Show Summary Details
Page of

Diabetes 

Diabetes
Chapter:
Diabetes
DOI:
10.1093/med/9780199234073.003.0005
Page of

PRINTED FROM OXFORD MEDICINE ONLINE (www.oxfordmedicine.com). © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 13 December 2018

  • Introduction [link]

  • Primary prevention

    • Primary prevention: lifestyle intervention [link]

    • Primary prevention: metformin vs lifestyle intervention [link]

  • Complications

    • Preventing complications: intensive treatment [link]

    • Preventing complications: intensive glycaemic control [link]

    • Preventing complications: perindopril and indapamide [link]

    • Preventing complications: control of blood pressure [link]

    • Preventing complications: multifactorial intervention [link]

    • Preventing complications: statins [link]

  • Nephropathy

    • Nephropathy: ACE inhibitors [link]

    • Nephropathy: angiotensin II receptor antagonists [link]

  • Gestational diabetes

    • Gestational diabetes: screening and treatment [link]

Introduction

Arateus, the 2nd century Alexandrian physician, first coined the term ‘diabetes’ when he encountered a case of a patient complaining of excessive urination. For many centuries, this prominent symptom (polyuria) led to diabetes being considered as a disease solely of the kidneys. This theory has since been refuted by research uncovering its multifactorial aetio-pathogenesis, with complications affecting multiple organ systems; it goes far beyond being an endocrine disorder of the pancreas, and requires consideration of other systems, including cardiovascular (cerebrovascular and peripheral vascular disease), renal (nephropathy), ophthalmic (retinopathy), neurological (neuropathy—both peripheral and autonomic), orthopaedic (diabetic foot, Charcot neuroarthropathy), psychiatric (depression, anxiety), infectious (respiratory, skin, urinary tract), dermatological (ulcers, dermopathy, necrobiosis lipoidica diabeticorum, bullosa diabeticorum) … the list goes on. Indeed, it is fair to say that ‘if you fully understand diabetes and its complications, then you truly understand medicine!’

Although this assertion may appear slightly exaggerated, there remains a substantial truth in its central thesis. In fact, it can be suggested that diabetes has become modern medicine’s equivalent of syphilis, the multi-system nature of which took some time to be fully appreciated. In Western practice, about one in four patients admitted to hospital have diabetes. Its high prevalence and incidence, coupled with costly multi-organ complications, have been reflected by an unparalleled surge in diabetes-related research over the last decade. In this era of evidence-based medical practice, one may have to dedicate a considerable amount of time to keep abreast with new knowledge, concepts, and innovation. With this wealth of available resources, it can, unfortunately, be frustratingly arduous for a clinician to find the time to understand the most reliable evidence that underpins the practice of diabetology. This chapter will focus on key clinical trials that have had a significant impact on our understanding and/or treatment over the past decade.

Primary prevention: lifestyle intervention

DPS (Finnish Diabetes Prevention Study):

Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance.

Authors:

Tuomilehto J, Lindstrom J, Eriksson J et al.

Reference:

N Engl J Med (2001) 344, 1343–50.

Study design:

RCT.

Evidence level:

1b.

Key message

This landmark RCT study shows that type 2 diabetes mellitus (DM) can be prevented in high-risk patients through lifestyle changes.

Impact

In view of the rising incidence of type 2 DM, lifestyle changes are currently advocated to reduce the burden of diabetes. Today, healthy diet, weight reduction in overweight people, and regular moderate physical exercise are important lifestyle measures in diabetes prevention.

Aims

Lifestyle factors are key determinants in the development of type 2 DM, especially in those at high risk. People with impaired glucose tolerance (IGT) have derangements in glucose metabolism, and are thus at high risk of developing type 2 DM. This study was conducted to determine the feasibility and effect of lifestyle changes designed to prevent or delay the onset of type 2 DM in people with IGT.

Methods

Patients:

522 patients at five centres in Finland.

Inclusion criteria:

  • IGT (defined as fasting blood glucose <7.8mmol/L and 7.8–11.1mmol/L, 2h after a 75g glucose load in an oral glucose tolerance test, OGTT);

  • Overweight (defined as BMI ≥25);

  • Age 40–65y.

Exclusion criteria:

  • Diagnosis of DM;

  • Chronic disease and unlikely to survive >6y;

  • Psychological and physical impairments deemed likely to interfere with the study.

Groups:

  • Control: oral advice and written information on diet and exercise;

  • Intervention: detailed individualised dietary and exercise advice with preset goals and guidance on how these can be achieved.

Primary endpoint:

Development of DM (according to 1985 World Health Organisation criteria).

Follow-up:

Control group: Annually. Intervention group: Seven sessions in the first year, then three sessions annually for the rest of the study period, for nutritional advice. Study started in 1993, last patient assigned in 1998. Initially planned to run for 6y after this, but was terminated in 2000 when an interim analysis showed that the intervention was clearly beneficial.

Results

Primary endpoint

Intervention

Control

p

Cumulative incidence of DM

11%

23%

<0.001

Discussion

Previous studies, notably the Da Qing trial (Diabetes Care (1997) 20, 537–44) and the Malmo 6-year feasibility study (Diabetologia (1991) 34, 891–8), had shown that lifestyle interventions prevented or delayed the onset of type 2 DM in people with IGT. However, neither of these studies were randomised. The DPS, a carefully designed RCT, also showed that lifestyle intervention was clearly beneficial. The NNT by lifestyle intervention for 1y to prevent one case of incident type 2 DM in people with IGT was 22. In an extended F/U of the study participants for 7y, the authors subsequently showed that the benefits of this intervention were maintained. These results have been confirmed on a larger scale by an equivalent USA-based study, the Diabetes Prevention Program (DPP).

