Predictive clinical parameters for the hemoglobin A1c-lowering effect of vildagliptin in Japanese patients with type 2 diabetes

• Published in: Diabetology international Vol. 5; no. 4; pp. 229 - 233
• Main Authors: Bando, Yukihiro, Yamada, Masayuki, Aoki, Keiko, Kanehara, Hideo, Hisada, Azusa, Okafuji, Kazuhiro, Toya, Daisyu, Tanaka, Nobuyoshi
• Format: Journal Article
• Language: English
• Published: Tokyo Springer Japan 01.12.2014
• Subjects: Diabetes; Endocrinology; Medicine; Medicine & Public Health; Metabolic Diseases; Original Article; Type 2 diabetes; HbA1c; Vildagliptin
• ISSN: 2190-1678; 2190-1686
• DOI: 10.1007/s13340-014-0161-3

Aims This study assessed the independent predictors of the hemoglobin A1c (HbA1c)-lowering effect of vildagliptin in Japanese patients with type 2 diabetes. Materials and methods The data of a total of 155 patients with type 2 diabetes who had been taking vildagliptin 50 or 100 mg once or twice daily for inadequate glycemic control for at least 12 weeks regardless of oral hypoglycemic agents (OHAs) were retrieved from medical records. Spearman’s rank correlation coefficient method was applied to determine the correlations between sets of two independent continuous variables. Multiple linear regression analysis was used to search for independent predictors of the reduction in HbA1c levels after 12 weeks of vildagliptin treatment (ΔHbA1c). Results Among all subjects, ΔHbA1c was positively correlated with baseline HbA1c ( r  = 0.549, P  < 0.0001). Multiple linear regression analysis using baseline variables among all patients also revealed that the independent factors contributing to ΔHbA1c were prescribing method (drug naïve or add-on vs. switch from other OHAs; P  < 0.0001) and baseline HbA1c ( P  < 0.0001), without being significantly influenced by body mass index, diabetes duration, or age. Conclusions Our analysis provides evidence that baseline HbA1c and prescribing method contribute to the HbA1c-lowering effect of vildagliptin in Japanese patients with type 2 diabetes. Long-term analysis of a larger population is warranted to confirm these findings.