lications 25 (2011) 202–207WWW.JDCJOURNAL.COM

Journal of Diabetes and Its Comp

The importance of glycemic control: how low should we go with HbA1c?Start early, go safe, go low☆

Katrien Benhalimaa, Eberhard Standlb, Chantal Mathieua,⁎

aDepartment of Endocrinology, University Hospital Gasthuisberg, Leuven, BelgiumbMunich Diabetes Research Institute, Munich Helmholtz Center, Munich, Germany

Received 12 January 2010; received in revised form 15 March 2010; accepted 29 March 2010

Abstract

Epidemiologic data indicate a continuous relationship between hemoglobin A1c (HbA1c) and risk for microvascular and macrovascularcomplications of diabetes. Intensive glycemic control reduces risk of microvascular complications in Type 1 and Type 2 diabetes, and long-term treatment and follow-up studies have shown that initial intensive control is associated with reduced cardiovascular risk. Recentintervention trials in older, high-risk patients with Type 2 diabetes have not shown a benefit of intensive control in reducing cardiovascularrisk over a rather short-term follow-up period of up to 5 years, with some data indicating that intensive control accompanied byhypoglycemia is detrimental in patients with high cardiovascular risk. Indeed, hypoglycemia with current antidiabetic agents—primarilyinsulin and sulphonylureas—is the main limiting factor in achieving desirable levels of glycemic control. Still, the goal in treating both Type1 and Type 2 diabetes should be to safely get HbA1c as close to normal as possible. In Type 2 diabetes, this goal should be tempered for thetime being in patients with shorter life expectancy or co-existing cardiovascular disease or other co-morbidities, in whom a target of 7.0–7.5% may be advisable until we can demonstrate that lower targets in such patients can be safely achieved. Newer agents with lower risk ofhypoglycemia—e.g., insulin analogues, incretin mimetics and incretin enhancers—may form an integral component of strategies for safelyachieving lower HbA1c levels.© 2011 Published by Elsevier Inc.

Keywords: Diabetes; Hypoglycemia; HbA1c; Microvascular complications; Cardiovascular complications

1. Introduction

Patients with diabetes are at increased risk of microvas-cular complications and cardiovascular disease comparedwith the general population. Compared to those with Type 2diabetes, patients with Type 1 diabetes have lower rates ofsuch risk factors as obesity, hypertension, and dyslipidemia,

☆ Conflicts of interests/disclosures: Dr. Benhalima has no conflicts ofinterest. Dr. Standl has been an advisor/lecturer/investigator for activitiessponsored by Astra-Zeneca, Bayer-Schering, BMS, Johnson & Johnson,Merck, MSD, Novartis and Novo Nordisk. Dr. Mathieu is/has been advisorfor Eli Lilly, Merck, Sharp & Dohme, Novartis, Novo Nordisk and SanofiAventis. Editorial assistance provided by BioScience Communications.

⁎ Corresponding author. Department of Endocrinology, UZ Gasthuis-berg, Herestraat 49, B-3000 Leuven, Belgium. Tel.: +32 16 34 69 94; fax:+32 16 34 69 89.

E-mail address: chantal.mathieu@uzleuven.be (C. Mathieu).

1056-8727/$ – see front matter © 2011 Published by Elsevier Inc.doi:10.1016/j.jdiacomp.2010.03.002

and their elevated lifetime cardiovascular risk appears to bemore directly related to hyperglycemia. Cardiovascular riskin patients with Type 2 diabetes reflects both the effects ofhyperglycemia and a high frequency of such additional riskfactors. Available evidence indicates that any increase inhemoglobin A1c (HbA1c) above normal levels is associatedwith increased risk of microvascular and cardiovascularcomplications (Fig. 1) (Khaw et al., 2004; Krzentowski,Zhang, Albert, & Lefèbvre, 2004; Stratton et al., 2000),providing the rationale for treatment to reduce glucose tonormal or near-normal levels. Initial trials of intensiveglycemic control showed benefits in reducing microvascularcomplications in both Type 1 (The Diabetes Control andComplications Trial Research Group, 1993) and Type 2diabetes (UK Prospective Diabetes Study (UKPDS) Group,1998a, 1998b), with inconclusive evidence regardingprevention of cardiovascular events. Three recently reported

Fig. 1. (A) Relationship between updated mean HbA1c and risk for diabeticcomplications in patients with newly diagnosed Type 2 diabetes in theprospective observational UKPDS-35 study. Each 1% decrease in meanHbA1c was associated with risk reductions of 21% for any diabetes endpoint(Pb.0001), 21% for diabetes-related mortality (Pb0.0001), 14% formyocardial infarction (MI) (Pb.0001) and 37% for microvascularcomplications (Pb.0001) (Stratton et al., 2000). (B) Association between a1% increase in HbA1c and risk for coronary heart disease, cardiovasculardeath and all-cause mortality among more than 10,000 men and women aged45–79 years from the general population in the European ProspectiveInvestigation into Cancer in Norfolk study (Khaw et al., 2004).

