Lin Yu-Jie, Liu Peter Pin-Sung, Huang Hui-Kai, Liu Hwan-Wun, Loh Ching-Hui, Chuang Lee-Ming, Chang Chia-Hsuin, Yeh Jih-I
School of Medicine, Tzu Chi University, Hualien, Taiwan.
Department of Aging and Community Medicine, Hualien Tzu Chi General Hospital, Hualien, Taiwan.
Sci Rep. 2025 Aug 13;15(1):29694. doi: 10.1038/s41598-025-13254-2.
The optimal timing for initiating insulin therapy in patients with type 2 diabetes (T2D) remains uncertain and varies among clinical guidelines. This retrospective cohort study analyzed claims data to include insulin-naïve T2D patients aged ≥ 20 years who intensified treatment using either insulin or non-insulin therapies between 2012 and 2021. Cox proportional hazards models were applied to estimate hazard ratios (HRs) for major adverse cardiovascular events (MACE) and all-cause mortality, with adjustments for sex, age interval, index year interval, number of outpatient visits, number of inpatient admissions, diabetes duration, Charlson Comorbidity Index (CCI), Diabetes Complication Severity Index (DCSI), the number of prior antidiabetic medications, medications for hypertension, hyperlipidemia, antiplatelets, and anticoagulants. Subgroup analyses stratified by sex, age, the number of prior antidiabetic medications and types of insulin were also performed. Compared to non-insulin intensification, insulin therapy was associated with significantly higher risks of MACE and all-cause mortality. The adjusted HRs (95% confidence intervals [CIs]) were 2.78 (2.64-2.92) for MACE and 4.74 (4.63-4.85) for all-cause mortality. Subgroup analyses revealed consistently elevated risks across all patient groups, with the smallest risk increases observed in patients who had previously used three non-insulin drugs before initiating insulin therapy (HRs for MACE: 2.62 [2.13-3.22]; all-cause mortality: 3.05 [2.68-3.49]). Among insulin types, long-acting insulin was associated with the lowest risk increases (MACE HR: 1.34 [1.19-1.51]; all-cause mortality HR: 1.90 [1.78-2.03]). In conclusion, treatment intensification with insulin was linked to increased risks of MACE and all-cause mortality. The lowest risks were observed in patients initiating long-acting insulin following prior therapy with three non-insulin drugs. These findings highlight the need for careful patient evaluation and individualized decision-making when initiating insulin therapy in T2D management.
2型糖尿病（T2D）患者开始胰岛素治疗的最佳时机仍不确定，且不同临床指南的建议有所不同。这项回顾性队列研究分析了索赔数据，纳入了2012年至2021年间年龄≥20岁、开始使用胰岛素或非胰岛素疗法强化治疗的初治T2D患者。应用Cox比例风险模型估计主要不良心血管事件（MACE）和全因死亡率的风险比（HR），并对性别、年龄区间、索引年份区间、门诊就诊次数、住院次数、糖尿病病程、Charlson合并症指数（CCI）、糖尿病并发症严重程度指数（DCSI）、既往抗糖尿病药物数量、高血压用药、高脂血症用药、抗血小板药物和抗凝药物进行了调整。还按性别、年龄、既往抗糖尿病药物数量和胰岛素类型进行了亚组分析。与非胰岛素强化治疗相比，胰岛素治疗与MACE和全因死亡率的风险显著升高相关。调整后的HR（95%置信区间[CI]），MACE为2.78（2.64 - 2.92），全因死亡率为4.74（4.63 - 4.85）。亚组分析显示，所有患者组的风险持续升高，在开始胰岛素治疗前曾使用三种非胰岛素药物的患者中，风险增加最小（MACE的HR：2.62[2.13 - 3.22]；全因死亡率：3.05[2.68 - 3.49]）。在胰岛素类型中，长效胰岛素的风险增加最低（MACE的HR：1.34[1.19 - 1.51]；全因死亡率的HR：1.90[1.78 - 2.03]）。总之，胰岛素强化治疗与MACE和全因死亡率风险增加有关。在先前使用三种非胰岛素药物后开始使用长效胰岛素的患者中观察到的风险最低。这些发现凸显了在T2D管理中开始胰岛素治疗时仔细评估患者和个体化决策的必要性。
