Diabetes mellitus is not a single disease entity, but a heterogenous group of disorders with a striking diversity of etiopathogenetic mechanisms as well as clinical manifestations. Lack of a known genetic marker for the disease(s) and variable influences of environmental factors on the expression of a putative diabetic genome have resulted in considerable debate over its etiology.
Over the past few years, systematic epidemiologic studies, along with knowledge gained from a close association of certain human-leukocyte-antigens with the diabetic diathesis and possible role of host-immune factors, and gene-virus interaction have led to considerable advancement in the understanding of the disease-complex.
Pending the availability of definite genetic marker(s), we propose a new, tentative classification based on the etiologic mechanisms. We also suggest that the term "prediabetes" be abandoned as a prospective entity, since as presently employed, this connotation carries a risk probability no different than the terms like prehypertension or precoronary thrombosis.
(Arch Intern Med 137:461-469, 1977)