Problems

  • The staff involved in this trial were not blinded to the intervention/control groups.

  • The glucose levels used to define IGT in this study included subjects who today would be classified as having type 2 DM. This raises the question as to whether this could be truly classified as a ‘prevention’ (as opposed to a ‘delayed onset’) study.

Primary prevention: metformin vs lifestyle intervention

DPP (Diabetes Prevention Program):

Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin.

Authors:

The diabetes prevention research group.

Reference:

N Engl J Med (2002) 346, 393–402.

Study design:

RCT.

Evidence level:

1b.

Key message

In patients with impaired glucose tolerance (IGT), intensive lifestyle modification and to a lesser extent, metformin, significantly reduce the incidence of type 2 diabetes mellitus (DM).

Impact

Lifestyle changes are now universally recommended not only to prevent the development of DM, but also to improve overall cardiovascular health, particularly in those at high risk.

Aims

Other studies, such as the Da Qing trial and the almost contemporary Finnish Diabetes Prevention Study, had shown lifestyle interventions to be effective in preventing or delaying the onset of DM in high-risk subjects. This larger study aimed to test the veracity of these findings across varied ethnic groups in the USA, using newly defined American Diabetes Association (ADA) 1997 criteria for the diagnosis of DM and IGT. It specifically sought to determine whether lifestyle intervention and the biguanide, metformin, were both effective in preventing DM and which of them was more effective.

Methods

Patients:

3234 non-diabetic subjects at 27 centres in the USA.

Inclusion criteria:

  • Age ≥ 25y;

  • BMI ≥24 (22 in Asians);

  • IGT: fasting plasma glucose (FPG) 5.3–6.9mmol/L and 7.8–11.0 at 2h post-75g glucose load in an oral glucose tolerance test (OGTT).

Groups:

  • Metformin (850mg bd) plus standard lifestyle recommendations (n=1073);

  • Placebo (bd) plus standard lifestyle recommendations (n=1082);

  • Individualised intensive programme of lifestyle modification: target being to lose 7% body weight and to do 150min of exercise per wk (n=1079).

Primary endpoint:

Diabetes (FPG ≥ 7.8mmol/L; or ≥ 11.0mmol/L, 2h post-75g OGTT).

Follow-up:

Quarterly in metformin and placebo groups to assess adherence; monthly in the intensive lifestyle group. All groups had F/U for 2y.

Results

Primary endpoint

Lifestyle

Metformin

Placebo

Incidence of DM per 100 person-years

4.8

7.8

11.0

  • p<0.001 for each comparison;

  • Risk reduction: lifestyle vs placebo 58%; metformin vs placebo 31%; lifestyle vs metformin 39%.

Discussion

This trial confirmed that lifestyle modification in a large and high-risk multiracial cohort with IGT effectively prevented or delayed the onset of type 2 DM. The oral antidiabetic agent, metformin, was also effective in preventing DM in people with IGT, but was inferior to lifestyle modification; its effect was most marked in people with higher baseline BMI and blood glucose levels. The benefits of various pharmacological agents have been examined in other studies. In the STOP-NIDDM study (Diabetes Care (1998) 21, 1720–5), arcabose was shown to reduce the incidence of DM (number to treat (NNT) 22), while in the TRIPOD study (Controlled Clinical Trials (1998) 19, 217–31), troglitazone reduced the incidence of DM in high-risk Hispanic women with gestational diabetes (NNT 6). However, this agent was noted to be hepatotoxic in an abandoned fourth arm of the DPP. The recent DREAM trial (Diabetes Care (2008) 31, 1007–14) used rosiglitazone/ramipril in people with IGT and/or impaired fasting glucose (IFG), concluding that for every 1000 people treated with rosiglitazone for about 3y, 144 cases of DM would be prevented at the expense of about four cases of congestive heart failure. In summary, whereas the evidence for lifestyle intervention in reducing the incidence of type 2 DM is striking, the same cannot be said of pharmacological interventions, with studies having produced mixed results.

Problems

  • This study investigated the preventive effect of a drug that is used to treat DM, and this raises the question as to whether the benefits observed in those on metformin were simply due to this drug’s glucose-lowering effect.

Preventing complications: intensive treatment

DCCT (Diabetes Control and Complications Trial):

The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus.

Authors:

Diabetes Control and Complications Trial Research Group.

Reference:

N Engl J Med (1993) 329, 977–86.

Study design:

RCT.

Evidence level:

1b.

Key message

‘Tight control’ of blood sugar levels with intensive insulin treatment delays the onset and slows the progression of diabetes-related complications.

Impact

It is now accepted that blood glucose control in type 1 DM should be optimised towards attaining DCCT targets: HbA1C <7.5% (or 003C<6.5% for those at increased risk of arterial disease).

Aims

Long-term microvascular and neurological complications cause major morbidity and mortality in type 1 diabetes mellitus (DM). The aim of this study was to compare the effects of two treatment regimes—intensive vs conventional—with regard to their effects on the development and progression of the early vascular and neurological complications.

