Comparison of Glycemic Variability in Chinese T2DM Patients Treated with Exenatide or Insulin Glargine: A Randomized Controlled Trial

Comparison of Glycemic Variability in Chinese T2DM Patients Treated with Exenatide or Insulin... Diabetes Ther (2018) 9:1253–1267 https://doi.org/10.1007/s13300-018-0412-6 ORIGINAL RESEARCH Comparison of Glycemic Variability in Chinese T2DM Patients Treated with Exenatide or Insulin Glargine: A Randomized Controlled Trial . . . . . Ting-Ting Yin Yan Bi Ping Li Shan-Mei Shen Xiao-Lu Xiong . . . . Li-Jun Gao Can Jiang Yan Wang Wen-Huan Feng Da-Long Zhu Received: February 13, 2018 / Published online: May 9, 2018 The Author(s) 2018 (N = 39) to either exenatide treatment or insu- ABSTRACT lin glargine treatment for 16 weeks. Glycemic variability was assessed using a CGMS; hemo- Introduction: Increasing the frequency of globin A1c (HbA1c), b-cell function, weight, blood glucose monitoring aids the evaluation of body mass index (BMI), and waist circumfer- glycemic variability and blood glucose control ence were also evaluated. by antidiabetic drugs. It remains unclear, how- Results: Mean blood glucose level, continuous ever, whether GLP-1 receptor agonists or basal overlapping net glycemic action, mean ampli- insulin has a better effect on glycemic variabil- tude of glycemic excursions, percentage of the ity in type 2 diabetes mellitus (T2DM) patients time that the blood glucose value who are inadequately controlled by metformin. was [ 10.0 mmol/L, and highest blood glucose We used a continuous glucose monitoring sys- level (P \ 0.01–0.05) significantly decreased in tem (CGMS) to compare patients on a GLP-1 both groups. Standard deviation of the mean receptor agonist with patients on basal insulin glucose value, largest amplitude of glycemic in terms of glycemic variability. excursions, and waist circumference signifi- Methods: This prospective randomized study cantly decreased for those treated with exe- assigned T2DM patients treated with metformin natide (P \ 0.05), while no changes were Enhanced content To view enhanced content for this observed with insulin glargine treatment. Per- article go to https://doi.org/10.6084/m9.figshare. centage of the time that the blood glucose value was [ 7.8 mmol/L decreased after insulin glargine use (P \ 0.05) but not with the exe- T.-T. Yin  L.-J. Gao  W.-H. Feng (&) natide intervention. Similar decreases in fasting Department of Endocrinology, Drum Tower Clinical Hospital, Medical School of Southeast University, blood glucose and HbA1c and increases in the Nanjing, China 1/homeostasis model assessment of insulin e-mail: fengwh501@163.com resistance, disposition index 30, and disposition index 120 were observed in both groups T.-T. Yin  Y. Bi  P. Li  S.-M. Shen  X.-L. Xiong L.-J. Gao  Y. Wang  W.-H. Feng  D.-L. Zhu (&) (P \ 0.01–0.05). Reductions in weight and BMI Department of Endocrinology, Drum Tower were greater with exenatide than with insulin Hospital Affiliated to Nanjing University Medical glargine treatment (P \ 0.05). School, Nanjing, China Conclusions: In overweight and obese patients e-mail: zhudalong@nju.edu.cn with T2DM inadequately controlled by met- C. Jiang formin, exenatide and insulin glargine have Department of Endocrinology, Jining No 1. People’s similar efficacies in terms of glycemic Hospital, Shandong, China 1254 Diabetes Ther (2018) 9:1253–1267 variability, HbA1c alleviation, and b-cell func- and a sulfonylurea found that the addition of tion, but exenatide has a greater effect on body exenatide minimized GV to a greater extent weight and BMI. than the addition of insulin glargine, based on SMBG results [18, 19]. A previous study and meta-analysis found Keywords: Basal insulin; Glucagon-like that liraglutide (1.8 and 1.2 mg), a long-term peptide-1 receptor agonist; Glucose GLP-1 receptor agonist, provided greater fluctuation; Type 2 diabetic mellitus reductions in HbA1c and fasting plasma glucose (FBG) than exenatide (10 lg b.i.d.), a short-term GLP-1 receptor agonist. Compared with insulin INTRODUCTION glargine, adding liraglutide or other long-term GLP-1 receptor agonists to metformin alone or For patients with type 2 diabetes mellitus (T2DM), increasing the frequency of blood with a sulfonylurea decreased body weight and led to significantly improved or at least nonin- glucose monitoring is beneficial for maintain- ing physiological blood glucose levels [1, 2]. ferior glycemic control [20–22]. In contrast to Western countries, exenatide is a commonly Self-monitoring of blood glucose (SMBG) and and widely used short-term GLP-1 receptor hemoglobin A1c (HBA1c) are the most com- monly used methods to assess short-term and agonist in China [23, 24]. It is unclear from the use of a CGMS, which is a reliable technology long-term blood glucose control, respectively [3–5]. However, these measurements are inade- for monitoring glycemic control, whether exe- natide or basal insulin has a better effect on GV quate for real-time blood glucose monitoring. A continuous glucose monitoring system (CGMS) for patients with T2DM and inadequate gly- cemic control who receive metformin. In the conveniently and accurately records real-time glycemic values and trends over multiple days work described in the present paper, we used a CGMS to compare the effects of exenatide by providing a large number of blood glucose recordings [2, 5–7]. A CGMS provides a detailed treatment with those of insulin glargine treat- depiction of glycemic variability (GV) and effi- ment on the GVs of overweight and obese T2DM patients with poor glycemic control ciently analyzes rates of change in glucose levels [7, 8]. Wide fluctuations in blood glucose levels using metformin. may lead to excessive glycosylation and oxida- tive stress and are associated with reduced METHODS endothelial function in patients with T2DM; these are key factors in the development of Subjects diabetic complications [9, 10]. Patients with glycemic lability often have long-standing dia- Thirty-nine T2DM patients with poor glycemic betes with both beta-cell function and counter- control after receiving metformin monotherapy regulatory hormone failure [11]. GV has were recruited from Drum Tower Hospital (af- recently been evaluated as a potential target for filiated with Nanjing University Medical School, diabetic complication interventions, and the China) and randomized to receive exenatide or GV record provided by a CGMS may improve insulin glargine treatment for 16 weeks. Ran- glycemic control [12–14]. dom numbers were provided by third-party Metformin, an oral antidiabetic drug, is the statisticians and block randomization was used. first-line treatment for T2DM. In patients with Inclusion criteria were as follows: (1) a T2DM and poorly controlled blood glucose stable metformin dose C 1.5 g/day over at least receiving metformin alone, glucagon-like pep- 8 weeks; (2) body mass index (BMI) C 24 kg/m ; tide-1 (GLP-1) receptor agonists and basal (3) HbA1c level between 7.0% and 10.0%; and insulin are used as optional antidiabetic drugs (4) age between 18 and 70 years. Exclusion cri- [15–17]. Studies involving patients with T2DM teria were as follows: (1) renal dysfunction, and poor glycemic control receiving metformin, defined as a serum creatinine level C 1.5 mg/dL; a sulfonylurea, or a combination of metformin Diabetes Ther (2018) 9:1253–1267 1255 (2) diseases causing acute or chronic hypoxia; peripheral blood glucose levels reached (3) liver dysfunction, defined as an alanine 6.1 mmol/L; the insulin glargine dose was not aminotransferase (ALT) or aspartate amino- adjusted over the final 12 weeks. GV was mea- transferase (AST) level C 3 times higher than sured at baseline and during the