So far the focus on the 1997 ADA and 1999 WHO revision of the diagnostic criteria has been on the impact of the revised diagnostic thresholds. Another often neglected but equally (or even more) important revision relates to the classification of patients. In 1985, patients were classified as having insulin-dependent (IDDM) and non-insulin-dependent (NIDDM) diabetes based on the underlying disease, that is, whether beta-cell dysfunction was reduced to a level where insulin was needed to survive without entering ketoacidosis (insulin-dependent diabetes) or whether the patient had diabetes based on insulin resistance (with or without associated beta-cell dysfunction) where the patient would survive without insulin, but where insulin could be necessary to maintain acceptable metabolic
Table 2. All cause excess mortality by fasting and 2 hour glucose in the DECODE study (Adopted from ref 11).
Fasting plasma glucos (mmol/L) <6.1 6.1-6.9 7.0-7.7 >7.0
control. From 1985 and onwards several clinical studies  as well as practical clinical experience demonstrated that a large proportion of patients characterised as non-insulin dependent would subsequently need insulin to maintain acceptable metabolic control. This often led to confusion with respect to classification of the individual patient, and increasingly patients were re-classified from NIDDM to IDDM. This clinical observation combined with a wish to establish a classification based on a combination of clinical stages and aetiological types  led WHO to abandon the terms IDDM and NIDDM and to reintroduce the terms Type 1 and Type 2 diabetes. This development was helped by the identification of several markers of autoimmunity linked to the destruction of beta cells such as islet cell antibodies (ICA), insulin auto- antibodies (IAA) and auto-antibodies to glutamic acid decarboxylase (anti-GAD). Consequently, the revised classification included as the two main groups
• Type 1 diabetes (beta-cell destruction, usually leading to absolute insulin deficiency). In this group 85-90% are antibody positive for at least one of the antibodies ICA, IAA or GAD, while a smaller group (10-15%) have total beta-cell destruction without any signs of autoimmunity. Within the group of patients with type 1 diabetes there is a smaller group that have antibodies, but are not insulin-requiring for survival at least for several years. These patients are characterised by a slower disease process and very slow loss of beta-cell function and this group is often referred to as latent autoimmune diabetes in adults (LADA).
• Type 2 diabetes (predominantly insulin resistant with relative insulin deficiency or predominantly an insulin-secretory defect with/without insulin resistance). These individuals consequently have a relative not an absolute insulin deficiency. At the same time this group of individuals have no other known specific aetiology. At present this group comprises 70-80% of all cases of diabetes (even more in some parts of the world), but given the fact that molecular biology combined with other scientific disciplines continuously identifies an increasing number of "specific types" this group will gradually diminish. Apart from this the specific types will not be discussed further in this chapter.
One problem related to the change in classification to an aetiological definition is that the diagnosis of a so-called type 1 process is based on measurement of autoimmune markers, which is not a part of routine clinical practice, and markers that currently have none or very limited impact on the treatment regiment for the individual patient. As a consequence of this, a patient with diabetes with considerable residual beta-cell mass and obviously not insulin requiring from a clinical point of view, but with an ongoing autoimmune process, will be diagnosed as having type 2 diabetes unless admitted to a centre where measurement of auto-antibodies for some reason (typical research) is a part of routine clinical practice. In this case, in real life, the classification of the patient would therefore reflect the centre at which the patient is treated, not the underlying disease process. This would clinically be a minor problem, but with 5-15% of patients with type 2 diabetes being antibody positive, and given that treatment guidelines differ and type 1 diabetes patients are treated centrally while type 2 diabetic patients are treated in general practice, this would have tremendous impact on the organisation of the health care system if all patients had antibodies measured and subsequently were remitted accordingly.
Another problem that has not been solved is that even in the general population with normal glucose tolerance following an OGTT 2-5% are antibody positive [16,17]. This would suggest that some antibody positive individuals with clinical T2DM are truly type 2 diabetic where antibody positivity reflects a "by chance finding" and not necessarily an ongoing autoimmune disease process.
Was this article helpful?