Diabetes mellitus in Kabuki syndrome 1 on a background of post-transplant diabetes mellitus.

Summary: Kabuki syndrome is a genetic disorder characterised by distinctive facial features, developmental delays, and multisystem congenital anomalies. Endocrine complications such as premature thelarche and short stature are common, whereas disorders of glycaemic control are less frequent. We describe a 23-year-old white female referred to the diabetes clinic for hyperglycaemia during haemodialysis. She was subsequently diagnosed with Kabuki syndrome based on characteristic clinical features, confirmed by detecting a heterozygous pathogenic variant in KMT2D. She was known to have had multiple congenital anomalies at birth, including complex congenital heart disease and a single dysplastic ectopic kidney, and received a cadaveric transplanted kidney at the age of 13. She had hyperglycaemia consistent with post-transplant diabetes mellitus (DM) and was started on insulin. Examination at the time revealed truncal obesity. She developed acute graft rejection and graft failure 14 months post-transplant and she was started on haemodialysis. Her blood glucose levels normalised post-graft explant, but she was hyperglycaemic again during haemodialysis at the age of 23. Given her clinical phenotype, negative diabetes antibodies and normal pancreas on ultrasound, she was assumed to have type 2 DM and achieved good glycaemic control with gliclazide. Learning points: Involve clinical genetics early in the investigative pathway of sick neonates born with multiple congenital anomalies to establish a diagnosis to direct medical care. Consider the possibility of Kabuki syndrome (KS) in the differential diagnoses in any neonate with normal karyotyping or microarray analysis and with multiple congenital anomalies (especially cardiac, renal, or skeletal), dysmorphic facial features, transient neonatal hypoglycaemia and failure to thrive. Consider the possibility of diabetes as an endocrine complication in KS patients who are obese or who have autoimmune disorders.

Glucocorticoid-induced hyperglycaemia in respiratory disease: a systematic review and meta-analysis.

The relative risk of glucocorticoid-induced hyperglycaemia is poorly quantified. We undertook a meta-analysis to estimate the association between glucocorticoid treatment and hyperglycaemia, overall and separately in individuals with and without diabetes and underlying respiratory disease. We searched electronic databases for clinical trials of adults randomized to either glucocorticoid treatment or placebo. Eight articles comprising 2121 participants were identified. We performed a random effects meta-analysis to determine relative risks for the associations between glucocorticoid use and both hyperglycaemia and starting hypoglycaemic therapy. In all individuals, the relative risk of hyperglycaemia comparing glucocorticoid treatment with placebo was 1.72 [95% confidence interval (CI) 1.50-2.04; p < .001]. The relative risks in individuals with and those without diabetes were 2.10 (95% CI 0.92-5.02; p = .079) and 1.50 (95% CI 0.79-2.86; p = .22), respectively. In all individuals, the relative risk of hyperglycaemia requiring initiation of hypoglycaemic therapy, comparing glucocorticoid treatment with placebo, was 1.73 (95% CI 1.40-2.14; p < .001). In conclusion, glucocorticoid therapy increases the risk of hyperglycaemia in all individuals with underlying respiratory disease but not when diabetic status is analysed separately.

Smoking and other factors associated with short-term partial remission of Type 1 diabetes in adults.

AIMS: The duration of partial remission of Type 1 diabetes is associated with the degree of initial metabolic disturbance and features of insulin resistance. Cigarette smoking decreases insulin sensitivity, but its influence on the length of remission is unknown. Therefore, this study assessed the relationship between cigarette smoking and duration of partial remission in adults with newly diagnosed Type 1 diabetes. METHODS: We recruited 149 patients (48 women and 101 men, aged 16-35 years, median age 25 years), admitted to a teaching hospital with newly diagnosed Type 1 diabetes and followed them for a median period of 1 year and 9 months. We introduced intensive insulin therapy in multiple injections (basal-bolus) in all patients. We defined partial remission as an insulin dose of ≤ 0.3 U/kg body weight/24 h, an HbA(1c) value < 53 mmol/mol (7.0%) and a random serum C-peptide concentration over 0.5 ng/ml. Cigarette smoking was determined by self-report. RESULTS: Of 149 patients, 68 (46%) fulfilled the criteria for partial remission at 1 year after diagnosis of diabetes. Fewer patients who were in partial remission at 1 year smoked (19/68, 28%) than did patients that were not in partial remission (41/81, 51%). In logistic regression analyses, non-smoking was associated with remission at 1 year independent of age, sex, HbA(1c) and presence of diabetic ketoacidosis, all measured at onset of diabetes (OR 3.32, 95% CI 1.42-7.75, P = 0.005). CONCLUSION: Relative to individuals in this study who smoked, those who did not smoke at diagnosis of Type 1 diabetes experienced a longer duration of partial remission.

