Intensive Glycemic Therapyin Patients With Type 2 Diabeteson b-BlockersDiabetes Care 2016;39:1818–1826 | DOI: 10.2337/dc16-0721

OBJECTIVE

Recent studies have suggested that b-blockers may decrease the adverse influ-ence of hypoglycemia and reduce hypoglycemia-associated cardiac arrhythmiasand death. We evaluated whether intensive glycemic therapy in patients withdiabetes receiving treatment with b-blockers showed beneficial effects for theprevention of cardiovascular events without increased mortality compared with astandard glycemic therapy.

RESEARCH DESIGN AND METHODS

We used Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial data toassess the risks of cardiovascular events, all-cause death, and cardiovasculardeath in patients with diabetes receiving treatment with b-blockers (n = 3,079)and not receiving treatment with b-blockers (n = 7,145) using Cox proportionalhazard models.

RESULTS

In patients receiving treatment with b-blockers, the cumulative event rates forcardiovascular events were significantly lower in the intensive therapy groupcompared with the standard therapy group (hazard ratio [HR] 0.81; 95% CI0.67–0.97; P = 0.02), whereas those rates in patients not receiving treatment withb-blockers were not significantly different (HR 0.92; 95% CI 0.78–1.09; P = 0.36).Conversely, the cumulative event rates for all-cause and cardiovascular deaths inpatients receiving treatment with b-blockers were not significantly different be-tween the standard therapy and intensive therapy groups (all-cause death: HR1.08; 95% CI 0.83–1.42; P = 0.54; cardiovascular death: HR 1.05; 95% CI 0.72–1.51;P = 0.79), whereas in patients not receiving treatment with b-blockers, the eventrates were significantly higher in the intensive therapy group compared with thestandard therapy group (all-cause death: HR 1.25; 95% CI 1.02–1.52; P = 0.02;cardiovascular death: HR 1.43; 95% CI 1.03–1.98; P = 0.03).

CONCLUSIONS

Intensive glycemic therapy may be effective in patients with type 2 diabetes re-ceiving treatment with b-blockers.

Glycemic control in patients with diabetes is necessary to prevent diabetes-relatedcomplications. Intensive glycemic control for patients with type 2 diabetes candecrease the risks for microvascular diseases, such as diabetic retinopathy and nephrop-athy (1), but the prevention of macrovascular diseases remains difficult. Recent large-scale randomized control trials (2–4) did not show the efficacy of intensive glycemic

1Department of Diabetes, Endocrinology, andMetabolism, Center Hospital, National Centerfor Global Health and Medicine, Tokyo, Japan2Department of Clinical Study and Informatics,Center for Clinical Sciences, National Center forGlobal Health and Medicine, Tokyo, Japan3Department of Public Health/Health Policy, TheUniversity of Tokyo, Tokyo, Japan4Department of Endocrinology and Diabetes,Saitama Medical University, Saitama, Japan

Corresponding author: Tetsuro Tsujimoto,ttsujimoto@hosp.ncgm.go.jp.

Received 2 April 2016 and accepted 4 July 2016.

This article contains Supplementary Data onlineat http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dc16-0721/-/DC1.

© 2016 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered. More infor-mation is available at http://www.diabetesjournals.org/content/license.

Tetsuro Tsujimoto,1 Takehiro Sugiyama,2,3

Mitsuhiko Noda,4 and Hiroshi Kajio1

1818 Diabetes Care Volume 39, October 2016

CARDIOVASC

ULA

RANDMETABOLICRISK

therapy for the prevention of cardiovas-cular events. In addition, the Action toControl Cardiovascular Risk in Diabetes(ACCORD) trial revealed that intensiveglycemic therapy can increase all-causeand cardiovascular mortalities (2). Apossible explanation for these results isthe fact that glucose-lowering therapymay increase the frequency of hypogly-cemic episodes, which in turn is associ-ated with increased risks for vascularevents and mortality (5,6). In fact, pa-tients with diabetes with severe hypo-glycemia face many critical problemssuch as severe hypertension, hypokale-mia, and QT prolongation, which couldlead to cardiovascular disease, fatal ar-rhythmia, and death (7,8).Recent studies have suggested that

b-blockers may prevent or decreasethe adverse influence of severe hypo-glycemia, such as severe hypertensionand hypokalemia, and may reduce se-vere hypoglycemia-associated cardiacarrhythmias and death (9,10). There-fore, the use of b-blockers may help toachieve maximum effects of glycemiccontrol, particularly in intensive glyce-mic therapy, due to a decrease in theadverse influence of severe hypoglyce-mia. Although b-blockers theoreticallypose a potential risk for the occurrenceand prolongation of severe hypogly-cemia (11), there is little evidence tosupport the assertion that b-blockersshould be routinely contraindicatedin patients with diabetes (12–18). Inthe current study, we assessed whetherintensive glycemic therapy in patientswith diabetes receiving treatment withb-blockers showed beneficial effectsfor the prevention of cardiovascularevents without increased mortality,compared with a standard glycemictherapy.

