目的 探討胰島素強化治療后2型糖尿?。═2DM）患者發(fā)生低血糖的危險因素以及血糖的監(jiān)控策略。方法 選取2019年3月—2021年3月宜賓市第二人民醫(yī)院確診為T2DM患者172例作為研究對象，采用計算機產(chǎn)生隨機數(shù)法以2∶1的比例分為訓練集（n＝115）和測試集（n＝57），患者均接受胰島素強化治療，依據(jù)訓練集患者治療后是否發(fā)生低血糖癥狀將其分為低血糖組（n＝35）和非低血糖組（n＝80）。采用單因素分析和多因素Logistic回歸分析對兩組患者的臨床資料進行統(tǒng)計分析。根據(jù)多因素分析結(jié)果利用R 3.6.0軟件建立模型，并對預測模型進行驗證。172例患者依據(jù)血糖監(jiān)測方法分為瞬感掃描式葡萄糖監(jiān)測系統(tǒng)（FGM）組（n＝102）和血糖儀組（n＝70）評估FGM對患者治療后血糖監(jiān)測價值，測定兩組患者治療前后的靜脈血糖以及指尖血糖指標，對比分析兩組患者的血糖控制水平、血糖波動水平、低血糖發(fā)生情況、血糖達標時間、胰島素用量等。結(jié)果 訓練集115例患者胰島素強化治療后35例患者出現(xiàn)低血糖癥狀，單因素分析和多因素Logistic回歸分析表明：病程、體質(zhì)量指數(shù)（BMI）、肌酐（Cre）為T2DM患者進行胰島素強化治療后出現(xiàn)低血糖的獨立危險因素，通過FGM監(jiān)測血糖為保護因素。根據(jù)多因素分析結(jié)果構(gòu)建列線圖預測模型并對模型進行驗證，訓練集和驗證集的一致性指數(shù)（C-index）分別為0.843（95%置信區(qū)間：0.747～0.893）、0.832（95%置信區(qū)間：0.725～0.881），表明該模型具有較強的預測能力。此外，應用FGM監(jiān)測的患者血糖達標所需時間比血糖儀監(jiān)測組更短，胰島素的用量更少，治療后患者24 h內(nèi)的血糖平均值以及血糖標準差更低，而且在治療后患者出現(xiàn)低血糖的概率也更低，差異具有統(tǒng)計學意義（P＜0.05）。結(jié)論 病程、BMI、Cre為T2DM患者進行胰島素強化治療后出現(xiàn)低血糖的獨立危險因素，F(xiàn)GM監(jiān)測血糖為保護因素。FGM相比于血糖儀監(jiān)測具有更好的臨床效果，可廣泛應用于臨床治療中。

Objective To explore the risk factors of hypoglycemia in patients with T2DM (type 2 diabetes) after intensive insulin therapy and the monitoring strategy of blood glucose. Methods A total of 172 patients with T2DM diagnosed in Yibin Second People's Hospital from March 2019 to March 2021 were selected as the research object. The computer-generated random number method was used to divide them into training set (n=115) and test set (n=57) in the ratio of 2:1. The patients were treated with intensive insulin therapy. According to whether the patients in the training set had hypoglycemic symptoms after treatment, they were divided into hypoglycemic group (n=35) and non hypoglycemic group (n=80). The clinical data of the two groups were statistically analyzed by univariate analysis and multivariate Logistic regression analysis. According to the results of multi factor analysis, the model is established by using R3.6.0 software, and the prediction model is verified. In addition, in order to evaluate the value of transient scanning glucose monitoring system (FGM) in patients' blood glucose monitoring after treatment, 172 patients were divided into FGM group (n=102) and blood glucose meter group (n=70) according to the blood glucose monitoring method. The venous blood glucose and fingertip blood glucose indexes of two groups were measured before and after treatment. The blood glucose control level, blood glucose fluctuation level, hypoglycemia occurrence, blood glucose standard time, insulin dosage and so on were compared and analyzed between two groups. Results After intensive insulin treatment, 35 patients had hypoglycemia symptoms. Univariate analysis and multivariate Logistic regression analysis showed that the course of disease, BMI and Cre were independent risk factors for hypoglycemia in T2DM patients after intensive insulin treatment, and blood glucose monitored by FGM was the protective factor. According to the results of multivariate analysis, the nomogram prediction model is constructed and verified. The C-index of training set and verification set are 0.843 (95%CI:0.747-0.893) and 0.832 (95%CI:0.725-0.881), respectively, indicating that the model has strong prediction ability. In addition, the time required for the blood glucose of patients monitored by FGM detection system to reach the standard was shorter than that of the blood glucose monitor group, the dosage of insulin was less, the average blood glucose value and blood glucose standard deviation within 24 hours after treatment were lower, and the probability of hypoglycemia was lower after treatment, the difference was statistically significant (P < 0.05). Conclusion Course of disease, BMI and Cre are independent risk factors for hypoglycemia in T2DM patients after intensive insulin treatment, and FGM monitoring blood glucose is a protective factor. Compared with blood glucose monitor, FGM has better clinical effect and can be widely used in clinical treatment.

R977