203K. Benhalima et al. / Journal of Diabetes and Its Complications 25 (2011) 202–207

trials in Type 2 diabetic patients showed no benefit ofintensive control in reducing cardiovascular risk over theshort term (The Action to Control Cardiovascular Risk inDiabetes Study Group, 2008; The ADVANCE CollaborativeGroup, 2008; Duckworth et al., 2009), whereas long-termfollow-up studies over 17 years and longer both in Type 1(The Diabetes Control and Complications Trial/Epidemiol-ogy of Diabetes Interventions and Complications (TheDiabetes Control and Complications Trial/Epidemiology ofDiabetes Interventions and Complications (DCCT/EDIC)Study Research Group, 2005) and Type 2 diabetes (Holman,Paul, Bethel, Matthews, & Neil, 2008) showed significantreductions in cardiovascular risk in patients who hadreceived intensive therapy. When all of these findings areincorporated into the debate over how low we should go withHbA1c, the conclusion that still beckons is that we should goas low as we safely can.

2. HbA1c reduction in Type 1 diabetes

The Diabetes Control and Complications Trial (DCCT),reported in 1993, demonstrated that intensive glycemiccontrol in patients with newly diagnosed Type 1 diabetes

produced a dramatic reduction in risk for microvascularcomplications—including retinopathy, nephropathy, andneuropathy—compared with conventional insulin treatmentover 6.5 years of follow-up (The Diabetes Control andComplications Trial Research Group, 1993). For example,intensive control significantly reduced risk of retinopathy by76% in the primary prevention cohort; risk of retinopathyprogression and development of proliferative or severe non-proliferative retinopathy decreased by 54% and 47%,respectively, in the secondary prevention cohort. For bothtreatment groups in the study, there was no HbA1c thresholdfor risk of retinopathy (The Diabetes Control and Complica-tions Trial Research Group, 1996).

There were few macrovascular events in the DCCT trial,limiting the ability to generate conclusions regardingpreventive benefit. When all major cardiovascular andperipheral vascular end points were combined, intensivecontrol was associated with a nonsignificant 41% reductionin risk for macrovascular disease during the study period.However, the significant benefit of this initial period ofintensive control in reducing cardiovascular risk emergedover long-term follow-up in the DCCT/Epidemiology ofDiabetes Interventions and Complications (EDIC) study(The Diabetes Control and Complications Trial/Epidemiol-ogy of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group, 2005). After 11 years offollow-up, subsequent to the conclusion of the initial study(approximately 17 years total), during which time thedifference in HbA1c between the initial intensive controland conventional control groups decreased to 0.2%, and theintensive control group had a significant 42% reduction inrisk for any cardiovascular disease (P=.02) and a 57%reduction in risk for nonfatal myocardial infarction, stroke,or cardiovascular mortality (P=.02). On risk factor analysis,this risk reduction was primarily associated with thereduction in HbA1c during DCCT trial treatment; althoughmicroalbuminuria and albuminuria were associated withincreased risk for cardiovascular disease, the associationbetween the initial HbA1c reduction and reduction incardiovascular risk remained significant after adjustmentfor these factors.

3. HbA1c reduction in Type 2 diabetes

As shown in Fig. 1, the prospective observationalUKPDS-35 study demonstrated a continuous relationshipbetween HbA1c and microvascular and macrovascular riskamong patients with newly diagnosed Type 2 diabetes. Anepidemiological extrapolation of this study showed that each1% reduction in mean updated HbA1c was associated withsignificant decreases in risk for any diabetes end point,diabetes-related mortality, myocardial infarction, and micro-vascular complications, with no threshold of risk beingobserved for any end point (Stratton et al., 2000). Risk forcomplications is shown by HbA1c strata of b6%, 6–b7%

204 K. Benhalima et al. / Journal of Diabetes and Its Complications 25 (2011) 202–207

and 7–b8% in Fig. 2, with such findings again indicating thathaving HbA1c as close to normal as possible reduces risk ofall complications.

