Download this issue

New advances in the treatment of type 2 diabetes

MEDICOGRAPHIA No. 100!

• Medicographia
  J. Servier, France

EDITORIAL

• ADVANCE: setting new standards for optimal management of patients with type 2 diabetes. ADVANCE : nouvelles normes pour la prise en charge optimale des patients atteints de diabète de type 2
  J. Chalmers

THEMED ARTICLES

• Major findings from ADVANCE: blood pressure–lowering arm
  N. R. Poulter
• Major findings from the ADVANCE study: glucose-lowering arm
  M. E. Cooper
• Implications of ADVANCE in the management of blood pressure in diabetic patients
  B. Williams
• Implications of ADVANCE in the management of glucose lowering in diabetic patients
  D. R. Matthews
• The burden of vascular disease in diabetes and hypertension: from micro- to macrovascular disease—the “bad loop”
  H. A. J. Struijker-Boudier
• Advanced glycation end products (AGEs) and their receptors (RAGEs) in diabetic vascular disease
  P. Marchetti
• How should future guidelines implement the results of ADVANCE?
  M. Marre

CONTROVERSIAL QUESTION

• Is microalbuminuria a marker for microangiopathy or macroangiopathy?
  M. Burnier – P. Fioretto – J. Gumprecht – G. Halaby – R. Unnikrishnan and V. Mohan – F. Puchulu – R. Roussel – G. Schernthaner – M. Shestakova – J.-G. Wang, Y. Li, and C.-S. Sheng

DIAMICRON MR – PRETERAX

• Management of type 2 diabetes: a multifactorial approach to a complex disease
  S. Laroche and S. Corda

INTERVIEW

• How can the results of ADVANCE be applied to daily clinical practice?
  J. Bringer

FOCUS

• Blood pressure, blood glucose, and diabetic renal disease
  C. E. Mogensen

UPDATE

• Will ADVANCE population genomic determinants improve upon
  biomarkers in predicting vascular complications of diabetes?
  P. Hamet and J. Tremblay

A TOUCH OF FRANCE

• Famous French diabetics
  C. Régnier
• Late Renoir, 1892-1919
  I. Spaak

Download this issue

by Dr Jacques Servier

With this issue of Medicographia, we are celebrating a very special—dual—anniversary: 30 years and No. 100, of a quarterly medical journal with a circulation of more than 15 000 copies in more than 130 countries. We take special pride, at Servier, in this anniversary. Just look around and count how many other medical journals of this kind—published by a pharmaceutical company and so highly respected by a demanding audience of hospital and community clinicians and institutional researchers—have been able to pass with flying colors this test of time. I personally know of none other. Medicographia stands alone and unchallenged.

We thought it was high time you, our readers, were asked, in an international survey carried out last year, about what you wanted to keep, and what changes would improve the journal. As your replies came in by fax and e-mail, we were both surprised and thrilled to discover that although you thought the journal could do with a facelift to make the layout lighter, simpler, clearer, more colorful, and with somewhat shorter articles, a larger font, and highlights to draw the attention to points of relevance, the very structure itself of Medicographia was still earning unreserved kudos, and that you wanted it to be kept as it was.

This wish met one of our core values more than half way. Indeed, Medicographia is but one example among many of how obsessed we are with excellence in the long term, whether this durability concerns our research and our discoveries, or our partnerships, our projects, our commitments. As a company, we have jealously preserved our independence by resisting the lure of the stock exchange and the tyranny of short-term results driven by quarterly dividends—a freedom that has recently, in this time of turmoil, acquired special meaning. It is that very freedom that enables us to commit to the long term, which is one of the key features of the “Servier style” that is most appreciated by those who have come to rely on our projects, our drugs, and our services to research, the medical community, and the patients who benefit from our therapeutic innovations.

The main reason for Medicographia’s longevity resides in the fact that it offers a unique form of partnership between Servier and field leaders worldwide in giving doctors across all sectors of medicine an overview on a given subject. Few journals currently offer state-of-the-art summaries aimed, not just at specialists in the field, but at all clinicians, allowing them to keep abreast of developments in fields other than their own and remain physicians of the whole person, rather than of a discrete “slice,” whether organ, disease, or specialty. Medicographia could be described as a two-way partnership, since its quality makes it an ambassador of Servier excellence, both to a readership that enjoys a broad sweep of information on a particular topic and to field leaders in university hospitals and public or private research institutions. For them, it serves as a tribune from which, alongside their peers, they can write papers that analyze developments within their specialty with complete editorial independence.

Another aim of the journal, in its own modest way, is to represent France in the eyes of international doctors who are more familiar with a world dominated by the English-language medical press. Medicographia editorials and abstracts are routinely bilingual, in French and English, and each issue features two cultural articles, each profusely illustrated, in the “Touch of France” section. One article touches on the history of medicine, based on a great figure in French medical history, while the other addresses broader topics, such as art, literature, and other aspects of France’s rich cultural heritage.

This is what Medicographia has been offering for 30 years, and, with its new layout and thanks to our readers’ unwavering support, will continue to offer for, we hope, many more years to come.

Yours truly,

Doctor Jacques Servier

Download this issue

by J. Chalmers, Australia John CHALMERS, FAA, MD, PhD, FRACP The George Institute University of Sydney Sydney, NSW AUSTRALIA

Address for correspondence: Prof John Chalmers, The George Institute for International Health, PO Box M201, Missenden Road, NSW 2050, Australia (e-mail: jchalmers@george.org.au) MEDICOGRAPHIA. 2009;31:217-222.

With completion of patient follow-up and publication of the main results of both its blood pressure–lowering arm¹ and its intensive glucose control arm,² the Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation (ADVANCE) takes over the mantle as standard bearer for the management of patients with type 2 diabetes in the 21st century. The United Kingdom Prospective Diabetes Study (UKPDS) was clearly the outstanding study on the management of type 2 diabetes in the 20th century. It established the benefit of blood pressure lowering in hypertensive patients with type 2 diabetes, demonstrating reductions in mortality and in macrovascular disease.sup>3 Similar findings were obtained from subgroup analyses in patients with type 2 diabetes from the HOT trial (Hypertension Optimal Treatment) trial⁴ and the MIcroalbuminuria, Cardiovascular, and Renal Outcomes–Heart Outcomes Prevention Evaluation (MICRO-HOPE).⁵ UKPDS also demonstrated that tighter glucose control, ie, lowering glycosylated hemoglobin (HbA1c) by about one percentage point, reduced the burden of microvascular disease, predominantly through reduction in retinal disease.⁶ UKPDS was unable to show any reduction in macrovascular disease during the course of the trial.⁶

ADVANCE now takes this further, by demonstrating that routine blood pressure lowering with the fixed combination of perindopril and indapamide in patients with type 2 diabetes, whether hypertensive or not, reduces mortality and prevents both coronary disease and diabetic nephropathy.¹ ADVANCE has also shown that more intensive glucose lowering with a regimen based on gliclazide MR (modified release) reduces major vascular events, predominantly as a result of reduction in microvascular disease, driven by substantial reduction in new or worsening diabetic nephropathy. While there was no significant reduction in macrovascular events, all-cause mortality, or cardiovascular death, the UKPDS confirmation that a sulfonylurea-insulin regimen has a legacy effect with late benefit in these outcomes⁷ suggests that the trend towards a benefit seen for these outcomes in ADVANCE, particularly in the last year of follow-up, should translate into real benefits over time.⁸

Exciting new analyses have confirmed that the joint effects of these two treatments— the fixed combination of perindopril and indapamide and the gliclazide MR–based intensive glucose control regimen—result in very substantial reductions in mortality and morbidity.⁹ This combined effect reduces all-cause mortality by one fifth, cardiovascular mortality by one quarter, and new or worsening nephropathy by one third.⁹ Thus, ADVANCE has set the standard and the pace for the new century.

ADVANCE: design and rationale

ADVANCE was conceived and planned, in 2000, to take us beyond UKPDS and beyond available evidence, by addressing two pressing problems in the management of patients with type 2 diabetes. The first problem related to the benefits of blood pressure lowering, since all recommendations and guidelines went beyond the evidence from UKPDS, to recommend that the blood pressure thresholds and targets for treatment should be around 130/80 mm Hg, well below the range established in UKPDS.³ The second problem concerned the benefits of tighter glucose control, since the major international guidelines again went beyond the evidence from UKPDS, advocating reductions in HbA1c to levels equal to or less than 6.5% or 7%. ADVANCE was therefore initiated as a randomized 2_2 factorial study with two arms.¹⁰ One arm was established as a double-blind comparison of the fixed combination of perindopril and indapamide compared with placebo in patients with type 2 diabetes, irrespective of initial blood pressure, thus enrolling both normotensive and hypertensive patients.^1,10 The other arm was established as an open, randomized comparison with a prospective, open, blinded end point (PROBE) design, of an intensive gliclazide MR–based glucose control regimen, targeting an HbA1c of 6.5% or less, against standard, guideline-based glucose control regimens.^2,10 For both sets of comparisons, the primary outcomes were defined as composites of major macrovascular disease and of major micro-vascular disease, analyzed jointly and separately.^1,2

The fixed combination of perindopril and indapamide produced significant reductions in cardiovascular morbidity and mortality, which translated into substantial absolute benefits. Thus, it can be estimated that 5 years of treatment with a single tablet of perindopril-indapamide once daily would prevent 1 death among every 79 patients so treated, 1 major vascular event among every 66 patients, 1 coronary event among every 75 patients, and 1 renal event among every 20 patients so treated.¹

Furthermore, these benefits were all obtained in a group of patients receiving comprehensive cardiovascular care, including blood pressure–lowering drugs in over 75% of all participants, oral hypoglycemic agents in over 90%, and statins and aspirin in over 50%.¹ As a result, the participants in ADVANCE had much lower risk profiles and event rates than those in the UKPDS, HOT, and MICRO-HOPE studies.^3-5 Importantly, all the benefits observed were consistent, whether the patients were hypertensive or not, and whatever concomitant therapies they were receiving, including inhibitors of the renin-angiotensin system.¹ Furthermore, the treatment was remarkably well tolerated, with adherence to randomized treatment similar in the active treatment group and the placebo- treated group.¹ It can be estimated that if all 250 million people alive with diabetes today were treated with one tablet of perindopril-indapamide daily, over 3 million lives would be saved over 5 years. It clearly behoves all physicians to consider treatment with this fixed-dose combination whenever they see a patient with type 2 diabetes.

ADVANCE: lessons and implications from treatment with the intensive gliclazide MR–based regimen

The gliclazide MR–based intensive glucose control regimen considerably reduced vascular disease, largely as a result of great reductions in microvascular and renal disease. These translated into considerable absolute benefits. It can be estimated that for every 52 patients undergoing intensive control with this regimen for 5 years, 1 major macrovascular or microvascular event would be averted.² Furthermore, 1 major renal event would be averted among every 20 patients with type 2 diabetes, so treated. These benefits were obtained with very acceptable rates of hypoglycemia—only 7 cases of severe hypoglycemia per 1000 patients per annum, and without any weight gain, on average, in the intensively treated group.