Methods

Patients:

1441 teenagers/young adults at 29 centres in the USA and Canada.

Inclusion criteria:

Age 13–39y, with type 1 DM for 1–15y; insulin dependence as defined by deficient C-peptide secretion.

Exclusion criteria:

Hypertension, hypercholesterolaemia, or other significant medical conditions; severe diabetes-related complications.

Groups:

  • Primary prevention cohort (n=726): Type 1 DM for 1–5y, with no complications (no retinopathy and urinary albumin excretion <40mg/24h);

  • Secondary prevention cohort (n=715): Type 1 DM for 1–15y, with mild to moderate non-proliferative retinopathy and urinary albumin excretion <200mg/24h.

  • Conventional therapy (CT): with urine or blood sugar tests and 1–2 insulin injections per d (n=378 primary, 342 secondary);

  • Intensive treatment (IT): initial hospitalisation for education and stabilisation, ≥ 4 blood sugar tests per d, insulin ≥3 per d by pump or injections (n=348 primary, 363 secondary).

Primary endpoints

  • Rate of onset of retinopathy in the primary prevention cohort;

  • Rate of progression of retinopathy in the secondary intervention cohort.

Follow-up:

CT group: three monthly F/U (target HbA1C <9.0%). IT group: monthly visits, frequent telephone calls by diabetes nurse educators (target HbA1C <6.0%). Care team comprised physician, nurse, dietician, and behavioural therapist. Mean F/U=6.5y (range 3–9y).

Results

Conventional

Intensive

p

Primary prevention group

Total incidence of retinopathy

91

23

<0.001

Reduction in retinopathy

76% (95% CI 62 to 85%)

Secondary prevention group

Total incidence of retinopathy

77

43

0.002

Slowed progression of retinopathy

54% (95% CI 39 to 66%)

Combined cohort: Mean HbA1c

231±55mg/dL

155±30mg/dL

<0.001

The intensively treated group (combined cohort) showed reduction in occurrence of three major microvascular complications in all subgroups:

  • 76% reduction in diabetic retinopathy;

  • 57% reduction in progression of kidney disease;

  • 60% reduction in progression of neuropathy.

Discussion

This study found differences between the intensive and conventional groups to be apparent within 3y. The curves showing the rate of development of complications between the two groups progressively diverged as the years went by, suggesting an overtly beneficial effect of intensive insulin therapy in type 1 DM. The chief adverse events associated with the intensively treated group were a 2–3 fold increase in severe hypoglycaemia and an average small weight gain (1kg).

Problems

  • Mean F/U was only 6.5y—short when considering long-term complications such as retinopathy. However, DCCT patients subsequently had >10y F/U in the EDIC study, which confirmed these findings.

  • Retinopathy: too few patients were affected by severe retinopathy or macular oedema or required laser treatment to notice a significant difference between groups. In the secondary prevention group, the incidence of progression of retinopathy in the first year was higher in the intensive group. This suggests patients who have suddenly improved glycaemic control should be watched carefully as they are at risk of initial worsening before benefit is obtained.

Preventing complications: intensive glycaemic control

UKPDS (United Kingdom Prospective Diabetes Study) 33:

Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes.

Authors:

UKPDS Group.

Reference:

Lancet (1998) 352, 837–53.

Study design:

RCT.

Evidence level:

1b.

Key message

HbA1C level and the life-threatening complications of DM can be reduced by more intensive management using existing treatments (UKPDS 33 and 34). Reducing blood pressure in these patients reduces the risk of fatal and non-fatal complications (UKPDS 38 and 39).

Impact

As perhaps the single most important study ever conducted in type 2 DM, this underpinned the need for aggressive treatment in maintaining good glycaemic control. It has been the most important driving force behind many international guidelines, including those of the UK’s National Institute of Health and Clinical Excellence (NICE), in setting targets for tight glycaemic control (HbA1C between 6.5% and 7.5%), based on the risk of microvascular and macrovascular complications.

Aims

Patients with type 2 DM are at risk of developing both microvascular and macrovascular complications in the setting of chronic hyperglycaemia. The aims of UKPDS were to determine whether: (a) intensive use of pharmacological therapy to lower blood sugar would result in clinical benefits (reduced cardiovascular and microvascular complications), and (b) the use of sulphonylureas, metformin, or insulin had specific therapeutic advantages or disadvantages (UKPDS 33 and 34).

Methods

Patients:

5102 recruited (randomised: 3867 UKPDS 33; 753 UKPDS 34), from 23 clinics in the UK (82% Caucasian; 10% Asian; 8% Afro-Caribbean).

Inclusion criteria:

Age 25–65y, with newly diagnosed type 2 DM and fasting plasma glucose (FPG) >6mmol/L on two mornings, 1–3wk apart.

Exclusion criteria:

Ketonuria >3mmol/L or serum creatinine >175 μmol/L; myocardial infarction (MI) in last y, angina, heart failure, major vascular event, malignant hypertension, contraindication to insulin treatment; retinopathy requiring laser treatment.

Groups:

Non-overweight patients (UKPDS 33):

After 3 months dietary run-in: Intensive (target FPG <6.0mmol/L):received insulin (30%) or sulphonylurea (40%) (n=2729); Conventional (target FPG <15.0mmol/L):diet only (30%) (n=1138).