As did the DCCT trial in patients with Type 1 diabetes, theUKPDS showed that intensive glycemic control significantlyreduced the risk of microvascular complications in newlydiagnosed patients with Type 2 diabetes. In the UKPDS-33trial reported in 1998, intensive control with sulphonylureasor insulin was associated with a 25% reduction in risk ofmicrovascular end points (P=.0099) over 10 years comparedwith conventional control (UK Prospective Diabetes Study[UKPDS] Group, 1998a, 1998b).

As was the case in the DCCT trial, evidence of benefit inpreventing cardiovascular events was not straightforward inUKPDS-33, with intensive glycemic control being associat-ed with a borderline significant 16% reduction in risk(P=.052) of myocardial infarction. However, as was also thecase in the DCCT/EDIC follow-up study, 10-year follow-upof the UKPDS-33 population after the end of the studyshowed emergence of significant preventive benefits (Hol-man et al., 2008). Despite the absence of any difference inHbA1c between the initial intensive control group and theinitial conventional control group after 1 year of follow-upafter the end of the initial trial, intensive control wasassociated with a significant 15% reduction in risk formyocardial infarction (P=.01) and a significant 13%reduction in risk for all-cause mortality (P=.007), with theoriginal significant reductions in risk being maintained overfollow-up for any diabetes-related end point (9% reduction,P=.04) and microvascular complications (24% reduction,P=.001). The portion of the UKPDS-34 trial comparingintensive control with metformin versus conventionalcontrol in newly diagnosed overweight patients had shownsignificant reductions in risk for any diabetes-relatedendpoint, all-cause mortality, and myocardial infarctionwith intensive control (UK Prospective Diabetes Study[UKPDS] Group, 1998a, 1998b), and these benefits weremaintained over a 10-year follow-up, with risk reductions of21% (P=.01), 27% (P=.002) and 33% (P=.005) for any

Fig. 2. Adjusted rates of diabetes-related morbidity and mortality, all-causemortality, myocardial infarction (MI), stroke, and microvascular diseaseaccording to HbA1c strata in the UKPDS-35 study in Type 2 diabetes.(Stratton et al., 2000)

diabetes-related endpoint, all-cause mortality and myocardi-al infarction, respectively (Holman et al., 2008).

The data from the follow-up studies in both Type 1 andType 2 diabetes indicate that intensive glycemic control soonafter diagnosis yields long-term benefits in reducingmacrovascular risk. This benefit is probably because of areduced longterm exposure to hyperglycaemia. A comple-mentary mechanism might be the notion of “metabolicmemory,” the idea that early glycemic control is rememberedin the different target organs (such as the heart, kidneys andeyes), resulting in a long-term “legacy” benefit of reducedHbA1c. It may be that increased oxidative stress causesirreversible damage in the absence of early control ofglycemia. Indeed, the recently reported findings in theobservational follow-up of the Steno-2 study confirm thelong-term benefits of initial intensive control of cardiovas-cular risk factors. The Steno-2 study showed that intensivemultifactorial intervention with strict glycemic control,blood pressure control and use of aspirin and lipid-loweringtherapy led to an approximately 50% reduction in micro-vascular complications and cardiovascular disease over 7.8years, including a 53% reduction in risk for cardiovasculardisease, compared with conventional control in patients withType 2 diabetes and microalbuminuria (Gæde et al., 2003).After an additional 5.5 years of observation, during whichthe difference in risk factor control narrowed primarily due toimproved control in the original conventional therapy group,the initial intensive intervention group had a 57% reductionin risk for cardiovascular mortality (P=.04; 13% absolutereduction), a 59% reduction in risk for cardiovascular events(Pb.001) and a 46% reduction in risk for any-cause mortality(P=.02; absolute reduction 20%) (Gæde, Lund-Andersen,Parving, & Pedersen, 2008). These findings both confirmthat early aggressive control of risk factors produces long-term benefits and suggest that early intensive multifactorialintervention in Type 2 diabetes may be imperative foroptimal reduction of cardiovascular risk in the long term.