One of the most important outcomes, given the early termination of the intensive glucose control arm of the Action to Control CardiOvascular Risk in Diabetes (ACCORD) trial on account of excess mortality,¹¹ was the trend toward a reduction in both all-cause mortality (7%; nonsignificant) and in cardiovascular death (12%; nonsignificant). There was no trend toward an increase in mortality in any subgroup of ADVANCE, including those most closely resembling the ACCORD cohort. The much greater safety of the ADVANCE regimen cannot be attributed to the level of the HbA1c achieved (a mean of 6.5%) since this was very similar in the two studies.^2,11 It must be associated with the very real differences in the treatment strategies used in these two trials.^2,11 The strategy in ADVANCE started with oral hypoglycemic agents in a progressive and incremental manner, beginning with gliclazide MR, once daily, then increasing the dose of gliclazide MR to the point that 70% of patients were receiving the maximum dose of 120 mg daily. Other hypoglycemic agents were added progressively, so that by the end of follow-up, 91% were on gliclazide MR, 67% on metformin, and only 17% on thiazolidinediones. Insulin was only added if these measures failed, beginning with basal, nocturnal insulin, and only moving to multiple injection insulin if really needed further down the track. Only 40% of patients were on insulin at the end of follow-up in the intensively treated group of ADVANCE, compared with 24% in the standard treatment group.² This should be contrasted with the much more aggressive strategy used in ACCORD, where the intention was to reduce HbA1c to a target of 6.0% as soon as possible, with the use of multiple oral hypoglycemic agents and insulin from the first few months. By the time the intensive glucose arm of ACCORD was terminated, after an average of 3.5 years of follow-up, over 90% were receiving thiazolidine- diones, over 90% were receiving metformin, most were on 4 or more agents, and 77% were receiving insulin, most often multidose.¹¹ Not surprisingly, the annual rate of severe hypoglycemia in the ACCORD study was 6 to 7 times greater than that seen in ADVANCE.^2,11

ADVANCE: conclusions

The great benefits observed in the separate and joint effects of the perindopril-indapamide intervention and the gliclazide MR–based regimen are founded on very practical and pragmatic strategies suited to everyday practice, and on the particularly appropriate properties of the study drugs for use in patients with type 2 diabetes. Patients with diabetes are notorious for not tolerating elevated blood pressure, largely because of progressive stiffening of large arteries coupled with progressive damage to small vessels, resulting in poor tissue perfusion. The properties of the fixed combination of perindopril and indapamide are made to order for this situation, with very effective lowering of blood pressure, coupled with a great efficacy in combating arterial stiffness and in enhancing tissue perfusion through the microcirculation.^12-16 Lowering blood glucose is clearly paramount in the treatment of patients with type 2 diabetes, and the gliclazide MR–based regimen used in ADVANCE has proved to be remarkably safe and effective in controlling the metabolic aspects of the disease as well as the renal and microvascular complications.² The modified-release formulation of the sulfonylurea gliclazide MR is at least as effective as other sulfonylureas in lowering blood sugar, but causes less hypoglycemia than most.¹⁷ This hypoglycemic agent formed an excellent base for a pragmatic glucose control strategy with progressive intensification in a manner well suited to clinical practice worldwide. The Steno 2 trial has confirmed the great value of a multifactorial approach to the control and prevention of the most serious complications of type 2 diabetes.¹⁸ The results of ADVANCE, the largest ever study of treatment in patients with diabetes, clearly justify a recommendation that the multifactorial regimen used in all patients with type 2 diabetes should include routine blood pressure lowering, regardless of initial blood pressure and intensive glucose control targeting an HbA1c around 6.5%. Furthermore, the outcomes observed with the two arms of ADVANCE suggest that the two strategies used in ADVANCE—routine blood pressure lowering with the fixed combination of perindopril and indapamide and intensive glucose lowering with a gliclazide MR–based regimen— should be considered for all patients with type 2 diabetes across the world. This would yield enormous benefits and avert millions of deaths, major vascular events, and cases of renal complications of diabetes among the 250 million people alive with this disease today.

References

ADVANCE : nouvelles normes pour la prise en charge optimale des patients atteints de diabète de type 2 par J. Chalmers, Australie

L’ achèvement du suivi des patients et la publication des principaux résultats des deux bras (abaissement de la pression artérielle1 et contrôle intensif de la glycémie2) de l’étude ADVANCE (Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation) a permis d’établir les normes de la prise en charge des diabétiques de type 2 pour le XXIe siècle. L’étude UKPDS (United Kingdom Prospective Diabetes Study) avait établi quant à elle les principes de la prise en charge du diabète de type 2 au XXe siècle. Elle avait démontré le bénéfice d’une réduction de la pression artérielle chez les patients hypertendus atteints de diabète de type 2, en mettant en évidence des réductions de la mortalité et des maladies macrovasculaires3. Des résultats similaires avaient été obtenus avec des analyses de sous-groupes effectuées chez des patients diabétiques de type 2 ayant participé à l’étude HOT (Hypertension Optimal Treatment)4 et à l’étude MICRO-HOPE (MIcroalbuminuria, Cardiovascular, and Renal Outcomes–Heart Outcomes Prevention Evaluation)5. L’étude UKPDS a également démontré qu’un contrôle plus étroit de la glycémie, c’est-à-dire une réduction de l’hémoglobine glycosylée (HbA1c) d’environ un point de pourcentage, permettait de diminuer l’impact des maladies microvasculaires, principalement par la réduction des pathologies rétiniennes.6 L’étude UKPDS n’a pas été en mesure de mettre en évidence une réduction des maladies macrovasculaires au cours de son déroulement6.

L’étude ADVANCE prend désormais le relais en démontrant qu’une réduction systématique de la pression artérielle par l’association fixe de perindopril et d’indapamide chez des patients atteints de diabète de type 2, qu’ils soient hypertendus ou non, permet de réduire la mortalité et de prévenir à la fois les maladies coronariennes et la néphropathie diabétique 1. L’étude ADVANCE a également démontré qu’un abaissement plus intense de la glycémie par un schéma thérapeutique à base de gliclazide MR (= modified release, à libération modifiée) minorait les événements vasculaires majeurs, principalement à la suite d’une réduction des pathologies microvasculaires, due en particulier à une diminution significative des nouveaux cas ou à des exacerbations de la néphropathie diabétique2. Malgré l’absence de réduction significative des événements macrovasculaires, de la mortalité de toute cause ou de la mortalité cardio-vasculaire, la confirmation apportée par l’étude UKPDS de l’existence d’un « legacy effect » (bénéfices à très long terme), obtenu grâce à un schéma thérapeutique associant un sulfamide hypoglycémiant et l’insuline7, suggère que la tendance vers l’obtention d’effets positifs observée dans les données de l’étude ADVANCE, en particulier au cours de la dernière année du suivi, devrait se traduire par l’obtention de résultats positifs concrets avec le temps8.

De nouvelles analyses extrêmement intéressantes ont confirmé que les effets conjoints de ces deux traitements – l’association fixe perindopril plus indapamide et le contrôle intensif de la glycémie reposant sur un traitement à base de gliclazide MR – permettent d’obtenir des réductions très substantielles de la mortalité et de la morbidité.9 Cet effet combiné réduit la mortalité de toute cause d’un cinquième, la mortalité cardio-vasculaire d’un quart et les nouveaux cas et les aggravations de néphropathie d’un tiers 9. Par conséquent, l’étude ADVANCE a fixé les normes ainsi que la marche à suivre pour ce nouveau siècle.

ADVANCE : schéma et justificatif

L’étude ADVANCE a été conçue et programmée en 2000, afin de prendre le relais de l’étude UKPDS et progresser audelà des données disponibles, en s’attaquant à deux problèmes aigus rencontrés dans la prise en charge des diabétiques de type 2. Le premier problème concerne les bénéfices d’un abaissement de la pression artérielle, dans la mesure où toutes les recommandations et toutes les directives, dépassant les faits recueillis au cours de l’étude UKPDS, ont recommandé des seuils et des valeurs thérapeutiques cibles pour la pression artérielle d’environ 130/80 mm Hg, c’est-àdire très inférieurs aux chiffres établis dans l’étude UKPDS3. Le second problème avait trait au bénéfice d’un contrôle plus strict de la glycémie, dans la mesure où les principales directives internationales allaient elles aussi au-delà des conclusions de l’étude UKPDS, recommandant des réductions de l’HbA1c à des valeurs inférieures ou égales à 6,5 % ou 7 %. L’étude ADVANCE a par conséquent été mise en oeuvre selon un schéma factoriel 2_2 comportant deux bras10. Dans l’un des bras, une comparaison en double aveugle a été effectuée entre une association fixe perindopril plus indapamide et un placebo chez des patients diabétiques de type 2, quelle que soit leur pression artérielle initiale, incluant par conséquent des sujets normotendus et hypertendus1,10. L’autre bras a comporté une comparaison ouverte randomisée, reposant sur un schéma prospectif, ouvert et en aveugle vis-à-vis du critère, entre un contrôle strict de la glycémie utilisant un schéma à base de gliclazide MR et visant une valeur d’HbA1c inférieure ou égale à 6,5 %, et des schémas de contrôle de la glycémie basés sur les directives standard 2,10. Pour les deux comparaisons, les critères primaires ont été des paramètres composites rassemblant les principales pathologies macrovasculaires et microvasculaires, analysés simultanément et séparément1,2.

ADVANCE : enseignements et conséquences du traitement par l’association fixe perindopril plus indapamide

L’association fixe de perindopril et d’indapamide a provoqué des réductions significatives de la morbidité et de la mortalité cardio-vasculaires qui se sont traduites par des bénéfices absolus significatifs. Par conséquent, il peut être estimé qu’un traitement de 5 ans par un seul comprimé de perindopril- indapamide une fois par jour permet de prévenir un décès pour 79 patients recevant ce type de traitement, un événement vasculaire majeur pour 66 patients, un événement coronaire pour 75 patients, et un événement rénal pour 20 patients ainsi traités1.

En outre, ces bénéfices ont tous été obtenus dans un groupe de patients recevant des soins cardio-vasculaires très suivis, comprenant notamment des antihypertenseurs chez plus de 75 % des participants, des hypoglycémiants oraux chez plus de 90 % de ces patients et des statines et de l’aspirine chez plus de 50 % d’entre eux1. Par conséquent, les participants de l’étude ADVANCE présentaient des profils de risque et des taux d’événements beaucoup plus favorables que ceux des études UKPDS, HOT et MICRO-HOPE3-5. Il est important de souligner que tous les bénéfices observés ont été constants, que les patients aient été hypertendus ou non, et quels qu’aient été les traitements concomitants administrés, notamment des inhibiteurs du système rénine-angiotensine1. En outre, le traitement a été remarquablement bien toléré, avec une observance du traitement randomisé similaire dans le groupe du médicament actif et le groupe placebo1. Par conséquent, si les 250 millions de personnes atteintes de diabète aujourd’hui étaient traitées par un comprimé de perindoprilindapamide une fois par jour, on estime que plus de 3 millions de vies pourraient être sauvées sur une période de 5 ans. Ces chiffres doivent inciter tous les médecins à envisager un traitement par cette association fixe chez tous les patients atteints de diabète de type 2 venant les consulter.

ADVANCE : enseignements et conséquences du traitement par le schéma intensif à base de gliclazide MR

Le schéma de contrôle intense de la glycémie reposant sur l’administration de gliclazide MR a entraîné des réductions considérables des maladies vasculaires, principalement à la suite d’importantes diminutions des pathologies microvasculaires et rénales. Ces réductions se sont traduites par des bénéfices absolus considérables. Il peut être estimé que pour 52 patients soumis à un contrôle intensif par ce schéma thérapeutique pendant 5 ans, un événement macrovasculaire ou microvasculaire majeur peut être prévenu2. En outre, un événement rénal majeur peut être évité pour 20 patients diabétiques de type 2 ainsi traités. Ces bénéfices ont été obtenus avec des taux très acceptables d’hypoglycémie – seulement 7 cas d’hypoglycémie sévère pour 1 000 patients par an, sans aucun gain de poids, en moyenne, dans le groupe traité intensivement.

L’un des résultats les plus importants, compte tenu de l’interruption prématurée du bras de contrôle intensif de la glycémie de l’étude ACCORD (Action to Control CardiOvascular Risk in Diabetes), due à une mortalité excessive11, a été la tendance vers une réduction de la mortalité de toute cause (7 % ; non significatif) et de la mortalité cardio-vasculaire (12 % ; non significatif). Il n’a été observé aucune tendance vers une augÉ mentation de la mortalité dans aucun des sous-groupes de l’étude ADVANCE, y compris les plus proches de la cohorte de l’étude ACCORD. La sécurité d’emploi très supérieure observée avec le schéma thérapeutique de l’étude ADVANCE ne peut pas être attribuée au niveau de l’HbA1c atteint (en moyenne 6,5 %), car il était très similaire dans les deux études2,11. Elle doit en revanche être associée aux différences marquées entre les stratégies thérapeutiques utilisées dans ces deux essais2,11. La stratégie de l’étude ADVANCE a débuté avec des hypoglycémiants oraux administrés de manière progressive et graduellement croissante, en commençant avec le gliclazide MR une fois par jour, puis en augmentant la posologie du gliclazide MR jusqu’à ce que 70 % des patients reçoivent la dose maximale de 120 mg/jour. D’autres hypoglycémiants ont été ajoutés progressivement de telle sorte qu’à la fin du suivi 91 % des patients étaient traités par le gliclazide MR, 67 % par la metformine, et seulement 17 % par les thiazolidinediones. L’insuline n’a été ajoutée qu’en cas d’insuffisance de résultats thérapeutiques, en commençant par le schéma insulinique de base nocturne, pour ne progresser vers de multiples injections d’insuline qu’en cas de besoin réel, puis en revenant à la dose initiale. Seulement 40 % des patients étaient sous insuline à la fin du suivi dans le groupe recevant le traitement intensif de l’étude ADVANCE, par rapport à 24 % dans le groupe du traitement standard2. Cette observation doit être mise en perspective avec la stratégie beaucoup plus agressive utilisée dans l’étude ACCORD, au cours de laquelle l’objectif était de réduire l’HbA1c à une valeur cible de 6,0 % dès que possible, par l’utilisation d’hypoglycémiants oraux et d’insuline dès les premiers mois. Lorsque le bras du contrôle intensif de la glycémie de l’étude ACCORD a été interrompu, après une moyenne de 3,5 ans de suivi, plus de 90 % des patients recevaient des thiazolidinediones, plus de 90 % des patients recevaient de la metformine, la plupart étaient traitées par au moins 4 médicaments, et 77 % recevaient de l’insuline, souvent en multidoses9. Il n’est pas surprenant que le taux annuel d’épisodes d’hypoglycémie sévère dans l’étude ACCORD ait été 6 à 7 fois supérieur à celui observé dans l’étude ADVANCE 2,11.