NB. It became clear that no oral agent could maintain the intensive group goal. Thus, combination therapy was used, mixing insulin or metformin with sulphonylureas. Those in the conventional group developing FPG >15mmol/L were secondarily randomised to sulphonylurea or insulin therapy.

Overweight patients (UKPDS 34):

Intensive: treatment with metformin (n=342); conventional: diet only (n=411).

NB. A secondary analysis compared 342 patients in metformin group with 951 patients allocated intensive control with chlorpropamide (n=265), glibenclamide (n=277), or insulin (n=409).

Primary endpoints:

21 endpoints defined, including: any DM–related complications, DM-related deaths, and all-cause mortality.

Follow-up:

3–4 monthly for the duration of the study (10y).

Results

Primary endpoints

Intensive

Conventional

p

RR (95% CI)

Any DM-related

35.3%

38.5%

0.03

0.88 (0.79 to 0.99)

DM-related death

10.4%

11.3%

0.3

0.90 (0.73 to 1.11)

All-cause mortality

17.9%

18.7%

0.4

0.94 (0.80 to 1.10)

Median HbA1C

7.0%

7.9%

<0.0001

MI

14.2%

16.3%

0.05

0.84 (0.71 to 1.00)

Microvascular

8.2%

10.6%

0.01

0.75 (0.60 to 0.93)

Single endpoints

Retinal photocoagulation

7.6%

10.3%

0.003

0.71 (0.53 to 0.96)

Vitreous haemorrhage

0.7%

0.9%

0.5

0.77 (0.28 to 2.11)

Blindness in one eye

2.9%

3.3%

0.4

0.84 (0.51- to 1.40)

Cataract extraction

5.5%

7.0%

0.05

0.76 (0.53 to 1.08)

Intensive metformin treatment decreased the risks of DM-related endpoints in overweight patients (32%, p=0.0023); DM-related death and all-cause mortality (36%, p=0.011); and MI (39%, p=0.010) vs conventional treatment. Oddly, the supplementary analysis of UKPDS 34 showed sulphonylurea-treated patients given metformin had significant increases in DM-related (96%) and all-cause deaths (60%) vs conventional therapy.

Discussion

This study showed the importance of good blood glucose control in reducing the incidence of microvascular complications. The method of control seemed less important.

Problems

  • The endpoints related to eye disease (diabetic retinopathy) were indicators of late disease, and a F/U of only 10y may not be sufficient to get an accurate idea of patients’ risk of developing these complications.

Preventing complications: perindopril and indapamide

ADVANCE (Action in Diabetes and VAscular disease: preterax and diamicroN MR Controlled Evaluation) study:

Effects of a fixed combination of perindopril and indapamide on macrovascular and microvascular outcomes in patients with type 2 diabetes mellitus.

Authors:

ADVANCE collaborative group.

Reference:

Lancet (2007) 370, 829–40.

Study design:

RCT.

Evidence level:

1b.

Key message

In patients with diabetes mellitus (DM), a fixed combination of perindopril-indapamide reduces the risk of death and macrovascular or microvascular events, independent of initial blood pressure (BP) level or concomitant medication.

Impact

The population sampled in this study was large and involved various ethnic groups, thus providing strong evidence for the use of this fixed combination in different settings across the world. The BP-lowering arm of this study ended in June 2007, and the blood glucose-lowering arm (the largest ever to date) in December 2007. Although the results of the latter have not yet been published, interim results have suggested no change in the risk of death from intensive blood glucose lowering. These results do not support those of the ACCORD study, which suggested increased mortality with intensive control. Action to Control Cardiovascular Risk in Diabetes (http://www.accordtrial.org/

Aims

Current recommendations for treating BP in patients with DM are based on arbitrary targets. This target-driven approach is usually resource-intensive, and may miss out patients who are not hypertensive by current definitions but who may still benefit from BP reduction. The BP arm of the ADVANCE trial was designed to verify whether ‘blind’ addition of a fixed combination of perindopril-indapamide to a diverse group of people with type 2 DM, irrespective of their initial BP or other treatment (including BP medication), was safe and effective in reducing vascular events. The blood glucose-lowering arm aimed to determine whether intensive blood glucose-lowering treatment would alter outcomes such as mortality (not yet published).

Methods

Patients:

11,140 patients at 215 centres across 20 countries in Asia, Australasia, Europe, and North America.

Inclusion criteria:

  • Age ≥55y;

  • Diagnosed with type 2 DM after the age of 30;

  • EITHER history of major cardiovascular (CV) disease OR one risk factor for CV disease.

Groups:

2×2 factorial design comparing:

  • Double-blind comparison of perindopril-indapamide (2mg/0.625mg, increased to 4mg/1.25mg to achieve target BP) vs matching placebo: (n=5569 vs 5571);

  • Open-label comparison of modified release gliclazide (30–120mg od) in an intensive regime vs standard glucose control guidelines.

Primary endpoint:

Composite of major macrovascular (CV death, non-fatal MI, non-fatal stroke) and microvascular events (nephropathy, retinopathy).

Secondary endpoints:

All-cause mortality; CV death; total coronary events; major cerebrovascular events; total cerebrovascular events; heart failure; and peripheral vascular disease.

Follow-up:

At 3, 4, and 6 months, then every 6 months until study end.

Results

  • After mean of 4.3y F/U, 73% of those assigned active treatment and 74% of those assigned control remained in the study.