RESEARCH DESIGN AND METHODS

Study DesignWe used ACCORD data to evaluatethe associations between the use ofb-blockers and cardiovascular events,and all-cause and cardiovascular mor-talities in patients with type 2 diabetes.The detailed design and description ofglycemia interventions of the ACCORDtrial have been reported previously(2,19–21). Briefly, the ACCORD trialwas sponsored by the National Heart,Lung, and Blood Institute (NHLBI) andwas conducted in 77 clinical centers

across the U.S. and Canada. In total,10,251 men and women between40 and 79 years of age with type 2diabetes, a glycated hemoglobin levelof $7.5%, and who either were be-tween 40 and 79 years of age and hadcardiovascular disease or were be-tween 55 and 79 years of age and hadalbuminuria, anatomical evidence ofsignificant atherosclerosis, left ven-tricular hypertrophy, or at least twoadditional risk factors for cardiovascu-lar disease (current smoker, obesity,hypertension, or dyslipidemia) wereincluded in the trial (2,19). Exclu-sion criteria included a BMI (weightin kilograms per square meters) of.45 kg/m2, an unwillingness to per-form home glucose monitoring or toinject insulin, frequent or recent seri-ous hypoglycemia, a serum creatininelevel of .1.5 mg/dL, or any other se-rious illness. All 10,251 patients wererandomly allocated into one of thetwo groups: one group of patients re-ceived comprehensive intensive glyce-mic therapy that targeted a glycatedhemoglobin level of ,6.0%; and theother group of patients received stan-dard glycemic therapy that targeted alevel of 7.0–7.9%. The medicationsused to achieve these targets werethe same in the two groups and in-cluded metformin, short-acting andlong-acting insulins, sulfonylureas,acarbose, meglitinides, thiazolidine-diones, and incretins. Patients were fol-lowed up at least every 4 months toensure that therapeutic goals weremet and maintained, and to monitorstudy outcomes and adverse effects.The study protocol was approved bythe ethics committee of each studycenter, and approved and monitoredby an independent data safety andmonitoring board. All participants pro-vided written informed consent. Be-cause of the increase in all-cause andcardiovascular mortalities, intensiveglycemic therapy was discontinued on6 February 2008 (2). Participants wereswitched to the standard regimen andwere followed up until 31 December2010. The occurrence of outcomes inthis study was maximally followed upfor 7 years. This study was approvedby the institutional review board ofthe National Center for Global Healthand Medicine, and NHLBI has approvedour use of the ACCORD data.

Outcome MeasurementsThe primary outcome in this study wasthe first occurrence of a cardiovascularevent, which included nonfatal myocar-dial infarction, unstable angina, nonfatalstroke, and cardiovascular death. Sec-ondary outcomes were all-cause death,cardiovascular death, and severe hypo-glycemia. Cardiovascular death was de-fined as presumed cardiovascular death;unexpected death; and death from amyocardial infarction, arrhythmia, con-gestive heart failure, stroke, and othercardiovascular diseases, including ab-dominal aortic aneurysm rupture andpulmonary emboli (19). Severe hypogly-cemia was defined as hypoglycemicevents with confirmed blood glucoselevels of ,50 mg/dL and requiringmedical assistance.

Statistical AnalysisDemographic data were presented asnumbers with proportions (percentage)or means with SDs. Continuous vari-ables were compared using Studentt tests, and categorical variables werecompared using x2 tests or Fisher ex-act tests, as appropriate. The study par-ticipants were first divided into twogroups according to their use or non-use of b-blockers. With the exceptionof severe hypoglycemia, the numberof events occurring within 1 year wassmall, and there were concerns regard-ing subject identification. Therefore,before we analyzed the data, follow-uptimes for all early events were trimmedto 1 year by the NHLBI. For cardiovascu-lar events and all-cause and cardiovas-cular deaths within 1 year, we comparedthe incidences of these events by inten-sive glycemic therapy with standard gly-cemic therapy in each patient receivingtreatment with and not receiving treat-ment with b-blockers. We analyzed haz-ard ratios (HRs) with 95% CIs in patientsreceiving intensive glycemic therapycompared with those receiving standardglycemic therapy by Cox proportionalhazard models, again for each patientreceiving treatment with and not receiv-ing treatment with b-blockers. Analysesof events before the treatment transi-tion were also performed. Kaplan-Meiersurvival curves were constructed for thecardiovascular events and all-cause andcardiovascular mortalities.

All statistical analyses were con-ducted using Stata software (version

care.diabetesjournals.org Tsujimoto and Associates 1819

14.1; StataCorp). P , 0.05 was consid-ered statistically significant for all tests.