last week. This the upper limit of normal; (4) history of car- study was registered with Clinical trials.gov (ID: diovascular disease (CVD) during the preceding NCT02325960). year; (5) proliferative retinopathy; (6) positive pregnancy test result, breast-feeding, or refusal Measurements to use appropriate contraceptive methods; (7) systemic corticosteroid therapy during the pre- Glycemic Variability ceding 2 months; (8) type 1 diabetes; and (9) use GV was assessed at baseline and at the end of of other experimental drugs over the preceding the study using a CGMS (Glod, Medtronic) for 1 month. All patients included in the study up to 72 h. The CGMS sensor was inserted into provided written informed consent. The study subcutaneous abdominal fat tissue, and the protocol was approved by the Research Ethics average electrical signal was recorded every Board of Drum Tower Hospital, which is affili- 5 min, yielding 288 glucose level measurements ated with Nanjing University Medical School per day and 864 data points for 3 consecutive (Protocol: AF/SQ-2014-072-01). The present days. According to the CGMS results, the fol- study was monitored by the Drug Clinical Trial lowing key parameters were calculated: (1) Agency Office of Drum Tower Hospital, which is mean blood glucose (MBG) level, which was affiliated with Nanjing University Medical estimated as the average value and standard School. Adverse events (AEs) were collected and deviation (SD) of the 288 data points recorded recorded in case report form, and serious during the 24 h of continuous glucose moni- adverse events (SAEs) were reported in written toring; (2) mean amplitude of glycemic excur- form to the Institutional Review Board of the sion (MAGE), determined by calculating the Drug Clinical Trial Agency Office and the arithmetic mean of the difference between Research Ethics Board of Drum Tower Hospital. consecutive peaks and nadirs if the difference was [ 1 SD of mean glucose; (3) mean absolute Endpoint value of daily differences, calculated as the mean absolute deviation of matched values The primary endpoint of this study was the measured during two consecutive 24-h periods change in GV during a 16-week follow-up per- of continuous glucose monitoring [25]; (4) the iod. Secondary endpoints included changes in difference between the highest and lowest levels the following factors at 16 weeks: blood glucose, of blood glucose, calculated as the difference HbA1c, weight, BMI, waist circumference, between the highest and lowest blood glucose insulin function, liver function, and lipid values during 24 h of continuous glucose profile. monitoring; (5) TBG [ 7.8 mmol/L and TBG [ 11.1 mmol/L, calculated as the percent- age of the time that the blood glucose value Protocol (TBG) was [ 7.8 and [ 11.1 mmol/L, respec- tively, during 24 h; and (6) TBG \ 3.9 mmol/L, This prospective, randomized, and parallel-de- calculated as the percentage of the time that the sign trial lasted 16 weeks. For the exenatide- blood glucose value was \ 3.9 mmol/L during treated patients, the initial dosage was 5 lg 24 h [26]. twice daily for 4 weeks, followed by a mainte- nance dose of 10 lg twice daily throughout the Standard Meal Tolerance Test trial. For patients receiving insulin glargine, the All patients enrolled in our study underwent a initial dose was 8 IU once daily, followed by a standard meal tolerance test at baseline and titrated dosage of C 2 IU every 3 days based on after 16 weeks of intervention. Plasma glucose FBG levels during the first 4 weeks until the 1256 Diabetes Ther (2018) 9:1253–1267 and insulin levels were measured 0, 30, 60, and exenatide and glargine, we can conservatively 120 min after meal ingestion. The 1/homeosta- estimate that the MAGE change from baseline sis model assessment of insulin resistance (1/ in the exenatide group is approximate HOMA-IR) index and the Matsuda insulin sen- 2.55 ± 3.15, so the sample size per group nee- sitivity index (ISI ) were used to measure insu- ded to provide 80% power was calculated. lin sensitivity. The basal homeostasis model Assuming a 10% subject drop-out rate, a total of assessment of insulin secretion (HOMA-b), 44 patients (22 per group) was targeted for early-phase total insulin area under the curve randomization. Data are presented as the divided by the total glucose area under the mean ± standard error (SE). The paired Stu- curve during the first 30 min of the standard dent’s t test was used to analyze the pre- and meal tolerance test (InsAUC30/GluAUC30), and post-intervention differences within each the total-phase total insulin area under the group. Analysis of covariance (ANCOVA) was curve divided by the total glucose area under used to test for differences between the inter- the curve during the 120 min of the oral glucose vention groups after adjusting for baseline. tolerance test (InsAUC120/GluAUC120) were Pearson’s correlation analysis was conducted to used to calculate insulin release. The disposition determine the association between the variables index was used to express b-cell function and GV and HBA1c. P values of \ 0.05 were [27–29]. Values were calculated using the fol- considered significant. lowing formulae: 1/HOMA-IR = 1/(Ins 9 Glu / 0 0 22.5); ISI = 10,000/(Glu 9 Ins 9 average M 0 0 Compliance with Ethics Guidelines glucose 75-g oral glucose tolerance test [OGTT] 1/2 9 average insulin OGTT) ; HOMA-b = (20 9 All procedures performed in studies involving Ins )/(Glu - 3.5); InsAUC30/luAUC30 = (Ins 0 0 0 human participants were done in accordance ? Ins )/(Glu ? Glu ); InsAUC120/GluAUC120 30 0 30 with the ethical standards of the institutional =(Ins ? 4 9 Ins ? 3 9 Ins )/(Glu ? 4 9 0 30 120 0 and/or national research committee, as well as Glu ? 3 9 Glu ); DI30 = (InsAUC30/ 30 120 with the 1964 Helsinki Declaration and its later GluAUC30) 9 ISI ; and DI120 = (InsAUC120/ amendments or comparable ethical standards. GluAUC120) 9 ISI [30]. Informed consent was obtained from all par- ticipants included in the study. Biochemical Analyses Fasting serum HBA1c, ALT, AST, total choles- RESULTS terol (TC), high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol Baseline Characteristics (LDL), triglyceride (TG), uric acid, and crea- tinine levels were measured at baseline and after 16 weeks of intervention. Out of the 52 patients treated with metformin monotherapy, 3 were not willing to participate in the trial and 4 did not meet the inclusion Statistical Analysis criteria. In total, 45 patients were randomized into the two treatment groups: 22 received Statistical analyses were performed using the exenatide therapy and 23 received insulin glar- SPSS version 18.0 software (SPSS Inc., Chicago, gine therapy. Three patients in each group IL, USA) with a significance level of a = 0.05. A refused to undergo continuous glucose moni- previous study demonstrated that insulin glar- toring again after 16 weeks of intervention. gine treatment results in a MAGE change from Therefore, 39 patients completed the study baseline at week 36 of - 0.3 ± 1.3 [31], while (Fig. 1). The baseline characteristics of the exenatide used twice daily results in an patients in the two treatment groups, including approximate MAGE change from baseline at age and diabetes duration, were well matched week 16 of - 2.91 ± 3.15 [32]. Using SMBG data (Table 1). [33] from a head-to-head study between Diabetes Ther (2018) 9:1253–1267 1257 measured at 30 min (P = 0.009) using a stan- dard meal tolerance test was found to be sig- nificantly decreased with insulin glargine treatment (Table 2). The exenatide treatment group exhibited larger decreases in FBG (- 0.17 mmol/L; 95% CI - 1.64 to 1.30 mmol/ L) and HbA1c (- 0.16%; 95% CI - 0.67% to 0.35%) than