Maternal waist to hip ratio is a risk factor for macrosomia.

OBJECTIVE: Fetal growth during pregnancy may be affected by the metabolic activity and distribution of fat stores in women. This study investigates the association between waist to hip ratio (WHR) as a measure of the distribution of adiposity in primiparous mothers living in Avon, England, and macrosomia in their offspring. DESIGN: Prospective historical cohort study. SETTING: The Avon Longitudinal Study of Parents and Children (ALSPAC) prospective cohort study in Avon, UK. POPULATION: A cohort of 3083 primiparous women with a term singleton delivery with expected dates of delivery from 1 April 1991 to 31 December 1992. METHODS: The distribution of WHR was categorised into quartiles. We compared the second, third and fourth quartiles against the first (reference) quartile with respect to whether the mother delivered a macrosomic newborn. We controlled for maternal age, gestational age, body mass index (BMI), marital status and racial group using multivariate logistic regression. MAIN OUTCOME MEASURES: Macrosomia defined in three ways: birthweight ≥ 4000 g; birthweight ≥ 4500 g; large for gestational age (LGA: ≥ 95th percentile of birth weight adjusted for sex and gestational age). RESULTS: Waist to hip ratios in the third and fourth quartiles were associated with a higher odds of delivering a macrosomic infant, defined as a birthweight ≥ 4000 g (third quartile, OR 1.59, 95% CI 1.12-2.26; fourth quartile, OR 1.69, 95% CI 1.18-2.42) or as LGA (≥95th percentile of the cohort; third quartile, OR 1.77, 95% CI 1.10-2.85; fourth quartile, OR 1.78, 95% CI 1.09-2.91). When defined as a birthweight ≥ 4500 g, the fourth quartile was associated with increased odds of macrosomia (OR 2.74, 95% CI 1.05-7.16). Odds ratios after adjustment for confounding factors followed a similar pattern. CONCLUSION: Independent of confounding factors, women with increased WHRs were significantly more likely to give birth to macrosomic newborns.

Association between alcohol consumption and diabetic retinopathy and visual acuity-the AdRem Study.

AIMS: We investigated the association between alcohol consumption and diabetic retinopathy and deterioration of visual acuity in individuals with Type 2 diabetes. METHODS: We conducted a cohort analysis of 1239 participants with Type 2 diabetes aged 55-81 years enrolled in the AdRem study, a sub-study of the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE) trial. Current and past consumption of wine, spirits and beer was measured by self-report. Moderate and heavy alcohol consumption was defined as 1-14 and >14 drinks/week, respectively. Diabetic retinopathy, measured by mydriatic stereoscopic seven-field retinal photography, was defined by a 2-step progression in the Early Treatment of Diabetic Retinopathy Study (ETDRS) score or the presence of any retinal vascular lesions. Deterioration of visual acuity was defined by a decrease of two lines in best vision in either eye, measured corrected, or through a pinhole using a Snellen chart. RESULTS: In a mean follow-up of 5.5 years, we identified 182 participants with a 2-step progression in the ETDRS score, 640 participants with the presence of any retinal vascular lesions and 693 participants with a deterioration of visual acuity. Current moderate consumption of alcohol, compared with no current consumption, was not associated with presence or progression of diabetic retinopathy; however, it was associated with higher risk of deterioration of visual acuity (multivariable-adjusted OR 1.83; 95% CI 1.34-2.48; P<0.001). CONCLUSIONS: Alcohol consumption is associated with increased risk of deterioration of visual acuity, but not with retinopathy in individuals with Type 2 diabetes.

Association between glycated haemoglobin and the risk of lower extremity amputation in patients with diabetes mellitus-review and meta-analysis.

