目的观察胰岛素泵强化治疗对新诊断T2DM患者长期白蛋白尿发生风险的影响。方法收集2000年1月至2008年1月于北京军区总医院内分泌科住院的新诊断T2DM患者583例,按住院期间治疗方式分为胰岛素泵强化治疗(CSII)组275例和口服药物降糖治疗(OHA)组308例。于2014年1月至2015年6月进行随访,复查糖尿病及其并发症相关指标。比较两组白蛋白尿发生率及尿白蛋白/肌酐(UAlb/Cr)。按UAlb/Cr采取分层抽样法从两组各抽出60例行72 h动态血糖监测(CGMS),比较两组间血糖波动参数,并探讨UAlb/Cr与血糖波动的相关性。结果随访时,两组BMI、BP、TG、TC、LDL-C、HbA_1c比较,差异无统计学意义(P>0.05);CSII组UAlb/Cr水平低于OHA组(P<0.05)。随访时,CSII组UAlb/Cr 30~299 mg/24 h和UAlb/Cr≥300 mg/24h发生率均低于OHA组(P<0.05);CSII组日内平均血糖波动幅度(MAGE)、全天血糖标准差(SDBG)、平均餐后血糖波动幅度(MP—PGE)及最大血糖波动幅度(LAGE)均低于OHA组(P<0.05)。Pearson相关分析表明,UAlb/Cr水平与MAGE、SDBG、MPPGE、LAGE呈正相关(r=0.227、0.103、0.326、0.156,P<0.05)。多元逐步回归分析显示,病程、SBP、TG、MAGE是UAlb/Cr的影响因素。结论早期胰岛素泵强化治疗可降低新诊断T2DM患者长期白蛋白尿的发生风险,血糖波动可能参与其中。 Objective To observe the effect of intensive insulin therapy on the risk of long-term albuminuria in newly diagnosed T2DM patients. Methods A total of 583 newly diagnosed T2DM patients hospitalized in Department of Endocrinology of General Hospital of Beijing Military Command from January 2000 to January 2008 were collected and divided into 275 cases of insulin pump intensive treatment (CSII) group and oral hypoglycemic treatment (OHA) group of 308 cases. From January 2014 to June 2015, follow-up was conducted to review the related indicators of diabetes and its complications. The incidence of albuminuria and urinary albumin / creatinine (UAb / Cr) were compared between the two groups. According to the UAlb / Cr stratified sampling method, 60 patients underwent 72-h continuous glucose monitoring (CGMS), and the blood glucose fluctuation parameters were compared between the two groups. The correlation between UAlb / Cr and blood glucose was also discussed. Results There was no significant difference in BMI, BP, TG, TC, LDL-C and HbA 1c between the two groups (P> 0.05). The level of UAlb / Cr in CSII group was lower than that in OHA group (P <0.05). At follow-up, the incidence of UAIb / Cr 30 ~ 299 mg / 24 h and UAlb / Cr ≥ 300 mg / 24h in CSII group was lower than that in OHA group (P <0.05) The standard deviation of blood glucose (SDBG), average postprandial blood glucose fluctuation (MP-PGE) and maximum blood glucose fluctuation (LAGE) were lower than those in OHA group (P <0.05). Pearson correlation analysis showed that UAlb / Cr levels were positively correlated with MAGE, SDBG, MPPGE and LAGE (r = 0.227,0.103,0.326,0.156, P <0.05). Multiple stepwise regression analysis showed that duration, SBP, TG and MAGE were the influencing factors of UAlb / Cr. Conclusions Early insulin pump therapy can reduce the risk of long-term albuminuria in newly diagnosed T2DM patients. The fluctuation of blood glucose may be involved.