Scholarlink

Aims <p>To evaluate whether continuing versus reducing metformin at imeglimin initiation is associated with 24-week changes in HbA1c and body weight in routine practice, and to describe 48-week outcomes.</p> Methods <p>Single-center retrospective study of adults with type 2 diabetes who started imeglimin 2000&#xa0;mg/day (September 2022–August 2024). For the primary 24-week efficacy analysis, metformin users who tolerated and continued imeglimin through week 24 were classified as metformin continuation (no change) or metformin reduction (≥ 250&#xa0;mg/day decrease, including discontinuation). The primary endpoint was 24-week ΔHbA1c. Associations were assessed with nonparametric tests and ANCOVA adjusting for age, sex, diabetes duration, and baseline HbA1c; sensitivity models additionally adjusted for baseline metformin dose and major concomitant glucose-lowering drug classes (SGLT2i, GLP-1RA, insulin, and sulfonylureas).</p> Results <p>Seventy metformin users were included (continuation n = 34; reduction n = 36). At 24&#xa0;weeks, HbA1c decreased by − 0.6% versus − 0.2% (median difference − 0.5%, 95% CI − 0.6 to − 0.3), and body weight by − 1.2&#xa0;kg versus − 0.4&#xa0;kg (median difference − 0.9&#xa0;kg). The absolute metformin dose reduction correlated with ΔHbA1c (Spearman ρ = 0.52). Metformin continuation remained associated with greater adjusted HbA1c reduction in sensitivity ANCOVA (adjusted difference − 0.57%, 95% CI − 0.88 to − 0.26; <i>P</i> = 0.0006). In a 48-week cohort (n = 65) continuing imeglimin, metformin dose reduction remained associated with ΔHbA1c in multivariable analysis.</p> Conclusions <p>Continuing metformin at imeglimin initiation was associated with greater improvements in HbA1c and body weight than dose reduction. Findings are associative and may reflect residual confounding.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Effect of metformin dose reduction versus continuation at imeglimin initiation on glycemic control in type 2 diabetes: a retrospective real-world analysis

  • 期刊论文

  • 发表日期：

    2026年2月9日

查看全文
  • Yoshiro Fushimi,
  • Tomohiko Kimura,
  • Junpei Sanada,
  • Takashi Ito,
  • Masato Kubo,
  • Kazunori Dan,
  • Yui Okamoto,
  • Hideyuki Iwamoto,
  • Yuichiro Iwamoto,
  • Masashi Shimoda,
  • Shuhei Nakanishi,
  • Kohei Kaku,
  • Hideaki Kaneto

摘要

Aims

To evaluate whether continuing versus reducing metformin at imeglimin initiation is associated with 24-week changes in HbA1c and body weight in routine practice, and to describe 48-week outcomes.

Methods

Single-center retrospective study of adults with type 2 diabetes who started imeglimin 2000 mg/day (September 2022–August 2024). For the primary 24-week efficacy analysis, metformin users who tolerated and continued imeglimin through week 24 were classified as metformin continuation (no change) or metformin reduction (≥ 250 mg/day decrease, including discontinuation). The primary endpoint was 24-week ΔHbA1c. Associations were assessed with nonparametric tests and ANCOVA adjusting for age, sex, diabetes duration, and baseline HbA1c; sensitivity models additionally adjusted for baseline metformin dose and major concomitant glucose-lowering drug classes (SGLT2i, GLP-1RA, insulin, and sulfonylureas).

Results

Seventy metformin users were included (continuation n = 34; reduction n = 36). At 24 weeks, HbA1c decreased by − 0.6% versus − 0.2% (median difference − 0.5%, 95% CI − 0.6 to − 0.3), and body weight by − 1.2 kg versus − 0.4 kg (median difference − 0.9 kg). The absolute metformin dose reduction correlated with ΔHbA1c (Spearman ρ = 0.52). Metformin continuation remained associated with greater adjusted HbA1c reduction in sensitivity ANCOVA (adjusted difference − 0.57%, 95% CI − 0.88 to − 0.26; P = 0.0006). In a 48-week cohort (n = 65) continuing imeglimin, metformin dose reduction remained associated with ΔHbA1c in multivariable analysis.

Conclusions

Continuing metformin at imeglimin initiation was associated with greater improvements in HbA1c and body weight than dose reduction. Findings are associative and may reflect residual confounding.