AIMS/HYPOTHESIS: Diabetes increases the risk of lower extremity amputation (LEA). Although epidemiological studies report positive associations between glycaemia and LEA, the magnitude of the risk is not adequately quantified and clinical trials to date have not provided conclusive evidence about glucose lowering and LEA risk. We synthesised the available prospective epidemiological data on the association between glycaemia measured by HbA(1c) and the risk of LEA in individuals with diabetes. METHODS: We searched electronic databases and reference lists of relevant articles. We considered prospective epidemiological studies that had measured HbA(1c) level and assessed LEA as an outcome among diabetic individuals without acute foot ulcerations or previous history of amputation. Of 2,548 citations identified, we included 14 studies comprising 94,640 participants and 1,227 LEA cases. We abstracted data using standardised forms and obtained data from investigators when required. Data included characteristics of study populations, HbA(1c) assay methods, outcome and covariates. Study-specific relative risk estimates were pooled using random-effects model meta-analysis; heterogeneity was explored with meta-regression analyses. RESULTS: The overall RR for LEA was 1.26 (95% CI 1.16-1.36) for each percentage point increase in HbA(1c). There was considerable heterogeneity across studies (I (2) 76%, 67-86%; p < 0.001), which was not accounted for by recorded study characteristics. The estimated RR was 1.44 (95% CI 1.25-1.65) for type 2 diabetes and 1.18 (95% CI 1.02-1.38) for type 1 diabetes; however, the difference was not statistically significant (p = 0.09). We found no strong evidence for publication bias. CONCLUSIONS/INTERPRETATION: There is a substantial increase in risk of LEA associated with glycaemia in individuals with diabetes. In the absence of conclusive evidence from trials, this paper provides further epidemiological support for glucose-lowering as a strategy to reduce amputation in a population without acute foot ulceration or former amputation; it also provides disease modellers with estimates to assess the overall burden of hyperglycaemia.

Apolipoproteins in diabetes dyslipidaemia in South Asians with young adult-onset diabetes: distribution, associations and patterns.

BACKGROUND: Apolipoproteins B (apoB) and AI (apoAI) are strong predictors of cardiovascular disease (CVD). We describe apolipoprotein distributions and their associations with lipids and diabetes subtype in diabetic young adult South Asians. METHODS: In 995 subjects with diabetes, we measured fasting total cholesterol (TC), high-density lipoprotein cholesterol (HDLC), triglycerides (TG), apoB and apoAI, glycosylated haemoglobin (HbA1c) and glutamic acid decarboxylase antibodies (GADA). Low-density lipoprotein cholesterol (LDLC) and non-HDLC (NHDLC) were calculated. We compared values in subjects aged 15-50 y from the United States National Health and Nutrition Examination Survey (NHANES). RESULTS: Median age and duration of diabetes were 38 (range 14-45) and 4 (0-24) y. Men had significantly higher TC, TG, NHDLC, TC/HDLC, apoB/AI and NHDLC/apoB, and lower apoAI than women. Compared with the reference group, patients with type 1 diabetes had lower TG, apoB:apoAI and HDLC:apoAI, and higher HDLC and apoAI. Patients with type 2 diabetes had higher TG, TC, LDLC, NHDLC, TC:HDL, apoB, apoAI and apoB:apoAI, and lower HDLC, LDLC:apoB and HDLC:apoAI. Among patients with type 2 diabetes, 54% had high apoB (>1.2 g/L) and 33% also had high TG (>1.5 mmol/L). Measures of obesity (body mass index and waist circumference) were weakly correlated with lipid and apoprotein parameters, suggesting a modest contribution to dyslipidaemia. CONCLUSIONS: A large proportion of young adult Sri Lankan patients with type 2 diabetes has a low LDLC:apoB and high apoB and/or TG, suggesting that these patients are at increased risk of CVD.

Association of C-reactive protein with type 2 diabetes: prospective analysis and meta-analysis.

AIMS/HYPOTHESIS: We examined the association between serum C-reactive protein (CRP) and incident diabetes in a prospective study, and added these data to a literature-based meta-analysis to explore potential sources of heterogeneity between studies. METHODS: We analysed a case-control study nested within the European Prospective Investigation of Cancer (EPIC)-Norfolk cohort, including 293 incident diabetes cases and 708 controls. We combined 16 published studies on CRP and incident diabetes in a random-effect meta-analysis. RESULTS: In the EPIC-Norfolk cohort, serum CRP was associated with a higher risk of diabetes after adjusting for age, sex, BMI, family history of diabetes, smoking and physical activity (OR 1.49, comparing the extreme thirds of CRP distribution [95% CI 1.03-2.15], p = 0.03). However, the association was completely attenuated after further adjustment for WHR, serum gamma-glutamyltransferase and serum adiponectin (OR 1.00; 95% CI 0.66-1.51, p = 1.0). In a meta-analysis of 16 published studies with 3,920 incident diabetes cases and 24,914 controls, the RR was 1.72 (95% CI 1.54-1.92), comparing the extreme thirds of CRP distribution, with substantial heterogeneity between studies (I (2) = 52.8%, p = 0.007). CONCLUSIONS/INTERPRETATION: Initial evidence of association between CRP and incident diabetes was confounded by central adiposity, markers of liver dysfunction and adiponectin in the primary analysis. Despite an overall positive association in the meta-analysis, considerable heterogeneity existed between studies. The degree of adjustment for central adiposity and baseline glycaemia explained some of this heterogeneity and suggests that CRP may not be an independent risk factor for type 2 diabetes.

