Ideal preventive strategies for both type 1 and type 2 diabetes should range from focusing efforts on individuals identified as being at risk for developing diabetes to broader, more group- or population-based strategies. For the practicing healthcare professional, it is the individualized strategies that are sought. The prevention or delay of diabetes not only would alleviate the individual of the burden of disease but also could decrease associated morbidity and mortality. Preventive strategies would differ depending on the type of diabetes. Given the increasing incidence and prevalence of diabetes, development of safe and cost-effective interventions to reduce the risk of diabetes are needed to help decrease the burden of diabetes on individuals and the healthcare system.
Reducing the Risk of Developing Type 1 Diabetes
Type 1 diabetes is a chronic autoimmune condition characterized by destruction of pancreatic beta cells. The causes are multifactorial, with both genetic and environmental factors playing a part. There is a long preclinical period before the onset of overt symptoms, which may be amenable to therapeutic intervention to prevent disease. The exact nature of causative environmental factors continues to be debated. Immunotherapeutic interventions continue to be the main focus of disease prevention. Enhancement of “regulatory” immune mechanisms currently shows the most promise in terms of slowing the progression of the disease and preserving beta cell mass (secondary prevention).
Two major trials of interventions to prevent or delay the onset of type 1 diabetes have been completed. The European Nicotinamide Diabetes Intervention Trial (ENDIT), a randomized, double-blind, placebo-controlled trial of high-dose nicotinamide therapy, recruited first-degree relatives of people who were <20 years old when diagnosed with type 1 diabetes, were islet cell antibody positive, were <40 years of age and had a normal oral glucose tolerance test (OGTT) result. Although nicotinamide had proved protective in animal studies, no effect was observed in ENDIT during the 5-year trial period (1). The Diabetes Prevention Trial-Type 1 (DPT-1) studied the efficacy of low-dose insulin injections in high-risk (projected 5-year risk of >50%) first-degree relatives of subjects with type 1 diabetes. Overall, the insulin treatments had no effect (2), but in a subset of participants with high levels of insulin autoantibodies, a delay, and perhaps a reduction, in the incidence of type 1 diabetes was observed (3). A third large trial, the ongoing Trial to Reduce IDDM in the Genetically at Risk (TRIGR) study, is investigating the effect of excluding cow’s milk protein and replacing it with hydrolyzed formula milk in genetically at-risk infants until 6 to 8 months of age. Preliminary data suggest there were fewer autoantibody-positive children at 10 years (4), but data on the overt development of diabetes by age 10 will not be available until 2017.
A second strategy is to try and halt, at the time of diagnosis, the immune-mediated destruction of beta cells so as to preserve any residual capacity to produce insulin. Progress in the field has been appropriately slow due to important ethical considerations. Namely, side effects from excessive immunosuppression/modulation cannot be tolerated because of the reasonable life expectancy with insulin substitution therapy.
As safe and effective preventive therapies for type 1 diabetes have not yet been identified, any attempts to prevent type 1 diabetes should be undertaken only within the confines of formal research protocols.
Reducing the Risk of Developing Type 2 Diabetes
Preventing type 2 diabetes may result in significant public health benefits, including lower rates of cardiovascular disease (CVD), renal failure, blindness and premature mortality. An epidemiological analysis projected that if all diabetes could be avoided in white American males through effective primary prevention, the risk of all-cause and cardiovascular mortality in the entire population could be reduced by up to 6.2% and 9.0%, respectively (5). Data from the US indicate that 28% of cardiovascular expenditures are attributable to diabetes (6). Primary approaches to preventing diabetes in a population include the following: 1) programs targeting high-risk individuals in the community (such as those with impaired glucose tolerance [IGT] or obesity); 2) programs targeting high-risk subgroups of the population, such as high-risk ethnic groups; and 3) programs for the general population, such as those designed to promote physical activity and healthy eating in adults or children (7–9). Prospective cohort studies have identified historical, physical and biochemical variables associated with the subsequent development of type 2 diabetes. These include older age, certain ethnic backgrounds, obesity (especially abdominal obesity), physical inactivity, history of gestational diabetes mellitus, overt coronary artery disease, high fasting insulin levels and IGT (10–12). Results of large, well-designed studies assessing lifestyle and pharmacological interventions in adults to prevent the progression from IGT to diabetes have been published. No pharmacological agent is approved for diabetes prevention in Canada.