RESULTS

Study ParticipantsThe characteristics of patients withtype 2 diabetes receiving treatmentwith and not receiving treatment withb-blockers are presented in Table 1.Among the study patients, 3,079 werereceiving treatment with b-blockers(standard glycemic therapy, n = 1,580;intensive glycemic therapy, n = 1,499)and 7,145 were not receiving treatmentwith b-blockers (standard glycemictherapy, n = 3,526; intensive glycemictherapy, n = 3,619). In patients receivingtreatment with and not receiving treat-ment with b-blockers, the characteris-tics were not significantly differentbetween those receiving standard glyce-mic therapy and those receiving inten-sive glycemic therapy. Characteristics,including age, sex, duration of diabetes,history of cardiovascular disease, smok-ing status, BMI, cholesterol and triglyc-eride levels, and estimated glomerularfiltration rate, were significantly differ-ent between patients receiving treat-ment with and not receiving treatmentwithb-blockers. The rate ofb-blocker usein patients with a history of coronaryheart disease and heart failure was 66%,and that in patientswithout such ahistorywas 21% (Supplementary Table 1), whichindicated guideline-compliant b-blockeruse for patients with a history of coronaryheart disease and/or heart failure.

Cardiovascular Events and All-Causeand Cardiovascular MortalitiesThe incidences of cardiovascular eventsand all-cause and cardiovascular deathswithin 1 year in all patients receiv-ing intensive and standard glycemictherapies and in those receiving treat-ment with and not receiving treat-ment with b-blockers are presentedin Fig. 1. In patients receiving treat-ment with and not receiving treat-ment with b-blockers, the incidences ofcardiovascular events, all-cause death,and cardiovascular death were not sig-nificantly different between patientsreceiving standard and intensive glyce-mic therapies in the first year. Kaplan-Meier survival curves and the eventrates for the following periods areshown in Fig. 2 and Table 2, respec-tively. The mean follow-up periods

(6SD) were 4.6 6 1.5 years in pa-tients not receiving treatment withb-blockers and 4.36 1.5 years in thosereceiving treatment with b-blockers.Consistent with previous studies onthe ACCORD trials (2,21), the cumu-lative event rates for cardiovascularevents were lower, and those for all-cause and cardiovascular deaths werehigher, in the intensive therapy groupcompared with the standard therapygroup (Fig. 2A, C, and E). In patients re-ceiving treatment with b-blockers, thecumulative event rates for cardiovascu-lar events were significantly lower in theintensive therapy group compared withthe standard therapy group (HR 0.81;95% CI 0.67–0.97; P = 0.02), whereasthose in patients not receiving treat-ment with b-blockers were not signif-icantly different (HR 0.92; 95% CI0.78–1.09; P = 0.36) (Fig. 2B). Con-versely, the cumulative event rates forall-cause and cardiovascular deaths inpatients receiving treatment withb-blockers were not significantly differ-ent between the standard and intensivetherapy groups (all-cause death: HR1.08; 95% CI 0.83–1.42; P = 0.54; cardio-vascular death: HR 1.05; 95% CI 0.72–1.51; P = 0.79), whereas in patients notreceiving treatment with b-blockers,the event rates were significantly higherin the intensive therapy group com-pared with the standard therapy group(all-cause death: HR 1.25; 95% CI 1.02–1.52; P = 0.02; cardiovascular death: HR1.43; 95% CI 1.03–1.98; P = 0.03) (Fig.2D and F, respectively). In patients re-ceiving treatment with and not receiv-ing treatment with b-blockers, thecumulative event rates for nonfatalmyocardial infarction and nonfatal strokewere nonsignificantly lower in the inten-sive therapy group compared with thestandard therapy group (Table 2). The cu-mulative event rates for fatal or hospital-ized congestive heart failure in patientsreceiving treatment with b-blockerswere not significantly different betweenthe standard and intensive therapygroups (HR 0.96; 95% CI 0.71–1.30; P =0.80), whereas in patients not receivingtreatment with b-blockers, the eventrates were higher in the intensive ther-apy group compared with the standardtherapy group (HR 1.23; 95% CI 0.92–1.64; P = 0.15).

Kaplan-Meier survival curves for car-diovascular events and all-cause and

cardiovascular deaths in intensive andstandard glycemic therapies before treat-ment transition in all patients and thosereceiving treatment with and not re-ceiving treatment with b-blockers areshown in Fig. 3. Similarly, we found thatthe cardiovascular event rate in patientswith diabetes receiving treatment withb-blockers was lower in the intensivetherapy group than in the standard ther-apy group (HR 0.81; 95% CI 0.66–1.01;P = 0.06). In addition, the cumulativeevent rates for all-cause and cardiovascu-lar deaths in patients receiving treatmentwith b-blockers were not significantlydifferent between the standard and inten-sive therapy groups (all-cause death: HR1.00; 95% CI 0.73–1.38; P = 0.96; cardiovas-cular death: HR 0.85; 95% CI 0.55–1.31; P =0.48), whereas in patients not receivingtreatment with b-blockers, the eventrates were significantly higher in the in-tensive therapy group compared with thestandard therapy group (all-cause death:HR 1.27; 95% CI 1.01–1.59; P = 0.03; car-diovascular death: HR 1.52; 95% CI 1.06–2.18; P = 0.02).