UKPDS-modelling of cardiovascular risk assessment and lifetime simulation of outcomes.

Although known principally as a clinical trial, the UK Prospective Diabetes Study (UKPDS) provided longitudinal data which helped define the natural history of cardiovascular complications in Type 2 diabetes. Using clinical, epidemiological, statistical and economics methods, UKPDS investigators developed mathematical models that helped define predictors (risk factors) for cardiovascular disease including angina, myocardial infarction, stroke, peripheral vascular disease and death in Type 2 diabetes. The UKPDS made clearer the contributions to risk of age, hyperglycaemia, elevated blood pressure, adverse blood lipids and smoking. Equations were developed, combined and incorporated into the UKPDS Risk Engine and the UKPDS Outcomes models. For example, the UKPDS risk engine-version 2-estimates that a white 62-year-old man with 11 years of Type 2 diabetes, a glycated haemoglobin of 8.3%, a systolic blood pressure of 145 mmHg and total and high-density lipoprotein cholesterol values of 5.8 and 1.1 mmol/l who did not smoke has a 33% chance of having overt coronary heart disease within 10 years. These models contribute to the estimation of risk and/or health outcomes adjusted for quality of life for use by, amongst others, clinicians, trialists, health planners, guideline developers and health economists.

Prevalence and projections of diabetes and pre-diabetes in adults in Sri Lanka--Sri Lanka Diabetes, Cardiovascular Study (SLDCS).

AIMS: To determine the prevalence of diabetes mellitus and pre-diabetes (impaired fasting glucose and impaired glucose tolerance) in adults in Sri Lanka. Projections for the year 2030 and factors associated with diabetes and pre-diabetes are also presented. METHODS: This cross-sectional study was conducted between 2005 and 2006. A nationally representative sample of 5000 adults aged >or= 18 years was selected by a multi-stage random cluster sampling technique. Fasting plasma glucose was tested in all participants and a 75-g oral glucose tolerance test was performed in non-diabetic subjects. Prevalence was estimated for those > 20 years of age. RESULTS: Response rate was 91% (n = 4532), males 40%, age 46.1 +/- 15.1 years (mean +/- standard deviation). The age-sex standardized prevalence (95% confidence interval) of diabetes for Sri Lankans aged >or= 20 years was 10.3% (9.4-11.2%) [males 9.8% (8.4-11.2%), females 10.9% (9.7-12.1%), P = 0.129). Thirty-six per cent (31.9-40.1%) of all diabetic subjects were previously undiagnosed. Diabetes prevalence was higher in the urban population compared with rural [16.4% (13.8-19.0%) vs. 8.7% (7.8-9.6%); P < 0.001]. The prevalence of overall, urban and rural pre-diabetes was 11.5% (10.5-12.5%), 13.6% (11.2-16.0%) and 11.0% (10.0-12.0%), respectively. Overall, 21.8% (20.5-23.1%) had some form of dysglycaemia. The projected diabetes prevalence for the year 2030 is 13.9%. Those with diabetes and pre-diabetes compared with normal glucose tolerance were older, physically inactive, frequently lived in urban areas and had a family history of diabetes. They had higher body mass index, waist circumference, waist-hip ratio, systolic/diastolic blood pressure, low-density lipoprotein cholesterol and triglycerides. Insulin was prescribed to 4.4% (2.7-6.1%) of all diabetic subjects. CONCLUSIONS: One in five adults in Sri Lanka has either diabetes or pre-diabetes and one-third of those with diabetes are undiagnosed.

Characteristics of adults with and without cystic fibrosis-related diabetes.