ADVANCE : conclusions

Les bénéfices importants obtenus par les effets séparés et simultanés de l’association perindopril-indapamide et du schéma à base de gliclazide MR sont fondés sur des stratégies pratiques et pragmatiques, adaptées à la pratique quotidienne, et sur les propriétés particulièrement appropriées de ces médicaments chez les patients atteints de diabète de type 2. Les patients diabétiques tolèrent mal l’hypertension artérielle, principalement à cause de la rigidification progressive des grandes artères associée à des lésions graduelles des petits vaisseaux entraînant une altération de la perfusion tissulaire. Les propriétés de l’association fixe perindopril plus indapamide sont particulièrement adaptées à cette situation, dans la mesure où elle entraîne une réduction efficace de la pression artérielle, tout en luttant efficacement contre la rigidité artérielle et en favorisant la perfusion tissulaire par la microcirculation12-16.

La réduction de la glycémie est essentielle dans le traitement des patients atteints de diabète de type 2, et le schéma à base de gliclazide MR utilisé dans l’étude ADVANCE s’est avéré remarquablement sûr et efficace dans le contrôle des aspects métaboliques de la maladie, ainsi que des complications rénales et microvasculaires2. La formulation à libération modifiée du gliclazide MR est au moins aussi efficace que les autres sulfamides hypoglycémiants pour la réduction de la glycémie, mais elle entraîne moins d’épisodes d’hypoglycémie que la plupart d’entre eux17. Ce sulfamide hypoglycémiant constitue une base excellente pour la stratégie pragmatique de contrôle de la glycémie reposant sur une intensification progressive, qui peut être instaurée d’une façon parfaitement adaptée à la pratique clinique à travers le monde.

L’étude Steno 2 a confirmé l’intérêt d’une approche multifactorielle pour la lutte contre les complications les plus graves du diabète de type 2 et leur prévention18. Les résultats de l’étude ADVANCE, la plus importante étude jamais réalisée sur le traitement des patients diabétiques, justifient parfaitement d’inclure dans le schéma multifactoriel utilisé chez tous les patients atteints de diabète de type 2 un abaissement systématique de la pression artérielle, quelle que soit sa valeur initiale, et un contrôle intensif de la glycémie, avec une valeur cible de l’HbA1c d’environ 6,5 %. En outre, les résultats observés dans les deux bras de l’étude ADVANCE suggèrent que les deux stratégies utilisées dans cet essai – abaissement systématique de la pression artérielle par l’association fixe perindopril plus indapamide et réduction intensive de la glycémie à l’aide d’un schéma thérapeutique à base de gliclazide MR – doivent être envisagées chez tous les patients atteints de diabète de type 2 à travers le monde. L’utilisation de ce traitement pourrait apporter des bénéfices considérables, empêcher des millions de décès, et prévenir les événements vasculaires majeurs et les complications rénales du diabète, chez les 250 millions de personnes aujourd’hui atteintes de cette maladie.

Download this issue

by N. R. Poulter, United Kingdom Neil R. POULTER, MBBS, MSc, FRCP International Centre for Circulatory Health (ICCH) Imperial College London UNITED KINGDOM

Background: Although guidelines throughout the world recommend lower blood pressure (BP) treatment thresholds and lower BP targets for patients with type 2 diabetes, the evidence base for so-doing is limited.
Methods: As part of a 2_2 factorial design, the Action in Diabetes and Vascular disease: PreterAx and DiamicroN Controlled Evaluation (ADVANCE) trial randomized 11 140 patients with type 2 diabetes, and at least 1 other prespecified determinant of cardiovascular (CV) risk, to receive perindopril (4 mg)/indapamide (1.25 mg) or placebo, irrespective of whether other antihypertensive medication was being used and irrespective of BP level.
Results: During an average follow-up of 4.3 years, BP was lowered, on average, by 5.6/2.2 mm Hg among those on perindopril/indapamide compared with placebo, to a level of 135/75 mm Hg. This BP reduction was associated with a significant 9% reduction (95% confidence interval [CI], 0.83-1.00; P=0.04) in the primary end point (major macrovascular or microvascular events), an 18% (0.68-0.98) reduction in CV mortality and a 14% (0.75-0.98) reduction in all-cause mortality. New or worsening nephropathy was reduced by 18% (0.68-1.01), but no effects were apparent on retinopathy. CV benefits were apparent across all subgroups, and occurred irrespective of baseline BP and/or background use of angiotensin-converting enzyme inhibition.
Conclusion: Patients with type 2 diabetes should be routinely considered for the addition of the type of treatment used in the ADVANCE trial, irrespective of their baseline BP levels.

Background and rationale

Intervention on major cardiovascular (CV) risk factors has traditionally been based on the absolute levels of the risk factors in question (eg, fasting plasma glucose >7 mmol/L or systolic blood pressure >140 mm Hg). More recently, the concept of intervention being based on, or at least influenced by, an estimate of total CV risk has been introduced into guidelines.^1,2

Prior to the conduct and publication of the Action in Diabetes and Vascular disease: PreterAx and DiamicroN Controlled Evaluation (ADVANCE) trial,³ two things relating to blood pressure (BP) among people with diabetes were clear. Firstly, there was a strong dose-response relationship between BP and risk of both macrovascular and microvascular events across the whole range of BP, irrespective of “hypertensive” status.^4,5 Secondly, findings were consistent from randomized trial data (albeit rel- atively limited in size) that greater BP lowering was superior to lesser BP lowering in terms of preventing major CV events.⁶ Interestingly, whilst guidelines among the world vary on several key issues relating to optimal BP management, all guidelines are consistent in suggesting a lower BP target (<130/80 mm Hg) for patients with type 2 diabetes.^1,2 Paradoxically, there are no good data, based on morbidity and mortality trials, to support these specifically lower BP levels recommended for diabetic patients and accepted throughout the world!

Figure 1. Trial design for the blood pressure–lowering arm of ADVANCE. The original power calculations were based on a total of 10 000 participants with type 2 diabetes, with an annual event rate of 3% and requiring an average follow-up of 4.5 years. All participants entered a preliminary 6-week open run-in phase, during which they received one tablet daily of perindopril 2 mg – indapamide 0.625 mg (Preterax).
Abbreviation: ADVANCE, Action in Diabetes and Vascular disease: PreterAx and DiamicroN Controlled Evaluation.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

Prior to the ADVANCE trial, data on BP lowering in diabetes were available from the results of the United Kingdom Prospective Diabetes Study (UKPDS),⁷ the Hypertension Optimal Treatment (HOT) trial,⁸ and the MIcroalbuminuria, Cardiovascular, and Renal Outcomes in the Heart Outcomes Prevention Evaluation (MICRO-HOPE) trial.⁹

UKPDS included a small BP-lowering limb comparing more versus less intensive BP lowering, but only among frankly hypertensive patients with diabetes. This trial also included a comparison of two different BP-lowering regimens, but it was underpowered and totally inadequate, and offered no useful information in this regard. In the small subgroup of diabetic hypertensive patients in HOT, a comparison of more versus less BP lowering (achieved with a regimen based on the dihydropyridine calcium channel blocker felodipine) showed a significant reduction in major CV events, despite only modest differential BP reduction. MICRO-HOPE compared the angiotensin- converting enzyme (ACE) inhibitor ramipril with placebo in a subgroup of about 3300 high-risk patients with type 2 diabetes.

ADVANCE extended these data in a larger database by adding additional BP-lowering therapy (with a single-pill combination of perindopril and indapamide) to whatever preventive therapies were being taken (including ACE inhibition). Other studies have shown that drugs which block the reninangiotensin system (RAS)—ACE inhibitors and angiotensin receptor blockers—appear to have beneficial effects on the development and progression of renal disease in patients with diabetes, and that these effects may be superior to those achieved by other BP-lowering agents for a similar level of BP reduction.¹⁰ These results, based on renal end points, are the major reason for the recommendation that a RAS-blocker should probably form part of whatever antihypertensive “cocktail” is provided for patients with diabetes.¹¹

The ADVANCE trial set out to add to the currently available data by evaluating whether the addition of further BP lowering with perindopril and indapamide would reduce the risk of major macrovascular and microvascular disease in patients with type 2 diabetes, irrespective of their baseline BP or whether they were already taking background ACE inhibitor treatment or not.

Design

Details of the design of the ADVANCE trial have been published previously.^3,12 However, in summary, ADVANCE had a 2_2 factorial design comparing additional active BP lowering versus placebo and comparing intensive versus standard glucose control (Figure 1).³

Patients with type 2 diabetes were eligible to join the trial if they were aged _55 years and had one or more of the following additional criteria for increased vascular risk—age _65 years, history of major macrovascular disease, history of major microvascular disease, diabetes diagnosed >10 years ago, or any other major risk factor (eg, current smoker, total cholesterol > 6.0 mmol/L, high-density lipoprotein [HDL] cholesterol <1.0 mmol/L). Following registration with the trial, potentially eligible participants underwent a 6-week runin treatment period with a single daily pill combination of perindopril 2 mg and indapamide 0.625 mg, which was added to whatever previous therapy was being taken. Patients with a compelling indication for an ACE inhibitor other than perindopril or for a thiazide/thiazide-like diuretic were ineligible. During the run-in period, these diuretics were withdrawn and ACE inhibitors other than perindopril were replaced by perindopril (up to 4 mg daily).

After the run-in period, those who had tolerated and adhered to trial therapy were randomized double-blind to receive either placebo or a single pill combination of perindopril 2 mg/indapamide 0.625 mg daily for 3 months, after which active therapy was doubled to perindopril 4mg/indapamide 1.25 mg daily. If, during the trial, a compelling indication to use a different ACE inhibitor and/or a thiazide/thiazide-like diuretic was found, open-label therapy was substituted for study treatment. Patients were seen at 3, 4, and 6 months after randomization and 6-monthly thereafter.

At each study visit, BP was recorded using standardized conditions and recording devices,3 and details of all/any end points were recorded. The primary outcome was major macrovascular disease (nonfatal stroke, nonfatal myocardial infarction [MI], or CV death) and/or major microvascular disease (new or worsening nephropathy or diabetic eye disease). Secondary outcomes included all those expected in a major trial of CV outcomes in diabetic patients.³ At 2 and 4 years of follow-up, a quality of life assessment, a mini mental state examination, and a retinal examination were made, and a urinary albumin-creatinine ratio was measured.

The trial was originally designed to detect a _16% greater reduction in macrovascular and in microvascular events, with 90% power at the “5% level,” between the active BP lowering and placebo groups. However, due to lower than expected event rates after approximately half the expected follow- up period had elapsed, it was agreed to extend the BP-lowering arm by an extra year and to evaluate macrovascular and microvascular events both jointly and separately.

Results

Of the 12 877 men and women from Europe, Canada, Asia, the Indian Subcontinent, and Australasia registered for the trial, 11 140 were randomized into the main trial. Baseline characteristics of those randomized are shown in Table I.

Table I. Baseline characteristics of randomized patients in ADVANCE.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

BP was lowered, on average, by 5.6/2.2 mm Hg with perindopril/ indapamide, compared with placebo. The combined primary outcome was significantly reduced by 9% (95% confidence interval [CI], 0% to 17%) in favor of active BP lowering (Figure 2).¹² Major macrovascular and microvascular events were nonsignificantly reduced by 8% (-4% to 19%) and 9% (–4% to 20%), respectively. The biggest single contributor to the beneficial effect of perindopril/indapamide on the primary outcome was a significant 18% (2% to 32%) reduction in cardiovascular death, which when allied to an absence of adverse effect on noncardiovascular death generated a 14% (2% to 25%) significant reduction in all-cause mortality.