  • Significant reduction in primary endpoint; separate reductions in macrovascular and microvascular complications were not independently significant.

Perindopril-indapamide (n=5569)

Placebo (n=5571)

RRR (95% CI)

p

Primary endpoint

861

938

9% (0 to 17)

0.04

Macrovascular

480

520

8% (–4 to 19)

0.2

Microvascular

439

477

9% (–4 to 20)

0.2

Discussion

This study showed that adding a fixed combination of perindopril-indapamide to a wide variety of people with type 2 DM reduced major vascular events (including death), independent of initial BP levels. In fact, for every 66 patients started on such combination over a 5y period, one would avoid at least one major vascular event, and for every 79 patients, one death would be averted. This may be of importance in the primary care setting as general practitioners may find this a better way of achieving target coronary heart disease reduction without needing specialist input. Many experts argue that it is the BP reduction that is important and not necessarily the particular agent(s) used.

Problem

  • The cost-effectiveness of such a strategy has yet to be established, and this must be taken into consideration before large-scale prescription of this fixed drug combination is widely accepted.

Preventing complications: control of blood pressure

HOT (Hypertension Optimal Treatment) trial:

Effects of intensive blood pressure-lowering and low-dose aspirin in patients with hypertension.

Authors:

Hansson L, Zanchetti A, Carruthers S et al.

Reference:

Lancet (1998) 351, 1755–62.

Study design:

RCT.

Evidence level:

1b.

Key message

Patients with diabetes mellitus (DM) and hypertension benefit from active optimal reduction of their BP to normotensive levels.

Impact

The recommended target diastolic BP (DBP) in patients with DM has been reduced to <80mmHg without any major concern about risk. Before this study, it was debated that BP reduction to normotensive levels was associated with excess cardiovascular (CV) risk. The HOT trial was the first RCT to dispel this misconception.

Aims

The CV effect of BP reduction in hypertensive individuals is a J-shaped relationship—there is a level below which subsequent BP reduction results in adverse CV events. It was generally assumed from observational studies that <140/90mmHg was the optimal threshold. However, this BP level is still above the normotensive range. This trial was designed to assess the risk/benefit of further BP reduction (towards normotensive levels) on CV events. Although this study was primarily aimed at the general hypertensive population, the greatest benefits had previously been observed amongst those with DM; this subgroup will be the main focus of this analysis.

Methods

Patients:

19,193 patients (1,501 with DM) from 26 countries in Europe, North and South America, and Asia.

Inclusion criteria:

  • Age 50–80y;

  • Diagnosis of hypertension;

  • DBP between 100–115mmHg.

Groups:

All three groups received 5mg long-acting calcium channel blocker (felodipine) with subsequent dose adjustments and addition of angiotensin-converting enzyme inhibitors (ACE-Is), beta (β)-blockers or diuretics, as per predefined protocol, to achieve target BP:

  • DBP target ≤ 90mmHg (n=6264; 501 with DM);

  • DBP target ≤ 85mmHg (n=6264; 501 with DM);

  • DBP target ≤ 80mmHg (n=6262; 499 with DM).

Primary endpoint:

Major CV events:

  • Fatal and non fatal MI;

  • Fatal and non-fatal stroke;

  • Other CV deaths.

Follow-up:

At baseline, then at 3 and 6 months; and subsequently every 6 months.

Results

Number of events in patients with DM

DBP ≤90mmHg

≤85mmHg

≤80mmHg

Major CV events

45

34

22

Major CV events, including silent MI

48

42

30

RR (95 % CI)

Major CV events

1.32 (0.84 to 2.06)

1.56 (0.91 to 2.67)

2.06 (1.24 to 3.44)

Major CV events, including silent MI

1.13 (0.75 to 1.71)

1.42 (0.89 to 2.26)

1.60 (1.02 to 2.53)

Discussion

These data clarified the need to reduce BP (especially diastolic) to optimal targets in patients with hypertension. The greatest benefits were observed in patients with DM. This study demonstrated that targets as low as <80mmHg could be achieved with no significant risk. This has been confirmed by subsequent controlled studies and has resulted in lowered BP targets for people with DM. The current National Institute of Health and Clinical Excellence (NICE) guidelines recommend a target of <140/80mmHg in the general population and even lower values in people with DM (depending on the presence or absence of proteinuria).

Problems

  • Renal endpoints were not examined in this study despite the fact that DM remains the commonest cause of end-stage renal disease in the Western world, and microalbuminuria (frequent in DM) is an independent risk factor for coronary heart disease.

  • More recent trials have shown that apart from BP lowering, ACE-Is and angiotensin receptor blockers are more effective in reducing renal outcomes in patients with DM, independent of their BP-lowering effect. These agents are now used as first-line BP agents in patients with DM and proteinuria, with calcium-channel blockers as second line.

Preventing complications: multifactorial intervention

Steno 2:

Intensified multifactorial intervention in patients with type 2 diabetes mellitus and microalbuminuria.

Authors:

Gaede P, Vedel P, Parving H et al.

Reference:

Lancet (1999) 353, 617–22.

Study design:

RCT.

Evidence level:

1b.

Key message

Addressing all diabetes-related co-morbidities using a multifactorial approach, and treating patients to targets is effective in reducing cardiovascular (CV) and microvascular complications.