AIMS: The prevalence of diabetes in adults with cystic fibrosis (CF) approximates 25%, yet few studies have defined risk factors. We examined the association between biochemical and clinical factors and CF-related diabetes. METHODS: We performed a study in adults with CF in 2004 in Cambridgeshire, UK. Of 160 individuals, 51 had diabetes (cases) on the basis of medical history or screening using a 75-g oral glucose tolerance test, and 107 did not have diabetes (control subjects); two were excluded. We used logistic regression to model the cross-sectional association between potential risk factors and diabetes. RESULTS: The mean age was 26 (16-58) years, mean body mass index (BMI) was 21 (16-28) kg/m(2), and mean forced expiratory volume in 1 s was 60 +/- 24% (mean +/- sd). All of the cases and 88% of control subjects had pancreatic insufficiency. Cases did not differ from control subjects with respect to age, sex, body mass index, or dose of oral pancreatic enzymes. Cases were more likely to have low serum magnesium, haemoglobin, and pulmonary function, and higher serum gamma-glutamyl transferase (GGT) activity, plasma fibrinogen levels, erythrocyte sedimentation rate, use of oral corticosteroids, and number of CF-related complications. In multivariate analyses, GGT, previous organ transplantation, plasma fibrinogen and the presence of CF-related complications were independently associated with diabetes, after controlling for corticosteroid use. CONCLUSIONS: These data confirm the high prevalence of diabetes in adults with CF, and identify plasma fibrinogen and GGT, and organ transplantation as factors independently associated with CF-related diabetes. A prospective study would clarify the nature of these associations.

Additive effects of glycaemia and blood pressure exposure on risk of complications in type 2 diabetes: a prospective observational study (UKPDS 75).

AIMS/HYPOTHESIS: The relative importance of glucose and blood pressure control in type 2 diabetes remains uncertain. We assessed interactive effects of glycaemia and systolic blood pressure (SBP) exposure on the risk of diabetic complications over time. SUBJECTS, MATERIALS AND METHODS: HbA(1c) and SBP, measured annually for a median of 10.4 years in 4,320 newly diagnosed type 2 diabetic patients from the UK Prospective Diabetes Study (UKPDS), were categorised as updated mean values <6.0, 6.0-6.9, 7.0-7.9 or > or =8.0%, and <130, 130-139, 140-149 or > or =150 mmHg, respectively. Clinical outcomes were UKPDS predefined composite endpoints. RESULTS: The incidence of the "any diabetes-related endpoint" in the lowest (HbA(1c) <6.0%, SBP <130 mmHg) and highest (HbA(1c) > or =8%, SBP > or =150 mmHg) category combinations was 15 and 82 per 1,000 person-years, respectively, and 24 and 120 per 1,000 person-years in a Poisson model adjusted to white Caucasian male sex, age 50 to 54 years and diabetes duration of 7.5 to 12.5 years. Updated mean HbA(1c) and SBP effects were additive in an adjusted proportional hazards model with risk reductions of 21% per 1% HbA(1c) decrement and 11% per 10 mmHg SBP decrement. Endpoint rates obtained in the 887 patients randomised in both the glycaemia and hypertension intervention trial arms were consistent with the observational data. Those allocated to both intensive glucose and tight blood pressure control policies had fewer events than those allocated to either policy alone or to neither (p for trend 0.024). CONCLUSIONS/INTERPRETATION: Risk of complications in type 2 diabetes is associated independently and additively with hyperglycaemia and hypertension. Intensive treatment of both these risk factors is required to minimise the incidence of complications.

Insulin sensitivity at diagnosis of Type 2 diabetes is not associated with subsequent cardiovascular disease (UKPDS 67).

AIMS: Insulin resistance is common in Type 2 diabetes which, in turn, is associated with a markedly increased risk of cardiovascular disease. Whether insulin sensitivity measured after diagnosis of diabetes is associated with incident cardiovascular disease was evaluated in this prospective study. METHODS: Three thousand five hundred and eighty-two subjects with newly diagnosed diabetes, recruited to the UK Prospective Diabetes Study (UKPDS), free of cardiovascular disease, and with complete information on insulin sensitivity and potential confounders, were followed prospectively to the first occurrence of (i) fatal or non-fatal myocardial infarction, MI (ii) fatal or non-fatal stroke, and (iii) coronary heart disease, CHD (fatal or non-fatal MI, sudden death or ischaemic heart disease). Insulin sensitivity was measured by Homeostatic Model Assessment (HOMA). RESULTS: Insulin sensitivity as measured by HOMA was not associated with subsequent MI, stroke, or CHD in univariate or multivariate models controlling for age, sex, ethnicity, HbA(1c), body mass index, plasma triglycerides, cholesterol and smoking. The hazard ratio associated with a doubling of insulin sensitivity with fatal or non-fatal MI in a multivariate model was 0.92 (95% confidence interval, CI, 0.80-1.05). These results were not changed by the exclusion of overweight patients randomized to metformin. DISCUSSION: Estimation of insulin sensitivity provides no additional useful information with respect to the risk of the first occurrence of cardiovascular disease in patients with newly diagnosed Type 2 diabetes. Among patients with Type 2 diabetes, insulin resistance is not a risk factor for cardiovascular disease.