the insulin glargine group, although there were no statistical differences between the two groups (Table 2). Following both treatments, increases in 1/HOMA-IR (exe- natide: from 0.35 ± 0.05 to 0.54 ± 0.08 lIU/ mL, mmol/L, P = 0.035; insulin glargine: from Fig. 1 Flow chart of study participants. Of the 52 0.28 ± 0.03 to 0.47 ± 0.06 lIU/mL, mmol/L, participants recruited for the present study, 3 were P = 0.017), ISI (exenatide: from 4.43 ± 0.71 to unwilling to participate and 4 did not meet the inclusion 7.13 ± 1.06 lIU/mL, mmol/L, P = 0.035), dis- criteria position index 30 (exenatide: from 41.94 ± 5.30 to 68.50 ± 11.06 mmol/L, Biomedical Parameters mmol/L, P = 0.008; insulin glargine: from 36.40 ± 4.40 to 60.39 ± 6.24 mmol/L, mmol/L, After intervention, body weight (D = - 3.51 kg, P = 0.001), and disposition index 120 (exe- P \ 0.001), BMI (D = - 1.22 kg/m , P \ 0.001), natide: from 61.31 ± 9.64 to and waist circumference (D = - 3.03 cm, 97.12 ± 17.30 mmol/L, mmol/L, P = 0.020; P = 0.012) significantly decreased in the exe- insulin glargine: from 52.71 ± 6.87 to natide treatment group but did not change in 88.33 ± 10.21 mmol/L, mmol/L, P = 0.006) the insulin glargine group. Larger reductions in were observed (Table 2); however, there were no body weight (- 3.96 kg; 95% CI -5.65 to statistical differences between the two groups. - 2.27 kg; P \ 0.001) and BMI (- 1.38 kg/m ; 95% CI - 1.99 to - 0.77 kg/m ; P \ 0.001) Continuous Glucose Monitoring occurred in the exenatide group compared with the insulin glargine group. Although TC Overall, blood glucose fluctuations significantly (P = 0.019) and LDL (P = 0.010) levels were improved after the exenatide and insulin glar- significantly reduced after exenatide interven- gine treatments (Fig. 2). There were similarly tion, and ALT (P = 0.027) levels were signifi- significant reductions in the MBG (exenatide cantly decreased after the insulin glargine D = - 1.25 ± 0.29 mmol/L, P \ 0.001; insulin treatment, there were no significant differences glargine D = - 1.83 ± 0.66 mmol/L, P = 0.007), between the two groups (Table 1). continuous overlapping net glycemic action (CONGA) (exenatide D = - 1.13 ± 0.30 mmol/ Glycemic Control and Insulin Function L, P \ 0.001; insulin glargine D = - 2.09 ± 0.70 mmol/L, P = 0.015), and MAGE (exenatide After 16 weeks of intervention, HbA1c was sig- D = - 1.20 ± 0.54 mmol/L, P = 0.001; insulin nificantly reduced by both exenatide treatment glargine D = - 1.47 ± 0.52 mmol/L, P = 0.001) and by insulin glargine treatment (exenatide values after the exenatide and insulin glargine 8.01 ± 0.21 vs 6.83 ± 0.25%, D = - 1.18%, interventions. There appeared to be smaller P \ 0.001; insulin glargine 8.35 ± 0.24 vs decreases in MBG (0.36 mmol/L; 95% CI - 1.00 7.14 ± 0.16%, D = - 1.21%, P \ 0.001; to 1.72 mmol/L), CONGA (0.45 mmol/L; 95% CI Table 2). Both exenatide and insulin glargine - 0.94 to 1.84 mmol/L), and MAGE (0.20 mmol/ significantly reduced FBG (P = 0.002, L; 95% CI - 1.00 to 1.41 mmol/L) in the exe- P = 0.007, respectively); the blood glucose level natide group when compared to the insulin 1258 Diabetes Ther (2018) 9:1253–1267 Table 1 Subject characteristics at baseline and after treatment Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin glargine Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) Mean (95% CI) Mean – SE Mean – SE Mean – SE Mean – SE Number (n)19 – – 20 – – – – – – Sex (male/female) 12/7 – – 10/10 – – – – – – Age 46.74 ± 2.31 – – 49.45 ± 2.17 – – – – – – Diabetic duration 6.37 ± 0.99 – – 4.35 ± 0.68 – – – – – – Weight (kg) 80.36 ± 2.35 76.85 ± 2.36 \ 0.001 74.75 ± 1.95 75.20 ± 1.98 0.263 - 3.51 0.45 - 3.96 \ 0.001 (- 5.09 to - 1.93) (- 0.37 to 1.27) (- 5.65 to - 2.27) BMI (kg/m ) 28.03 ± 0.50 26.81 ± 0.51 \ 0.001 27.08 ± 0.52 27.24 ± 0.55 0.258 - 1.22 0.17 - 1.38 \ 0.001 (- 1.78 to - 0.65) (- 0.13 to 0.47) (- 1.99 to - 0.77) Waist circumference 96.74 ± 1.50 93.71 ± 1.59 0.012 93.70 ± 1.52 93.30 ± 1.48 0.366 - 3.03 - 0.40 - 2.62 0.061 (cm) (- 5.30 to - 0.75) (- 1.30 to 0.50) (- 5.03 to - 0.22) ALT (U/L) 31.79 ± 4.33 31.56 ± 3.96 0.949 36.46 ± 4.82 28.83 ± 3.81 0.027 - 0.23 - 7.63 7.39 0.183 (- 7.75 to 7.28) (- 14.27 to - 0.99) (- 2.27 to 17.05) AST (U/L) 22.28 ± 2.13 21.97 ± 2.18 0.855 24.78 ± 2.15 22.39 ± 1.92 0.103 - 0.32 - 2.39 2.07 0.522 (- 3.92 to 3.29) (- 5.31 to 0.53) (- 2.37 to 6.51) TG (mmol/L) 1.75 ± 0.17 1.61 ± 0.23 0.286 1.70 ± 0.17 1.49 ± 0.19 0.295 - 0.15 - 0.21 0.07 0.744 (- 0.42 to 0.13) (- 0.63 to 0.20) (- 0.42 to 0.55) TC (mmol/L) 4.55 ± 0.31 4.04 ± 0.28 0.019 4.45 ± 0.23 4.30 ± 0.15 0.557 - 0.51 - 0.14 - 0.37 0.183 (- 0.93 to - 0.10) (- 0.64 to 0.35) (- 1.00 to 0.25) HDL (mmol/L) 1.04 ± 0.05 1.00 ± 0.06 0.324 1.03 ± 0.05 1.11 ± 0.07 0.147 - 0.04 0.09 - 0.12 0.083 (- 0.11 to 0.04) (- 0.03 to 0.21) (- 0.26 to 0.01) LDL (mmol/L) 2.42 ± 0.21 2.07 ± 0.20 0.010 2.36 ± 0.17 2.28 ± 0.11 0.617 - 0.35 - 0.09 - 0.26 0.153 (- 0.60 to - 0.10) (- 0.44 to 0.27) (- 0.69 to 0.16) P values of\ 0.05 were considered to indicate significant differences between the groups BMI body mass index, ALT alanine aminotransferase, AST aspartate aminotransferase, TG triglyceride, TC total cholesterol, LDL low-density lipoprotein cholesterol, HDL high-density lipoprotein cholesterol Diabetes Ther (2018) 9:1253–1267 1259 Table 2 Glucose control, insulin sensitivity, and b-cell function at baseline and after treatment Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) glargine Mean – SE Mean – SE Mean – SE Mean – SE Mean (95% CI) FBG (mmol/L) 8.65 ± 0.53 6.88 ± 0.40 0.002 8.83 ± 0.48 7.23 ± 0.39 0.007 - 1.77 - 1.60 - 0.17 0.580 (- 2.83 to - 0.72) (- 2.70 to - 0.50) (- 1.64 to 1.30) 30 min glucose 11.03 ± 0.74 9.06 ± 0.87 0.061 11.90 ± 0.68 10.08 ± 0.59 0.009 - 1.97 - 1.82 - 0.15 0.497 (mmol/L) (- 4.05 to 0.10) (- 3.13 to - 0.52) (- 2.48 to 2.19) 60 min glucose 14.27 ± 0.99 11.82 ± 1.03 0.080 14.70 ± 0.67 12.18 ± 0.66 0.091 - 2.45 - 2.52 0.08 0.844 (mmol/L) (- 5.23 to 0.33) (- 3.92 to - 1.12) (- 2.82 to 2.97) 120 min glucose 13.31 ± 0.88 11.71 ± 0.75 0.106 13.99 ± 0.90 12.77 ± 0.86 0.204 - 1.59 - 1.22 - 0.37 0.457 (mmol/L) (- 3.56 to 0.38) (- 3.17 to 0.74) (- 3.04 to 2.29) Fasting insulin 10.92 ± 1.85 10.03 ± 1.75 0.701 10.48 ± 1.02 9.62 ± 1.88 0.671 - 0.89 - 0.86 - 0.03 0.903 (uIU/mL) (- 5.71 to 3.93) (- 5.07 to 3.35) (- 6.22 to 6.15) 30 min insulin 19.69 ± 3.03 17.57 ± 3.96 0.523 18.47 ± 2.12 22.48 ± 4.83 0.388 - 2.12 4.01 - 6.13 0.298 (uIU/mL) (- 8.99 to 4.74) (- 5.57 to 13.59) (- 17.37 to 5.10) 60 min insulin 30.26 ± 5.48 33.86 ± 8.82 0.646 29.92 ± 3.53 33.99 ± 6.26 0.525 3.60 4.06 - 0.46 0.970 (uIU/mL) (- 12.64 to 19.84) (- 9.19 to 17.32) (- 20.78 to 19.85) 120 min insulin 35.65 ± 5.61 39.33 ± 8.87 0.549 38.38 ± 6.05 44.93 ± 8.61 0.436 3.67 6.54 - 2.87 0.756 (uIU/mL) (- 9.00 to 16.35) (- 10.81 to 23.90) (- 23.36 to 17.62) 1/HOMA-IR 0.35 ± 0.05 0.54 ± 0.08 0.035 0.28 ± 0.03 0.47 ± 0.06 0.017 0.19 0.19 0.00 0.718 (lIU/mL, (0.02 to 0.37) (0.04 to 0.33) (- 0.22 to 0.22) mmol/L) ISI (lIU/mL, 4.43 ± 0.71 7.13 ± 1.06 0.035 3.81 ± 0.54 5.29 ± 0.70 0.059 2.70 1.48 1.22 0.218 mmol/L) (0.22 to 5.19) (- 0.07 to 3.03) (- 1.62 to 4.06) HOMA-B 55.55 ± 11.55 120.79 ± 48.44 0.153 54.17 ± 10.90 84.09 ± 30.50 0.361 65.24 29.91 35.33 0.536 (IU/mol) (- 26.68 to 157.17) (- 37.74 to 97.57) (- 76.88 to 147.54) InsAUC30/ 12.09 ± 1.90 15.29 ± 3.32 0.181 11.20 ± 1.64 14.77 ± 3.20 0.237 3.20 3.58 - 0.37 0.898 GluAUC30 (- 1.65 to 8.05) (- 2.613 to 9.76) (- 7.86 to 7.11) (IU/mol) 1260 Diabetes Ther (2018) 9:1253–1267 Table 2 continued Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) glargine Mean – SE Mean – SE Mean – SE Mean – SE Mean (95% CI) InsAUC120/ 17.39 ± 3.17 22.03 ± 5.78 0.158 16.43 ± 2.73 21.94 ± 4.63 0.227 4.64 5.51 - 0.87 0.836 