Impact

Intensified multifactorial treatment has now become a cornerstone in the management of type 2 diabetes mellitus (DM) and associated co-morbidities. So-called treating-to-target is being recommended by UK national (National Institute of Health and Clinical Excellence, NICE) as well as various international guidelines.

Aims

Previous studies had shown the benefit of intensified control of modifiable risk factors (blood sugar, blood pressure, lipids, and smoking cessation) in reducing CV diseases, but none had addressed the effectiveness of such an intensified approach to target multiple risk factors at the same time. The Steno 2 study was conducted to assess the effectiveness of this multifactorial strategy in reducing complications in patients with type 2 DM.

Methods

Patients:

160 patients at 1 centre in Denmark.

Inclusion criteria:

Type 2 DM:

  • Age 40–65y;

  • Persistent microalbuminuria (albumin excretion rate (AER) 30–300mg/24h).

Exclusion criteria:

  • Serum C-peptide concentration <600pmol/L (6min post-injection of 1mg glucagon);

  • Pancreatic insufficiency or DM secondary to pancreatitis;

  • Alcohol abuse;

  • Non-diabetic kidney disease;

  • Malignancy;

  • Life-threatening disease with death probable within 4y.

Groups:

  • Intervention group: Specific dietary advice, smoking cessation, antihypertension treatment, vitamin supplements, 150mg aspirin, stepwise treatment of DM, treatment of raised cholesterol (n=80).

  • Conventional group: routine care (n=80).

Primary endpoint (CV arm):

Composite of death from CV causes, non-fatal myocardial infarction (MI), non-fatal stroke, revascularisation, and amputation from lower limb ischaemia.

Secondary endpoint (microvascular arm):

Nephropathy (median albumin excretion rate >300mg/24h in ≥1 of the two-yearly examinations); neuropathy and retinopathy.

Follow-up:

Every 3 months, for 8y.

Results

Primary endpoint:

  • The unadjusted HR of the composite primary endpoint in the intensive group relative to the conventional group was 0.47 (95% CI: 0.24 to 0.73; p=0.008).

  • After adjusting for diabetes control, age, sex, smoking, and baseline CV status, HR=0.47; 95% CI: 0.22 to 0.74; p=0.01).

Secondary endpoint

RR (intensive vs conventional)

95% CI

p

Nephropathy

0.39

0.17 to 0.87

0.003

Retinopathy

0.42

0.21 to 0.86

0.02

Autonomic neuropathy

0.37

0.18 to 0.79

0.002

Discussion

This study demonstrated that targeted, long-term, intensified intervention in people with type 2 DM and microalbuminuria reduced the rate of vascular events. The data suggest that the NNT over the 7.8y F/U period to prevent one CV event was 5. This study demonstrated the impact of targeted multifactorial therapy in the management of type 2 DM. Of particular note, almost every individual component assessed in the intervention group (BP, cholesterol, blood glucose, etc) has been shown in subsequent trials to be effective in reducing CV endpoints. Today, integrated multiple risk factor assessment and therapy has become the pinnacle of DM management.

Preventing complications: statins

CARDS (Collaborative AtoRvastatin Diabetes Study):

Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes.

Authors:

Colhoun H, Betteridge J, Durrington P et al.

Reference:

Lancet (2004) 364, 685–96.

Study design:

RCT.

Evidence level:

1b.

Key message

Atorvastatin 10mg daily is safe and effective in preventing the onset of first CV events in people with type 2 diabetes mellitus (DM) without high LDL cholesterol.

Impact

Statin therapy is now indicated for primary prevention in people with type 2 DM and one additional risk factor for CV disease, including those patients with ‘normal’ cholesterol levels.

Aims

Available evidence on statin use in primary prevention of CV disease in type 2 DM had been extrapolated from studies (e.g. Heart Protection Study) where DM was not the main focus, and this was reflected by the reluctance of those caring for people with DM to prescribe these agents. This trial sought to clarify whether atorvastatin 10mg was better than placebo in the primary prevention of CV diseases in patients with type 2 DM.

Methods

Patients:

2838 patients at 132 centres in the UK.

Inclusion criteria:

Type 2 DM:

  • Age 40–75y;

  • No previous history of CV disease;

  • LDL cholesterol ≤4.14mmol/L;

  • Triglycerides ≤6.79mmol/L;

  • One of: retinopathy/hypertension/albuminuria/current smoking.

Groups:

  • Atorvastatin 10mg od (n=1428);

  • Matching placebo od (n=1410).

Primary endpoint:

Time to first occurrence of any of:

  • Acute coronary heart disease (CHD) events (myocardial infarction, unstable angina, CHD death, resuscitated cardiac arrest);

  • Coronary revascularisation;

  • Stroke.

Secondary endpoints:

  • Total mortality;

  • Hospital-verified CV endpoint.

Follow-up:

Monthly for the first 3 months; then at 6 months; then every 6 months, for endpoints, BP measurement and weight.