Changes in lifestyle were assessed in the Finnish Diabetes Prevention Study (DPS) (13)and the Diabetes Prevention Program (DPP) (14). Dietary modification that targeted a low-calorie, low-fat, low-saturated fat, high-fibre diet and moderate-intensity physical activity of at least 150 minutes per week resulted in a moderate weight loss of approximately 5% of initial body weight. In both studies, the risk reduction for diabetes was 58% at 4 years. These studies included comprehensive, sustained programs to achieve these outcomes. On the basis of the observed benefits of lifestyle in the DPP, all participants were offered further lifestyle interventions for a median of 5.7 more years and benefits were sustained for up to 10 years (15).
In another lifestyle intervention trial, 458 Japanese males with IGT were randomly assigned in a 4:1 ratio to a standard intervention (n=356) or an intensive intervention (n=102) and followed for 4 years (16). Intensive treatment was associated with a 67.4% reduction in risk of diabetes (p<0.001). IGT and diabetes were diagnosed using a 100 g OGTT and the following diagnostic criteria: IGT =2-hour plasma glucose (2hPG) 8.8 to 13.1 mmol/L; diabetes = 2hPG ≥13.2 mmol/L (16). These levels corresponded to the 1980 World Health Organization (WHO) diagnostic criteria using a 75 g OGTT (17,18).
In a more recent trial, 641 overweight Japanese men (aged 30 to 60 years) with impaired fasting glucose (IFG) were randomized to either a frequent intervention group (n=311) or a control group (n=330) for 36 months. The frequent intervention group received individual instructions and follow-up support for lifestyle modification from medical staff 9 times. The control group received similar individual instructions 4 times at 12-month intervals during the same period. Results showed an incidence of type 2 diabetes of 12.2% in the frequent intervention group and 16.6% in the control group, with an adjusted hazard ratio (HR) in the frequent intervention group of 0.56 (95% confidence interval [CI] 0.36–0.87). Post hoc subgroup analyses showed the HR reduced to 0.41 (95% CI 0.24–0.69) among participants with IGT at baseline and to 0.24 (95% CI 0.12–0.48) among those with a baseline A1C level >5.6% (19).
A 20-year follow-up of the Chinese Da Qing Diabetes Prevention Trial showed that after 6 years of active lifestyle interventions vs. no treatment and an additional 14 years of passive follow-up, a 43% (95% CI 19–59) relative risk reduction for incident diabetes persisted, and vision-threatening retinopathy was reduced by 47% (95% CI 1–71). There were, however, no identified reductions in nephropathy, neuropathy, cardiovascular events or mortality (20,21).
Metformin was used in a second arm of the DPP (14). A dosage of 850 mg bid for an average of 2.8 years significantly decreased progression to diabetes by 31%. In the DPP population, metformin did not have any significant effect in the older age group (≥60 years) and in less obese subjects (body mass index [BMI] <35 kg/m 2 ). To determine whether the observed benefit was a transient pharmacological effect or was more sustained, a repeat OGTT was undertaken after a short washout period. The results of this study suggested that 26% of the diabetes prevention effect could be accounted for by the pharmacological action of metformin (which did not persist when the drug was stopped). After the washout, the incidence of diabetes was still reduced by 25% (22). The benefits persisted for up to 10 years (15).
The DPP Research Group published the results from the troglitazone arm, which was part of the original protocol (23). The drug was discontinued after a mean follow-up of 0.9 years due to liver toxicity. Troglitazone 400 mg once daily resulted in a relative risk reduction of 75% (p=0.02) during the short period of time. This effect was not sustained after discontinuation of troglitazone.
The Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medication (DREAM) trial randomized 5269 subjects with IGT and/or IFG, in a 2 × 2 factorial fashion, to ramipril (up to 15 mg/day) and/or rosiglitazone (8 mg/day) vs. placebo (24,25). Eligible subjects were 30 years old and not known to have CVD. The primary outcome of DREAM was a composite of development of diabetes or death. The conclusion of the DREAM investigators was that the results suggest that ramipril may have favourable effects on glucose metabolism, a finding that is consistent with other reports. However, not all trials have found such an association. For now, the routine use of ramipril for the express purpose of preventing diabetes is not indicated.” Treatment with rosiglitazone resulted in a 60% reduction in the primary composite outcome of diabetes or death (HR 0.40, 95% CI 0.35–0.46), primarily due to a 62% relative reduction in the risk of progression to diabetes (HR 0.38, 95% CI 0.33–0.44). Although the trial was not powered to provide a definitive estimate of the effect of rosiglitazone on CV outcomes, there was a trend toward an increase in risk of the CV composite outcome with rosiglitazone (HR 1.37, 95% CI 0.97–1.94) driven primarily by a significant increase in nonfatal congestive heart failure (HR 7.03, 95% CI 1.60–30.9; p=0.01).