GluAUC120 (- 2.02 to 11.29) (- 3.88 to 14.90) (- 11.64 to 9.89) (IU/mol) DI 30 (mmol/L, 41.94 ± 5.30 68.50 ± 11.06 0.008 36.40 ± 4.40 60.39 ± 6.24 0.001 26.57 23.99 2.57 0.872 mmol/L) (8.20 to 44.93) (12.71 to 35.27) (- 18.36 to 23.51) DI 120 (mmol/L, 61.31 ± 9.64 97.12 ± 17.30 0.020 52.71 ± 6.87 88.33 ± 10.21 0.006 35.81 35.62 0.19 0.949 mmol/L) (6.65 to 64.97) (12.24 to 59.00) (- 35.79 to 36.17) HbA1c (%) 8.01 ± 0.21 6.83 ± 0.25 \ 0.001 8.35 ± 0.24 7.14 ± 0.16 \ 0.001 - 1.21 - 1.05 - 0.16 0.537 (- 1.57 to - 0.85) (- 1.41 to - 0.69) (- 0.67 to 0.35) InsAUC30/GluAUC30 was calculated as the total insulin area under the curve divided by the total glucose area under the curve during the first 30 min of the 75-g oral glucose tolerance test (OGTT); InsAUC120/GluAUC120 was calculated as the total insulin area under the curve divided by the total glucose area under the curve during the 120 min of the OGTT P values of \ 0.05 were considered to indicate significant differences between the groups FBG fasting blood glucose, 1/HOMA-IR 1/homeostasis model assessment of insulin resistance, ISIM Matsuda insulin sensitivity index, HOMA-b basal homeostasis model assessment of insulin secretion, DI30 disposition index 30, DI120 disposition index 120, HbA1c hemoglobin A1c Diabetes Ther (2018) 9:1253–1267 1261 Fig. 2 Mean interstitial glucose values at baseline and after treatment with exenatide (a) and insulin glargine (b) glargine group, although there were no statisti- DISCUSSION cally significant differences. After exenatide intervention, there were significant reductions This study compared the effects of treatment in the standard deviation of the mean glucose with the short-term GLP-1 receptor agonist (D = 0.70 ± 0.24 mmol/L, P = 0.001) and largest exenatide and insulin glargine treatment on GV amplitude of glycemic excursions in overweight and obese patients with T2DM (D = - 2.32 ± 1.06 mmol/L, P = 0.004), while and poor glycemic control who were receiving no changes were observed after the insulin glar- metformin monotherapy. Importantly, our gine intervention (Table 3). results using a CGMS revealed that similar Both interventions significantly decreased improvements in GV, HBA1c, and b-cell func- the percentage of the time that TBG [ tion were achieved with exenatide treatment 10.0 mmol/L [exenatide D = - 14.79 ± 4.97%, and insulin glargine treatment, although P = 0.008; insulin glargine D = - 25.11 ± greater body weight and BMI reductions were 10.16%, P = 0.024 (Fig. 3b)] and the highest achieved with the exenatide treatment during blood glucose level [exenatide D = - 2.70 ± the 16-week intervention. After treatment, waist 0.72 mmol/L, P = 0.001; insulin glargine circumference was significantly decreased in the D = - 2.49 ± 1.16 mmol/L, P = 0.045 (Fig. 3e)]. exenatide group, but there was no statistically The percentage of the time that TBG was significant difference between the two groups. [ 7.8 mmol/L was significantly decreased in the Recent studies have suggested that glycemic insulin glargine treatment group (P = 0.030) fluctuation is a useful indicator of the effec- but not in the exenatide group (Fig. 3a). How- tiveness of blood glucose control and the pre- ever, there were no statistically significant dif- vention of diabetes complications [1, 2, 10]. ferences between the two groups. Less frequent monitoring of blood glucose levels using SMBG may lead to inaccurate GV assessment and could falsely suggest a lack of or Correlative Analysis a decrease in glycemic fluctuation [3, 5, 6]. A CGMS automatically records blood glucose levels every 5 min, achieving up to 288 mea- For all subjects, there were significant positive surements per day, which provide a better correlations between DHbA1c and the reduc- measure of the extent of glycemic fluctuation tions in MBG, CONGA, and Max BG (0 \ r \ 1; and overall trends [7, 8]. Therefore, we used a P \ 0.05; Table 4). CGMS in this study to monitor blood glucose. 1262 Diabetes Ther (2018) 9:1253–1267 Table 3 Glycemic variability at baseline and after treatment Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin glargine Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) Mean (95% CI) Mean – SE Mean – SE Mean – SE Mean – SE MBG (mmol/L) 9.77 ± 0.40 8.32 ± 0.44 \ 0.001 10.85 ± 0.54 9.04 ± 0.54 0.007 - 1.45 - 1.82 0.36 0.824 (- 2.06 to - 0.85) (- 3.01 to - 0.57) (- 1.00 to 1.72) SDBG (mmol/L) 1.87 ± 0.15 1.19 ± 0.13 0.001 2.02 ± 0.14 1.71 ± 0.16 0.187 - 0.68 - 0.31 - 0.37 0.022 (- 1.06 to - 0.31) (- 0.79 to 0.17) (- 0.96 to 0.21) MODD (mmol/L) 1.75 ± 0.21 1.52 ± 0.17 0.215 2.01 ± 0.18 1.82 ± 0.13 0.248 - 0.24 - 0.19 - 0.05 0.294 (- 0.63 to 0.16) (- 0.53 to 0.15) (- 0.53 to 0.44) MAGE (mmol/L) 4.51 ± 0.32 2.94 ± 0.33 0.001 5.62 ± 0.46 3.84 ± 0.39 0.001 - 1.57 - 1.78 0.20 0.322 (- 2.42 to - 0.73) (- 2.70 to - 0.86) (- 1.00 to 1.41) LAGE (mmol/L) 7.82 ± 0.59 5.18 ± 0.53 0.004 8.60 ± 0.63 6.97 ± 0.87 0.166 - 2.64 - 1.63 - 1.00 0.083 (- 4.29 to - 0.99) (- 4.01 to 0.75) (- 3.77 to 1.77) CONGA (mmol/L) 8.82 ± 0.39 7.53 ± 0.40 \ 0.001 9.55 ± 0.49 7.80 ± 0.47 0.015 - 1.30 - 1.75 0.45 0.973 (- 1.91 to - 0.68) (- 3.12 to - 0.38) (- 0.94 to 1.84) P values of\ 0.05 were considered to indicate statistically significant differences between the groups MBG mean blood glucose level, SDBG standard deviation of the mean glucose value, MODD absolute mean of daily differences, MAGE mean amplitude of glycemic excursions, LAGE largest amplitude of glycemic excursions, CONGA continuous overlapping net glycemic action Diabetes Ther (2018) 9:1253–1267 1263 Fig. 3 Effects of exenatide treatment and glargine treatment on the percentage of the time that the blood glucose value (TBG) was[7.8 mmol/L (a),[10.0 mmol/L (b), or[11.1 mmol/L (c) as well as the lowest (d) and the highest (e) blood glucose (BG) values. *P \ 0.05, 16 weeks vs baseline patients with suboptimally controlled diabetes Table 4 Correlations between glycemic variability and using metformin or a sulfonylurea [18, 19]. glycated hemoglobin However, the average daily risk range as well as DHBA1c the low and high blood glucose indices, rP indicative of the fluctuation in blood glucose levels, were based on SMBG and not on the DMBG (mmol/L) 0.498 0.001 more accurate CGMS [18, 19]. We found that DCONGA (mmol/L) 0.512 0.002 FBG and HbA1c were significantly reduced after exenatide and insulin glargine intervention. DMax BG (mmol/L) 0.439 0.006 Compared with baseline levels, glycemic fluc- P \ 0.05 was considered to indicate statistical significance tuation results calculated by the CGMS indi- MBG mean blood glucose level, CONGA continuous cated that both exenatide treatment and insulin overlapping net glycemic action, HbA1c hemoglobin A1c glargine treatment significantly decreased MBG, CONGA, MAGE, and the highest blood glucose value, as well as the percentage of time that the blood glucose was [ 10.0 mmol/L through the Glycemic fluctuation following the use of a day. Following the standard meal tolerance test, GLP-1 receptor agonist was compared with the insulin glargine significantly decreased blood GV observed after the introduction of basal glucose levels at 30 min; glucose levels at 60 and insulin. Results showed that although the exe- 120 min showed a downward trend after exe- natide and glargine treatments were both asso- natide intervention and insulin glargine inter- ciated with similar significant improvements in vention. The decrease in MAGE with insulin HbA1c from baseline, exenatide treatment glargine treatment might due to improved resulted in greater improvements in GV than b-cell function and insulin resistance associated insulin glargine treatment for previously with improved glycemic control. uncontrolled glucose levels when