Severe HypoglycemiaThe HRs for severe hypoglycemia arepresented in Supplementary Table 2.Although the event rates per year werehigher in patients receiving treatmentwith b-blockers compared with thosenot receiving treatment with b-blockers,HRs for severe hypoglycemia in the in-tensive therapy group compared withthe standard therapy group differedonly slightly between those receivingtreatment with b-blockers (HR 2.98;95% CI 2.19–4.06; P , 0.001) and notreceiving treatment with b-blockers(HR 2.87; 95% CI 2.27–3.62; P , 0.001).In the model that included the interac-tion term between the use of b-blockersand the type of therapy (intensive/stan-dard glycemic therapy), we found thatthe association between intensive glyce-mic therapy and the incidence of severehypoglycemia was not interacted by theuse of b-blockers (P for interactionterm = 0.86, data not shown).

CONCLUSIONS

In the current study, cardiovascular eventrates in patients receiving treatmentwith b-blockers were significantly lowerin the intensive therapy group com-pared with those in the standard ther-apy group. In addition, all-cause and

1820 b-Blockers in Patients With Diabetes Diabetes Care Volume 39, October 2016

Table

1—Base

linech

aracte

risticsofpatie

nts

with

type2diabeteswhoare

rece

ivingtre

atm

entwith

andnotrece

ivingtre

atm

entwith

b-b

lock

ers

Characteristics

b(2

)b(+)

b(2

)vs.b

(+)Pvalu

eAll

(N=7,145)

Standard

(n=3,526)

Inten

sive(n

=3,619)

Pvalu

e†

All

(N=3,079)

Standard

(n=1,580)

Inten

sive(n

=1,499)

Pvalu

e

Age

(years)62.6

(6.4)62.6

(6.4)62.6

(6.5)0.80

63.1(7.1)

63.1(7.2)

63.2(7.0)

0.62,0.001

Female

sex(%

)40.2

40.140.3

0.8934.8

34.734.8

0.93,0.001

Duratio

nofdiab

etes(years)

10.5(7.4)

10.7(7.4)

10.4(7.4)

0.1011.4

(8.0)11.3

(8.0)11.6

(8.0)0.23

,0.001

Histo

ryofcoronary

heart

disease

(%)*

14.013.4

14.60.14

43.743.3

44.10.65

,0.001

Histo

ryofheart

failure

(%)

2.72.6

2.90.40

9.79.7

9.60.95

,0.001

Histo

ryofstro

ke(%

)5.2

5.45.0

0.498.3

8.58.1

0.73,0.001

Race

andeth

nicity

(%)

White

61.561.0

62.00.37

64.365.7

62.90.10

0.007Black

19.218.9

19.50.54

18.718.0

19.40.32

0.58Hisp

anic

7.27.5

7.00.36

7.17.2

7.10.93

0.82Others

12.112.6

11.50.17

9.99.1

10.60.16

0.001

Educatio

nalattain

men

t(%

)Less

than

high

school

14.514.1

14.90.28

15.613.9

17.50.006

0.13High

school

25.925.6

26.10.64

27.729.1

26.20.07

0.05Somecollege

32.933.0

32.80.85

32.432.5

32.30.88

0.64College

degree

orhigh

er26.7

27.326.2

0.2624.2

24.524.0

0.720.008

Curren

tsm

okin

g(%

)12.6

12.412.9

0.5611.1

10.611.5

0.420.02

BMI(kg/m

2)32.0

(5.5)32.0

(5.4)32.0

(5.5)0.80

32.7(5.3)

32.7(5.3)

32.6(5.2)

0.47,0.001

Med

ications(%

)Insulin

33.134.2

32.10.06

39.940.0

39.80.92

,0.001

Sulfo

nylu

rea53.0

52.553.4

0.4154.6

53.955.4

0.420.12

Metfo

rmin

63.964.1

63.80.74

64.364.6

63.90.68

0.74ARB/ACE-I

67.267.5

66.90.62

74.774.4

75.10.69

,0.001

CCB

10.710.8

10.60.81

13.913.9

13.90.99

,0.001

Thiazid

e25.8

25.526.2

0.4832.0

32.631.3

0.43,0.001

Statin58.2

58.757.7

0.3876.4

75.677.1

0.33,0.001

Aspirin

50.249.9

50.40.70

65.164.3

66.00.33

,0.001

Systolic

bloodpressu

re(m

mHg)

136.0(15.9)

136.1(15.9)

135.9(15.9)

0.61136.5

(17.9)136.6

(17.7)136.3

(18.0)0.73

0.21

Glycated

hem

oglo

bin

(%)

8.3(1.0)

8.3(1.0)

8.3(1.0)

0.368.3

(1.0)8.3

(0.9)8.3

(1.0)0.75

0.69

Cholestero

l(mg/d

L)LD

L107.4

(33.1)107.4

(33.3)107.4

(32.9)0.94

98.1(31.9)

98.2(31.3)

97.9(32.6)

0.74,0.001

HDL

42.9(11.4)

42.9(11.2)

42.8(11.5)

0.5339.4

(10.4)39.6

(10.4)39.2

(10.3)0.36

,0.001

Triglyceride(m

g/dL)

194.8(122.2)

192.1(113.1)

181.5(114.8)

0.58195.8

(121.1)195.0

(120.5)196.6

(121.7)0.71

,0.001

Estimated

GFR

(mL/m

in/1.73m

2)§91.5

(22.5)91.8

(22.2)91.1

(22.8)0.16

87.1(22.5)

87.3(22.3)

87.0(22.7)

0.72,0.001

Data

arepresen

tedas

themean

(SD),u

nlessotherw

iseindicated

.Amongthe10,251stu

dypatien

ts,27didnothave

theinform

ationontheiruse

ofb

-blocke

rs.Glycated

hemoglo

bin:8.3%=67mmoL/m

oL.