Figure 2. Results on the combined primary outcome in ADVANCE (major macro- or microvascular events) for perindopril/indapamide versus placebo.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

Interestingly, nonfatal stroke and MI were unaffected by additional BP lowering as was new or worsening eye disease, whilst new or worsening nephropathy showed an 18% (–1% to 32%) reduction. A summary of the outcomes is shown in Figure 3¹² and the use of concomitant medications at baseline and by the end of follow-up are shown in Table II. When subgroups of patients were considered in relationship to the primary composite outcome (Figure 4, page 228)¹² or the development of microalbuminuria (Figure 5, page 229), no sign of heterogeneity was apparent.

Two critical findings among these subgroup analyses were that the benefits of perindopril/indapamide were equally large whether patients were receiving background ACE in- hibition or not, and that the benefits of BP lowering were equally large whether “hypertensive” or not and regardless of whether initial systolic BP was above or below 140 mm Hg. Indeed, the impact of baseline BP level, considered as a continuous variable, on the beneficial effects of perindopril/indapamide, was further and more robustly evaluated, and no sign of any interaction was apparent. That is to say, that right across the BP range, additional BP lowering generated similar relative benefits.

Figure 3. Efficacy of perindopril/indapamide versus placebo on primary and secondary end points in ADVANCE.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

Table II. Drug treatment being received by patients at registration visit* and end of follow-up in ADVANCE.
Abbreviations: ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BP, blood pressure; MR, modified release.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

ADVANCE in the context of previous trials

Table III shows that ADVANCE was larger than all previous relevant trials. In addition, by virtue of the inclusion criteria and, to a larger extent, the greater use of in-trial concomitant medications, the CV event rates experienced were much lower in ADVANCE than in UKPDS and MICROHOPE. This reflects the fact that current management of high-risk diabetic patients currently usually involves the use of ACE inhibition, statins, and aspirin—a situation which did not prevail at the time UKPDS or MICRO-HOPE were conducted. Importantly, the BP level achieved (136/73 mm Hg) is the lowest level achieved in any of the relevant trials— though some way short of the target level of <130/80 mm Hg recommended in all the world’s guidelines! While a “the lower, the better” approach for BP in patients with diabetes seems appropriate, it must be acknowledged that current recommendations are not truly evidence-based.

Table III. ADVANCE in context: comparative patient profiles for UKPDS, MICRO-HOPE, and ADVANCE.
Abbreviations: ACE, angiotensin-converting enzyme; BP, blood pressure; CV, cardiovascular; MICRO-HOPE, MIcroalbuminuria, Cardiovascular, and Renal Outcomes in the Heart Outcomes Prevention Evaluation; UKPDS, United Kingdom Prospective Diabetes Study.
Based on data from references 3, 7, 9, and 12.

Figure 4. Subgroup efficacy analysis on the primary composite end point in ADVANCE.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

Figure 5. Subgroup analysis of microalbuminuria events by age, sex, blood pressure, and HbA1c in ADVANCE.
Modified from reference 12: ADVANCE Collaborative Group. Lancet. 2007;370:829-840. Copyright © 2007, Elsevier, Ltd.

If the results of the effect of BP lowering on the individual macrovascular components evaluated in ADVANCE are compared with those predicted by the Blood Pressure Lowering Treatment Trialists’ Collaboration,¹³ some effects (eg, on stroke) are apparently rather smaller than what might be expected whilst others (CV mortality and total mortality), are larger than expected. Nevertheless, the observed effect sizes are all essentially compatible with those predicted. The small effect on fatal and nonfatal stokes (2% [--17% to 20%]) is, at first sight, surprising. It may be that the significantly greater in-trial use of calcium channel blockers (CCBs) in the placebo group—which was 10% higher—may have contributed to the relatively greater stroke protection in this group, which is in keeping with previous analyses that suggest CCBs may provide stroke protection beyond that expected by BP reduction alone.¹⁴ However, it should be remembered that in the Perindopril pROtection aGainst REcurrent Stroke Study (PROGRESS)¹⁵ and Poststroke Antihypertensive Treatment Study (PATS)¹⁶ trials, perindopril/indapamide and indapamide, respectively, have been shown to prevent stroke significantly, and hence chance along with differential CCB use may have contributed to the apparently small effect size in ADVANCE.

The lack of impact on the hard end points relating to new or worsening nephropathy (development of proliferative retinopathy, macular edema, retinal photocoagulation therapy, or diabetes-related blindness) is in sharp contrast to the results of UKPDS,⁷ in which large benefits of BP lowering were apparent (mainly regarding retinal photocoagulation). These disparities may reflect differences in terms of duration of the diabetes diagnosis (new-onset in UKPDS) or, more likely, the much lower BP levels at play in the 2 trials, and the much greater use of other CV protective agents in ADVANCE (Table III)

Implications of the ADVANCE results

The number needed to treat (NNT) for 5 years associated with the use of perindopril/indapamide in the context of the ADVANCE population was 66 to prevent 1 major macro- or microvascular event and 79 to prevent 1 death. Allied with the fact that the active BP-lowering therapy was tolerated as well as the placebo (adherence rates were 73% and 74%, respectively), a blanket policy to apply additional BP lowering to all patients with type 2 diabetes, irrespective of their BP level, seems reasonable. Two critical caveats need to be considered first, however.

Firstly, is it affordable? Cost-benefit analyses are in the process of being published, but given the massive health burden associated with the CV sequelae of type 2 diabetes, it is likely that the approach practiced in ADVANCE is costeffective.

Secondly, were the beneficial effects observed due to BP lowering alone, to the use of an ACE inhibitor plus a diuretic, or to the specific combination used? Prior prejudice determines the answers given by individuals to this question, and there is no definitive “true” answer. However, evidence is mounting that not all hypertensive agents are equal in terms of associated outcomes for any level of BP reduction,^14,17 and, pending evidence to the contrary, we should try to stick with the evidence base arising from relevant trials.

Clinicians and research workers are variably influenced by the effect of interventions on various surrogate renal end points, such as microalbuminuria. However, most diabetologists and renal physicians will be suitably impressed by the large benefits of perindopril/indapamide on microalbuminuria observed in ADVANCE. Various measures of proteinuria are undoubtedly associated with CV events (as was the case in ADVANCE), presumably reflecting generalized endothelial dysfunction and increased cardiovascular risk. The greater effect on these end points observed in ADVANCE compared with the size of effect on individual macrovascular events (except CV mortality) may well reflect a time lag which prevents the realization of CV benefits in the relatively short follow-up period of less than 5 years.

Conclusions

The BP-lowering arm of the ADVANCE trial provides the best evidence to date that modest BP lowering is beneficial in terms of reducing the risk of critical CV events and renal end points in a broad range of patients with established type 2 diabetes, irrespective of their baseline BP levels. The significant 9% reduction in major vascular events and 18% reduction in CV mortality, plus an 18% reduction in new or worsening nephropathy and a 14% reduction in coronary events, were apparent in all major subgroups of patients, whether already taking ACE inhibitors or not, and irrespective of being hypertensive or not. Critically, for a potentially generalizable recommendation, the agents used to lower BP in ADVANCE were very well tolerated and NNTs to prevent major CV events or deaths were modest.

It therefore seems reasonable to conclude that all patients with type 2 diabetes should be considered for treatment with the type of intervention used in ADVANCE, ie, the combination of perindopril and indapamide, in addition to whatever other agents are already being taken and irrespective of the patient’s BP level.

References

Download this issue

by M. E. Cooper, Australia Mark E. COOPER, MB, BS, PhD, FRACP Diabetes Division, Baker IDI Heart and Diabetes Institute Melbourne, Victoria AUSTRALIA

In the glucose arm of the Action in Diabetes and Vascular disease: PreterAx and DiamicroN Controlled Evaluation (ADVANCE) study, subjects were randomized to either intensified glycemic control (n=5571) involving the use of gliclazide modified release (MR) and other drugs, as required, or to standard management (n=5569). In the intensified glycemic control arm, the strategy was to reduce glucose levels with the aim of reaching a glycated hemoglobin ≤5%. In this group, after maximizing the dose of gliclazide MR, there was sequential addition and/or an increase in the dose of metformin, thiazolidinediones, acarbose, or insulin. At the end of the follow-up, mean HbA1c was 6.5% and 7.3% in the intensive and standard groups, respectively. There was a statistically significant 10% decrease in the primary end point, a composite of micro- and macrovascular complications. This effect on the primary end point appeared to be related predominantly to the 14% decrease in microvascular complications and, in particular, to the reduction in renal, rather than retinal, events. These findings emphasize the role of intensified glycemic control in reducing the major burden of diabetes, its vascular complications. It is anticipated that further follow-up of these subjects over the next few years will allow us to determine if the beneficial effects seen with respect to renal disease ultimately translate into reduced cardiovascular events and a decrease in overall mortality.

Rationale of the ADVANCE study

Although the link between hyperglycemia and complications has been clearly demonstrated in type 1 diabetes, primarily from the findings of the Diabetes Control and Complications Trial (DCCT) and its follow-up study, the Epidemiology of Diabetes Interventions and Complications (EDIC) trial,¹ the issue has remained unresolved with respect to type 2 diabetes. At the time of the design and commencement of the Action in Diabetes and Vascular disease: PreterAx and DiamicroN Controlled Evaluation (ADVANCE) study, the best data available came from the United Kingdom Prospective Diabetes Study (UKPDS), which demonstrated that intensified glycemic control with various drugs, including sulphonylureas and insulin (achieving a reduction in HbA1c of 0.9% when compared with conventional treatment in newly diagnosed type 2 diabetic subjects), led to reduced microvascular complications.² With an achieved average HbA1c of 7% in UKPDS, there was a tendency towards, but neither a statistically significant decrease in macrovascular events nor a decrease in mortality. These findings led to most, but not all, international guidelines to suggest an HbA1c of _6.5% or 7% as the appropriate target for glucose-lowering treatment in type 2 diabetes.^3,4

Design of the ADVANCE study

The ADVANCE study has made great progress in helping us to determine what is the right approach for managing hyperglycemia in type 2 diabetes. The ADVANCE study was an investigator- initiated multinational trial. Within the glucose arm of the study, subjects were randomized to either intensified glycemic control (n=5571) involving the use of gliclazide modified release (MR) and other drugs, as required, or to standard management (n=5569). In the intensified glycemic control arm, the strategy was to reduce glucose levels with the aim of reaching a glycated hemoglobin ≤6.5%. In this group, after maximizing the dose of gliclazide MR, there was sequential addition and/or an increase in the dose of metformin, thiazolidinediones, acarbose, or insulin. The intensified glycemic control group also had more frequent follow-up visits and were encouraged to be more active with respect to home blood-glucose monitoring. All analyses in this trial were based on the intention to treat principle.

Results of the ADVANCE study

The median duration of follow-up was 5 years, and, at baseline, both groups had similar characteristics, including a baseline HbA1c of 7.5%. At the end of the follow-up, mean HbA1c was 6.5% and 7.3% in the intensive and standard groups, respectively (Figure 1). In terms of the various glucose- lowering treatments, the intensive group were taking more drugs, and, in particular in this group, insulin was prescribed in over 40% of subjects versus 24% in the standard group. The glucose arm of the study achieved its primary end point, a statistically significant 10% decrease in a composite of micro- and macrovascular complications.⁵ There was no evidence of an interaction between the blood pressure and glucose interventions for this primary outcome. This effect on the primary end point appeared to be related predominantly to the 14% decrease in microvascular complications and, in particular, to the 21% reduction in renal, rather than the nonsignificant 5% decrease in retinal, events. There was no significant effect on major macrovascular events, including no statistical decrease in cardiovascular mortality. Importantly, there was no evidence of an increase in total mortality. Indeed, there was a nonstatistical 7% decrease in mortality. No effects were seen on nonfatal myocardial infarction or stroke.

Figure 1. Glucose control at baseline and during follow-up, according to glucose-control strategy.
Data are shown for mean glycated hemoglobin. The average difference between the intensive-control group and the standard-control group for the follow-up period was 0.67 percentage points (95% confidence interval [CI], 0.64 to 0.70) for glycated hemoglobin.
Reproduced from reference 5: Patel A, MacMahon S, Chalmers J, et al. N Engl J Med. 2008;358:2560-2572. Copyright © 2008, Massachusetts Medical Society.

