Current smoking is an independent risk factor for new-onset diabetes mellitus during high-dose glucocorticoid treatment

摘要

Aims/Introduction: Although high-dose glucocorticoids have been reported to cause new-onset diabetes mellitus (glucocorticoid-induced diabetes mellitus), its risk factors have remained to be determined. We investigated the risk factors related to glucocorticoid-induced diabetes mellitus diagnosed within 2 months after the high-dose treatment (newly treated with an initial high dose of > 20 mg prednisolone (PSL) equivalent per day for at least more than 6 months) in collagen vascular diseases. Methods: A total of 2,631 patients with collagen vascular diseases was registered between 1986 and 2006 in the Chiba-Shimoshizu Rheumatic Cohort. We analyzed 681 patients newly treated with high-dose glucocorticoid who did not have diabetes mellitus and/or its previous diagnosis (age: 46.3 +/- 16.7 years, PSL dose: 40.0 +/- 14.1 mg/day). Glucocorticoid-induced diabetes mellitus was diagnosed by two or more glucose measurements in patients with fasting glycaemia >= 7 mmol/L and 120 minutes post-load glycaemia >= 11.1 mmol/L. Results: Glucocorticoid-induced diabetes mellitus was observed in 26.3% of patients, and the glucocorticoid-induced diabetes mellitus group had higher age, higher BMI, lower rates of females and systemic lupus erythematosus, higher rates of smoking, alcohol use, and microscopic polyangiitis. Multivariate logistic regression analysis demonstrated that the risk of glucocorticoid-induced diabetes mellitus was independently higher in every 10-year increment of initial age with adjusted odds ratio (OR) 1.556 (95% confidence interval: 1.359 - 1.783), in every 1 kg/m(2) increment of BMI with OR 1.062 (1.002 - 1.124), in current smoking with OR 1.664 (1.057 - 2.622), and in every 10 mg increment of initial dose of prednisolone with OR 1.250 (1.074 - 1.454). Conclusions: High-dose glucocorticoids caused diabetes mellitus with high prevalence within a short period, and current smokers should be considered at higher risk of glucocorticoid-induced diabetes mellitus in addition to age, BMI, and initial dose.