While these findings support intensive control of glycemiainitiated as soon as possible after diagnosis of Type 2diabetes, the debate over targets in lowering glucose has beenfueled by the recent reporting of three large-scale interventiontrials—the ACCORD (Action to Control CardiovascularRisk in Diabetes), ADVANCE (Action in Diabetes andVascular Disease: Preterax and Diamicron Modified ReleaseControlled Evaluation) and VADT (Veterans Affairs Diabe-tes Trial) trials (The Action to Control Cardiovascular Risk inDiabetes Study Group, 2008; The ADVANCE CollaborativeGroup, 2008; Duckworth et al., 2009)—conducted todetermine whether reducing HbA1c to near-normal levelscan reduce macrovascular complications. These trialscompared intensive and standard glycemic control in high-risk populations in which high event rates were expected:patients were on average 60 years or older and had a meanduration of diabetes of approximately 8–12 years; 30–40%had cardiovascular disease at entry. Median follow-up in thetrials was 3.5–5.6 years.

205K. Benhalima et al. / Journal of Diabetes and Its Complications 25 (2011) 202–207

While the conclusions from these trials are relevant for thestudied populations, care should be exercised in applyingthem to the broader population of patients with Type 2diabetes, including lower-risk patients. None of the trialsshowed benefit in reducing cardiovascular risk for intensiveversus standard treatment. The ADVANCE study showed asignificant 10% reduction in combined incidence of majormicrovascular and macrovascular events over 5 years thatprimarily reflected reduced risk of nephropathy (TheADVANCE Collaborative Group, 2008), supporting therole of intensive control in reducing microvascular risk. Datafrom the VADT trial indicated an alarming relationshipbetween sudden death and severe hypoglycemia within theprior 3 months of study treatment, re-emphasizing thedangers of hypoglycemia as an adverse effect of glucose-lowering treatment (Duckworth et al., 2009; Abraira, 2008).The ACCORD trial was stopped prematurely after 3.5 yearsdue to increased mortality in the intensive control group (TheAction to Control Cardiovascular Risk in Diabetes StudyGroup, 2008); this group was also characterized by greateruse of insulin alone and in combination with oral antidiabeticagents, significantly greater weight gain, and significantlygreater frequency of severe hypoglycemia compared with thestandard treatment group.

Despite numerous post hoc analyses and speculations(Del Prato, 2009; Lebovitz, 2008; Cukierman-Yaffe et al.,2009), the causes of increased mortality in the intensivecontrol group in the ACCORD trial remain unclear. It isunlikely that the achieved HbA1c is an explanation for thefinding, because similar median HbA1c levels were achievedin the ACCORD and ADVANCE studies (6.4% and 6.5%,respectively). However, as noted, findings in these studiesshould heighten awareness of the possibility that hypogly-cemia in high-risk patients may be an important factor inmasking the potential benefits of strict glycemic control inthe short term. Indeed, findings in these trials have reneweddiscussion regarding the detrimental effect of hypoglycemiaon the already ischemic heart. Many studies have shown thatcardiovascular disease can be precipitated by the hemody-namic changes associated with severe hypoglycemia(Cukierman-Yaffe et al., 2009; Tattersall & Gill, 1991;Ewing, Boland, Neilson, Cho, & Clarke, 1991). The counter-regulatory responses to hypoglycemia, including sympathet-ic-adrenal activation and catecholamine release, may havehazardous effects including increased risk of localizedischemia and electrolyte disturbances with cardiac arrhyth-mias, and risk for these events may be greater among patientswith preexisting endothelial dysfunction.

Long-term studies of glycemic control, such as theUKPDS, PROactive, ADVANCE, VADT and ACCORDstudies, have been completed, and several meta-analyseshave recently been published that examine the results ofthese important studies and the over 30,000 patients withType 2 diabetes who were enrolled in them (Mannucci,Monami, Lamanna, Gori, & Marchionni, 2009; Ray et al.,2009; Standl, Muller, & Schnell, 2009). In the meta-analyses