Clinical relevance of the findings of the ADVANCE study

The major issue is the clinical relevance of these findings and how these will translate into routine care of type 2 diabetes. The population studied came from 20 different countries of diverse ethnic backgrounds, with a significant proportion of subjects from Asia. Furthermore, when one looks at the demographic characteristics of the individuals in this study, these reflect the clinical features of type 2 diabetic subjects currently managed throughout the world. Indeed, if one compares the age, gender, glycemic control, and blood pressure of the subjects in the ADVANCE study, their clinical characteristics are almost identical to those from recently reported studies of the French and Australian type 2 diabetic populations.^6,7 Thus, the conclusions from the ADVANCE study should be considered highly relevant for both specialists and family practitioners who treat type 2 diabetes.

Renal events in the glucose arm of the ADVANCE study

One of the most important findings from the ADVANCE study was the demonstration of reduced renal events in the inten- sified glucose control arm (Figure 2).⁵ Renal disease is itself a major cause of morbidity and mortality in the diabetic population. Furthermore, in Western countries, diabetic nephropathy remains the major cause of end-stage renal disease (ESRD), with diabetes being the primary cause of renal failure in over 50% of individuals currently entering renal replacement programs.⁸ There was also a reduction in new-onset microand macroalbuminuria with intensified glycemic control. This will clearly ultimately translate into reduced ESRD, and although in the ADVANCE study there was a greater than 30% decrease in the development of ESRD, this did not reach statistical significance since only a small number of subjects developed this very serious diabetic complication.

Figure 2. Relative effects of glucose-control strategy on all prespecified primary and secondary outcomes. The diamonds incorporate the point estimates, represented by the vertical dashed lines, and the 95% confidence intervals of the overall effects within categories; for subcategories, black squares represent point estimates (with the area of the square proportional to the number of events), and horizontal lines represent 95% confidence intervals. The hazard ratios and relative risk reductions are given for intensive glucose control as compared with standard glucose control.
Reproduced from reference 5: Patel A, MacMahon S, Chalmers J, et al. N Engl J Med. 2008;358:2560-2572. Copyright © 2008, Massachusetts Medical Society.

Renal and cardiovascular end points in the ADVANCE study

It is well known that there is a close association between renal and cardiovascular disease in diabetes and that both micro- and macroalbuminuria have been reported to be linked to the subsequent development of macrovascular disease. Indeed, this association was also observed in the ADVANCE study population when assessing the link between microand macroalbuminuria at baseline and in the subsequent development not only of renal, but also of cardiovascular events. Thus, although the ADVANCE study did not demonstrate a decrease in cardiovascular mortality or events over the relatively short period of 5 years, one cannot exclude the pos- sibility that longer follow-up would demonstrate a benefit on macrovascular disease. There are such precedents in the literature, including, firstly, the recent report on the longer follow- up of the UKPDS cohort, where the initial borderline benefit seen with intensified glycemic control on cardiovascular disease has now been shown to be very clear after a much longer period of follow-up, despite subjects returning back to usual care without ongoing, intensified, glucose-lowering management.⁹ Secondly, in the Steno 2 study, where multifactorial intervention with not only intensified glycemic control, but also active treatment with antihypertensive and lipidlowering therapy was included, the initial evaluation after 4 years of active treatment revealed no evidence of a decrease in macrovascular, but a reduction in microvascular, events.¹⁰ However, at the 8-year time point, there was a decrease in macrovascular disease¹¹ that, interestingly, translated into a decrease in mortality 5 years later, despite subjects returning to usual care after 8 years in the study.¹² These findings emphasize the importance of either ongoing follow-up of subjects in these relatively short duration clinical trials involving cardiovascular end points or considering trials which involve active interventions for more than 5 years. Thus, it is critical to continue to monitor cardiovascular events and mortality in the subjects from the ADVANCE study, as is planned.

ADVANCE findings with respect to those from the ACCORD study

The findings of the ADVANCE study need to be considered in the light of the results of the Action to Control CardiOvascular Risk in Diabetes (ACCORD) study,¹³ published at the same time, as well as the as the more recently reported findings of the smaller Veterans Affairs Diabetes Trial (VADT) study.^14,15 The ACCORD study also included more than 10 000 type 2 diabetic subjects, although all subjects in that study were recruited from the US and Canada.¹³ The glucose arm of that trial was prematurely terminated due to an unexpected 22% increase in total mortality, primarily reflecting the reported increase in cardiovascular mortality in that study. This increase in mortality remains unexplained, but was clearly not observed in the ADVANCE study. Furthermore, as yet, the results on microvascular events have not been reported from the ACCORD trial.

There are a number of key differences between the ACCORD and ADVANCE studies. In the ACCORD study, subjects had a higher baseline HbA1c, a higher body mass index (BMI), and lower blood pressure than in the ADVANCE study. Thus, the investigators from the ADVANCE study have explored whether subjects with higher baseline HbA1c, BMI >30, and lower blood pressure behaved differently to the other subjects in terms of response to intensified glycemic control. However, even in the subjects from the ADVANCE trial that matched the ACCORD cohort more closely, no evidence of increased cardiovascular or total mortality was observed, nor did these individuals behave differently in terms of their clinical response to more intensive glucose-lowering management.⁵

In the ACCORD study, the intensive glucose-lowering strategy was a much more aggressive approach with reductions in HbA1c of between 1% and 1.5% achieved within 6 months of starting active treatment.¹³ Multiple drugs were used with many subjects on up to 4 glucose-lowering drugs within the first year of the study. Indeed, a very high proportion of the intensified glucose control group in that study ended up on treatment with the thiazolidinedione rosiglitazone and/or insulin. This approach clearly comes at a price—increased hypoglycemia and weight gain. In the ADVANCE study, the glucose- lowering strategy was much more gradual than in the ACCORD trial, with the maximal HbA1c reduction not seen till at least 2 to 3 years after the commencement of the ADVANCE study. This approach was not associated with significant weight gain, and the rates of hypoglycemia were much lower than observed in the ACCORD study. Although the ACCORD investigators have suggested that the marked increase in hypoglycemia does not explain the increase in mortality seen with aggressive glucose-lowering in that study,¹³ one cannot be sure as to the impact of intermittent, often unrecognized hypoglycemia in diabetic subjects with silent cardiovascular disease. Clearly, this is a potential area of exciting research, and, in particular, the link between hypoglycemia (including asymptomatic episodes), the sympathetic nervous system, and cardiovascular events needs further examination.¹⁶

Tissue remodeling in diabetes

As outlined previously, the ADVANCE study did not demonstrate a decrease in cardiovascular disease, although there appears to be a trend, albeit modest and not statistically significant after 5 years of treatment.⁵ It remains unexplained as to why no statistically significant beneficial effect was observed on macrovascular disease in this study. This may not only relate to the short duration of the study, but could reflect underlying pathological processes within the diabetic vasculature. Indeed, in the seminal studies involving pancreatic transplantation in type 1 diabetes, renal morphological injury took up to 10 years to reverse, despite a decade of normoglycemia.¹⁷ The rates of remodeling within organs, such as the kidney, and within blood vessels may be rather slow and, thus, take many years of improved glycemic control for endorgan injury to be reversed.

Advanced glycation

One of the key biochemical mechanisms implicated in diabetic complications, a result of chronic hyperglycemia, is known as advanced glycation.¹⁸ In blood vessels, these advanced glycation end products (AGEs) accumulate and may persist, despite improved glycemic control. These chemical moieties interact with specific receptors, such as the receptor for advanced glycation end products (RAGE), to induce vascular inflammation and promote atherosclerosis.¹⁹ Thus, it may be necessary to not only reduce glucose levels, but to directly inhibit the advanced glycation pathway. Such an approach is currently under active clinical investigation, with preclinical studies in models of diabetic complications providing promising results.^20,21 It should also be noted that angiotensin-converting enzyme inhibitors, such as perindopril, can directly influence various components of the AGE/RAGE pathway²² and could explain the added benefits seen in the subjects from the intensive glycemic control group in ADVANCE, who also had more aggressive antihypertensive treatment with a perindopril- based regimen.

Hyperglycemic memory

Another possibility for the lack of dramatic effect of glucose lowering on macrovascular disease may relate to a phenomenon known as “hyperglycemic memory”. It has previously been reported in type 1 diabetes in the DCCT/EDIC studies and more recently in type 2 diabetes in the longer term follow-up of UKPDS that there are sustained effects on the vasculature as a result of prior levels of metabolic control.^1,9 For example, in both the DCCT/EDIC and UKPDS studies, despite subjects returning to usual management of their hyperglycemia, these individuals continued to benefit from their previous period of better metabolic control.^1,9 The underlying explanation for this phenomenon, known as either “hyperglycemic memory” or the “legacy effect,” remains unexplained,²³ but recently it has been suggested that epigenetic mechanisms may be involved,^24,25 possibly with reactive oxygen species acting as key intermediates.^24,26 This is likely to be an active area of ongoing research with the molecular mechanisms responsible for conferring hyperglycemic memory being increasingly delineated.

Summary

ADVANCE provides us with new evidence emphasizing the role of intensive glucose control in reducing diabetic complications, in particular, nephropathy. Thus, the current guidelines suggesting that clinicians should aim to achieve an HbA1c of 6.5% or 7% appear sensible and can be justified by the more recent data obtained from trials like ADVANCE5 and the recent long-term follow-up from UKPDS.⁹ The practical and gentle strategy of reducing glucose levels in order to achieve, on average, an HbA1c of 6.5%, using regimens involving agents such as gliclazide as first-line therapy with subsequent use of other oral agents and, ultimately, often insulin, appears to be feasible, as has been demonstrated in a typical group of type 2 diabetic subjects in ADVANCE. This strategy, of course, must be considered in conjunction with important lifestyle measures, including diet (often with an emphasis on weight loss) as well as increased exercise. Such individuals could be managed in general or specialist practice, and it is likely that this strategy will not be associated with major side effects, such as severe hypoglycemia or excessive weight gain.

References

Download this issue

by B. Williams,United Kingdom Bryan WILLIAMS, MD, FRCP, FAHA Leicester Royal Infirmary Leicester UNITED KINGDOM

Epidemiological studies in the 1970s established that high blood pressure (BP) is both common and a major risk factor for macrovascular and microvascular disease in people with type 2 diabetes. However, uncertainty remained about the benefits and safety of BP lowering in people with diabetes. Previous studies, such as the United Kingdom Prospective Diabetes Study (UKPDS) in the late 1990s, demonstrated that lowering BP is very effective at reducing cardiovascular disease (CVD) and microvascular disease risk and, as a consequence, mortality. Questions remained, however, about whether there was a specific threshold below which BP lowering would be ineffective at reducing risk. The Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation (ADVANCE) study was designed with this key question in mind and set about lowering BP in people with diabetes, irrespective of their baseline BP, to determine whether this strategy would safely further reduce risk. The strategy employed was the addition of a combination of perindopril/indapamide, added to usual care, as part of a factorial study design that also examined the impact of “more versus less” glucose lowering. The BP-lowering arm of the ADVANCE study demonstrated that further BP lowering with perindopril/indapamide significantly reduced the risk of a combined end point of macrovascular and microvascular events and total mortality. This is a significant advance as it suggests that this treatment strategy will further reduce the risk for people with type 2 diabetes, irrespective of their prior treatment or baseline BP. This finding has important implications for people with type 2 diabetes and is a further step forward in improving treatment strategies for these patients beyond existing therapies to reduce their CVD and microvascular risk.

Some 25 years ago as a young doctor, I became interested in blood pressure (BP) in people with diabetes when I noted three things that intrigued me. Firstly, that the pathology associated with diabetes seemed to be primarily a vascular problem, and that the pathology looked remarkably similar to that seen in nondiabetic people with severe hypertension. Secondly, that people with diabetes invariably seemed to be hypertensive, often severely so. And, thirdly, that the doctors caring for patients with type 2 diabetes seemed less concerned about measuring and treating their blood pressure than I was! We now know much more, and the approach to detection and treatment of blood pressure has changed dramatically.

People with diabetes often develop hypertension. In younger people with type 1 diabetes, an increase in BP often signals the early development of diabetic nephropathy. In older patients with type 2 diabetes, hypertension—defined as a BP _140/90 mm Hg—is at least twice as common as in an agematched nondiabetic population, affecting approximately 80% of people with type 2 diabetes. It is also important to note that the characteristics of hypertension are very different in people with type 2 diabetes. In these patients, the progressive agerelated rise in systolic BP occurs earlier. This means that many people with type 2 diabetes develop systolic hypertension at least 10 years earlier than the general population, and there is an associated early widening of pulse pressure indicative of accelerated aging and stiffening of the large arteries. This arterial stiffening is important because it renders BP more resistant to treatment. Furthermore, subtle autonomic dysfunction occurs in the majority of these patients that disturbs the normal circadian rhythm of BP regulation. In particular, there is a blunting of the usual nocturnal dip in BP during sleep. This means that, for any given level of office BP, people with type 2 diabetes invariably have a higher 24-hour BP load. To compound matters, blood flow autoregulation is also impaired in people with diabetes. This means that any increase in circulatory pressures is more readily transferred to the delicate microcirculation, which thereby explains, in large part, the development of devastating microvascular disease in these patients. These observations are fundamental because they underscore why the detection and treatment of hypertension in people with diabetes is especially important in protecting both the macro- and microvasculature. Mindful of the fact that the BP load is elevated and the microcirculatory defenses are impaired, it also provides a rationale for advocating more aggressive BP lowering.