combined Other long-term GLP-1 receptor agonists with metformin and sulfonylurea therapy. that only require once-daily or once a week Improvements in GV were also observed in 1264 Diabetes Ther (2018) 9:1253–1267 injection have become widely used or are insulin glargine facilitate glycemic control and beginning to find wider use in China. The long- reduce GV, thus aiding diabetes management. term GLP-1 receptor agonists liraglutide and The main AEs reported in the exenatide- semaglutide have shown beneficial effects on treated group were gastrointestinal intolerance/ CVD and mortality-related CVD in T2DM appetite suppression (17 subjects), nausea patients, although there would appear to be no (which relieved over time; 3 subjects), and difference in the effects of once-weekly exe- abdominal distension (4 subjects). No patients natide, the other long-term GLP-1 receptor in the group treated with insulin glargine suf- agonist, or lixisenatide, a short-term GLP-1 fered gastrointestinal intolerance, hypo- receptor agonist [34–36]. HbA1c and FBG glycemia, or other AEs. No hypoglycemia events showed greater reductions with liraglutide occurred in these patients, which may be treatment rather than exenatide treatment because prandial insulin and drugs stimulating [23, 24], and significantly greater improvements insulin secretion were not used. No participants in HbA1c were seen with liraglutide and withdrew from the study due to these AEs, and semaglutide than with insulin glargine [20–22]. no SAEs occurred in the present study. Whether the beneficial effects of liraglutide and There are a number of limitations of our semaglutide for CVD were the result of greater study. Only 39 patients were enrolled, and this decreases in GV requires further study. small sample size may have had a negative Regarding insulin resistance and b-cell func- impact on the statistical analysis. It might affect tion, the 1/HOMA-IR, disposition index 30, and the comparison of waist circumference between disposition index 120 improved with either the two intervention groups, for greater reduc- exenatide treatment or insulin glargine treat- tion of waist circumference in exenatide group ment. Only the ISI improved to a greater might be confirmed after enlarging the sample extent with exenatide treatment than with size. The results of the present study could be insulin glargine treatment. These results indi- further confirmed by expanding the sample cate that b-cell function and insulin resistance size, using a blinded design, testing long-term improved following a 16-week course of exe- GLP-1 receptor analogists, and potentially natide or insulin glargine treatment. developing a multicenter research study. In Overweight and obese patients in the coop- addition, the study period was 16 weeks, which erative meta-analysis group of the China Obe- may not have been sufficient to assess addi- sity Task Force were enrolled in our study [25]. tional benefits relating to the GV. Since a fixed The results showed that exenatide treatment dosage of exenatide was given in the exenatide resulted in weight loss while insulin glargine group after 4 weeks and the dosage of the treatment did not. Standardized weight is an insulin glargine was only titrated during the important part of integrated T2DM manage- first four weeks, rising FBG levels in some ment because weight loss is associated with insulin-glargine-treated patients during the fol- improvements in b-cell function and insulin lowing 12 weeks may have affected glucose action, it makes long-term glycemic control control in the insulin glargine group. management easier, and it may favorably affect other comorbid diseases such as dyslipidemia, CONCLUSIONS hypertension, and hyperuricemia [15]. An increasing waist circumference contributes to Our study showed that comparable improve- central obesity and is an independent risk factor ments in GV, HBA1c, and b-cell function can be for T2DM, dyslipidemia, hypertension, and achieved in overweight and obese T2DM CVD [26, 37]. In the present study, waist cir- patients with inadequate glycemic control who cumference only decreased in the exenatide- were receiving metformin monotherapy by treated group. combining that treatment with either exenatide An important finding from our study was a or insulin glargine. Compared to insulin glar- positive association between GV and HbA1c gine, exenatide had a superior impact on body level, which may indicate that exenatide and Diabetes Ther (2018) 9:1253–1267 1265 weight and BMI. Waist circumference was sig- Compliance with Ethics Guidelines. All nificantly reduced with exenatide treatment, procedures performed in studies involving but there was no statistically significant differ- human participants were in accordance with ence from the insulin glargine treatment. the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed ACKNOWLEDGEMENTS consent was obtained from all participants included in the study. We thank the participants of the study. Data Availability. The datasets analyzed for Funding. This work was supported by grants the current analysis are available from the cor- from the Investigator Initiated-Sponsored Clin- responding author on reasonable request. ical Research from AstraZeneca Investment (China) Co., Ltd. (ISSEXEN0034), the National Open Access. This article is distributed Natural Science Foundation of China under the terms of the Creative Commons (81570736, 81570737), Medical and Health Attribution-NonCommercial 4.0 International Research Projects of Nanjing Health Bureau in License (http://creativecommons.org/licenses/ Jiangsu Province of China (YKK14055), Project by-nc/4.0/), which permits any non- of Standardized Diagnosis and Treatment of Key commercial use, distribution, and reproduction Diseases in Jiangsu Province of China in any medium, provided you give appropriate (2015604), China Diabetes Young Scientific credit to the original author(s) and the source, Talent Research Project (2017-N-05), and Nan- provide a link to the Creative Commons license, jing University Central University Basic Scien- and indicate if changes were made. tific Research (14380296). Drum Tower Hospital (affiliated with Nanjing University Medical School, China) provided sponsorship for article processing charges. Our study sponsor did not REFERENCES fund the journal’s article processing charges. All authors had full access to all of the data in this 1. Siegelaar SE, Holleman F, Hoekstra JB, DeVries JH. study and take complete responsibility for the Glucose variability; does it matter? Endocr Rev. 2010;31(2):171–82. integrity of the data and accuracy of the data analysis. 2. Jin YP, Su XF, Yin GP, et al. Blood glucose fluctua- tions in hemodialysis patients with end stage dia- Editorial Assistance. Editage, a brand of betic nephropathy. J Diabetes Complicat. 2015;29(3):395–9. Cactus Communications, offered professional English language editing. 3. Lee MK, Lee KH, Yoo SH, Park CY. Impact of initial active engagement in self-monitoring with a tele- Authorship. All the named authors meet the monitoring device on glycemic control among International Committee of Medical Journal patients with type 2 diabetes. Sci Rep. 2017;7(1):3866. Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of 4. Tildesley HD, Mazanderani AB, Ross SA. Effect of the work as a whole, and have given their Internet therapeutic intervention on A1C levels in approval for this version to be published. patients with type 2 diabetes treated with insulin. Diabetes Care. 2010;33(8):1738–40. Disclosures. Ting-Ting Yin, Yan Bi, Ping Li, 5. Clar C, Barnard K, Cummins E, Royle P, Waugh N, Shan-Mei Shen, Xiao-Lu Xiong, Li-Jun Gao, Can Aberdeen Health Technology Assessment Group. Jiang, Yan Wang, Wen-Huan Feng, and Da- Self-monitoring of blood glucose in type 2 diabetes: Long Zhu have nothing to disclose. systematic review. Health Technol Assess. 