In the Actos Now for the Prevention of Diabetes (ACT NOW) study, 602 high-risk IGT subjects were randomized to receive pioglitazone or placebo and were followed for 2.4 years. Pioglitazone decreased the conversion of IGT to type 2 diabetes by 72% (p<0.00001) (26).
Despite the favourable effects of thiazolidinediones on delaying the development of type 2 diabetes, the multiple potential adverse effects of this class of medication makes it difficult to recommend their widespread use in IFG or IGT.
The combination of metformin 500 mg twice daily and rosiglitazone 2 mg twice daily was found to reduce the progression to diabetes by 66% (95% CI 41–80) among 103 people with IGT compared to 104 people randomized to placebo over a median of 3.9 years (27).
The Study to Prevent Non-Insulin Dependent Diabetes (STOP-NIDDM) used acarbose at a dosage of 100 mg tid in a 5-year study with a mean follow-up of 3.3 years (28). Overall, there was a 25% reduction in the risk of progression to diabetes when the diagnosis was based on 1 OGGT and a 36% reduction in the risk of progression to diabetes when the diagnosis was based on 2 consecutive OGTTs. This beneficial effect was not affected by age or BMI. However, when the drug was discontinued, the effect of acarbose did not persist (28). In this IGT population, acarbose treatment was also associated with a 49% reduction in CV events (p=0.032) and a 50% reduction in the progression of carotid intima-media thickness (29,30). In another trial, 1780 Japanese patients with IGT were randomly assigned to oral voglibose 0.2 mg three times per day (n=897) or placebo (n=883). Results showed that, over a mean of 48.1 weeks, voglibose was better than placebo at reducing the progression to type 2 diabetes (5.6% vs. 11.9%; HR 0.595, 95% CI 0.433–0.818; (p=0.0014). More subjects in the voglibose group achieved normoglycemia than in the placebo group (66.8% vs. 51.5%; HR 1.539, 95% CI 1.357–1.746; p<0.0001).
The Xenical in the Prevention of Diabetes in Obese Subjects (XENDOS) study examined the effect of orlistat in combination with an intensive lifestyle modification program (diet and exercise) on the prevention of diabetes in 3305 obese individuals (31). Subjects were randomized to orlistat 120 mg or placebo tid with meals for 4 years. Weight loss was observed in both groups, but the orlistat group lost significantly more (5.8 vs. 3 kg; p<0.001). Compared to placebo, orlistat treatment was associated with a further 37% reduction in the incidence of diabetes. However, 2 important methodological limitations affect the interpretation of these results. First, there was a very high dropout rate—48% in the orlistat group and 66% in the placebo group. Second, the last observation carried forward was used for analysis, which is generally not favoured for prevention or survival studies. Nonetheless, the significant weight loss would be expected to decrease the risk of diabetes as already shown in the DPS and the DPP.
Liraglutide has been shown to effectively prevent IGT conversion to type 2 diabetes and cause reversion to normoglycemia (32). In a 20-week study, liraglutide was administered to 564 obese individuals who did not have diabetes, 31% of whom had IGT. Subjects were randomized to 1 of 4 liraglutide doses (1.2 mg, 1.8 mg, 2.4 mg, or 3.0 mg; n=90–95) or to placebo (n=98), or to orlistat (120 mg; n=95) three times daily. A1C was reduced by 0.14% to 0.24%. The prevalence of prediabetes decreased by 84% to 96% with liraglutide 1.8 mg, 2.4 mg and 3.0 mg doses.
Diabetes prevention in high-risk ethnicities
Ethnic groups, such as South Asians, Hispanics and Aboriginals, are at very high risk for and have a high prevalence of type 2 diabetes (12%–15% in the Western world) (33,34).
The reasons for this are multifactorial and include genetic susceptibility, altered fat distribution (more visceral fat with greater insulin resistance) and higher prevalence of metabolic syndrome. Many of them develop diabetes at a younger age and often have complications at the time of diagnosis due to long-standing, preexistent diabetes. As a result, there may be a benefit of delaying the onset of diabetes in this population. The Indian Diabetes Prevention Programme showed similar results in preventing diabetes with both lifestyle and pharmacological interventions where the progression to diabetes from IGT was quite high (55%) over 3 years (35). The Indian Diabetes Prevention Programme shows that lifestyle modification and metformin prevent type 2 diabetes in Asian subjects with IGT (IDPP-1).
This approach of prevention may lead to cost savings, fewer complications and lower morbidity, but it remains to be proven with hard clinical endpoints. Lifestyle measures not only reduce the risk of diabetes but have other health benefits, so the overall benefit is positive with little harm. One must keep in mind that the measures of prevention must be delivered in a culturally sensitive manner to these populations.