Importance of diabetes and hypertension as cardiovascular disease risk factors

We now recognize that an elevated BP is a major risk factor for macrovascular and microvascular complications in people with diabetes and a major contributor to associated premature morbidity and mortality. In 1974, the Framingham study was the first major study to highlight the fact that cardiovascular disease (CVD) risk was elevated in people with diabetes.1 This was the result of two important factors: (i) people with diabetes had a greater likelihood of abnormalities in major risk factors, such as lipids and blood pressure; and (ii) that, even when these risk factors were accounted for, their risk was magnified. Since then, there has been some controversy as to the magnitude of the excess CVD risk associated with type 2 diabetes, but no dispute about the fact that it is elevated when compared to the nondiabetic population. Moreover, elevation in CVD risk occurs at levels of glycemia below that required for the diagnosis of diabetes, suggesting that control of glycemia per se is unlikely to be a very effective strategy in reducing CVD risk. Another important observation from early epidemiological studies noted that when people with type 2 diabetes develop coronary heart disease, heart failure, or stroke, their prognosis was worse than that of the nondiabetic population. This latter point is especially important because it highlights the special importance of primary prevention in people with type 2 diabetes.

With regard to BP, it was recognized in the Framingham study that the prevalence of hypertension was higher in people with type 2 diabetes. However, hypertension at the time was defined at a higher threshold (_160/90 mm Hg). There was also a much greater focus on diastolic pressure at the time, and even though systolic pressure was commonly elevated, it was often ignored in BP classification. This means that the true prevalence of hypertension was often underestimated in the early surveys. In 1993, data from the United Kingdom Prospective Diabetes Study (UKPDS) reported a strikingly high prevalence of hypertension—defined as a systolic BP _160 mm Hg and/or a diastolic BP _90 mm Hg—in patients (40%) at the time of diagnosis of type 2 diabetes, with a higher prevalence among women and increasing prevalence with age.2-4 The thresholds for the diagnosis of hypertension used in UKPDS were much higher than current thresholds. Consequently, by modern criteria (BP _140/90 mm Hg), almost 80% would have been designated hypertensive. They also noted that hypertensive patients were more likely to be obese and that they already had almost double the risk of established cardiovascular complications at the time of diagnosis of type 2 diabetes when compared with those without hypertension.

Alarmingly, most of the hypertension was unrecognized and untreated.3 This hypertensive cohort within UKPDS became the basis of a key substudy: the Hypertension in Diabetes Study (HDS). HDS was one of the first prospective evaluations of the association between BP and clinical outcomes within a major clinical outcomes trial. Within 5 years of commencing HDS, the strong association between BP and clinical outcomes was reported by the group for patients with type 2 diabetes.3 After a median follow-up of only 4.6 years, it had become clear that hypertension was a major risk factor for cardiovascular morbidity and mortality in this population. The authors speculated that “antihypertensive therapy may provide greater benefit in this high risk group than in the general population.”3

One of the problems bedeviling treatment was a lack of clinical trial data demonstrating the safety and efficacy of BP lowering in people with diabetes. This was a key question. Today, many would regard the answer as obvious, but at the time there was concern that BP lowering might be poorly tolerated by people with diabetes and that it could lead to critical ischemia in vital organs, compounded by autonomic dysfunction. Moreover, the “presence of diabetes” was often an exclusion criteria for early BP-lowering trials, thus little data existed upon which to base treatment recommendations. Consequently, the detection and treatment of hypertension was often poor in people with diabetes and considered less important than glucose control.

The emergence of blood pressure–lowering trials in people with diabetes

The UKPDS study had originally been established to study the impact of improved glycemic control on clinical outcomes in people with type 2 diabetes. The investigators, recognizing the high prevalence of hypertension in their patients, embedded HDS within UKPDS. I recently reviewed the impact of this study on the treatment of hypertension in people with diabetes, in some detail.4 HDS addressed a key question: would more intensive versus less intensive blood pressure lowering improve clinical outcomes in people with type 2 diabetes?5 HDS started in 1987 and recruited 1148 hypertensive type 2 diabetic patients from the 4297 patients recruited into UKPDS. Of note was the fact that these patients were newly diagnosed and had not previously had BP treatment. The patients were allocated to treatment with either “less tight control of BP,” aiming for a BP <180/105 mm Hg, or “tight control of BP,” aiming for a BP of <150/85 mm Hg. The very fact that, in the early 1990s, it was considered reasonable to randomize people with diabetes to what is now considered grade III hypertension, is testimony to the lack of evidence that existed and, consequently, the low profile of BP control in the routine care of people with type 2 diabetes, at the time.

After a median of 8.4 years of follow-up, the mean blood pressure during follow-up was significantly reduced in the group assigned tight blood pressure control (144/82 mm Hg) compared with the group assigned to less tight control (154/87 mm Hg).5 This 10/5 mm Hg difference in BP was associated with a reduction in diabetes-related end points of a quarter, reductions in deaths related to diabetes of almost a third, a reduction in stroke of almost a half and in heart failure of more than a half, and a reduction in microvascular disease of about a third (mainly delayed progression of retinopathy). The number needed to treat (NNT) to prevent one major complication over 10 years was only 6 patients, while the NNT to prevent a diabetes-related death was only 15 patients. The treatment strategy was safe and well tolerated, and the result was unequivocal. Moreover, the treatment benefit from BP lowering was greater than anticipated from the epidemiological association between BP and risk in people with type 2 diabetes (Table I). This, perhaps, reflects the enhanced vulnerability of patients with diabetes to pressure-mediated cardiovascular disease (see below).

Table I. Relationship between predicted benefit of BP lowering and actual benefit of BP lowering in the UKPDS. Based on data from reference 5.
Abbreviations: BP, blood pressure; UKPDS, United Kingdom Prospective Diabetes Study.

Just prior to publication of UKPDS and HDS in 1998, the diabetes subgroup (n=583) from the Systolic Hypertension in the Elderly Programme (SHEP) had reported a reduction in half of cardiovascular events in an elderly population experiencing a similar improvement in BP control over 5 years of treatment.6 However, this had less impact than the UKPDS data that identified that hypertension was the norm and almost invariably present in these patients, rather than the exception. UKPDS demonstrated that hypertension was dangerous and markedly increased the risk of the premature development of diabetic complications and that, left untreated, ultimately contributed to premature mortality. It not only showed for the first time that, for BP, “lower was better” in improving the outcomes in people with type 2 diabetes, but that lower was safe. The study also showed that almost all patients require multiple drugs in combination to achieve improved BP control and better clinical outcomes.

Early indicators of the relative importance of blood pressure versus glycemic control in type 2 diabetes

Glycemic control has for many years been the main focus of treatment for people with type 2 diabetes. UKPDS allowed the relative impact of BP lowering versus intensified glycemic control on clinical outcomes to be compared prospectively. As indicated above, intensified BP control was impressive at reducing major diabetes end points, including significant reduction in diabetes-related death, stroke, and microvascular disease, and tends to be of benefit in all end points. This is despite the small size of the BP-lowering study. The benefits of BP lowering appeared more impressive than those resulting from the intensified glycemic control strategy in UKPDS, even on microvascular end points, for which the latter had been strongly advocated. I highlight this comparison not to suggest that glycemic control is unimportant, but, rather, because of its importance in changing perceptions about the importance of BP control in protecting people with type 2 diabetes from microvascular and macrovascular damage.

The ADVANCE of knowledge regarding blood pressure lowering in people with diabetes

Many questions remained after the emergence of the early evidence that BP lowering was both safe and very efficacious at reducing CVD events and microvascular disease in people with diabetes. Should BP be lowered even further than the average of 144/82 mm Hg achieved in the tight control group of HDS in UKPDS? Is the definition of hypertension, in any case, arbitrary, and would people with diabetes benefit from BP lowering, whatever their baseline BP? Moreover, does it matter what drug we use to lower BP in people with type 2 diabetes? A meta-analysis of trials that followed UKPDS concluded that BP lowering per se is the dominant means of achieving macro- and microvascular benefits, and that achieving a lower BP does appear to be better—even more so than in the nondiabetic hypertensive population.7 This, in part, relates to the fact that people with diabetes are at much higher cardiovascular risk and, thus, stand to gain more absolute benefit from BP lowering. It may also relate to the fact that people with diabetes are especially vulnerable to hypertensive injury.8 Data also suggest that blockade of the reninangiotensin system (RAS) “adds value,” over and above the BP lowering they produce, in preventing cardiovascular events in people with diabetes,9,10 although this has not been confirmed by all studies.11 One important clinical outcome where there appears to be a clear advantage of RAS blockade in people with type 2 diabetes is on renal end points (albuminuria and progression of renal disease).12-14 This has prompted international guidelines to conclude that RAS blockade should be part of the cocktail of treatments used to lower BP in people with type 2 diabetes. Thus, modern treatment strategies usually include RAS blockade, along with a calcium channel blocker and/or a thiazide-type diuretic.

However, uncertainty has remained about “how low to go?” Guidelines were making recommendations to aim for lower BP targets in people with diabetes, but evidence to support these recommendations was lacking. There was a clear need to push the boundaries beyond those explored by UKPDS and to take BP into uncharted territories and evaluate the “lower is better” hypothesis. This was the purpose of the recent Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation (ADVANCE) trial that tested the “lower is better” hypothesis by measuring the effect of further BP reductions (achieved with an angiotensin-converting enzyme [ACE] inhibitor [perindopril]/thiazide-type diuretic [indapamide] combination versus placebo, on top of conventional therapy) on vascular events in 11 140 people with type 2 diabetes.15 An important aspect of this study was that the BP-lowering combination therapy (perindopril/indapamide) was added: (i) irrespective of baseline blood pressure levels— that is, it was added even if patients would have been considered “normotensive” by conventional criteria; and (ii) irrespective of whether the patients were already receiving other BP-lowering drugs, including ACE inhibitors.

The mean age of the ADVANCE population was 66 years, of whom 57% were male. About a third had a prior history of major macrovascular disease and a tenth had microvascular disease. The patient profile in ADVANCE is shown in Table II (page 243). In addition, glycemic control was excellent throughout the study. A large proportion (68%) were being treated for hypertension at the time, with a baseline BP of approximately 145/81 mm Hg. Importantly, the standard deviation around this mean baseline was approximately 22/11 mm Hg, indicating that many patients had substantially lower baseline BPs. Over a mean follow-up of 4.3 years, additional perindopril/ indapamide therapy was associated with a lower BP (–5.6/–2.2 mm Hg) versus placebo treatment. This modest BP reduction reflects that fact that patients in the placebo group received additional add-on therapy for their BP (83% received additional BP-lowering drugs in the placebo group, versus 74% in the perindopril/indapamide group), as often happens in clinical trials when active BP-lowering therapy is compared to placebo treatment. The additional BP lowering associated with perindopril/indapamide was associated with a significant 9% risk reduction in major macrovascular and microvascular events (Figure 1), a significant 18% reduction in cardiovascular death, and a 14% reduction in all cause mortality (Figure 2). Importantly, there was no evidence that the benefits of additional BP-lowering therapy differed according to initial blood pressure level—in other words, even those within the lowest BP strata at baseline experienced a similar relative risk reduction to those in the highest BP strata. Moreover, the lowest BP strata included patients whose BP was already below the currently recommended treatment target for type 2 diabetes (<130/80 mm Hg). It is also noteworthy that the ADVANCE result was achieved in a population with high concomitant use of statins (almost half of patients by study end), antiplatelet drugs (more than half of patients by study end), and background ACE inhibition, and with good glycemic control, in both arms of the trial. With regard to background ACE inhibition, those already receiving ACE inhibition at baseline were standardized to treatment with perindopril, thus, by the study end, about half of all patients were receiving perindopril in addition to placebo or additional perindopril/indapamide.

Table II. Patient characteristics at baseline* in the ADVANCE study.
Abbreviations: ADVANCE, Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation; HDL, high-density lipoprotein.
Reproduced from reference 15: ADVANCE Collaborative Group. Lancet. 2007;370:829–840. Copyright © 2007, Elsevier Ltd.