ACE-I,an

giotensin

-convertin

genzym

einhibito

rs;ARB,an

giotensin

IIreceptorblockers;b

(2),p

atientsnotreceivin

gtreatm

entwith

b-blockers;b

(+),patien

tsreceivin

gtreatm

entwith

b-blocke

rs;CCB,

calcium

chan

nelb

locke

rs;GFR

,glo

meru

larfiltratio

nrate;

Inten

sive,inten

sivetherap

ygro

up;Stan

dard

,stan

dard

therap

ygro

up.*C

oronary

heart

disease

was

defined

asmyocard

ialinfarctio

norangin

apecto

ris;†Pvalu

ewas

calculated

bycomparin

gvariab

lesininten

sivetherap

ywith

those

instan

dard

therap

y;§Theestim

atedGFR

was

calculated

usin

gthefollowingModificatio

nofD

ietinRen

alDisease

(MDRD)

Studyeq

uatio

n:estim

atedGFR

(mL/m

in/1.73

m2)=175

3(seru

mcreatin

inein

mg/d

L) 21.1543

(agein

years) 20.2033

(0.742forfem

ale)3

(1.212forAfrican

American

).

care.diabetesjournals.org Tsujimoto and Associates 1821

cardiovascular mortalities in patientsreceiving treatment with b-blockers werenot significantly different between the in-tensive and standard therapy groups. Incontrast, in patients not receiving treat-ment with b-blockers, the cardiovascu-lar event rate did not differ significantlybetween intensive and standard glyce-mic therapies, whereas all-cause and

cardiovascularmortalitieswere significantlyhigher in the intensive therapy group.

A recent study on the ACCORD trial(21) reported that ischemic heart diseasewas less frequent in the intensive ther-apy group compared with the stan-dard therapy group. Considering theresults in the current study, the bene-ficial effects of intensive glycemic

therapy might be due to the efficacyin patients receiving treatment withb-blockers. In addition, although a pre-vious report on the ACCORD trial (2)demonstrated that intensive glycemictherapy was associated with an in-creased risk of all-cause and cardiovas-cular deaths, the current study foundthat the adverse effects of intensive gly-cemic therapy might be attributed tothe effects of not receiving treatmentwith b-blockers. Some reports have in-dicated that severe hypoglycemia, ofwhich intensive glycemic therapy wasassociated with higher risks, was asso-ciated with an increased risk of car-diovascular disease and death (5,6).One possible reason for the associationbetween severe hypoglycemia andcardiovascular events is that severehypoglycemia can lead to activation ofthe sympathoadrenal system and the re-lease of counter-regulatory hormones,resulting in significant hemodynamicchanges, hypokalemia, and QT pro-longation (7,22). Based on pathophys-iological mechanisms, a prior use ofb-blockers may prevent the adverseinfluence of the hypersecretion ofcatecholamines induced by severe hy-poglycemia, and that may reduce thenumber of vascular events, cardiac ar-rhythmias, and deaths due to severehypoglycemia (7,10). Indeed, the anal-yses of patients receiving treatmentwithb-blockers showed that the cardio-vascular event rate was significantlylower in the intensive therapy groupcompared with the standard therapygroup. In addition, although data onarrhythmogenic cardiac mortality wasnot assessed, the number of all-causeand cardiovascular mortalities did notincrease in the intensive therapy groupin patients receiving treatment withb-blockers, which is different from theresults of those not receiving treatmentwith b-blockers. The HRs for severe hy-poglycemia in the intensive therapygroup compared with the standard ther-apy group were not very different be-tween patients receiving treatmentwith and not receiving treatment withb-blockers, and the association be-tween intensive glycemic therapy andthe incidence of severe hypoglyce-mia was not interacted by the use ofb-blockers. Therefore, the decreasedrisk of cardiovascular events in patientsreceiving treatment with b-blockers

Figure 1—Incidences of cardiovascular events and all-cause and cardiovascular deaths within1 year in intensive and standard glycemic therapies in all patients and in those receiving treat-ment with and not receiving treatment with b-blockers. Incidence of cardiovascular events (A),all-cause death (B), and cardiovascular death (C). b (2), patients not receiving treatment withb-blockers; b (+), patients receiving treatment with b-blockers; Intensive, intensive therapy;Standard, standard therapy.

