National Evidence-Based Guideline For Blood Glucose Control In Type 2 .

Section 1: Blood Glucose ControlQuestionWhat is the effect of improving blood glucose control ona) Microvascular complications (retinopathy, neuropathy, nephropathy)b) Macrovascular complications (heart disease, stroke, peripheral vascular disease)c) Quality of lifeRecommendationBlood glucose control should be optimised because of its beneficial effects on thedevelopment and progression of microvascular complications. (Grade A)Evidence Statements Improving blood glucose control in people with type 2 diabetes reduces the developmentor progression of microvascular complications.Level of Evidence I No clear independent effect of improving blood glucose control on macrovascularcomplications has been demonstrated in people with type 2 diabetes.Level of Evidence I The effect of tight blood glucose control on premature mortality in people with type 2diabetes remains uncertain.Level of Evidence I There is an association between blood glucose control and quality of life in people withtype 2 diabetes.Level of Evidence IIType 2 Diabetes Guideline7Blood Glucose Control, July 2009

Background – Improving blood glucose control in people with type 2 diabetesType 2 diabetes is associated with reduced life expectancy, significant morbidity due to thespecific diabetes related microvascular complications (retinopathy, nephropathy andneuropathy), and the increased risk of macrovascular complications (ischemic heart disease,stroke and peripheral vascular disease). The development of these complications impacts onquality of life.In Australia, type 2 diabetes results in premature death and irreversible long termcomplications including myocardial infarction, stroke, retinopathy and blindness, renaldisease requiring dialysis or transplantation, neuropathy, foot ulcer, amputation, and erectiledysfunction.In 2004, diabetes was among the top ten leading causes of death being the direct cause of2.7% of deaths in Australia, and being associated with another 6% of deaths (AustralianBureau of Statistics, 2006). Cardiovascular disease is the major cause of death in peoplewith diabetes, accounting for approximately 50% of all fatalities (International DiabetesFederation, 2006). In 2005, diabetes was associated with cause of death in nearly 11,900Australian deaths or 9% of all deaths that year. Approximately half of these deaths involvedCHD (48%), stroke (16%), and PVD in 6% of diabetes deaths (Diabetes: Australian Facts2008).Over 81,000 hospitalisations occurred in 2004-05 where both diabetes and CHD werepresent, which accounted for 15.3% of all diabetes hospitalisations. In the same years, strokefrom diabetes amounted to 2.2% of all diabetes hospitalisations and peripheral vasculardisease accounted for 5.9% of all diabetes hospitalisations (Diabetes Australian Facts,2008).Age at diagnosis has an important influence on the occurrence of outcomes. People whowere older at diagnosis had more complications at baseline (1997). However, a recent studyreported an increased inherent susceptibility to retinopathy with earlier onset diabetes(Wong et al., 2008). Even after adjusting for glycaemic exposure, age of diagnosis was anindependent predictor of long term retinopathy. Furthermore, young adults with early-onsetdiabetes are at a much greater risk of CVD relative to matched controls (Hillier and Pedula,2003). Hanefeld et al (1996) found that all-cause mortality in newly diagnosed type 2diabetes followed for 12 years was increased 5.1-fold in males and 7-fold in women aged36-45 and 2-fold in males and 3.5-fold in women aged 46-55 years.While there is evidence in the general population that mortality from heart disease isdecreasing, the pattern in people with diabetes is different. In representative cohorts ofpeople with and without diabetes followed for 8 to 9 years from the First National Healthand Nutrition Examination Survey (NHANES I) and the NHANES I Follow-up Survey(NHEFS), there was a 36.4% decline in age-adjusted heart disease mortality in men withoutType 2 Diabetes Guideline8Blood Glucose Control, July 2009

diabetes compared with a 13.1% decline in men with diabetes for the two periods. Forwomen, the situation was worse with a decline of 27% in non-diabetic women, but anincrease of 23% in diabetic women (Gu et al., 1999).Over the past decade intervention studies have examined the effect of lowering bloodglucose levels in people with type 2 diabetes. This section examines the evidence of therelationship between blood glucose control and diabetes vascular complications and theimpact on quality of life.Type 2 Diabetes Guideline9Blood Glucose Control, July 2009