Figure 1. Primary outcome in the ADVANCE trial: reduction in combined primary end point of macrovascular and microvascular events. The vertical lines indicate the 2- and 4-year time points in the trial at which data were collected for microvascular end points (albuminuria and retinal photography).
Abbreviations: ADVANCE, Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation.
Reproduced from reference 15: ADVANCE Collaborative Group. Lancet. 2007; 370:829–840. Copyright © 2007, Elsevier Ltd.

Figure 2. Prespecified outcome of “all-cause mortality” in the ADVANCE trial.
Abbreviations: ADVANCE, Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation.
Reproduced from reference 15: ADVANCE Collaborative Group. Lancet. 2007; 370:829–840. Copyright © 2007, Elsevier Ltd.

It is also worth noting that because the background use of perindopril was so extensive, one of the major differences between the treatment arms was the use of indapamide, a thiazide- like diuretic. This is important when one considers the resistance to using diuretics in people with diabetes in the past. The data from SHEP,6 the Antihypertensive and Lipid-Lowering treatment to prevent Heart Attack Trial (ALLHAT),11 and now ADVANCE15 attest to the effectiveness of thiazide-type diuretics at reducing BP and major CVD and microvascular events in people with diabetes. In modern therapy for BP lowering in type 2 diabetes, thiazide-type diuretics are usually used alongside RAS blockade, as in the ADVANCE study.

The ADVANCE trial is very important because it extends our understanding of the importance of further BP lowering in patients with type 2 diabetes. The findings extend the original findings of UKPDS into new territories and provide strong evidence to support the safety, tolerability, and efficacy of a “lower is better” philosophy for BP control in people with diabetes. Indeed, in my recent review of key trials in hypertension, which considered the ADVANCE trial, I suggested that the modern treatment goal for hypertension in people with diabetes should be “the lowest pressure the patient will tolerate without an adverse impact on function”.16 Of course, this conclusion does warrant confirmation, and further information of the impact of a more intensive BP-lowering strategy will come from the ACCORD study (the Action to Control Cardiovascular Risk in Diabetes) in the US.17 ACCORD is testing whether lowering BP to normal (<120 mm Hg systolic) reduces stroke and heart disease risk compared with the level usually targeted in current clinical practice, ie, one below the current definition of hypertension (<140 mm Hg systolic) in people with type 2 diabetes.

Conclusions

The story of hypertension in diabetes is a young one. The first substantive epidemiological clues to its high prevalence and clinical importance first emerged less than 40 years ago. Since then, we have learned much about the pathophysiology of qualitative and quantitative BP disturbances that characterize people with type 2 diabetes. Changes in clinical practice have been driven by evidence from clinical trials, which began with UKPDS and which have continued (see the studies highlighted above) up until ADVANCE, from which the latest data come. In addition to BP, it is also clear that the high CVD risk and microvascular disease burden experienced by people with type 2 diabetes is best addressed by multifactorial intervention based on improved BP control, on better lipid management with statins, on antiplatelet drugs, and on improved glycemic control, which is best highlighted by studies from the Steno centre.18,19 Mindful of the almost obsessive focus of diabetes care on glucose control over the years, it is somewhat ironic that it is this aspect of intervention which has the less convincing evidence base and which remains the most controversial with regard to clinical outcomes. It is also clear that following the spectacular pace of change in the treatment of people with type 2 diabetes and, especially, the overall improvement in their CVD risk management, future trials will be more challenging. In this context, the data from the ADVANCE BP study are all the more remarkable in showing that we can still do more to improve the survival of an otherwise welltreated population.

References

Keywords: blood pressure; microvascular disease; macrovascular disease; diabetes; blood glucose control; epidemiological study; Preterax; Diamicron

Download this issue

by D. R. Matthews,United Kingdom David R. MATTHEWS, MA, DPhil, BM, BCh, FRCP Oxford Centre for Diabetes, Endocrinology and Metabolism National Institute for Health Research, Oxford Biomedical Research Centre, Oxford UNITED KINGDOM

Major studies have addressed the problem of whether improved glycemic control would improve the outcome in type 2 diabetes. Two large studies from the 20th century caused controversy: the Universities Group Diabetes Program (UGDP) raised a suggestion that tolbutamide might not be efficacious, and the United Kingdom Prospective Diabetes Study (UKPDS) had an equivocal outcome for myocardial infarction. In recent years, a number of landmark studies, including PROspective pioglitAzone Clinical Trial In macroVascular Events (PROACTIVE), Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation (ADVANCE), Action to Control CardiOvascular Risk in Diabetes (ACCORD), Veterans Affairs Diabetes Trial (VADT), Steno 2, and the United Kingdom Prospective Diabetes Study-PostTrial Monitoring (UKPDS-PTM) have all reported. Ongoing trials, such as Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of glycemia in Diabetes (RECORD) and Outcome Reduction with an Initial Glargine Intervention (ORIGIN), may add to our knowledge. The outcomes from these trials have given some answers. We are now convinced that early intervention and intensive control is worthwhile. However, there are still unanswered questions, including questions of interpretation and uncertainties about the choice of agents.

Background

Type 2 diabetes is now pandemic. In the World Health Organization report of 1997 the world estimate for diabetes was 140 million people. The estimates from current epidemiology suggest that this number will increase to 200 million people by the year 2010, and to 300 million by 2025. The problem is not one confined to country, race, or geographical location. Type 2 diabetes pervades societies in countries as diverse as Mexico, China, Japan, and the Ukraine.

The challenge for health care in the 21st century is one of provision of appropriate health care and establishing therapeutic regimes which optimize the outcome of metabolic control in the short term and minimize tissue damage, including retinopathy, neuropathy, nephropathy, stroke, and heart disease, in the longer term. Epidemiological studies from a wide range of data sources suggest that relative risks for ischemic heart disease are probably 3 times those of nondiabetic individuals, and that overall all-cause morbidity from diabetes-associated pathology is twice that of the nondiabetic age-matched population.

What is appropriate treatment for type 2 diabetes? What should our goals and targets be? How aggressively should we be treating hyperglycemia? Are there targets and goals to which we should adhere or simply aspire?

Trials of glycemic control

Throughout the 1950s and 1960s there had been a growing awareness that diabetes complications—both of microvascular and macrovascular origin—were presenting the greatest challenge to quality of life and to longevity in type 2 diabetes. Clinical acumen and observation had demonstrated that allowing glycosuria and very high blood glucose levels led to poor quality of life, but some physicians, up until late into the 20th century, were still thinking that perhaps poor control had an advantage in terms of weight loss. It was certainly cheaper than intensive therapy.

The Universities Group Diabetes Program (UGDP)¹ was the first to attempt to answer the question using a controlled trial approach: launched in 1960, this placebo-controlled, multi- center clinical trial aimed to determine which, if any, of the treatments for type 2 diabetes was efficacious. Although the differences seen in the cumulative total mortality were not statistically significant, a subgroup analysis suggested that cardiac deaths occurred more frequently in the tolbutamide group. The investigators terminated this limb of the study. However, the randomization was significantly skewed in that, at baseline, there were 30% more ECG abnormalities, 40% more angina, and 90% more hypercholesterolemia in the tolbutamide group.² So, randomization had failed to deliver equipoise in the outcome.

The United Kingdom Prospective Diabetes Study (UKPDS) was established to give a definitive answer to the glycemic control controversy as well as an attempt to answer important questions about class of agents used to achieve control.³ The UKPDS had stringent aims for glycemia, but allowed for fasting glucose to rise to 15 mmol/L before adding new therapies. Because it aimed to address the questions about which therapy should be used, the glycemia rose progressively throughout the trial. At the closeout, the results showed that intensive glucose control was efficacious in reducing many complications, but the results for myocardial infarction were borderline with a statistical value for efficacy of P=0.052.^4,5 The world divided into the statistical fundamentalists, who used this as evidence against the hypothesis that macrovascular disease could be prevented by good glycemic control, and into those who pointed out that the odds of glycemic control not being important were 1 in 19.2 as opposed to 1 in 20.

The UKPDS used an intensive versus conventional treatment for blood-glucose control and achieved HbA1c in the study population of 7% in the intensive groups compared with 7.9% HbA1c in the conventional group. The totality of diabetes related end points was reduced by 12% (relative risk [RR], 0.88; 95% confidence interval [CI], 0.79-0.99).⁵ The contribution to the risk reduction was both from microvascular complications (RR, 0.75; 95% CI, 0.60-0.93) and myocardial infarction (RR, 0.84; 95% CI, 0.71-1.0).⁶ So questions remained— especially the question of how low should one aim in glycemic control. We did not know the extent to which more aggressive glucose control would decrease macrovascular or microvascular disease in conventionally normotensive diabetic patients. Nor did we understand the extent to which there might be differences in the outcomes between different racial groups. So, the question remained open. With new agents and new enthusiasm to demonstrate how diabetes should best be treated, a series of trials were initiated.

In 2005, PROspective pioglitAzone Clinical Trial In macro- Vascular Events (PROACTIVE)⁷ reported its results: it was a prospective, randomized controlled trial of 5238 patients with type 2 diabetes who had evidence of macrovascular disease. The median follow-up was just under 3 years. Patients were assigned to pioglitazone or placebo taken in addition to their glucose-lowering drugs and other medications. The primary end point was a composite of cardiovascular disease, including surgical intervention in the coronary or leg arteries and amputation above the ankle. The outcome of this was not significant (P=0.095). However, the “main secondary end point” was the composite of all-cause mortality, nonfatal myocardial infarction, and stroke: this showed a significant favorable response to pioglitazone (P=0.027).

So, the trial was marred by the known problem of selecting a primary combined outcome which involved not only the onset of new pathology, but the surgical interventions relating to pathology. Combining outcomes may increase the event count, but can do so at the expense of specificity.

Then, in 2007, Nissen et al⁸ produced a meta-analysis which seemed to demonstrate that rosiglitazone might have an ad- verse effect on cardiovascular outcome. That meta-analysis has been criticized,⁹ especially on the basis that the metaanalysis was not a comprehensive search for all studies that might yield evidence about rosiglitazone’s cardiovascular effects and that studies were combined on the basis of a lack of statistical heterogeneity, despite variability in study design and outcome assessment. Diamond et al⁹ concluded that the risk for myocardial infarction and death from cardiovascular disease for diabetic patients taking rosiglitazone was uncertain, and stated that “neither increased nor decreased risk is established”. One should certainly not regard such combinatorial analysis as being evidentially as good as a randomized trial. At this point, Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of glycemia in Diabetes (RECORD), an intervention trial using rosiglitazone, was under pressure to produce an early interim analysis,¹⁰ and this reported that, in a total of 217 patients in the rosiglitazone group and 202 patients in the control group, the hazard ratio for hospitalization or death from cardiovascular cause was 1.¹¹ (95% CI, 0.93- 1.32) and there were more patients with heart failure in the rosiglitazone group than in the control group (hazard ratio [HR], 2.15; 95% CI, 1.30-3.57). So, it remains to be demonstrated that the use of rosiglitazone is not associated with cardiovascular risk. Then, in the summer of 2008,³ cardiovascular disease (CVD) trials reported at the American Diabetes Association. These were the Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation (ADVANCE),¹¹ the Action to Control CardiOvascular Risk in Diabetes (ACCORD),¹² and the Veterans Affairs Diabetes Trial (VADT)¹³ trials. ACCORD produced a startling headline result that mortality was worse in the group that was intensively treated to lower HbA1c toward 6%. At 1 year, stable median glycated hemoglobin levels of 6.4% and 7.5% were achieved in the intensive-therapy group and the standard-therapy group, respectively. During follow-up, the primary outcome occurred in 352 patients in the intensive-therapy group, compared with 371 in the standard- therapy group (HR, 0.90; 95% CI, 0.78-1.04; P=0.16). However, 257 patients in the intensive-therapy group died, compared with 203 patients in the standard-therapy group (HR, 1.22; 95% CI, 1.01-1.46; P=0.04).¹² This then raised the question again about the main result from UKPDS. Is intensive glucose lowering actually harmful? Nevertheless, one cannot compare the patients in ACCORD directly with those recruited into UKPDS (Figure 1). UKPDS recruited “healthy” newonset type 2 diabetes patients (serious disease of any kind was a contraindication). In ACCORD, patients were 10 years into established diabetes and were selected for preexisting cardiovascular disease or specific risk factors. Making sudden changes in glycemia in such patients may not be a smart therapeutic idea. In fact, in this trial, the reports show that the majority of the glucose-lowering effect was already achieved within the first 4 months, by which time the median HbA1c was 6.6%. Although there was no explicit evidence that hypoglycemia was the precipitating cause of death, it remains the highest suspect for the increased death rate. Hypoglycemia rates were 3 times higher in the intensively treated group, and death precludes a contemporaneous measurement of blood glucose. Many of the patients were receiving rosiglitazone (91% in the intensive and 57% in the standard therapy arm, respectively). The excess mortality was not simply cardiovascular, but hypoglycemia can cause falls or aspiration at night, leading to pneumonia. In the elderly, any significant medical event may be seriously life-threatening.