1822 b-Blockers in Patients With Diabetes Diabetes Care Volume 39, October 2016

might be due to the protective effects ofb-blockers after the occurrence of se-vere hypoglycemia. Intensive glycemictherapy may be a preferable strategyto prevent cardiovascular events inpatients receiving treatment withb-blockers, which were essential fortreating underlying diseases, such ascoronary heart disease and heart fail-ure, compared with standard glyce-mic therapy. The beneficial effects ofb-blockers on cardiovascular eventsmay be observed in the high cardiovas-cular risk in patients with type 2 diabe-tes, partly because these patients maybe at a higher risk of severe hypoglycemiaand severe hypoglycemia-associated

adverse events, and b-blockers may al-leviate the damage from hypoglycemia-associated adverse events. It has beenknown that the use of b-blockers canbe a risk factor for hypoglycemia andhypoglycemia unawareness, presum-ably because of diminished or absentearly warning signs (11). However, therewas little evidence to support the asser-tion that b-blockers should be routinelycontraindicated in patients with diabe-tes as they have minimal clinical ef-fects on hypoglycemia unawarenessand recovery (12–16). Further studiesare needed to evaluate whether the useof b-blockers in patients with diabetesshows beneficial or adverse effects.

This study has several limitations.First, this was a post hoc analysis ofthe ACCORD trial, and residual con-founding might still be present. In addi-tion, although the current study waslarge scale and evidence based, andhad a robust subgroup representation,our findings may not be applicable toother patients with diabetes. Second,we were only able to analyze data thatearly events had been trimmed to1 year. The number of events prior to1 year was very small, and there wereconcerns regarding subject identifica-tion. Therefore, before we analyzedthe data, follow-up times for all earlyevents were trimmed to 1 year by theNHLBI. Although the incidences of car-diovascular events, all-cause death, andcardiovascular death in patients receiv-ing treatment with and not receivingtreatment with b-blockers were not sig-nificantly different between patientsreceiving standard and intensive glyce-mic therapies in a first year, anotherstudy is needed to confirm these results.However, we believe that the currentstudy provides extremely importantinformation regarding glycemic con-trol and diabetes management. Third,adherence to medication, includingb-blockers, might influence the studyoutcomes. Poor adherence tomedicationregimens is common and contributesto substantially worse cardiovascularoutcomes (23). However, the averagerate of adherence in clinical trials is usuallyhigh, owing to the monitoring programsand to the selection of the patients. Be-cause follow-up visits in patients in theACCORD trial, including a blood pressuretrial,wereconducted forat least4months,the rate of adherence to medicationsmight be remarkably high. Fourth, wecould not assess whether the differenttypes of b-blockers, such as cardioselec-tive and nonselectiveb-blockers, had sim-ilar effects on cardiovascular events anddeath. Although theb-blockers exert theireffects by competitively inhibiting cat-echolamines from binding to b-receptors,eachb-blocker has different characteristicswith respect to the cardioselectivity, phar-macokinetics, intrinsic sympathomimeticactivity, and a-adrenergic blocking activ-ity. Thus, further studies are needed toclarify which types of b-blockers aremore beneficial.

In conclusion, this study showedthat the cardiovascular event rate in

Figure 2—Kaplan-Meier survival curves for cardiovascular events and all-cause and cardiovas-cular deaths in intensive and standard glycemic therapies in all patients and in those receivingtreatment with and not receiving treatment with b-blockers. Rates of freedom from cardiovas-cular events (A and B), all-cause death (C and D), and cardiovascular death (E and F). b (2),patients not receiving treatment with b-blockers; b (+), patients receiving treatment withb-blockers; Intensive, intensive therapy; Standard, standard therapy.

care.diabetesjournals.org Tsujimoto and Associates 1823

Table

2—Cardiova

sculareve

nts

andall-cau

seandca

rdiova

sculardea

thin

patients

withtype2diabetesrece

ivingtreatmentwithandnotrece

ivingtreatm

entwithb-b

lock

ers

Even

t

All

b-blockers(2

)b-blockers(+)

Stan

dard

Intensive

Pvalue

Stan

dard

Intensive

Pvalue

Stan

dard

Intensive

Pvalue

Cardiovasculareven

tsNo.ofpatients/totaln

o.

559/4,912

489/4,915

300/3,386

284/3,472

257/1,509

205/1,433

Even

trate

per

year

(%)

3.0

2.6

2.3

2.1

4.9

3.9

HR(95%

CI)

1.00

(ref)

0.86(0.76–

0.97)

0.02

1.00(ref)

0.92

(0.78–

1.09)

0.36

1.00(ref)

0.81(0.67–

0.97)

0.02

Death

from

anycause

No.ofpatients/totaln

o.

279/5,070

328/5,066

173/3,493

218/3,576

105/1,560

110/1,480

Even

trate

per

year

(%)

1.4

1.6

1.2

1.5

1.8

1.9

HR(95%

CI)

1.00

(ref)

1.18(1.01–

1.39)

0.04

1.00(ref)

1.25

(1.02–

1.52)

0.02

1.00(ref)

1.08(0.83–

1.42)

0.54

Death

from

cardiovascularcauses

No.ofpatients/totaln

o.