2010;14(12):1–140. 1266 Diabetes Ther (2018) 9:1253–1267 6. Malanda UL, Bot SD, Nijpels G. Self-monitoring of 17. Buse JB, Bergenstal RM, Glass LC, et al. Use of twice- blood glucose in noninsulin-using type 2 diabetic daily exenatide in basal insulin-treated patients patients: it is time to face the evidence. Diabetes with type 2 diabetes: a randomized, controlled trial. Care. 2013;36(1):176–8. Ann Intern Med. 2011;154(2):103–12. 7. Levy JC, Davies MJ, Holman RR, Group TS. Con- 18. Barnett AH, Burger J, Johns D, et al. Tolerability and tinuous glucose monitoring detected hypogly- efficacy of exenatide and titrated insulin glargine in caemia in the Treating to Target in Type 2 Diabetes adult patients with type 2 diabetes previously Trial (4-T). Diabetes Res Clin Pract. 2017;131:161–8. uncontrolled with metformin or a sulfonylurea: a multinational, randomized, open-label, two-period, 8. Wei Q, Sun Z, Yang Y, Yu H, Ding H, Wang S. Effect crossover noninferiority trial. Clin Ther. of a CGMS and SMBG on maternal and neonatal 2007;29(11):2333–48. outcomes in gestational diabetes mellitus: a ran- domized controlled trial. Sci Rep. 2016;6:19920. 19. McCall AL, Cox DJ, Brodows R, Crean J, Johns D, Kovatchev B. Reduced daily risk of glycemic vari- 9. Brownlee M. The pathobiology of diabetic compli- ability: comparison of exenatide with insulin glar- gine. Diabetes Technol Therap. 2009;11(6):339–44. cations: a unifying mechanism. Diabetes. 2005;54(6):1615–25. 20. Aroda VR, Bain SC, Cariou B, et al. Efficacy and 10. Di Flaviani A, Picconi F, Di Stefano P, et al. Impact safety of once-weekly semaglutide versus once-daily of glycemic and blood pressure variability on sur- insulin glargine as add-on to metformin (with or rogate measures of cardiovascular outcomes in type without sulfonylureas) in insulin-naive patients 2 diabetic patients. Diabetes Care. 2011;34(7): with type 2 diabetes (SUSTAIN 4): a randomised, 1605–9. open-label, parallel-group, multicentre, multina- tional, phase 3a trial. Lancet Diabetes Endocrinol. 11. Murata GH, Duckworth WC, Shah JH, Wendel CS, 2017;5(5):355–66. Hoffman RM. Sources of glucose variability in insulin-treated type 2 diabetes: the Diabetes Out- 21. Russell-Jones D, Vaag A, Schmitz O, et al. Liraglu- comes in Veterans Study (DOVES). Clin Endocrinol. tide vs insulin glargine and placebo in combination 2004;60(4):451–6. with metformin and sulfonylurea therapy in type 2 diabetes mellitus (LEAD-5 met ? SU): a randomised 12. Figueira FR, Umpierre D, Casali KR, et al. Aerobic controlled trial. Diabetologia. 2009;52(10): and combined exercise sessions reduce glucose 2046–55. variability in type 2 diabetes: crossover randomized trial. PLoS One. 2013;8(3):e57733. 22. Diamant M, Van Gaal L, Stranks S, et al. Once weekly exenatide compared with insulin glargine 13. Cohen O, Korner A, Chlup R, et al. Improved gly- titrated to target in patients with type 2 diabetes cemic control through continuous glucose sensor- (DURATION-3): an open-label randomised trial. augmented insulin pump therapy: prospective Lancet. 2010;375(9733):2234–43. results from a community and academic practice patient registry. J Diabetes Sci Technol. 2009;3(4): 23. Twigg SM, Daja MM, O’Leary BA, Adena MA. Once- 804–11. daily liraglutide (1.2 mg) compared with twice- daily exenatide (10 mug) in the treatment of type 2 14. Peterson K, Zapletalova J, Kudlova P, et al. Benefits diabetes patients: an indirect treatment comparison of three-month continuous glucose monitoring for meta-analysis. J Diabetes. 2016;8(6):866–76. persons with diabetes using insulin pumps and sensors. Biomed Pap Med Fac Univ Palacky Olo- 24. Meier JJ, Rosenstock J, Hincelin-Mery A, et al. mouc Czechoslovakia. 2009;153(1):47–51. Contrasting effects of lixisenatide and liraglutide on postprandial glycemic control, gastric emptying, 15. Garber AJ, Abrahamson MJ, Barzilay JI, et al. Con- and safety parameters in patients with type 2 dia- sensus Statement by the American Association of betes on optimized insulin glargine with or without Clinical Endocrinologists and American College of metformin: a randomized, open-label trial. Diabetes Endocrinology on the Comprehensive Type 2 Dia- Care. 2015;38(7):1263–73. betes Management Algorithm—2017 Executive Summary. Endocr Pract. 2017;23(2):207–38. 25. Zhou BF, Cooperative Meta-Analysis Group of the Working Group on Obesity in China. Predictive values of body mass index and waist circumference 16. Eng C, Kramer CK, Zinman B, Retnakaran R. Glu- cagon-like peptide-1 receptor agonist and basal for risk factors of certain related diseases in Chinese insulin combination treatment for the manage- adults—study on optimal cut-off points of body ment of type 2 diabetes: a systematic review and mass index and waist circumference in Chinese meta-analysis. Lancet. 2014;384(9961):2228–34. adults. Biomed Environ Sci. 2002;15(1):83–96. Diabetes Ther (2018) 9:1253–1267 1267 26. Murphy KG. Editorial overview: endocrine and and beta-cell function in early type 2 diabetes. Acta metabolic diseases: waistline weapons: new thera- Diabetol. 2013;50(4):587–95. peutic avenues for the treatment of obesity and metabolic disease. Curr Opin Pharmacol. 32. Irace C, Fiorentino R, Carallo C, Scavelli F, Gnasso 2015;25:iv–vi. A. Exenatide improves glycemic variability assessed by continuous glucose monitoring in subjects with 27. Li C, Yang H, Tong G, et al. Correlations between type 2 diabetes. Diabetes Technol Therap. A1c, fasting glucose, 2 h postload glucose, and beta- 2011;13(12):1261–3. cell function in the Chinese population. Acta Dia- betol. 2014;51(4):601–8. 33. Heine RJ, Van Gaal LF, Johns D, et al. Exenatide versus insulin glargine in patients with subopti- 28. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, mally controlled type 2 diabetes: a randomized Treacher DF, Turner RC. Homeostasis model trial. Ann Intern Med. 2005;143(8):559–69. assessment: insulin resistance and beta-cell func- tion from fasting plasma glucose and insulin con- 34. Marso SP, Daniels GH, Brown-Frandsen K, et al. centrations in man. Diabetologia. 1985;28(7): Liraglutide and cardiovascular outcomes in type 2 412–9. diabetes. N Engl J Med. 2016;375(4):311–22. 29. Matsuda M, DeFronzo RA. Insulin sensitivity indi- 35. Holman RR, Bethel MA, Mentz RJ, et al. Effects of ces obtained from oral glucose tolerance testing: once-weekly exenatide on cardiovascular outcomes comparison with the euglycemic insulin clamp. in type 2 diabetes. N Engl J Med. 2017;377(13): Diabetes Care. 1999;22(9):1462–70. 1228–39. 30. Stancakova A, Javorsky M, Kuulasmaa T, Haffner 36. Pfeffer MA, Claggett B, Diaz R, et al. Lixisenatide in SM, Kuusisto J, Laakso M. Changes in insulin sen- patients with type 2 diabetes and acute coronary sitivity and insulin release in relation to glycemia syndrome. N Engl J Med. 2015;373(23):2247–57. and glucose tolerance in 6,414 Finnish men. Dia- betes. 2009;58(5):1212–21. 37. Cui R, Qi Z, Zhou L, Li Z, Li Q, Zhang J. Evaluation of serum lipid profile, body mass index, and waist- 31. Pistrosch F, Kohler C, Schaper F, Landgraf W, Forst line in Chinese patients with type 2 diabetes mel- T, Hanefeld M. Effects of insulin glargine versus litus. Clin Interv Aging. 2016;11:445–52. metformin on glycemic variability, microvascular http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diabetes Therapy Springer Journals

Comparison of Glycemic Variability in Chinese T2DM Patients Treated with Exenatide or Insulin Glargine: A Randomized Controlled Trial

Free
15 pages
Loading next page...