of all five studies by Mannucci et al. (2009) and Ray et al.(2009), it was found that intensified treatment in patientswith Type 2 diabetes led to a significant reduction in theoccurrence of non-fatal myocardial infarction (17%) andcoronary heart disease (15%) but did not have an affect onthe occurrence of stroke and cardiovascular mortality. Thesemeta-analyses found that intensive glycemic control signif-icantly reduced coronary events without an increased risk ofdeath but were associated with weight gain and an increase inhypoglycemic risk in patients with Type 2 diabetes. In themeta-analysis from Standl et al. (2009), which onlyexamined the VADT, ADVANCE and ACCORD studies,cardiovascular events were reduced by approximately 10–15% per 1% of absolute reduction of HbA1c. Comparingonly the ACCORD and ADVANCE studies led to the beliefthat the overaggressive treatment enforced in the ACCORDstudy, which targeted an HbA1c goal of ≤6.0%, may havehad the unintended result of increasing the mortality rate insome patients. In these two studies, the target HbA1c levelswere b6.0% (ACCORD) vs ≤ 6.5% (ADVANCE) and thehazard ratios were 1.22 vs 0.93, respectively. These threemeta-analyses point to the effect that the speed and extent ofHbA1c reduction can have on mortality rates in differentpopulations, especially those with pre-existing macrovascu-lar disease and Type 2 diabetes (Mannucci et al., 2009; Rayet al., 2009; Standl et al., 2009.

This is in agreement with a recent cohort study of patientsaged 50 years or older with Type 2 diabetes, who showedthat not only high, but also low mean HbA1c values wereassociated with increased all cause mortality and cardiacevents. The lowest risk was seen at a HbA1c of about 7.5%(Currie et al., 2010). Considering the fact that theseobservations were even stronger when studying the insulin-treated subgroup, hypoglycemia as cause for this excessmorbidity and mortality comes into the picture.

4. How low should we go?

The answer to the question “How low should we go?” inHbA1c lowering would seem to be “As close to normal aspossible” when considering epidemiologic data showing therelationship between HbA1c and microvascular and macro-vascular risk, as well as clinical trial data showing reductionof microvascular risk with intensive glycemic control in bothType 1 and Type 2 diabetes. Existing clinical trial data alsoclearly support the benefits of intensive control initiated soonafter diagnosis in reducing cardiovascular risk over the longterm in Type 1 diabetes. It seems clear that we should aim tolower HbA1c as much as possible as soon as possible afterdiagnosis in patients with Type 1 diabetes.

But how should the data from the recently reported studiesin Type 2 diabetes be integrated into recommendationsregarding treatment targets? The failure of the ACCORD,ADVANCE, and VADT trials to show any benefit ofintensive glycemic control in reducing cardiovascular risk

206 K. Benhalima et al. / Journal of Diabetes and Its Complications 25 (2011) 202–207

may imply that such disadvantages as weight gain andincreased frequency of severe hypoglycemia counterbalancepotential advantages in the short-term in patients with moreadvanced Type 2 diabetes. The UKPDS-33 follow-up study,on the other hand, showed that intensive control started soonafter diagnosis and sustained over many years leads tobenefits in reducing microvascular and macrovascularcomplications, as well as overall mortality risk over thelong term. It may be that, in addition to indicating elevatedrisk of poor outcome associated with hypoglycemia and otheradverse effects of antidiabetic agents in higher-risk patients,the three recent intervention studies highlight the importanceof the stage of disease at which intervention is initiated inpredicting potential benefit. Indeed, the aforementionedtrials, as well as their subset analyses, suggest a significantbenefit of intensive glycemic control in reducing cardiovas-cular disease risk in patients with shorter duration of diabetes,lower HbA1c at entry and/or absence of known cardiovas-cular disease at entry. And finally, assessing those studies inseveral meta-analyses together with other randomized trialsachieving at a difference of HbA1c lowering of at least 0.5%clearly showed a significant reduction of especially myocar-dial infarction in general and at least a neutral effect onmortality over the mean observation period of 5 years.

Taken together, these considerations suggest thatintensive glycemic control soon after the diagnosis ofType 2 diabetes has a positive effect on cardiovascular risk,although—as indicated by the findings in the UKPDS-33and Steno-2 follow-up studies—the benefit may becomeapparent only after 10 or more years of follow-up.

Treatment should be started as soon as possible afterdiagnosis. In patients with a good life expectancy (eg, N10years), shorter duration of diabetes, and no preexistingcardiovascular complications, we should aim at the lowestpossible HbA1c. For the present, less-stringent HbA1ctargets (N7.0–7.5%) may be advisable in patients withshorter life expectancy (eg, older patients, those with co-morbidities) and those with established cardiovascularcomplications. Individualisation of care is important. Thebest practice is to agree with the patient what the mostappropriate HbA1c target is in that particular setting. Thiswill depend on a whole range of patient and diseasecharacteristics. It is equally important to agree then with thepatient what the best way is to achive this and what measuresof success might be.

