Low prevalence of islet autoimmunity in adult diabetes and low predictive value of islet autoantibodies in the general adult population of northern Italy
E. Bosi et al., Low prevalence of islet autoimmunity in adult diabetes and low predictive value of islet autoantibodies in the general adult population of northern Italy, DIABETOLOG, 42(7), 1999, pp. 840-844
Aims/hypothesis. To assess the prevalence of islet autoimmunity in adult-on
set diabetes mellitus and the predictive value of islet autoantibodies in t
he general adult population of northern Italy.
Methods. A sample of 2076 people aged 40 years or more participating in the
population-based Cremona Study and classified in 1990 as having diabetes m
ellitus, impaired and normal glucose tolerance according to WHO criteria af
ter an oral glucose tolerance test, were tested for antibodies to glutamic
acid decarboxylase and IA-2.
Results, Increased concentrations of glutamic acid decarboxylase antibodies
were found in 4 (2.8%) of 143 participants with known diabetes and none of
50 with previously unknown diabetes, 1 (0.65%) of 153 with impaired and 18
(1.0%) of 1718 with normal glucose tolerance. The increased prevalence of
these antibodies in subjects with known diabetes was not statistically sign
ificant. Protein tyrosine phosphatase IA-2-antibodies were found in only fo
ur subjects, two of whom also had glutamic acid decarboxylase antibodies, a
ll with normal glucose tolerance. After 8 years of follow-up, none of 21 no
n-diabetic subjects with either glutamic acid decarboxylase or IA-2-antibod
ies had developed diabetes and only a slight deterioration from normal to i
mpaired fasting glucose was observed in 3 of 15 subjects with previous norm
al glucose tolerance.
Conclusion/interpretation. This study has shown that in northern Italy the
prevalence of adult autoimmune diabetes in the general adult population is
0.19% (95% CI 0.85-0.5); that autoimmune diabetes represents only a minorit
y of all cases of adult diabetes; and that islet autoantibodies are not a h
igh-risk factor for diabetes development in adults with normal glucose tole
rance over 8 years of follow-up.