Results

Number of patients with an event

Placebo

Atorvastatin

HR

p

Primary endpoint

127

83

0.63

0.001

Acute coronary event

77

51

0.64

Coronary revascularisation

34

24

0.69

Stroke

39

21

0.52

Secondary endpoint

Death from any cause

82

61

0.73

0.06

Primary endpoint

189

134

0.68

0.001

Discussion

Diabetes management was initially focused on blood sugar control until the recognition that most deaths in these patients were due to CV disease. Since then, hypertension management has also been considered integral to diabetes care. Patients with DM present with dyslipidaemia where HDL cholesterol is low, triglycerides are raised. and the classical culprit, LDL cholesterol, may be normal—a different lipid profile from patients in whom statins were previously found to be beneficial. This study demonstrated, that in patients with at least one other risk factor for CV disease, statin therapy reduced death and CV endpoints. The NNT over 4y was 27 patients to prevent one major vascular event. The benefits of statin therapy were so significant that the study was terminated 2y prior to its expected end date. In a recent paper (Diabetes Care (2006) 29, 2378–84), Neil et al. demonstrated the safety and efficacy of a similar approach in elderly people aged 65–75y.

Nephropathy: ACE inhibitors

The effect of angiotensin-converting enzyme inhibition on diabetic nephropathy.

Authors:

Lewis E, Hunsicker L, Bain R et al.

Reference:

N Engl J Med (1993) 329, 1456–62.

Study design:

RCT.

Evidence level:

1b.

Key message

ACE inhibitors (ACE-Is) provide additional benefit beyond blood pressure (BP) control in slowing diabetic nephropathy.

Impact

ACE-Is are the first choice antihypertensive in patients with diabetes mellitus (DM), especially those with proteinuria.

Aims

The progression of diabetic nephropathy is due to haemodynamic rather than metabolic factors. Prior to this trial, it was known that strict BP control could slow the progression, but it remained unclear whether ACE-Is could have additional benefits, independent of their antihypertensive effect. This study aimed to resolve the issue.

Methods

Patients:

409 patients at 30 nephrology centres in the USA.

Inclusion criteria:

Diabetic nephropathy:

  • Age 18–49y;

  • Type 1 DM for ≥7y with onset before age 30y;

  • Diabetic retinopathy;

  • Urinary protein excretion >0.5 g/d;

  • Serum creatinine <221 µmol/L.

Groups:

Both groups had target BP <140/90 mmHg:

  • Captopril (n=207);

  • Placebo (n=202).

Primary endpoint:

Time to doubling of baseline serum creatinine.

Secondary endpoints:

  • Time to death, dialysis, or transplantation;

  • Hyperkalaemia.

Follow-up:

  • At 2wk, 1 month, and then every 3 months until the patient died, started dialysis, or had renal transplant.

  • Median F/U=3y.

Results

Primary endpoint

Captopril

Placebo

p

Doubling of baseline creatinine

25

43

0.007

Secondary endpoints

Death

8

14

Not reported

Death, dialysis, or transplantation

23

42

0.006

Hyperkalaemia

3

0

Not reported

Discussion

This study confirmed the central role of ACE-Is in the treatment of diabetic nephropathy. It was the first study to clearly demonstrate that they had additional benefits beyond simple lowering of BP.

Problems

  • In current practice, the BP control would no longer be considered optimal, so the additional benefits of ACE-Is if BP was already <130/80mmHg might not be as apparent.

  • The rate of progression of renal disease in the control group was higher than expected from the contemporary literature. However, as this was an RCT, effects causing this accelerated decline should have impacted equally on both groups.

  • The patients were quite highly selected, but are still probably representative of the general type 1 DM population.

Nephropathy: angiotensin II receptor antagonists

RENAAL (Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan) study:

Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy.

Authors:

Brenner B, Cooper ME, Zeeuw D et al.

Reference:

N Engl J Med (2001) 345, 861–9.

Study design:

RCT.

Evidence level:

1b.

Key message

The angiotensin receptor blocker (ARB), losartan, reduces renal endpoints, independent of its blood pressure (BP)-lowering effect in patients with type 2 diabetes mellitus (DM) and advanced nephropathy.

Impact

ARBs have become the first-line agents in lowering BP in patients with type 2 DM and nephropathy. This has also made ARBs (together with angiotensin-converting enzyme inhibitors, ACE-Is), first-line agents in patients with DM and microalbuminuria or proteinuria, even in the absence of hypertension.

Aims

DM is the commonest cause of end-stage renal disease (ESRD). Interruption of the renin angiotensin system is known to slow the progression of renal disease in type 1 DM. However, type 2 DM is the most common form of diabetes. This trial was designed to assess the role of the ARB, losartan, in patients with type 2 DM and nephropathy. It specifically sought to determine whether there was any benefit independent of its BP-lowering effect.

Methods

Patients:

1513 patients at 250 centres in 28 countries.

Inclusion criteria:

  • Age 31–70y;

  • Type 2 DM;

  • Nephropathy: ratio of urinary albumin to urinary creatinine ≥300 (or urinary protein excretion ≥0.5g/d) and serum creatinine values between 1.3–3.0mg/dL (115–265µmol/L).

Primary endpoint:

Time to ESRD, death, or doubling of baseline serum creatinine.

Secondary endpoints:

  • Composite of morbidity/mortality from cardiovascular (CV) causes;

  • Changes in the level of proteinuria;

  • Rate of progression of renal disease (inverse of serum creatinine).

Follow-up:

At 1 month (titration of dose of losartan to 100mg), at 3 months (to titrate other antihypertensive medications to achieve BP ≤140/90), and subsequently every 3 months (to assess endpoints) up to a mean F/U of 3.4y.

