 603 (681) 628 (681)Week
24/28 635 (601) 748 (610) 311 (434) 849 (711) 865 (672)Change 75
(729) 174 (730) 2211 (646) 246 (878) 237 (851)Correlation 0.155*
0.132* 0.304* 0.179 0.023Total glucose AUCG (mg z h z dL21)Baseline
5,003 (1,233) 5,126 (1,241) 4,649 (1,139) 5,425 (1,162) 5,308
(1,239)Week 24/28 3,555 (786.4) 3,596 (779) 3,437 (798) 3,795 (905)
3,814 (775)Change 21,448 (1,213) 21,530 (1,204) 21,212 (1,208)
21,631 (1,269) 21,494 (1,215)Correlation 0.543* 0.531* 0.566*
0.589* 0.547*Basal contribution (%)(AUCB [AUCB + AUCP]) 3
100Baseline 78.2 (22.1) 79.1 (21.3) 75.6 (24.2) 80.3 (21.2) 78.2
(23.0)Week 24/28 42.6 (39.3) 33.7 (36.0) 67.9 (37.2) 35.9 (37.5)
35.3 (37.1)Change 235.7 (41.9) 245.4 (38.3) 27.7 (39.2) 244.3
(39.3) 243.0 (40.6)Correlation 0.121* 0.150* 20.014 0.126
0.216Values are mean (SD) unless otherwise noted. *P ,0.0001. P
,0.001 between change in glucose prole parameter and change in A1C
from baseline to end point.P . 0.05.care.diabetesjournals.org
DIABETES CARE,VOLUME 34, DECEMBER 2011 2511Riddle and
Associatescompared with the other forms of therapy,theincidencesof
symptomatichypogly-cemiaandglucoseconrmedsymptom-atichypoglycemiaweregreaterwiththeothertreatmentsthanwithbasal
insulin(65.1vs. 59.5%, P=0.0390; 42.2vs.31.5%, P,0.0001).
Theincidenceofseveresymptomatichypoglycemiawaslowandnot
clearlydifferent betweenmethods of intensication (2.5 vs.
1.2%,forotherregimensvs. basal insulin, re-spectively;
P=0.0589)(SupplementaryTable
2).CONCLUSIONSdInthispopulationofpatients withtype
2diabeteswhore-quiredintensicationof antihyperglyce-mic therapy,
the contribution from BHGto total hyperglycemic exposure was
uni-formlyhigh(7680%) acrosstheob-servedrangeof A1Clevelsat
baseline.After intensication of treatment leadingto lower mean
levels of A1C, there was asmaller (but important)
contributionfromBHGandgreateronefromPPHG.Alterationof the
contributions fromBHGand PPHGwas especially prominentwhen basal
insulin was used. For all pa-tients treated with basal insulin, an
aver-ageof 34%contributionfromresidualBHGwaspresent,
incontrastwith68%after treatment intensication with
otheragents.Afteradditionofbasalinsulin,amodest tendency toward
higher residualBHG with increasing A1C category at endpoint was
found, whereas this pattern didnot emergeaftertreatment
withotheragents.Theseobservations suggest the formof treatment
usedbypatients canbeamore signicant factor affecting the
con-tribution to hyperglycemia frombasalversus postprandial glucose
elevationsthantheobservedA1Clevel alone. Toillustratethis point,
withoral therapiesaloneatbaseline,participantswithA1C,8.0% (mean
7.6%) had 76% contribu-tionfromBHG. Afterintensicationoftreatment
withbasal insulin, peoplewithA1C7.57.9%(mean7.7%) had34%
contribution from BHG. After inten-sication with premixed insulin,
prandialinsulin, oradditional oral therapyindi-viduals withA1Cinthe
same range(mean 7.7%) had 68%contributionfrom BHG.Several aspects
of the methods we usedcould have contributed to
differencesbetweenthese ndingsandearlier ones.The present analysis
included a large pop-ulation studied both before and after
addi-tional treatment and used seven-pointFigure 2dA: The
seven-point glucose proles for patients on basal insulin versus
other treat-ments at week 24 or 28. B: Relative contributions of
BHG and PPHG to overall hypoglycemia byA1C category after 24 or 28
weeks of basal insulin treatment. C: Relative contributions of
BHGand PPHGtooverall hyperglycemia by A1C category after 24 or
28weeks of treatment withinsulin lispro, premixed insulin, or OAD
intensication.2512 DIABETES CARE,VOLUME 34, DECEMBER 2011
care.diabetesjournals.orgBasal versus postprandial
hyperglycemiaglucoseprolestoestimateglycemicex-posure during a full
24-h interval.Whereas some earlier studies did not in-clude basal
values lower than 110 mg/dL(6.1 mmol/L) in calculations of
hypergly-cemia, in keeping with denitions offasting hyperglycemia
that were then cur-rent (17), we considerednormal fasting lev-els
to be ,100 mg/dL (5.6 mmol/L) (9,10)andelevations above this level
to contributetoBHGexposure.Thegeneralizabilityofourndings is
limited by the fact that, atbaseline,
thestudypopulationincludedneither patients with A1C already