119/5,005

146/4,999

61/3,448

88/3,526

57/1,540

58/1,463

Even

trate

per

year

(%)

0.6

0.7

0.4

0.6

0.4

0.6

HR(95%

CI)

1.00

(ref)

1.24(0.97–

1.52)

0.09

1.00(ref)

1.43

(1.03–

1.98)

0.03

1.00(ref)

1.05(0.72–

1.51)

0.79

Nonfatalm

yocardialinfarction

No.ofpatients/totaln

o.

268/4,914

225/4,925

127/3,476

147/3,391

120/1,506

98/1,439

Even

trate

per

year

(%)

1.4

1.2

1.1

0.9

2.1

1.8

HR(95%

CI)

1.00

(ref)

0.83(0.70–

0.99)

0.04

1.00(ref)

0.84

(0.67–

1.07)

0.18

1.00(ref)

0.84(0.64–

1.09)

0.20

Nonfatalstroke

No.ofpatients/totaln

o.

79/4,970

58/4,958

49/3,426

39/3,498

30/1,527

19/1,450

Even

trate

per

year

(%)

0.4

0.3

0.4

0.3

0.5

0.3

HR(95%

CI)

1.00

(ref)

0.73(0.52–

1.03)

0.07

1.00(ref)

0.78

(0.51–

1.20)

0.27

1.00(ref)

0.65(0.36–

1.16)

0.15

Fatalo

rhospitalized

congestiveheartfailure

No.ofpatients/totaln

o.

173/4,955

184/4,937

84/3,421

104/3,494

87/1,517

80/1,434

Even

trate

per

year

(%)

0.9

1.0

0.6

0.8

1.5

1.5

HR(95%

CI)

1.00

(ref)

1.07(0.87–

1.32)

0.49

1.00(ref)

1.23

(0.92–

1.64)

0.15

1.00(ref)

0.96(0.71–

1.30)

0.80

All,allpatients;b

-blockers(2

),patientsnotreceivingtreatm

entwithb-blockers;b-blockers(+),patientsreceivingtreatm

entwithb-blockers;Intensive,intensive

therap

ygroup;ref,referen

cevalue;Stan

dard,

stan

dardtherap

ygroup.

1824 b-Blockers in Patients With Diabetes Diabetes Care Volume 39, October 2016

patients with diabetes receiving treat-ment with b-blockers was significantlylower in the intensive therapy groupcompared with the standard therapygroup. In addition, all-cause and car-diovascular mortalities in those pa-tients not receiving treatment withb-blockers were significantly higherin the intensive therapy group. Inten-sive glycemic therapy may be effec-tive in patients with type 2 diabeteswho are receiving treatment withb-blockers.

Acknowledgments. This article was preparedusing ACCORD Research Materials obtained

from the NHLBI Biologic Specimen and DataRepository Information Coordinating Centerand does not necessarily reflect the opinionsor views of the ACCORD Study or the NHLBI.Funding.This researchwas supportedbyGrant-in-Aid for ScientificResearch (KAKENHI) fromtheJapan Society for the Promotion of Science (no.26860701).Duality of Interest.M.N. has received speakerhonoraria from Sanofi, Mitsubishi TanabePharma, Daiichi Sankyo, Eli Lilly Japan, MSD,Sanwa Kagaku Kenkyusho, Ono PharmaceuticalCo. Ltd, TakedaPharmaceutical Co. Ltd, Astellas,Kowa Pharmaceutical Co. Ltd, Novo NordiskPharma Ltd, AstraZeneca, and Johnson & JohnsonK.K. and research grants from Takeda Pharma-ceutical Co. Ltd, Mitsubishi Tanabe Pharma,and AstraZeneca. No other potential conflictsof interest relevant to this article werereported.

Author Contributions. T.T. contributed tothe study concept and design; data acquisition,analysis, and interpretation; and statistical anal-ysis anddrafted themanuscript. T.S. contributedto the data acquisition, analysis, and interpre-tation and statistical analysis and drafted themanuscript. M.N. and H.K. drafted the manu-script. T.T. is the guarantor of this work and, assuch, had full access to all the data in the studyand takes responsibility for the integrity of thedata and the accuracy of the data analysis.