Diabetes associated with atypical antipsychotic treatment may be severe but reversible: case report.

OBJECTIVE: To draw attention to severe presentations of atypical neuroleptic related diabetes and to document that a marked degree of remission can take place after drug withdrawal. METHOD: We describe two patients who presented with diabetic ketoacidosis after treatment with quetiapine and risperidone, respectively. RESULTS: Both patients were negative for islet cell antibodies. They both required treatment with insulin, one in very high dosage, but their insulin requirements fell progressively after the atypical antipsychotic was withdrawn. After several months, neither patient required antidiabetic treatment. CONCLUSIONS: Atypical antipsychotic-induced diabetes does not always take a "type 2" presentation in which weight gain and insulin resistance are implicated. Sometimes the presentation is with diabetic ketoacidosis, requiring insulin treatment, which can nevertheless be reversible.

The UKPDS risk engine: a model for the risk of coronary heart disease in Type II diabetes (UKPDS 56).

A definitive model for predicting absolute risk of coronary heart disease (CHD) in male and female people with Type II diabetes is not yet available. This paper provides an equation for estimating the risk of new CHD events in people with Type II diabetes, based on data from 4540 U.K. Prospective Diabetes Study male and female patients. Unlike previously published risk equations, the model is diabetes-specific and incorporates glycaemia, systolic blood pressure and lipid levels as risk factors, in addition to age, sex, ethnic group, smoking status and time since diagnosis of diabetes. All variables included in the final model were statistically significant (P<0.001, except smoking for which P=0.0013) in likelihood ratio testing. This model provides the estimates of CHD risk required by current guidelines for the primary prevention of CHD in Type II diabetes.

Cardiovascular risk reduction in diabetes: underemphasised and overdue. Messages from major trials.

Diabetes markedly increases the risk of coronary artery disease and death, but is underrecognised as a cardiovascular risk factor, despite the existence of effective treatments. Because patients with diabetes are at high risk for coronary disease, they have more to gain from prevention. There is evidence from clinical trials that select cardiovascular therapies may work better in diabetes, beyond their expected benefit, and may prevent diabetes itself.

Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study.

OBJECTIVE: To determine the relation between exposure to glycaemia over time and the risk of macrovascular or microvascular complications in patients with type 2 diabetes. DESIGN: Prospective observational study. SETTING: 23 hospital based clinics in England, Scotland, and Northern Ireland. PARTICIPANTS: 4585 white, Asian Indian, and Afro-Caribbean UKPDS patients, whether randomised or not to treatment, were included in analyses of incidence; of these, 3642 were included in analyses of relative risk. OUTCOME MEASURES: Primary predefined aggregate clinical outcomes: any end point or deaths related to diabetes and all cause mortality. Secondary aggregate outcomes: myocardial infarction, stroke, amputation (including death from peripheral vascular disease), and microvascular disease (predominantly retinal photo-coagulation). Single end points: non-fatal heart failure and cataract extraction. Risk reduction associated with a 1% reduction in updated mean HbA(1c) adjusted for possible confounders at diagnosis of diabetes. RESULTS: The incidence of clinical complications was significantly associated with glycaemia. Each 1% reduction in updated mean HbA(1c) was associated with reductions in risk of 21% for any end point related to diabetes (95% confidence interval 17% to 24%, P<0.0001), 21% for deaths related to diabetes (15% to 27%, P<0.0001), 14% for myocardial infarction (8% to 21%, P<0.0001), and 37% for microvascular complications (33% to 41%, P<0.0001). No threshold of risk was observed for any end point. CONCLUSIONS: In patients with type 2 diabetes the risk of diabetic complications was strongly associated with previous hyperglycaemia. Any reduction in HbA(1c) is likely to reduce the risk of complications, with the lowest risk being in those with HbA(1c) values in the normal range (<6.0%).