摘要目的 评价亚洲2型糖尿病(T2DM)患者服用达格列净的安全性. 方法 检索有关数据库与临床试验注册网站从建库至2017年7月27日收录的文献,筛选亚洲T2DM患者应用达格列净的RCT.以应用达格列净的患者为达格列净组,应用安慰剂或其他降糖药的患者为对照组.主要安全性终点为心血管安全性、低血糖、生殖系统感染和泌尿系统感染;次要安全性终点为肿瘤、肾盂肾炎、肾衰竭、骨折和截肢等.使用RevMan 5.3和STATA12.1软件进行Meta分析.计数资料效应指标为风险比(RR)及95%置信区间(CI);计量资料效应指标为均数差(MD)及95% CI. 结果 纳入分析的RCT共9项,涉及1 857例患者,其中达格列净组1 180例,对照组677例[安慰剂组619例,二肽基肽酶4(DPP4)抑制剂组58例].Meta分析结果显示,达格列净组与安慰剂组主要不良心血管事件[0.58% (4/684)比0.99% (4/406),RR＝0.59,95% CI:0.17 ～ 2.02,P＝0.40)]、非致死性心肌梗死[0.20% (1/510)比0.63% (2/319),RR＝0.43,95% CI:0.08 ～2.30,P＝0.33]、非致死性卒中[0.52% (3/574)比0.57% (2/352),RR＝0.76,95% CI:0.18～3.24,P＝0.71]、低血糖[5.79% (65/1 122)比8.08% (50/619),RR＝0.92,95% CI :0.67～1.28,P＝0.64]和泌尿系统感染[3.48% (39/1 122)比3.39% (21/619),RR＝1.03,95% CI:0.61～1.73,P＝0.91]发生率的差异均无统计学意义,但达格列净组生殖系统感染的发生率高于安慰剂组,差异有统计学意义[2.08% (23/1 105)比0.33% (2/611),RR＝3.42,95% CI:1.21～9.70,P＝0.02].亚组分析结果显示,达格列净10 mg组生殖系统感染发生率高于安慰剂组,差异有统计学意义[2.30% (13/564)比0.36% (2/551),RR＝4.56,95% CI:1.32～15.78,P＝0.02],不同剂量达格列净与其他主要终点事件发生率无明显相关.达格列净组血尿酸水平较基线水平降低的程度强于安慰剂组,差异有统计学意义[MD＝ －15.47 μmol/L,95% CI:－30.35～ －0.60,P＝0.04).达格列净组与安慰剂组发生肿瘤、肾脏不良事件、骨折、截肢的风险相当,差异均无统计学意义(均P＞0.05).达格列净组与DPP4抑制剂组估计肾小球滤过率较基线水平下降的差异亦无统计学意义[MD＝1.70 ml/(min1.73m2),95% CI:－7.79～11.19,P＝0.73].结论 亚洲T2DM患者服用达格列净的安全性较好.临床实践中应警惕该药致生殖系统感染的可能.

abstractsObjective To evaluate the safety of dapagliflozin in Asian patients with type 2 diabetes mellitus (T2DM ). Methods Randomized controlled trials (RCTs ) which evaluated the safety of dapagliflozin in Asian patients with T2DM from related databases and clinical trial registries from establishment to July 27,2017 were searched. The patients who treated with dapagliflozin enrolled the dapagliflozin group. The patients who received placebo or other hypoglycemic agents enrolled the control group. The primary safety endpoints included cardiovascular safety,hypoglycemia,genital tract infection and urinary tract infection. The secondary safety endpoints included tumor,pyelonephritis,renal failure, fracture,amputation,etc. Meta-analysis was conducted using RevMan 5.3 and Stata 12.1 software. Risk ratio (RR)and 95% confidence intervals (CI)were calculated for dichotomous outcomes. Mean difference (MD)and 95% CI were calculated for continuous outcomes. Results A total of 9 RCTs involving 1 857 Asian T2DM patients enrolled the study. Of 1 857 patients,1 180 patients were in the dapagliflozin group and 677 patients were in the control group [the placebo group:619 patients;the dipeptidyl peptidase-4 (DPP4 ) inhibitor group:58 patients ]. The results of Meta-analysis showed that the differences of incidences of major adverse cardiovascular events (MACE ),non-fatal myocardial infarction,non-fatal stroke,hypoglycemia,and urinary tract infection between the dapagliflozin group and the placebo group were all no statistical significance[0.58% (4/684)vs. 0.99% (4/406),RR＝0.59,95% CI:0.17-2.02, P＝0.40],[0.20% (1/510)vs. 0.63% (2/319),RR ＝0.43,95% CI:0.08-2.30,P ＝0.33], [0.52% (3/574)vs. 0.57% (2/352),RR＝0.76,95% CI:0.18-3.24,P＝0.71],[5.79% (65/1 122)vs. 8.08% (50/619),RR＝0.92,95% CI:0.67-1.28,P＝0.64],and [3.48% (39/1 122) vs. 3.39% (21/619),RR＝1.03,95% CI:0.61-1.73,P＝0.91],respectively. The incidence of genital tract infection in the dapagliflozin group was higher than that in the placebo group,but the difference was statistically significant [2.08% (23/1 105)vs. 0.33% (2/611),RR＝3.42,95% CI:1.21-9.70,P＝0.02 ]. The result of subgroup analysis showed that the incidence of genital tract infection in the dapagliflozin 10 mg group was higher than that in the placebo group,the difference was statistically significant [2.30% (13/564)vs. 0.36% (2/551),RR＝4.56,95CI:1.32-15.78,P＝0.02]. There were no significant correlations of other major endpoint events between the dapagliflozin groups in different dosage and the placebo group. The level of serum uric acid,compared with the baseline in the dapagliflozin group,decreased more than that in the placebo group,the difference was statistically significant (MD ＝－15.47 μmol/L,95% CI:－30.35-－0.60,P ＝0.04). There were no statistical significances of the incidences of tumor,renal adverse events,fracture,and amputation between the dapagliflozin group and the placebo group (all P＞0.05). The level of estimated glomerular filtration rate,compared with the baseline in the dapagliflozin group,decreased similar to that in the DPP4 inhibitor group,the difference was not statistically significant [MD ＝ 1. 70 ml/(min1. 73 m2),95% CI:－ 7. 79-11. 19,P ＝ 0. 73 ]. Conclusions Dapagliflozin is relatively safe in the treatment of Asian patients with type 2 diabetes. It should be vigilant for the probability of genital tract infection induced by dapagliflozin in the clinical practice.