Gamble John-Michael, Thomas Jamie M, Twells Laurie K, Midodzi William K, Majumdar Sumit R
aSchool of Pharmacy bFaculty of Medicine, Memorial University of Newfoundland, St. John's, Newfoundland and Labrador, A1B 3V6 cDivision of General Internal Medicine, Department of Medicine, University of Alberta, Edmonton, Alberta, T6G 2B7, Canada.
Medicine (Baltimore). 2016 Jun;95(26):e3995. doi: 10.1097/MD.0000000000003995.
There is limited comparative effectiveness evidence to guide approaches to managing diabetes in individuals failing metformin monotherapy. Our aim was to compare the incidence of all-cause mortality and major adverse cardiovascular events (MACEs) among new metformin monotherapy users initiating a dipeptidyl-peptidase-4 inhibitor (DPP4i), glucagon-like peptide-1 receptor agonist (GLP-1RA), sulfonylurea (SU), thiazolidinedione, or insulin.We conducted a cohort study using the UK-based Clinical Practice Research Datalink. Participants included a cohort of 38,233 new users of metformin monotherapy who initiated a 2nd antidiabetic agent between January 1, 2007 and December 31, 2012 with follow-up until death, disenrollment, therapy discontinuation, or study end-date. A subcohort of 21,848 patients with linked hospital episode statistics (HES) and Office of National Statistics (ONS) data were studied to include MACE and cardiovascular-related death. The primary exposure contrasts, defined a priori, were initiation of a DPP4i versus an SU and initiation of a GLP-1RA versus an SU following metformin monotherapy. Cox proportional hazards models were used to assess the relative differences in time to mortality and MACE between exposure contrasts, adjusting for important baseline patient factors and comedications used during follow-up.The main study cohort consisted of 6213 (16%) patients who initiated a DPP4i, 25,916 initiated an SU (68%), 4437 (12%) initiated a TZD, 487 (1%) initiated a GLP-1RA, 804 (2%) initiated insulin, and 376 (1%) initiated a miscellaneous agent as their 2nd antidiabetic agent. Mean age was 62 years, 59% were male, and mean glycated hemoglobin was 8.8% (92.6 mmol/mol). Median follow-up was 2.7 years (interquartile range 1.3-4.2). Mortality rates were 8.2 deaths/1000 person-years for DPP4i and 19.1 deaths/1000 person-years for SU initiators. Adjusted hazards ratio (aHR) for mortality in DPP4i versus SU initiators = 0.58, 95% CI 0.46 to 0.73, P < 0.001. MACE rates were 19.1/1000 person-years for DPP4i initiators, 15.9/1000 person-years for GLP1-RA initiators versus 33.1/1000 person-years for SU initiators (aHR: DPP4i vs SU initiators = 0.64, 95%CI 0.52-0.80; GLP1RA vs SU initiators = 0.73, 95% CI 0.34-1.55).In this cohort of metformin monotherapy users, 2nd-line DPP4i use was associated with a 42% relative reduction in all-cause mortality and 36% reduction in MACE versus SUs, the most common 2nd-line therapy in our study. GLP-1RAs were not associated with adverse events in this cohort.
对于二甲双胍单药治疗效果不佳的糖尿病患者，目前指导治疗方案的比较有效性证据有限。我们的目的是比较新开始二甲双胍单药治疗的患者在加用二肽基肽酶-4抑制剂（DPP4i）、胰高血糖素样肽-1受体激动剂（GLP-1RA）、磺脲类药物（SU）、噻唑烷二酮类药物或胰岛素后全因死亡率和主要不良心血管事件（MACE）的发生率。我们利用英国临床实践研究数据链进行了一项队列研究。参与者包括38233名新开始二甲双胍单药治疗的患者队列，他们在2007年1月1日至2012年12月31日期间开始使用第二种抗糖尿病药物，并随访至死亡、退出研究、治疗中断或研究结束日期。对21848名与医院病历统计数据（HES）和国家统计局（ONS）数据相关联的患者亚组进行研究，以纳入MACE和心血管相关死亡情况。预先定义的主要暴露对比为二甲双胍单药治疗后开始使用DPP4i与SU的对比，以及开始使用GLP-1RA与SU的对比。采用Cox比例风险模型评估暴露对比组之间死亡率和MACE发生时间的相对差异，并对重要的基线患者因素和随访期间使用的合并用药进行调整。主要研究队列包括6213名（16%）开始使用DPP4i的患者、25916名（68%）开始使用SU的患者、4437名（12%）开始使用噻唑烷二酮类药物的患者、487名（1%）开始使用GLP-1RA的患者、804名（2%）开始使用胰岛素的患者以及376名（1%）开始使用其他药物作为第二种抗糖尿病药物的患者。平均年龄为62岁，59%为男性，平均糖化血红蛋白为8.8%（92.6 mmol/mol）。中位随访时间为2.7年（四分位间距1.3 - 4.2年）。DPP4i使用者的死亡率为8.2例/1000人年，SU使用者的死亡率为19.1例/1000人年。DPP4i使用者与SU使用者相比，调整后的风险比（aHR）为死亡率=0.58，95%置信区间为从0.46至0.73，P<0.001。DPP4i使用者的MACE发生率为19.1/1000人年，GLP1-RA使用者为15.9/1000人年，而SU使用者为33.1/1000人年（aHR：DPP4i与SU使用者相比=0.64，95%置信区间为0.52 - 0.80；GLP1RA与SU使用者相比=0.73，95%置信区间为0.34 - 1.55）。在这个二甲双胍单药治疗使用者队列中，与我们研究中最常用的二线治疗药物SU相比，二线使用DPP4i与全因死亡率相对降低42%以及MACE降低36%相关。在这个队列中，GLP-1RA与不良事件无关。