Evidence – Improving blood glucose control in people with type 2 diabetes andcomplicationsImproving blood glucose control in people with type 2 diabetes reduces thedevelopment or progression of microvascular complicationsA number of systematic reviews have examined the relationship between blood glucosecontrol and long term complications in people with type 2 diabetes (Gaster and Hirsch,1998; O'Connor et al., 1998; Vaaler, 2000; Woolf et al., 2000). These studies concluded thatimproved glycaemic control can reduce retinopathy, renal disease and neuropathy in peoplewith type 2 diabetes are largely based on the results of the Kumamoto study in Japan and theUKPDS study.The Kumamoto study (Ohkubo et al., 1995) was a prospective study conducted in 110 nonobese insulin-requiring Japanese people with type 2 diabetes. Subjects included 55 peoplewithout evidence of retinopathy or urinary albumin excretion 30 mg/24 h at baseline(primary prevention cohort) and 55 who showed “simple” retinopathy and urinary albuminexcretion 300 mg/24 h at baseline (secondary prevention cohort). Participants wererandomly allocated to intensive treatment with multiple insulin injections ( 3) orconventional treatment with 1-2 injections daily. The intensive treatment group achieved amean HbA1c of 7.1% and the conventionally treated group a mean HbA1c of 9.4% duringthe 6-year study period. There were significantly less people in the multiple injection groupcompared with the conventional group who developed retinopathy in the primary (7.7 vs32% respectively, p 0.04) and secondary prevention (19.2 vs 44% respectively, p 0.05)groups. Similar results were found for primary prevention of nephropathy (7.7 vs 28%respectively, p 0.03) and secondary prevention of nephropathy (11.5 vs 32% respectively,p 0.04). The odds ratio (OR and 95% CI) for the development or progression ofnephropathy was 0.26 (CI 0.09-0.76) and the number needed to treat was 5 (CI 4-19).Intensive glycaemic control delayed the onset and the progression of the early stages ofretinopathy, nephropathy and neuropathy in people with type 2 diabetes. These results wereconfirmed in an 8-year follow-up report of the Kumamoto study (Shichiri et al., 2000) whereintensive glycaemic control in people with type 2 diabetes effectively continued to delay theonset and progression of microvascular complications including retinopathy, nephropathy,and neuropathy.The UKPDS was a randomised controlled trial which compared the effects of intensiveblood-glucose control with either sulphonylurea or insulin and conventional treatment on therisk of microvascular and macrovascular complications in 3,867 people with newlydiagnosed type 2 diabetes (median age 54 years) (UKPDS Study Group, 1998). After 3months of diet treatment, subjects were randomly assigned to an intensive treatment policywith a sulphonylurea (chlorpropamide, glibenclamide, or glipizide) or with insulin, or toconventional treatment. The aim in the intensive group was to achieve a fasting plasmaglucose (FPG) of than 6 mmol/L while in the conventional group, the aim was the bestType 2 Diabetes Guideline200910Blood Glucose Control, July

achievable FPG with diet alone and pharmacotherapy was added only if there werehyperglycaemic symptoms or the FPG was 15 mmol/L. Aggregate endpoints were anydiabetes-related endpoint (sudden death, death from hyperglycaemia or hypoglycaemia, fatalor non-fatal myocardial infarction, angina, heart failure, stroke, renal failure, amputation (ofat least one digit), vitreous hemorrhage, retinopathy requiring photocoagulation, blindness inone eye, or cataract extraction); diabetes-related death (death from myocardial infarction,stroke, peripheral vascular disease, renal disease, hyperglycaemia or hypoglycaemia, andsudden death); all-cause mortality. Over 10 years, the median HbA1c was 7.0% in theintensive group compared with 7.9% in the conventional group – an 11% reduction(p 0.0001). There was no difference in HbA1c among agents in the intensive group.Compared with the conventional group, the risk in the intensive group was 12% lower (95%CI 1–21, p 0.03) for any diabetes-related endpoint; 10% lower (–11 to 27, p 0.34) forany diabetes-related death; and 6% lower (–10 to 20, p 0.4) for all-cause mortality. Mostof the risk reduction in the any diabetes-related aggregate endpoint was due to a 25% riskreduction (7–40, p 0.01) in microvascular endpoints, including the need for retinalphotocoagulation. There was no difference for any of the three aggregate endpoints betweenthe three intensive agents (chlorpropamide, glibenclamide, or insulin. In a 10-year followup, Holman et al (2008) monitored 3,277 people from the UKPDS. Over the first five yearsparticipants were asked to attend annual UKPDS clinics, but with no attempt made tomaintain their previously assigned therapies. Annual questionnaires were used to followparticipants who were able to attend the clinics, and all people in years 6 to 10 were assessedvia questionnaires. At 10 years, there was a 24% risk reduction in the sulphonylurea–insulingroup for microvascular disease (p 0.001) compared with the conventional group. Thisdifference was noted despite a lack of difference in HbA1c levels between the intensive andcontrol groups following the conclusion to the randomized intervention in 1997, suggestinga continuing legacy effect of the prior period of improved blood glucose control. In themetformin-treated group, no significant risk reduction was observed for microvasculardisease compared with the conventional group.The first randomised control trial to examine the effects of different agents in diabetesoutcomes (Smelo, 1971) – the University Group Diabetes Program (UGDP) – is invariablyexcluded from these reviews because it did not produce significant differences in glucosecontrol in some treatment arms and lacked statistical power (Woolf et al., 2000).Furthermore, the study pre-dated the use of glycated haemoglobin to measure longer termdiabetes control. This UGDP involved the recruitment of 823 people with newly diagnoseddiabetes, with a mean age of 53 years, who were randomised to treatment with placebo,tolbutamide, a fixed amount of daily insulin and a variable dose of insulin. Subjects werefollowed up for 5 years. There was little evidence showing that insulin treatment was anybetter than diet alone in changing the course of vascular complications in stable type 2diabetes. The only difference in outcomes observed in this study was increasedcardiovascular mortality in the tolbutamide treated group.Type 2 Diabetes Guideline200911Blood Glucose Control, July

