Diabetes: a time for excitement—and concern

These are exciting, and exasperating, times for people interested in diabetes. On the one hand, a tremendous volume of research is underway assessing both new prevention and new treatment protocols; on the other, the incidence (and associated mortality and morbidity) of this disease continues to rise with little sign of abating. Indeed, by 2010 the world’s diabetic population will probably have doubled from an estimated 110 million in 1994 to 221 million in 2010.¹ What have we learnt recently about the epidemiology, causation, and prevention of this condition?

Prevalence varies widely by ethnic group and country (adult rates range from <2% in rural Bantu people in Tanzania to nearly 50% in US Pima Indians and South Pacific Naurauns¹). Rates are also relatively high in “transplanted populations,” such as Asians in Europe² and African Americans.³ The projected increase in rates, however, is universal. Though this rise is mainly due to type 2 diabetes, alarming increases in type 1 have also been observed in many studies. For example, a recent report from Allegheny County, USA, showed for the first time that rates for type 1 diabetes in adolescent African Americans now exceed those in whites.⁴ Whether this relates to an increasing incidence of type 2 diabetes among adolescents and young adults elsewhere is unclear,⁵,⁶ as the distinction between type 1 and type 2 is often blurred. Many reports now show that 10-15% of phenotypic type 2 diabetic subjects have autoantibodies to GAD (glutamic acid decarboxylase) and thus may have an incomplete type 1 autoimmune process.⁷

The diagnosis and classification of diabetes has historically been controversial, and not until 1979 was some worldwide consensus achieved. Even then, the American National Diabetes Data Group⁸ and World Health Organisation criteria⁹ were not identical. In an attempt to resolve the confusion, the American Diabetes Association last year proposed a revised classification and diagnostic criteria.¹⁰ These proposals do away with the familiar terms non-insulin dependent and insulin dependent, replacing them with type 1 (β cell defect, usually autoimmune) and type 2 (insulin resistance with an insulin secretory defect), thus shifting the focus from mode of treatment to aetiology. Perhaps more importantly, the committee also changed the diagnostic criteria, lowering the fasting plasma glucose criterion to ⩾7.0 mmol/l and no longer recommending the use of the oral glucose tolerance test—which is little used in practice to diagnose diabetes.¹¹ The hope is that these changes will make the diagnosis easier, and thus more likely to be made—a commendable objective as a third to a half of all cases are undiagnosed.³ Furthermore, the 7 mmol/l fasting glucose cut off corresponds reasonably well to the two hour cut off value in the oral glucose tolerance test and, more importantly, to the future incidence of diabetic complications.¹⁰ The impact of these changes remains to be fully assessed, but two conclusions are emerging: in many populations fewer people will have diagnosable diabetes than before; and, disturbingly, there is less than 50% correspondence between the two sets of criteria.¹²

Apart from any influence of changing criteria and more frequent diagnostic testing, why is the incidence of type 2 diabetes rising? Three primary risk factors for type 2 diabetes are well established—genes, obesity, and activity. Genetic (natural) selection alone is unlikely to explain the increasing incidence as type 2 diabetes usually occurs after the reproductive years. Nevertheless, inheriting a metabolic profile which enhances survival through the reproductive years but which may eventually decompensate, leading to older onset diabetes, could still be important. Such a scenario might explain the association of low birth weight with subsequent diabetes¹³—that is, the low birth weight child may have inherited a metabolism enabling survival in an adverse intrauterine environment.¹⁴ Undoubtedly, inheriting a predisposition to diabetes is an important necessary prerequisite—and in a few cases, linked to specific genetic mutations, it may even be a sufficient cause.¹⁰

The alarmingly high prevalence of diabetes among Pima Indians and Naurauns provides further important clues to causation. Until recently such people were hunter-gatherers and probably acquired an insulin sensitive metabolism favouring fat storage at times of plenty but would not necessarily require a similar degree of insulin sensitivity in muscle tissue, where glucose entry to cells might be largely stimulated by high activity levels.¹¹ With Westernisation, a plentiful supply of energy dense food has been accompanied by a reduction in activity. Both factors may therefore cause the previously favourable metabolic profile seen in “survivors” to become a handicap: the “thrifty genotype rendered detrimental by ‘progress.’”¹⁵ Reaven has recently further developed this “muscle resistance-thrifty genotype” hypothesis by suggesting that muscle insulin resistance will favour survival by preserving muscle protein,¹⁶ thus enhancing the ability to hunt and gather.¹⁷

Similar factors probably underlie the increase of type 2 diabetes in most societies—for example, whites in the US, where the increase has been accompanied by an increase in obesity rates.¹⁸ These changes probably underlie the so called “insulin resistance syndrome,” which for many people is likely to increase both the incidence of diabetes and, through its multiple risk factor associations, cardiovascular risk.¹⁹ Epidemiological data support the roles of obesity,²⁰ fat intake,²¹ and low activity ²² as risk factors for diabetes and have led to a wave of trials to prevent type 2 diabetes by lifestyle or drug interventions.²³,²⁴ Some encouragement to these initiatives is given by the recent Da-Qing trial in China²⁵ and the positive metabolic effects seen when urbanised aborigines returned to a more traditional lifestyle.²⁶ Prevention trials are also underway in type 1 diabetes.²⁷,²⁸

While we await the results of these primary prevention trials, attention focuses on the treatment of diabetes to prevent both microvascular and macrovascular complications, for diabetes is the leading cause of blindness, renal failure, and amputation in middle aged US adults²⁹ and arguably the most prevalent risk factor for heart disease.³⁰ The diabetes control and complication trial established the value of intensive control of blood glucose in preventing the retinal, renal, and neuropathic complications of type 1 diabetes.³¹ Implementing such a regimen in routine clinical practice remains a challenge, however, while its applicability to type 2 diabetes is uncertain.

Furthermore, though microvascular complications are generally strongly related to cumulative glycaemic exposure,³² the same is not so clear for macrovascular complications. Nevertheless, the largest follow up ever of diabetic subjects, those screened in the MRFIT study, shows the power of standard cardiovascular risk factors, which accounted for two thirds of the excess deaths from cardiovascular disease.³⁰ This hope for prevention is further bolstered by the 4S study,³³ which showed the benefit of cholesterol lowering in type 2 diabetes. Little, however, is known about the value of blood pressure control on the risk of complications, other than nephropathy. In the context of these developments and uncertainties the long awaited results of the multifaceted United Kingdom prospective diabetes study now emerge and add valuable new data on both glycaemic³⁴ and blood pressure control (pp  703, 713, 720).^35–37 It is, indeed, an exciting time.

Editorial p 693, Papers pp 703, 713, 720