,7.0%who might have relatively greater contribu-tionfromPPHG,
norindividualswhosetreatmentwaslimitedtolifestyle. Thus,the
presentndings do not address
ques-tionssuchaswhetherPPHGusuallyex-ceedsBHGearlyinthecourseoftype2diabetesandwhetherinitial
pharmaco-therapyshouldtargetPPHGratherthanBHG. Our ndingsalsodonot
provideinsight into the relative effects of BHG
ver-susPPHGonmedicaloutcomes.Finally,useofseven-pointself-measuredglucoseproleshascertainlimitations,includinglack
of direct assessment of glucose levelsovernight
andvariabilityresultingfromday-to-daydifferencesincollectiontimeand
eating patterns in individual patients.Therefore, additional
studies using contin-uousglucosemonitoringinwell-denedpopulations
will be of interest (18).The present ndings have
clinicalimplications. They support the view thatfor most patients
not achieving A1Clevels,7.0%withoral therapies, targetingBHG with
basal insulin or other methodsof treatment (as
proposedbycurrenttreatmentguidelines)(19)isamorede-sirablerst
option than targeting PPHG.Moreover, with the methods used in
thesestudies, neitherBHGnorPPHGisrou-tinely normalized with a
single interven-tion, sothat additional treatment withcurrent or
future methods will be helpfulfor many patients.Insummary, this
analysis of 1,699patientsshowsthatwhenA1Cishigherthan 7.0% despite
diet and oral therapy,BHG dominates hyperglycemic
exposureoverawiderangeofA1Cvalues. Itex-pands the original concept
of Monnieret al. (1,2) by showing that intensicationof
antihyperglycemic therapy changes therelative contribution of
BHGversusPPHG, depending on the main effects oftheformof
treatmentused. Whentreat-ment is intensied with basal insulin,
BHGis markedly reduced yet still accounts forabout one-thirdof
hyperglycemic exposurewhen close to A1C targets and
potentiallymaybereducedfurther. Normalizationof
glycemicexposurewill requireatten-tion to both BHG and
PPHG.AcknowledgmentsdThisstudywasfundedbysano-aventisU.S. Editorial
support wasprovidedbyTraci A. Stuve(Embryon, Inc.,Somerville, NJ, a
division of Advanced HealthMedia).M.R. has received grant support
fromAmylin, GlaxoSmithKline, Eli Lilly and Com-pany, and
sanofi-aventis U.S. and honoraria forconsultingfromAmylin, Eli
LillyandCom-pany, Pfizer, Roche, and sanofi-aventis U.S. G.U.has
received research support from the Amer-ican Diabetes Association,
Baxter, Eli Lilly andCompany, National Institutes of Health,
sanofi-aventis U.S., and Takeda. A.D. is an employee
ofsanofi-aventis U.S. R.Z. is an employee ofMedpace. J.R. has
served on scientic advisoryboards
andreceivedhonorarium/consultingfees from Amylin, Boehringer
Ingelheim,Daiichi Sankyo, Eli Lilly and Company, Glaxo-SmithKline,
Johnson&Johnson, MannKind,Novartis, Novo Nordisk, Pfizer,
Roche, sanofi-aventis U.S., and Takeda. J.R. has also
receivedgrants/researchsupport fromMerck, Pfizer,sanofi-aventis
U.S., Novo Nordisk, Roche, Bris-tol-Myers Squibb, Eli Lilly and
Company,Forest, GlaxoSmithKline, Takeda, Novartis,AstraZeneca,
Amylin, Johnson&Johnson,Daiichi Sankyo, MannKind,
andBoehringerIngelheim. Nootherpotential conflictsof in-terest
relevant to this article were reported.M.R., G.U., A.D., R.Z., and
J.R. analyzed thedata, contributedtodiscussions, andwrote,reviewed,
and edited the manuscript.This study was presented in poster form
atthe70thScienticSessionsof theAmericanDiabetes Association,
Orlando, Florida, 2529June 2010, and at the 46thAnnual Meeting of
theEuropean Association for the Study of Diabetes,Stockholm,
Sweden, 2024 September 2010.References1.
MonnierL,LapinskiH,ColetteC. Con-tributions of fasting and
postprandialplasmaglucoseincrementstotheoveralldiurnal
hyperglycemiaof type2diabeticpatients:
variationswithincreasinglevelsof HbA(1c). Diabetes Care
2003;26:8818852. Monnier L, Colette C, Dunseath GJ,Owens DR. The
loss of postprandial gly-cemic controlprecedes stepwise
deterio-ration of fasting with worsening diabetes.Diabetes Care
2007;30:2632693. Peter R, Luzio SD, Dunseath G, PauvadayV,
MustafaN, Owens DR. RelationshipbetweenHbA1c andindices of
glucosetolerance derived froma standardizedmeal test
innewlydiagnosedtreatmentnaive subjects with Type 2 diabetes.
