Introduction and Objective: Adult-onset autoimmune diabetes (AD), driven by immune-mediated β-cell destruction, leads to early failure of oral antidiabetic agents and rapid insulin requirement. Fourlanos et al. (2006) proposed a five-item clinical score (age <50 years, BMI<25 kg/m², personal history of autoimmune disease, family history of autoimmune disease, and acute hyperglycemic symptoms at diagnosis) to identify patients with latent autoimmune diabetes in adults (LADA) who were positive for GAD65 antibodies and did not require insulin within the first 6 months after diagnosis. This study aimed to evaluate the performance of the original score to identify AD and compare it with a simplified versionMethods: Cross-sectional study including 1,000 adults with newly diagnosed diabetes from 12 primary care centers (55.9% men; mean age 62.8±10.8 years). GAD65 and IA-2 antibodies were measured, and clinical, laboratory, and anthropometric variables were collected. The original Fourlanos score (1 point per item; range 0-5) was applied to identify adult-onset AD. Discrimination was assessed using the area under the ROC curve (AUC). A reduced model was obtained by logistic regression, excluding family history (p=0.572). Calibration was assessed with the Hosmer-Lemeshow test. Two points (according to β-coefficient weight) were assigned to each variable except age <50 years (1 point), yielding a range of 0-7. AUC and optimal cut-off were calculated, and both scores were compared using DeLong testResults: Original Fourlanos score yielded an AUC=0.685 (95% CI 0.615-0.755). The reduced score achieved an AUC=0.694 (95% CI 0.664-0.722); DeLong’s test: Z=1.888, p=0.059. The optimal cut-off for the reduced score was ≥2 points (sensitivity 64%, specificity 67%, NPV 95.8%). Hosmer-Lemeshow: 1.655, p=0.647. Overall, 35.3% scored ≥2Conclusion: Simplified score for predicting adult-onset AD showed discriminative performance comparable to the original model, with no statistically significant differences, and should be more practical for routine use
M. Salinero-Fort: None. P. Vich-Pérez: None. B. Taulero-Escalera: None.
Source link
Leave a Reply