 
/lp/springer_journal/comparison-of-glycemic-variability-in-chinese-t2dm-patients-treated-7y22ZrMC1i
Publisher
Springer Healthcare
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Internal Medicine; Diabetes; Cardiology; Endocrinology
ISSN
1869-6953
eISSN
1869-6961
D.O.I.
10.1007/s13300-018-0412-6
Publisher site
See Article on Publisher Site

Abstract

Diabetes Ther (2018) 9:1253–1267 https://doi.org/10.1007/s13300-018-0412-6 ORIGINAL RESEARCH Comparison of Glycemic Variability in Chinese T2DM Patients Treated with Exenatide or Insulin Glargine: A Randomized Controlled Trial . . . . . Ting-Ting Yin Yan Bi Ping Li Shan-Mei Shen Xiao-Lu Xiong . . . . Li-Jun Gao Can Jiang Yan Wang Wen-Huan Feng Da-Long Zhu Received: February 13, 2018 / Published online: May 9, 2018 The Author(s) 2018 (N = 39) to either exenatide treatment or insu- ABSTRACT lin glargine treatment for 16 weeks. Glycemic variability was assessed using a CGMS; hemo- Introduction: Increasing the frequency of globin A1c (HbA1c), b-cell function, weight, blood glucose monitoring aids the evaluation of body mass index (BMI), and waist circumfer- glycemic variability and blood glucose control ence were also evaluated. by antidiabetic drugs. It remains unclear, how- Results: Mean blood glucose level, continuous ever, whether GLP-1 receptor agonists or basal overlapping net glycemic action, mean ampli- insulin has a better effect on glycemic variabil- tude of glycemic excursions, percentage of the ity in type 2 diabetes mellitus (T2DM) patients time that the blood glucose value who are inadequately controlled by metformin. was [ 10.0 mmol/L, and highest blood glucose We used a continuous glucose monitoring sys- level (P \ 0.01–0.05) significantly decreased in tem (CGMS) to compare patients on a GLP-1 both groups. Standard deviation of the mean receptor agonist with patients on basal insulin glucose value, largest amplitude of glycemic in terms of glycemic variability. excursions, and waist circumference signifi- Methods: This prospective randomized study cantly decreased for those treated with exe- assigned T2DM patients treated with metformin natide (P \ 0.05), while no changes were Enhanced content To view enhanced content for this observed with insulin glargine treatment. Per- article go to https://doi.org/10.6084/m9.figshare. centage of the time that the blood glucose value was [ 7.8 mmol/L decreased after insulin glargine use (P \ 0.05) but not with the exe- T.-T. Yin  L.-J. Gao  W.-H. Feng (&) natide intervention. Similar decreases in fasting Department of Endocrinology, Drum Tower Clinical Hospital, Medical School of Southeast University, blood glucose and HbA1c and increases in the Nanjing, China 1/homeostasis model assessment of insulin e-mail: fengwh501@163.com resistance, disposition index 30, and disposition index 120 were observed in both groups T.-T. Yin  Y. Bi  P. Li  S.-M. Shen  X.-L. Xiong L.-J. Gao  Y. Wang  W.-H. Feng  D.-L. Zhu (&) (P \ 0.01–0.05). Reductions in weight and BMI Department of Endocrinology, Drum Tower were greater with exenatide than with insulin Hospital Affiliated to Nanjing University Medical glargine treatment (P \ 0.05). School, Nanjing, China Conclusions: In overweight and obese patients e-mail: zhudalong@nju.edu.cn with T2DM inadequately controlled by met- C. Jiang formin, exenatide and insulin glargine have Department of Endocrinology, Jining No 1. People’s similar efficacies in terms of glycemic Hospital, Shandong, China 1254 Diabetes Ther (2018) 9:1253–1267 variability, HbA1c alleviation, and b-cell func- and a sulfonylurea found that the addition of tion, but exenatide has a greater effect on body exenatide minimized GV to a greater extent weight and BMI. than the addition of insulin glargine, based on SMBG results [18, 19]. A previous study and meta-analysis found Keywords: Basal insulin; Glucagon-like that liraglutide (1.8 and 1.2 mg), a long-term peptide-1 receptor agonist; Glucose GLP-1 receptor agonist, provided greater fluctuation; Type 2 diabetic mellitus reductions in HbA1c and fasting plasma glucose (FBG) than exenatide (10 lg b.i.d.), a short-term GLP-1 receptor agonist. Compared with insulin INTRODUCTION glargine, adding liraglutide or other long-term GLP-1 receptor agonists to metformin alone or For patients with type 2 diabetes mellitus (T2DM), increasing the frequency of blood with a sulfonylurea decreased body weight and led to significantly improved or at least nonin- glucose monitoring is beneficial for maintain- ing physiological blood glucose levels [1, 2]. ferior glycemic control [20–22]. In contrast to Western countries, exenatide is a commonly Self-monitoring of blood glucose (SMBG) and and widely used short-term GLP-1 receptor hemoglobin A1c (HBA1c) are the most com- monly used methods to assess short-term and agonist in China [23, 24]. It is unclear from the use of a CGMS, which is a reliable technology long-term blood glucose control, respectively [3–5]. However, these measurements are inade- for monitoring glycemic control, whether exe- natide or basal insulin has a better effect on GV quate for real-time blood glucose monitoring. A continuous glucose monitoring system (CGMS) for patients with T2DM and inadequate gly- cemic control who receive metformin. In the conveniently and accurately records real-time glycemic values and trends over multiple days work described in the present paper, we used a CGMS to compare the effects of exenatide by providing a large number of blood glucose recordings [2, 5–7]. A CGMS provides a detailed treatment with those of insulin glargine treat- depiction of glycemic variability (GV) and effi- ment on the GVs of overweight and obese T2DM patients with poor glycemic control ciently analyzes rates of change in glucose levels [7, 8]. Wide fluctuations in blood glucose levels using metformin. may lead to excessive glycosylation and oxida- tive stress and are associated with reduced METHODS endothelial function in patients with T2DM; these are key factors in the development of Subjects diabetic complications [9, 10]. Patients with glycemic lability often have long-standing dia- Thirty-nine T2DM patients with poor glycemic betes with both beta-cell function and counter- control after receiving metformin monotherapy regulatory hormone failure [11]. GV has were recruited from Drum Tower Hospital (af- recently been evaluated as a potential target for filiated with Nanjing University Medical School, diabetic complication interventions, and the China) and randomized to receive exenatide or GV record provided by a CGMS may improve insulin glargine treatment for 16 weeks. Ran- glycemic control [12–14]. dom numbers were provided by third-party Metformin, an oral antidiabetic drug, is the statisticians and block randomization was used. first-line treatment for T2DM. In patients with Inclusion criteria were as follows: (1) a T2DM and poorly controlled blood glucose stable metformin dose C 1.5 g/day over at least receiving metformin alone, glucagon-like pep- 8 weeks; (2) body mass index (BMI) C 24 kg/m ; tide-1 (GLP-1) receptor agonists and basal (3) HbA1c level between 7.0% and 10.0%; and insulin are used as optional antidiabetic drugs (4) age between 18 and 70 years. Exclusion