It has been proposed that inflammation or infection may contribute to the d
evelopment of type 2 diabetes. We examined whether serum gamma globulin, a
nonspecific measure of the humoral immune system, predicted changes in gluc
ose tolerance in 2,530 members of the Pima Indian population, a group with
a marked predisposition to type 2 diabetes. Cross-sectionally, gamma globul
in,was positively related to age (r = 0.08, P < 0.0005), BMI (r = 0.09; P <
0.0001), and female sex (P < 0.0001). Gamma globulin concentrations were f
amilial, being positively correlated among siblings (r = 0.23; P < 0.0001)
and between parents and their children (mother/child: r = 0.17, P < 0.0001;
father/child: r = 0.25, P < 0.0001). Gamma globulin concentrations were hi
gher with greater degrees of American Indian heritage (P < 0.004, with adju
stment for age, sex, and BMI) and in the presence of a family history of ty
pe 2 diabetes (P < 0.04). Higher gamma globulin levels predicted risk of di
abetes. In univariate analysis, a 1 SD difference in gamma globulin was ass
ociated with a 20%, higher incidence of diabetes in those who were normal g
lucose tolerant at baseline (hazard rate ratio 1.20 [CI 1.11-1.30]; P < 0.0
001) and remained as a significant predictor of diabetes, even when control
led for effects of sex, BMI,and 2-h glucose as additional predictors (hazar
d rate ratio for 1 SD difference in gamma globulin, 1.14 [1.05-1.24]; P = 0
.002). Gamma globulin was also associated in univariate analysis with later
development of impaired glucose tolerance (IGT) (hazard rate ratio 1.15 [1
.07-1.23]; P < 0.0001), but not with the transition from IGT to diabetes (h
azard rate ratio 1.04 [0.90-1.20]; P = 0.6). Thus, gamma globulin levels pr
edict increased risk of diabetes in the Pima population. We suggest that im
mune function or activation may play a role in the development of type 2 di
abetes.