The feasibility study for the Veterans Affairs Cooperative Study on Glycaemic Control andComplications in type 2 diabetes (VA-CSDM) study was conducted in 153 men with type 2diabetes from five medical centres (mean age 60 years, mean duration of diabetes 7.8 years)(Emanuele et al., 1996). Participants were randomly assigned to receive standard insulintreatment (one injection in the morning) or intensified treatment (a stepped plan startingwith one insulin injection in the evening glipizide up to multiple daily injections withoutglipizide) and were prospectively followed for a mean of 27 months. By 6 months theintensive blood glucose control group achieved a mean HbA1c of 7.1% (9.8% at baseline)while the standard treatment group had reached a baseline HbA1c of 9.5%. This level wasmaintained throughout the study period. The difference in HbA1c in the two groups was2.1% (p 0.001). Over this relatively short study period intensive therapy was notassociated with either a worsening or improvement in retinopathy. This result was notunexpected and was consistent with results of the UKPDS where a difference in retinopathywith intensive treatment was not observed until after 3 years (UKPDS Study Group, 1998).The effect of improved diabetes control on neuropathy has also been reported. TheKumamoto study reported that lowering blood glucose increased median nerve conductionvelocity in the intensively treated group compared with the conventional group and reducedthe arm vibration threshold (p 0.05 for both), but other physiological measures wereunaffected (Ohkubo et al., 1995). A follow up report of the Kumamoto study with 8 years ofobservation confirmed these findings and reported a significant deterioration in theseparameters in the conventionally treated group (Shichiri et al., 2000). In the UKPDS therewas no effect on the incidence of absent ankle and knee reflexes; however, although thenumber of subjects was small, abnormal biothesiometer ( 25 V) was significantly less inthe intensive policy group (p 0.0052) after 15 years of follow up (UKPDS Study Group,1998).A short-term study in 54 Japanese people with type 2 diabetes (mean age 49 years, meanduration of diabetes 10 years) was conducted to assess the reversibility of autonomic nervefunction in relation to improved glycaemic control (Isotani and Fukumoto, 2000). Thesubjects were admitted to hospital for 4 weeks and placed on a strict diet and treatment.HbA1c improved from 9.9% to 8.6% and was associated with a significant improvement indark-adapted pupillary area, an indicator of autonomic neuropathy, suggesting thatautonomic neuropathy could be improved by rapid improvement in diabetes control.An American study of 780 people with type 1 and type 2 diabetes examined the risk of deathor renal failure with long-term intensive diabetes treatment (Hellman et al., 1997). Thegroup was split into two groups: those with a longer duration of intensive therapy (medianduration 11 y, group 1), and 571 subjects with a shorter duration of intensive therapy(median duration 1 y, group 2). The intensive treatment involved maintaining FPGbetween 3.9 and 6.3 mmol/L and having an HbA1c 6.4%, seeing a physician an average offive times a year, regular telephone contact with a diabetes educator and participating in adiabetes education program. Intensive insulin therapy ( three injections a day) was alsoType 2 Diabetes Guideline200912Blood Glucose Control, July

used in half of this group. Of the study population 113 people with type 2 diabetesmaintained the intensive therapy long term (median 11 y) and these were compared withthe 377 people with type 2 diabetes who only achieved this short term (median 1 y).Baseline HbA1c in the two groups was 10.5% and 10.9% respectively. Overall, despite thegreater number of people with a higher initial comorbidity, group 1 subjects had asignificant reduction in mortality (25.9 vs 33.3%, p 0.05). People aged 65 y were lesslikely to die in the intensive group than in the control group (17.2 vs 29.7%, p 0.04). In theintensive group those who were on more intensive insulin therapy had a lower mortality,cardiac specific mortality and cardiac mortality with renal comorbidity than those whoreceived conventional insulin therapy (p 0.02, p 0.001 and p 0.02, respectively). Therewas no significant difference between the total number of renal events (dialysis,transplantation or death) in the two groups in people with type 2 diabetes who were under 65years.Intensified blood glucose control appears to reduce the incidence of albuminuria. In theKumamoto study the cumulative development and progression in nephropathy (defined byincrease in urinary albumin excretion) after 6 years was 7.7% in the intensively treatedgroup and 28% in the conventionally treated group in the primary prevention group(p 0.03) and 11.5% and 32% respectively in the secondary prevention group (p 0.04)(Ohkubo et al., 1995). The UKPDS observed a lower incidence of microalbuminuria withintensified treatment which became statistically significant after three years and a lowerincidence of gross proteinuria and increased plasma creatinine within nine years of followup (relative risk reduction 17%, 33%, and 60%, respectively). The incidence rates of renalfailure and death from renal disease did not differ significantly between the groups, but theabsolute number of cases was small. (Levin et al., 2000) reported the effect of intensifiedinsulin trea

Practice Point (s) - including experts' consensus in absence of gradable evidence Evidence Statements - supporting the recommendations Background - to issues for the guideline Evidence - detailing and interpreting the key findings Evidence tables - summarising the evidence ratings for the articles reviewed