5. Conclusion

At present, the limitations in achieving low HbA1c arelargely imposed by the side effects of existing medications.An additional consideration, as seen from the ACCORDstudy, is that the rate and target level of HbA1c reductionshould be tailored to each patient. For many widely useddrugs, particularly sulphonylureas and insulin, the majorconcerns in this regard are hypoglycemia and weight gain.

The introduction of insulin analogues with a similar potentialfor glycemic control but a lower risk for hypoglycemiacompared with human insulin was an important step in thetreatment of patients with Type 1 diabetes and those withType 2 diabetes on insulin therapy. The development of newantidiabetic agents based on the incretin system, such as theincretin mimetics (eg, exenatide, liraglutide) and dipeptidylpeptidase-4 (DPP-4) inhibitors (e.g., sitagliptin, vildagliptin,saxagliptin), present new options and potential strategies forsafely achieving glucose targets. These drugs reducehyperglycemia in a glucose-dependent manner and thuspose little risk of hypoglycemia; they are also associated withweight loss (incretin mimetics) or a weight neutral effect(DPP-4 inhibitors).

The goal in treating hyperglycemia in diabetes is to starttreatment early and reduce HbA1c to as close to normal as issafely possible. We have the tools to begin to devisestrategies to accomplish this objective in the form of lifestyleintervention and antidiabetic agents that avoid the risksassociated with hypoglycemia and weight gain—e.g.,metformin and incretin-based agents. The overall goal intreating diabetes is to reduce risk of microvascular andmacrovascular complications, and glycemic control is animportant component of efforts to achieve this goal. Optimalprevention of diabetic complications likely requires amultifactorial intervention strategy—eg, glycemic, bloodpressure and lipid control—initiated early in the course ofdisease. The UKPDS, PROactive, ADVANCE, VADT andACCORD studies have yielded large amounts of useful datafor consideration, but further randomized trials are urgentlyneeded to investigate the best approach to lowering HbA1clevels in patients with Type 2 diabetes who also have pre-existing macrovascular disease.

References

Abraira, C. (2008). Veterans Affairs Diabetes Trial: background, design, riskfactor control, glycemic separation, and hypoglycemia. Presented at:American Diabetes Association–68th Annual Scientific Sessions; June6-11, San Francisco, CA.

Cukierman-Yaffe, T., Gerstein, H. C., Williamson, J. D., Lazar, R. M.,Lovato, L., Miller, M. E., Coker, L. H., Murray, A., Sullivan, M. D.,Marcovina, S. M., & Launer, L. J., for the Action to ControlCardiovascular Risk in Diabetes-Memory in Diabetes (ACCORD-MIND) Investigators. (2009). Relationship between baseline glycemiccontrol and cognitive function in individuals with Type 2 diabetes andother cardiovascular risk factors: the action to control cardiovascular riskin diabetes-memory in diabetes (ACCORD-MIND) trial. Diabetes Care,32(2), 221−226.

Currie, C. J., Peters, J. R., Tynan, A., Evans, M., Heine, R. J., Bracco, O. L.,Zagar, T., & Poole, C.D. (2010). Survival as a function of HbA1c in peoplewith Type 2 diabetes: a retrospective cohort study. Lancet, 375, 481−489.

Del Prato, S. (2009). Megatrials in Type 2 diabetes. From excitement tofrustration? Diabetologia, 52(7), 1219−1226.

Duckworth, W., Abraira, C., Moritz, T., Reda, D., Emanuele, N., Reaven,P. D., Zieve, F. J., Marks, J., Davis, S. N., Hayward, R., Warren, S. R.,Goldman, S., McCarren, M., Vitek, M. E., Henderson, W. G., & Huang,G. D., for the VADT Investigators. (2009). Glucose control and vascularcomplications in veterans with Type 2 diabetes. New England Journal ofMedicine, 360(2), 129−139.

207K. Benhalima et al. / Journal of Diabetes and Its Complications 25 (2011) 202–207

Ewing, D. J., Boland, O., Neilson, J. M. M., Cho, C. G., & Clarke, B. F.(1991). Autonomic neuropathy, QT interval lengthening, and unexpect-ed deaths in male diabetic patients. Diabetologia, 34(3), 182−185.

Gæde, P., Lund-Andersen, H., Parving, H. H., & Pedersen, O. (2008). Effectof a multifactorial intervention on mortality in Type 2 diabetes. NewEngland Journal of Medicine, 358(6), 580−591.