Dia-bet Med 2006;23:9909954. Peter R, Dunseath G, Luzio SD,
ChudleighR, ChoudhurySR, Owens DR.
Relativeandabsolutecontributionsof postpran-dial and fasting plasma
glucose to daytimehyperglycaemiaandHbA(1c)insubjectswithType2
diabetes. DiabetMed 2009;26:9749805. Kikuchi K, NezuU, ShirakawaJ,
et al.Correlationsof
fastingandpostprandialbloodglucoseincrementstotheoveralldiurnal
hyperglycemic status intype 2diabetic patients: variations with
levels ofHbA1c. Endocr J 2010;57:2592666. Woerle HJ, Neumann C,
Zschau S, et al.Impact of fastingandpostprandial gly-cemia on
overall glycemic control in type2diabetes. Importance of
postprandialglycemia to achieve target HbA1c levels.Diabetes Res
Clin Pract 2007;77:2802857. Schernthaner G, Guerci B, Gallwitz
B,etal. Impactofpostprandial andfastingglucose concentrations
onHbA1cinpatientswith type 2 diabetes. Diabetes Metab
2010;36:3893948. Wang J-S, Tu S-T, Lee I-T, et al.
Contri-butionofpostprandial
glucosetoexcesshyperglycaemiainAsiantype2diabeticpatients using
continuous glucose moni-toring. Diabetes Metab Res Rev
2011;27:79849. GenuthS, Alberti KG, Bennett P, et al.;Expert
Committee on the DiagnosisandClassicationof
DiabetesMellitus.Follow-upreport onthediagnosisof di-abetes
mellitus. Diabetes Care 2003;26:3160316710. American Diabetes
Association. Standardsof medical care indiabetesd2011. Di-abetes
Care 2011;34(Suppl. 1):S11S6111. Gerstein HC, Yale J-F, Harris SB,
Issa M,Stewart JA, Dempsey E. Arandomizedtrial of adding insulin
glargine vs. avoid-ance of insulin in people with Type 2 di-abetes
on either no oral glucose-loweringagents or submaximal doses of
metforminand/or sulphonylureas.
TheCanadianINSIGHT(ImplementingNewStrategieswith Insulin Glargine
for HyperglycaemiaTreatment) Study. Diabet Med2006;23:73674212.
Riddle MC, Rosenstock J, Gerich J; InsulinGlargine
4002StudyInvestigators. TheTreat-to-Target Trial:
randomizedaddi-tion of glargine or human NPH insulin tooral
therapyof type2diabeticpatients.Diabetes Care 2003;26:3080308613.
Standl E, Maxeiner S, Raptis S; HOE901/4009 Study Group. Once-daily
insulinglargine administrationinthe morningcompared to bedtime in
combination withmorning glimepiride in patients with type2diabetes:
anassessment of treatmentflexibility. HormMetab Res
2006;38:17217714. JankaHU, PleweG, RiddleMC, Kliebe-Frisch C,
Schweitzer MA, Yki-Jrvinen H.Comparison of basal insulin added to
oralcare.diabetesjournals.org DIABETES CARE,VOLUME 34, DECEMBER
2011 2513Riddle and Associatesagents versus twice-daily premixed
in-sulinas initial insulintherapyfortype2diabetes. Diabetes Care
2005;28:25425915. Bretzel RG, Nuber U, Landgraf W, OwensDR,
BradleyC,LinnT.Once-dailybasalinsulinglargineversus
thrice-dailypran-dial insulinlisproinpeoplewithtype2diabetes on
oral hypoglycaemic agents(APOLLO): an open randomised con-trolled
trial. Lancet 2008;371:1073108416. Yki-Jrvinen H, Kauppinen-Mkelin
R,TiikkainenM, etal. InsulinglargineorNPHcombined with metformin in
type 2diabetes: the LANMET study. Diabetologia2006;49:44245117.
Expert Committee on the Diagnosis andClassicationof
DiabetesMellitus. Re-port of the Expert Committee ontheDiagnosis
andClassicationof DiabetesMellitus. Diabetes Care
2003;26(Suppl.1):S5S2018. MonnierL, ColetteC, OwensD. Post-prandial
and basal glucose in type 2 di-abetes: assessment and respective
impacts.Diabetes Technol Ther 2011;13(Suppl. 1):S25S3219. Nathan
DM, Buse JB, Davidson MB, et al.;AmericanDiabetes Association;
EuropeanAssociation for the Study of Diabetes. Med-ical management
of hyperglycemia in type 2diabetes: a consensus algorithm for the
ini-tiation and adjustment of therapy: a consen-sus statement of
the American DiabetesAssociationandtheEuropeanAssociationfor the
Studyof Diabetes. Diabetes Care2009;32:1932032514 DIABETES
CARE,VOLUME 34, DECEMBER 2011 care.diabetesjournals.orgBasal versus
postprandial hyperglycemia
