Evaluation of Family History of Diabetes and the Development of Cystic Fibrosis-Related Diabetes in a Clinical Cystic Fibrosis Cohort.

OBJECTIVES

Knowledge about risk factors for cystic fibrosis (CF) related diabetes (CFRD) are incomplete. Age, lower body mass index (BMI), female sex, pancreatic insufficiency (PI), cystic fibrosis transmembrane conductance regulator (CFTR) genotype severity and diabetes family history (FH) increase the risk. Pre-CFTR modulator studies depended on small samples and family-reported survey data. We evaluated clinically collected family history from a modulator-treated CF cohort.

METHODS

We conducted a retrospective, single-center observation utilizing clinician-collected, diabetes FH and age, sex, BMI, CFTR genotype and PI variables. With CFRD as the dependent variable, we estimated odds ratios (ORs) and 95% confidence intervals (CIs) for each FH variable in separate logistic regression models. The model was then adjusted for age, sex, genotype category, PI, and BMI category. We used a modified Poisson regression with robust standard error to avoid logistic regression inflated effect estimates and provide easily interpretable relative risk estimates (incidence rate ratios, IRRs).

RESULTS

Type 2 diabetes FH was associated with a 157% increased CFRD risk (IRR = 2.57, CI = 1.53-4.32, P< .001). Type 1 diabetes FH was insignificant. CFRD FH was associated with a 92% increase in CFRD risk (IRR = 1.92, CI = 1.15-3.21, P= .013). PI predicted CFRD (OR = 3.99, CI = 1.35-11.81, P= .012). Each additional age decade increased CFRD odds by 46% (OR = 1.46, CI = 1.18-1.81, P= .001). Sex, genotype category, and BMI were not associated with CFRD.

CONCLUSION

Type 2 diabetes and CFRD family history double CFRD risk. PI quadruples CFRD odds. Each age decade increased CFRD odds by 46%, but sex, genotype category, and BMI were not associated. Clinicians should obtain diabetes family health information in CF.