Gæde, P., Vedel, P., Larsen, N., Jensen, G. V., Parving, H. H., & Pedersen, O.(2003). Multifactorial intervention and cardiovascular disease in patientswith Type 2 diabetes.NewEngland Journal of Medicine, 348(5), 383−393.

Holman, R. R., Paul, S. K., Bethel, M. A., Matthews, D. R., & Neil, H. A.W. (2008). 10-year follow-up of intensive glucose control in Type 2diabetes. New England Journal of Medicine, 359(15), 1577−1589.

Khaw, K. T., Wareham, N., Bingham, S., Luben, R., Welch, A., & Day, N.(2004). Association of hemoglobin A1c with cardiovascular disease andmortality in adults: the European prospective investigation into cancer inNorfolk. Annals of Internal Medicine, 141(6), 413−420.

Krzentowski, G., Zhang, L., Albert, A., & Lefèbvre, P. (2004). Another lookat the implications of the DCCT study. Annals of Endocrinology (Paris),65(5), 429−435.

Lebovitz, H. E. (2008). Glycaemic control and vascular complications inType 2 diabetes: new observations and clinical significance. Journal ofthe Indian Medical Association, 106(11), 724−726.

Mannucci, E., Monami, M., Lamanna, C., Gori, F., & Marchionni, N.(2009). Prevention of cardiovascular disease through glycemic controlin Type 2 diabetes: A meta-analysis of randomized clinical trials.Nutrition, Metabolism, and Cardiovascular Disease, 19(9), 604−612.

Ray, K. K., Seshasai, S. R. K., Wijesuriya, S., Sivakumaran, R., Nethercott,S., Preiss, D., Erqou, S., & Sattar, N. (2009). Effect of intensive controlof glucose on cardiovascular outcomes and death in patients withdiabetes mellitus: a meta-analysis of randomized controlled trials. TheLancet, 373(9677), 1765−1772.

Standl, E., Muller, M., & Schnell, O. (2009). The impact of glucose-lowering therapy on cardiovascular outcomes. Best Practice andResearch: Clinical Endocrinology and Metabolism, 23(3), 401−411.

Stratton, I. M., Adler, A. I., Neil, H. A. W., Matthews, D. R., Manley, S.E., Cull, C. A., Hadden, D., Turner, R. C., & Holman, R. R. (2000).Association of glycaemia with macrovascular and microvascularcomplications of Type 2 diabetes (UKPDS 35): prospectiveobservational study. British Medical Journal, 321(7258), 405−412.

Tattersall, R. B., & Gill, G. V. (1991). Unexplained deaths of Type 1diabetic patients. Diabetic Medicine, 8(1), 49−58.

The Action to Control Cardiovascular Risk in Diabetes Study Group.(2008). Effects of intensive glucose lowering in Type 2 diabetes. NewEngland Journal of Medicine, 358(24), 2545−2559.

The ADVANCE Collaborative Group. (2008). Intensive blood glucosecontrol and vascular outcomes in patients with Type 2 diabetes. NewEngland Journal of Medicine, 358(24), 2560−2572.

The Diabetes Control and Complications Trial/Epidemiology of DiabetesInterventions and Complications (DCCT/EDIC) Study Research Group.(2005). Intensive diabetes treatment and cardiovascular disease inpatients with Type 1 diabetes. New England Journal of Medicine, 353(25), 2643−2653.

The Diabetes Control and Complications Trial Research Group. (1993). Theeffect of intensive treatment of diabetes on the development andprogression of long-term complications in insulin-dependent diabetesmellitus. New England Journal of Medicine, 329(14), 977−986.

The Diabetes Control and Complications Trial Research Group. (1996). Theabsence of a glycemic threshold for the development of long-termcomplications: the perspective of the Diabetes Control and Complica-tions Trial. Diabetes, 45(10), 1289−1298.

UK Prospective Diabetes Study (UKPDS) Group. (1998a). Effect of intensiveblood-glucose control with metformin on complications in overweightpatients with Type 2 diabetes (UKPDS 34). The Lancet, 352(9131),854−865.

UK Prospective Diabetes Study (UKPDS) Group. (1998b). Intensive blood-glucose control with sulphonylureas or insulin compared withconventional treatment and risk of complications in patients with Type2 diabetes (UKPDS 33). The Lancet, 352(9131), 837−853.