References1. UK Prospective Diabetes Study (UKPDS)Group. Intensive blood-glucose control with sul-phonylureas or insulin compared with conven-tional treatment and risk of complications inpatients with type 2 diabetes (UKPDS 33). Lan-cet 1998;352:837–8532. Gerstein HC, Miller ME, Byington RP, et al.;Action to Control Cardiovascular Risk in Diabe-tes Study Group. Effects of intensive glucoselowering in type 2 diabetes. N Engl J Med2008;358:2545–25593. Duckworth W, Abraira C, Moritz T, et al.;VADT Investigators. Glucose control and vascu-lar complications in veterans with type 2 diabe-tes. N Engl J Med 2009;360:129–1394. Patel A, MacMahon S, Chalmers J, et al.;ADVANCE Collaborative Group. Intensive bloodglucose control and vascular outcomes in pa-tients with type 2 diabetes. N Engl J Med 2008;358:2560–25725. Zoungas S, Patel A, Chalmers J, et al.; ADVANCECollaborative Group. Severe hypoglycemia andrisks of vascular events and death. N Engl J Med2010;363:1410–14186. Bonds DE, Miller ME, Bergenstal RM, et al.The association between symptomatic, severehypoglycaemia andmortality in type 2 diabetes:retrospective epidemiological analysis of theACCORD study. BMJ 2010;340:b49097. Tsujimoto T, Yamamoto-Honda R, Kajio H,et al. Vital signs, QT prolongation, and newlydiagnosed cardiovascular disease duringsevere hypoglycemia in type 1 and type 2 dia-betic patients. Diabetes Care 2014;37:217–2258. Feldman-Billard S, Massin P, Meas T,Guillausseau PJ, Heron E. Hypoglycemia-induced blood pressure elevation in patientswith diabetes. Arch InternMed 2010;170:829–8319. Tsujimoto T, Yamamoto-Honda R, Kajio H, et al.Effectiveness of prior use of beta-blockers for pre-venting adverse influences of severe hypoglycemiain patients with diabetes: an observational study.Medicine (Baltimore) 2015;94:e162910. Reno CM, Daphna-Iken D, Chen YS,VanderWeele J, Jethi K, Fisher SJ. Severe hypo-glycemia-induced lethal cardiac arrhythmias aremediated by sympathoadrenal activation. Dia-betes 2013;62:3570–358111. RevenoWS, RosenbaumH. Propranolol andhypoglycaemia. Lancet 1968;1:92012. Kerr D, MacDonald IA, Heller SR, TattersallRB. Beta-adrenoceptor blockade and hypogly-caemia. A randomised, double-blind, placebocontrolled comparison of metoprolol CR, ateno-lol and propranolol LA in normal subjects. Br JClin Pharmacol 1990;29:685–69313. Molnar GW, Read RC. Propranolol enhance-ment of hypoglycemic sweating. Clin PharmacolTher 1974;15:490–496

Figure 3—Kaplan-Meier survival curves for cardiovascular events and all-cause and cardiovas-cular deaths in intensive and standard glycemic therapies before treatment transition in allpatients and in those receiving treatment with and not receiving treatment with b-blockers.Rates of freedom from cardiovascular events (A and B), all-cause death (C and D), and cardio-vascular death (E and F ). b (2), patients not receiving treatment with b-blockers; b (+),patients receiving treatment with b-blockers; Intensive, intensive therapy; Standard, stan-dard therapy.

care.diabetesjournals.org Tsujimoto and Associates 1825

14. Viberti GC, Keen H, Bloom SR. Beta block-ade and diabetes mellitus: effect of oxprenololand metoprolol on the metabolic, cardiovascu-lar, and hormonal response to insulin-inducedhypoglycemia in normal subjects. Metabolism1980;29:866–87215. Barnett AH, Leslie D, Watkins PJ. Can insulin-treated diabetics be given beta-adrenergicblocking drugs? BMJ 1980;280:976–97816. Shorr RI, RayWA, Daugherty JR, GriffinMR. An-tihypertensives and the risk of serious hypoglycemiain older persons using insulin or sulfonylureas.JAMA 1997;278:40–4317. Thamer M, Ray NF, Taylor T. Associationbetween antihypertensive drug use and

hypoglycemia: a case-control study of diabeticusers of insulin or sulfonylureas. Clin Ther1999;21:1387–140018. UK Prospective Diabetes Study Group. Effi-cacy of atenolol and captopril in reducing risk ofmacrovascular andmicrovascular complicationsin type 2 diabetes: UKPDS 39. BMJ 1998;317:713–72019. Buse JB, Bigger JT, Byington RP, et al.;ACCORD Study Group. Action to Control Cardio-vascular Risk in Diabetes (ACCORD) trial: designand methods. Am J Cardiol 2007;99(12A):21i–33i20. Gerstein HC, Riddle MC, Kendall DM, et al.;ACCORD Study Group. Glycemia treatmentstrategies in the Action to Control Cardiovascular

Risk in Diabetes (ACCORD) trial. Am J Cardiol 2007;99(12A):34i–43i21. Gerstein HC, Miller ME, Ismail-Beigi F,et al.; ACCORD Study Group. Effects of inten-sive glycaemic control on ischaemic heart dis-ease: analysis of data from the randomised,controlled ACCORD trial. Lancet 2014;384:1936–194122. Chow E, Bernjak A, Williams S, et al. Risk ofcardiac arrhythmias during hypoglycemia in pa-tients with type 2 diabetes and cardiovascularrisk. Diabetes 2014;63:1738–174723. Osterberg L, Blaschke T. Adherence tomedication. N Engl J Med 2005;353:487–497

1826 b-Blockers in Patients With Diabetes Diabetes Care Volume 39, October 2016