Figure 1. Schematic of HbA1c in the study populations of UKPDS, ACCORD, and ADVANCE. The schematic shows how these studies are not directly comparable in terms of their populations, especially with regard to duration of diabetes at recruitment. Grey lines are the control groups. Left group of four lines all relate to UKPDS: grey = conventional treatment; green = glibenclamide; blue = chlorpropamide; and red = insulin. Right group of lines relate to: ACCORD, solid lines (——); and ADVANCE, dashed lines (—-) [red = intensive group for both studies].
Abbreviations: ACCORD, Action to Control CardiOvascular Risk in Diabetes; ADVANCE, Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation; UKPDS, United Kingdom Prospective Diabetes Study.

ADVANCE¹¹ was the largest trial of cardiovascular disease in type 2 diabetes to date, recruiting 11 140 patients with type 2 diabetes randomized to standard or intensive glucose control, with the aim of using gliclazide modified release (MR) plus other drugs, as required, to achieve an HbA1c value of 6.5% or less. After a median of 5 years of follow-up, the mean HbA1c in the intensive-control group had achieved 6.5% compared with 7.3% in the control group. In the intensive group, there was a reduced incidence in the combined end point of major macrovascular and microvascular events (HR, 0.90; 95% CI, 0.82-0.98; P=0.01) as well as that of major microvascular events (9.4% vs 10.9%; HR, 0.86; 95% CI, 0.77-0.97; P=0.01), primarily because of a reduction in the incidence of nephropathy (HR, 0.79; 95% CI, 0.66-0.93; P=0.006). However, there were no significant effects of the type of glucose control on major macrovascular events (HR with intensive con- trol, 0.94; 95% CI, 0.84-1.06; P=0.32), death from cardiovascular causes (HR with intensive control, 0.88; 95% CI, 0.74- 1.04; P=0.12), or death from any cause (HR with intensive control, 0.93; 95% CI, 0.83-1.06; P=0.28).

Comparisons between the trials

The effects shown in the ADVANCE trial were mainly attributable to a 21% relative reduction in nephropathy, but unlike in the ACCORD trial, there was no signal that achieving the target of 6.5% gradually over 4 years had any detrimental cardiovascular effects nor any increased mortality. How can one explain the differences between these outcomes?

Figure 2. Summary of the evidence for glycemia trials in type 2 diabetes. The lines show the stylized glycemia over time, and the boxes summarize the evidence from the trials.
Abbreviations: ACCORD, Action to Control CardiOvascular Risk in Diabetes; ADVANCE, Action in Diabetes and Vascular disease: PreterAx and DiamicroN MR Controlled Evaluation; PROACTIVE, PROspective pioglitAzone Clinical Trial In macroVascular Events; RECORD, Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of glycemia in Diabetes; UKPDS, United Kingdom Prospective Diabetes Study. UKPDS-PTM, United Kingdom Prospective Diabetes Study-PostTrial Monitoring.

ADVANCE used gliclazide (mainly gliclazide modified release) to achieve a lower glycemia in the intensive control group, and this contrasts with ACCORD where there was a high usage of rosiglitazone (in both arms) and insulin and sulphonylurea, in combination. In ACCORD, the glycemic targets were aggressively pursued, and the control of glycemia over time did not deteriorate (Figure 2). By contrast in ADVANCE, the target of HbA1c below 6.5% was achieved only progressively over a period of 4 years—a much slower rate than that of the ACCORD patients, and the totality of the updated mean difference was much less. On the other hand, the duration of diabetes was similar (8 years). So, the differences between the 2 trials was a marked difference in rate of achievement of target glycemia, a very high hypoglycemia rate in ACCORD (a nearly 4 times greater rate than in ADVANCE), and a clear difference in the choice of agents for the 2 trials. ACCORD suggests that intensive glycemic control achieved fast and late in diabetes using multiple agents might not be wise. ADVANCE suggests that achieving such targets over several years is not contraindicated, and there may be gains to be achieved in the prevention of renal disease.

The VADT trial also presented its results at the American Diabetes Association meeting, though they were not then available in print. The trial ran for about 6 years, but the numbers of patients were few, and it was not surprising that there was no differential in the cardiovascular outcome. Essentially, the trial was under powered—1792 subjects¹⁴ followed for “5 to 7” years.¹³ This was a strange error to make in view of the known longevity of UKPDS in terms of patient years required to produce a meaningful answer. A subgroup analysis of calcification in VADT<¹⁵ suggested that this was a clear marker of CVD risk—though these data do not help us decide on the generality of CVD risk reduction.

UKPDS long-term follow-up

In 2008, the glycemic control debate was galvanized not only by the presentation of the outcomes of seminal trials, but also by the publication and simultaneous presentation of the United Kingdom Prospective Diabetes Study-PostTrial Monitoring (UKPDS-PTM).¹⁶ This examined the outcome of the patients in UKPDS 10 years after the trial had finished. It examined the question of whether the effects of being in the intensively controlled group would dissipate with time. After the trial, everyone was given advice about intensive control. The 3277 patients remaining in the trial were asked to attend annual UKPDS clinics for 5 years, but no attempts were made to maintain their previously assigned therapies. Annual questionnaires were used to follow patients who were unable to attend the clinics, and all patients in years 6 to 10 were assessed through questionnaires. For the years that followed their inclusion in the trial, no glycemic differences were strived for, nor seen. The null hypothesis was that with no differences in treatment, the differences in outcome would be lost. But, far from there being a diminution of the glycemic trial effect over the 10 years, the effects of lower incidence of pathology were maintained, and with the advent of more events, the statistical probabilities of error declined. In the sulfonylurea-insulin group, relative reductions in risk persisted at 10 years for any diabetes-related end point (9%, P=0.04) and microvascular disease (24%, P=0.001), and risk reductions for myocardial infarction (15%, P=0.01) and death from any cause (13%, P=0.007) emerged over time, as more events occurred (Figure 3). In the metformin group, significant risk reductions persisted for any diabetes-related end point (21%, P=0.01), myocardial infarction (33%, P=0.005), and death from any cause (27%, P=0.002). So, despite an early loss of glycemic differences, a continued reduction in microvascular risk and emergent risk reductions for myocardial infarction and death from any cause were observed during 10 years of posttrial follow-up.

Figure 3. Hazard ratios (intensive vs conventional treatment) for poststudy monitoring in UKPDS. During this period, the glycemic control of the two groups converged, so that there was no significant difference in glycemic exposure. Upper panel shows data for any diabetes-related end point and lower panel for myocardial infarction.
Abbreviations: UKPDS, United Kingdom Prospective Diabetes Study.

Clinical implementation

How do these data help in making decisions about what we should do in clinical practice? It is paradoxical that it has taken the UKPDS-PTM 20-year duration trial and a 10-year posttrial follow-up to establish that early intervention in glycemic control is worthwhile. Is it worth intervening after 8 or 10 years of indifferent control? The answer seems to be “yes,” but we need to add particular caveats. ACCORD teaches us that very sudden changes in glycemia in the elderly may do more harm than good, while ADVANCE suggests that even if cardiovascular disease cannot be diminished in the short term, then one can at least get gains from less renal pathology.

What agents should one use? The peroxisome proliferatoractivated receptor ã (PPARã) agonist pioglitazone has emerged as efficacious in the light of the PROACTIVE trial, but rosiglitazone still has to be proved to be effective in cardiovascular outcomes. The evidence tends to point in the other direction.

What about sulphonylureas? Despite the debates dating back to the UGDP, and the fear that â-cell failure might be affected, the sulfonylureas have continued to stand the test of time. There was no detrimental signal from UKPDS or from UKPDSPTM,¹⁶ and ADVANCE shows that gliclazide modified release can be an effective late intervention. And what about insulin? UKPDS used insulin as one of its randomized interventions, and there was no suggestion that this policy was unsafe or had detrimental outcomes. What about the new analogue insulins? The Outcome Reduction with an Initial Glargine Intervention (ORIGIN) trial¹⁷ will report soon on CV outcome in a trial of people aged over 50 years with evidence of cardiovascular disease and with impaired fasting glucose, impaired glucose tolerance, or newly detected or established diabetes randomized to glargine or standard care to achieve fasting glucose of 5.3 mmol/L or less in the intensive group.

What about hypoglycemia? Here, I think we are closer to real answers. ACCORD had a very high rate of hypoglycemia, and there are many rational reasons to suppose this to be dangerous in the elderly. So, we need to take new stock of this as a real risk to life as well as to quality of life. Hypoglycemia in the elderly may lead to internal pathology (the myocardium is unlikely to function well) and will certainly cause threatening events related to falls, aspiration pneumonia, accidents, forgetfulness, and other significant risks.

Conclusions

All the trial data suggest that hyperglycemia is a risk for cardiovascular disease and should be lowered if possible. The targets for glycemia are for HbA1c lower than 7.5%, and may be nearer 6.5%, if achieved slowly and without dangerous hypoglycemia.

Finally, one should remember that diabetes cannot be treated simply as a disease of abnormal glucose. The trial data are strongly indicative that lipids and blood pressure¹⁸ should be treated in parallel, and the Steno 2^19,20 trial—a trial of multiple intervention care-packages—suggests real benefits from this approach.

References

1. Meinert CL, Knatterud GL, Prout TE, Klimt CR. A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. II. Mortality results. Diabetes. 1970;19(suppl):789-830.
2. Leibel B. An analysis of the University Group Diabetes Study Program: data results and conclusions. Can Med Assoc J. 1971;105:292-294.
3. UKPDS Group. UK Prospective Diabetes Study VIII: Study design, progress and performance. Diabetologia. 1991;34:877-890.
4. UKPDS Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352:854-865.
5. UKPDS Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352:837-853.
6. Stratton IM, Adler AI, Neil HA, et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ. 2000;321:405-412.
7. Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROACTIVE Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet. 2005;366:1279-1289.
8. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med. 2007;356:2457-2471.
9. Diamond GA, Bax L, Kaul S. Uncertain effects of rosiglitazone on the risk for myocardial infarction and cardiovascular death. Ann Intern Med. 2007;147: 578-581.
10. Yamasaki Y, Kawamori R, Wasada T, et al; AD-4833 Glucose Clamp Study Group. Pioglitazone (AD-4833) ameliorates insulin resistance in patients with NIDDM. Tohoku J Exp Med. 1997;183:173-183.
11. Patel A, MacMahon S, Chalmers J, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008;358: 2560-2572.
12. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545-2559.
13. Abraira C, Duckworth W, McCarren M, et al. Design of the cooperative study on glycemic control and complications in diabetes mellitus type 2: Veterans Affairs Diabetes Trial. J Diabetes Complications. 2003;17:314-322.
14. Duckworth WC, McCarren M, Abraira C. Control of cardiovascular risk factors in the Veterans Affairs Diabetes Trial in advanced type 2 diabetes. Endocr Pract. 2006;12(suppl 1):85-88.
15. Reaven PD, Emanuele N, Moritz T, et al. Proliferative diabetic retinopathy in type 2 diabetes is related to coronary artery calcium in the Veterans Affairs Diabetes Trial (VADT). Diabetes Care. 2008;31:952-957.
16. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359:1577- 1589.
17. Gerstein H, Yusuf S, Riddle MC, Ryden L, Bosch J; Origin Trial Investigators. Rationale, design, and baseline characteristics for a large international trial of cardiovascular disease prevention in people with dysglycemia: the ORIGIN Trial (Outcome Reduction with an Initial Glargine Intervention). Am Heart J. 2008;155: 26-32.
18. UKPDS Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998; 352:854-865.
19. Gaede P, Valentine WJ, Palmer AJ, et al. Cost-effectiveness of intensified versus conventional multifactorial intervention in type 2 diabetes: Results and projections from the Steno-2 study. Diabetes Care. 2008;31:1510-1515.
20. Gaede P, Pedersen O. Multi-targeted and aggressive treatment of patients with type 2 diabetes at high risk: what are we waiting for? Horm Metab Res. 2005; 37(suppl 1):76-82.