Results

Losartan (n=751)

Placebo (n=762)

p

RRR (95 % CI)

Composite primary endpoint

327

359

0.02

16 (2 to 28)

Doubling of serum creatinine

162

198

0.006

25 (8 to 39)

ESRD

147

194

0.002

28 (11 to 42)

Death

158

155

0.9

−2 (−27 to 19)

Discussion

Previous trials had shown that ARBs (irbesartan) reduced the progression from microalbuminuria to overt diabetic nephropathy in patients with DM and hypertension (IRMA. N Engl J Med (2001) 345, 870–8). However, none had examined their effects on overt nephropathy. This study extended the scope of ARBs by demonstrating their usefulness in reducing renal endpoints in hypertensive patients with DM and nephropathy. It also confirmed previous suggestions that ARBs could reduce CV endpoints (data not shown) in these patients, although the study was not sufficiently powered to specifically examine this endpoint.

Problems

  • This trial was terminated prematurely as concurrent evidence from the HOPE trial (N Engl J Med. 2000; 342(3):145–53) suggested that ACE-Is (closely related in mechanism of action) were effective at reducing CV endpoints in patients with renal impairment, including those with DM. ACE-I therapy was one of the exclusion criteria in the RENAAL study. Hence, the steering committee voted to discontinue the study. However, it is being argued that had the study continued to its intended 4.5y mean F/U period, this would only have increased the power and heightened the strength of an association that had already been proven to exist.

  • Although analyses were done on an intention-to-treat basis, there still remained a worryingly high proportion of patients who discontinued the study from both the placebo (53.5%) and losartan (46.5%) groups.

Gestational diabetes: screening and treatment

ACHOIS (Australian CarboHydrate Intolerance Study in pregnancy) trial:

Effect of treatment of gestational diabetes mellitus on pregnancy outcomes.

Authors:

Crowther C, Hiller J, Moss J et al.

Reference:

N Engl J Med (2005) 352, 2477–86.

Study design:

RCT.

Evidence level:

1b.

Key message

Routine screening for and intensive treatment of gestational diabetes in women reduces the rates of serious perinatal complications.

Impact

Although uncertainty remains as to the cut-off levels for glucose intolerance in pregnancy (or gestational diabetes), it is now standard, internationally recommended practice to screen at-risk women, and intensively manage gestational diabetes with diet, glucose monitoring, and insulin, as needed: this is reflected in the current American Diabetes Association, American College of Obstetrics and Gynaecology, and National Institute of Health and Clinical Excellence (NICE) guidelines.

Aims

The risks associated with gestational diabetes are well known, but the benefits of screening and treatment of this condition were uncertain. The ACHOIS trial was designed to determine whether treatment of women with gestational diabetes reduced the risk of perinatal complications.

Methods

Patients:

1000 patients at 14 Australian and four UK centres.

Inclusion criteria:

Women with singleton or twin pregnancy:

  • Between 16–30wk gestation;

  • ≥1 risk factors for gestational diabetes on selective screening or a positive 50g oral glucose challenge test (OGGT, glucose levels 1h after challenge ≥7.8mmol/L.);

  • 75g OGGT at 24–34wk gestation with fasting venous plasma glucose (FPG) <7.8mmol/L and 2h value of 7.8–11.0mmol/L post-load.

Exclusion criteria:

More severe glucose impairment.

Groups:

  • Intervention group: Individualised dietary advice, monitoring of blood glucose, insulin therapy with dose readjustments (n=490).

  • Routine-care group: Care left to the discretion of the attending clinician (n=510).

Primary endpoints:

  • Infants: Composite measure of serious perinatal complications: death, shoulder dystocia, bone fracture, and nerve palsy; admission to neonatal nursery; jaundice requiring phototherapy.

  • Mothers: Need for induction of labour and caesarean section.

Secondary endpoints:

  • Infants: Gestational age at birth.

  • Mothers: Number of prenatal visits, mode of birth, weight gain, number of antenatal admissions, presence or absence of pregnancy-induced hypertension.

Follow-up:

Until birth.

Results

Primary endpoints

Intervention

Routine care

p

Any serious perinatal complications

7 (1%)

23 (4%)

0.01

Admission to neonatal nursery

357 (71%)

321 (61%)

0.01

Induction of labour

189 (39%)

150 (29%)

<0.001

Discussion

This RCT showed that treatment of women with gestational diabetes—including dietary advice, blood glucose monitoring, and insulin therapy—reduced the rate of serious perinatal outcomes. In addition, the infants born to mothers in the intervention group were less likely to be large for gestational age or macrosomic, and were not more likely to be small for gestational age. Earlier studies had shown that infants with macrosomia or large for gestational age were more likely to develop diabetes mellitus (DM) or glucose intolerance, and that female infants would go on to develop gestational diabetes. Therefore, it is tempting to speculate that intensive treatment of gestational diabetes could reduce the risk of infants developing DM in later life. The health gains demonstrated in this study were obtained without significant harm (such as hypoglycaemia).

Problem

  • There is controversy as to which diagnostic test is best in screening for gestational diabetes, and whether screening should be routine or selective. However, in a recent paper, van Leeuwen et al (Diabetes Care 2007; 30:2779-84) showed that the 50g glucose challenge test performed better than conventional fasting blood glucose as a screening test. This makes it hard to translate the findings of this study into practical recommendations as the cost-effectiveness of intensive management would depend on proper identification of patients and targeted individualised therapy.