School of Pharmacy, Sungkyunkwan University, Suwon, South Korea.
Department of Biohealth Regulatory Science, Sungkyunkwan University, Suwon, South Korea.
BMC Med. 2023 Feb 10;21(1):47. doi: 10.1186/s12916-023-02765-2.
Impaired respiratory function remains underrecognized in patients with type 2 diabetes (T2D), despite common pulmonary impairment. Meanwhile, there is little data available on the respiratory effects of sodium glucose cotransporter 2 inhibitors (SGLT2i). Hence, we examined the association between SGLT2i use and the risk of adverse respiratory events in a real-world setting.
We conducted a population-based, nationwide cohort study using an active-comparator new-user design and nationwide claims data of South Korea from January 2015 to December 2020. Among individuals aged 18 years or older, propensity score matching was done to match each new user of SGLT2is with dipeptidyl peptidase 4 inhibitors (DPP4is), with patients followed up according to an as-treated definition. The primary outcome was respiratory events, a composite endpoint of acute pulmonary edema, acute respiratory distress syndrome (ARDS), pneumonia, and respiratory failure. Secondary outcomes were the individual components of the primary outcome and in-hospital death. Cox models were used to estimate hazard ratios (HRs) and 95% CIs.
Of 205,534 patient pairs in the propensity score matched cohort, the mean age of the entire cohort was 53.8 years and 59% were men, with a median follow-up of 0.66 years; all baseline covariates achieved balance between the two groups. Incidence rates for overall respiratory events were 4.54 and 7.54 per 1000 person-years among SGLT2i and DPP4i users, respectively, corresponding to a rate difference of 3 less events per 1000 person-years (95% CI - 3.44 to - 2.55). HRs (95% CIs) were 0.60 (0.55 to 0.64) for the composite respiratory endpoint, 0.35 (0.23 to 0.55) for acute pulmonary edema, 0.44 (0.18 to 1.05) for ARDS, 0.61 (0.56 to 0.66) for pneumonia, 0.49 (0.31 to 0.76) for respiratory failure, and 0.46 (0.41 to 0.51) for in-hospital death. Similar trends were found across individual SGLT2is, subgroup analyses of age, sex, history of comorbidities, and a range of sensitivity analyses.
These findings suggest a lower risk of adverse respiratory events associated with patients with T2D initiating SGLT2is versus DPP4is. This real-world evidence helps inform patients, clinicians, and guideline writers regarding the respiratory effects of SGLT2i in routine practice.
尽管存在常见的肺部损伤，2 型糖尿病（T2D）患者的呼吸功能受损仍未得到充分认识。同时，关于钠-葡萄糖共转运蛋白 2 抑制剂（SGLT2i）对呼吸影响的数据很少。因此，我们在真实环境中研究了 SGLT2i 使用与不良呼吸事件风险之间的关联。
我们使用基于人群的全国性队列研究，采用活性对照新用户设计和韩国全国范围内的索赔数据，时间范围为 2015 年 1 月至 2020 年 12 月。在年龄为 18 岁或以上的人群中，我们采用倾向评分匹配来匹配每个 SGLT2i 的新使用者与二肽基肽酶 4 抑制剂（DPP4i）的使用者，按照治疗定义对患者进行随访。主要结局是呼吸事件，急性肺水肿、急性呼吸窘迫综合征（ARDS）、肺炎和呼吸衰竭的综合终点。次要结局是主要结局的各个组成部分和住院内死亡。使用 Cox 模型估计风险比（HRs）和 95%置信区间（CIs）。
在倾向评分匹配队列中，共有 205534 对患者，整个队列的平均年龄为 53.8 岁，59%为男性，中位随访时间为 0.66 年；两组所有基线协变量均达到平衡。SGLT2i 和 DPP4i 使用者的总体呼吸事件发生率分别为每 1000 人年 4.54 次和 7.54 次，相应的每 1000 人年发生率差异为 3 次事件（95%CI：-3.44 至-2.55）。复合呼吸终点的 HRs（95%CI）为 0.60（0.55 至 0.64），急性肺水肿为 0.35（0.23 至 0.55），ARDS 为 0.44（0.18 至 1.05），肺炎为 0.61（0.56 至 0.66），呼吸衰竭为 0.49（0.31 至 0.76），住院内死亡为 0.46（0.41 至 0.51）。在个别 SGLT2i 中、年龄、性别、合并症史的亚组分析以及一系列敏感性分析中，都发现了类似的趋势。
这些发现表明，与使用 DPP4i 的 T2D 患者相比，开始使用 SGLT2i 的患者发生不良呼吸事件的风险较低。这一真实世界的证据有助于告知患者、临床医生和指南制定者有关 SGLT2i 在常规实践中的呼吸影响。