cri- [15–17]. Studies involving patients with T2DM teria were as follows: (1) renal dysfunction, and poor glycemic control receiving metformin, defined as a serum creatinine level C 1.5 mg/dL; a sulfonylurea, or a combination of metformin Diabetes Ther (2018) 9:1253–1267 1255 (2) diseases causing acute or chronic hypoxia; peripheral blood glucose levels reached (3) liver dysfunction, defined as an alanine 6.1 mmol/L; the insulin glargine dose was not aminotransferase (ALT) or aspartate amino- adjusted over the final 12 weeks. GV was mea- transferase (AST) level C 3 times higher than sured at baseline and during the last week. This the upper limit of normal; (4) history of car- study was registered with Clinical trials.gov (ID: diovascular disease (CVD) during the preceding NCT02325960). year; (5) proliferative retinopathy; (6) positive pregnancy test result, breast-feeding, or refusal Measurements to use appropriate contraceptive methods; (7) systemic corticosteroid therapy during the pre- Glycemic Variability ceding 2 months; (8) type 1 diabetes; and (9) use GV was assessed at baseline and at the end of of other experimental drugs over the preceding the study using a CGMS (Glod, Medtronic) for 1 month. All patients included in the study up to 72 h. The CGMS sensor was inserted into provided written informed consent. The study subcutaneous abdominal fat tissue, and the protocol was approved by the Research Ethics average electrical signal was recorded every Board of Drum Tower Hospital, which is affili- 5 min, yielding 288 glucose level measurements ated with Nanjing University Medical School per day and 864 data points for 3 consecutive (Protocol: AF/SQ-2014-072-01). The present days. According to the CGMS results, the fol- study was monitored by the Drug Clinical Trial lowing key parameters were calculated: (1) Agency Office of Drum Tower Hospital, which is mean blood glucose (MBG) level, which was affiliated with Nanjing University Medical estimated as the average value and standard School. Adverse events (AEs) were collected and deviation (SD) of the 288 data points recorded recorded in case report form, and serious during the 24 h of continuous glucose moni- adverse events (SAEs) were reported in written toring; (2) mean amplitude of glycemic excur- form to the Institutional Review Board of the sion (MAGE), determined by calculating the Drug Clinical Trial Agency Office and the arithmetic mean of the difference between Research Ethics Board of Drum Tower Hospital. consecutive peaks and nadirs if the difference was [ 1 SD of mean glucose; (3) mean absolute Endpoint value of daily differences, calculated as the mean absolute deviation of matched values The primary endpoint of this study was the measured during two consecutive 24-h periods change in GV during a 16-week follow-up per- of continuous glucose monitoring [25]; (4) the iod. Secondary endpoints included changes in difference between the highest and lowest levels the following factors at 16 weeks: blood glucose, of blood glucose, calculated as the difference HbA1c, weight, BMI, waist circumference, between the highest and lowest blood glucose insulin function, liver function, and lipid values during 24 h of continuous glucose profile. monitoring; (5) TBG [ 7.8 mmol/L and TBG [ 11.1 mmol/L, calculated as the percent- age of the time that the blood glucose value Protocol (TBG) was [ 7.8 and [ 11.1 mmol/L, respec- tively, during 24 h; and (6) TBG \ 3.9 mmol/L, This prospective, randomized, and parallel-de- calculated as the percentage of the time that the sign trial lasted 16 weeks. For the exenatide- blood glucose value was \ 3.9 mmol/L during treated patients, the initial dosage was 5 lg 24 h [26]. twice daily for 4 weeks, followed by a mainte- nance dose of 10 lg twice daily throughout the Standard Meal Tolerance Test trial. For patients receiving insulin glargine, the All patients enrolled in our study underwent a initial dose was 8 IU once daily, followed by a standard meal tolerance test at baseline and titrated dosage of C 2 IU every 3 days based on after 16 weeks of intervention. Plasma glucose FBG levels during the first 4 weeks until the 1256 Diabetes Ther (2018) 9:1253–1267 and insulin levels were measured 0, 30, 60, and exenatide and glargine, we can conservatively 120 min after meal ingestion. The 1/homeosta- estimate that the MAGE change from baseline sis model assessment of insulin resistance (1/ in the exenatide group is approximate HOMA-IR) index and the Matsuda insulin sen- 2.55 ± 3.15, so the sample size per group nee- sitivity index (ISI ) were used to measure insu- ded to provide 80% power was calculated. lin sensitivity. The basal homeostasis model Assuming a 10% subject drop-out rate, a total of assessment of insulin secretion (HOMA-b), 44 patients (22 per group) was targeted for early-phase total insulin area under the curve randomization. Data are presented as the divided by the total glucose area under the mean ± standard error (SE). The paired Stu- curve during the first 30 min of the standard dent’s t test was used to analyze the pre- and meal tolerance test (InsAUC30/GluAUC30), and post-intervention differences within each the total-phase total insulin area under the group. Analysis of covariance (ANCOVA) was curve divided by the total glucose area under used to test for differences between the inter- the curve during the 120 min of the oral glucose vention groups after adjusting for baseline. tolerance test (InsAUC120/GluAUC120) were Pearson’s correlation analysis was conducted to used to calculate insulin release. The disposition determine the association between the variables index was used to express b-cell function and GV and HBA1c. P values of \ 0.05 were [27–29]. Values were calculated using the fol- considered significant. lowing formulae: 1/HOMA-IR = 1/(Ins 9 Glu / 0 0 22.5); ISI = 10,000/(Glu 9 Ins 9 average M 0 0 Compliance with Ethics Guidelines glucose 75-g oral glucose tolerance test [OGTT] 1/2 9 average insulin OGTT) ; HOMA-b = (20 9 All procedures performed in studies involving Ins )/(Glu - 3.5); InsAUC30/luAUC30 = (Ins 0 0 0 human participants were done in accordance ? Ins )/(Glu ? Glu ); InsAUC120/GluAUC120 30 0 30 with the ethical standards of the institutional =(Ins ? 4 9 Ins ? 3 9 Ins )/(Glu ? 4 9 0 30 120 0 and/or national research committee, as well as Glu ? 3 9 Glu ); DI30 = (InsAUC30/ 30 120 with the 1964 Helsinki Declaration and its later GluAUC30) 9 ISI ; and DI120 = (InsAUC120/ amendments or comparable ethical standards. GluAUC120) 9 ISI [30]. Informed consent was obtained from all par- ticipants included in the study. Biochemical Analyses Fasting serum HBA1c, ALT, AST, total choles- RESULTS terol (TC), high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol Baseline Characteristics (LDL), triglyceride (TG), uric acid, and crea- tinine levels were measured at baseline and after 16 weeks of intervention. Out of the 52 patients treated with metformin monotherapy, 3 were not willing to participate in the trial and 4 did not meet the inclusion Statistical Analysis criteria. In total, 45 patients were randomized into the two treatment groups: 22 received Statistical analyses were performed using the exenatide therapy and 23 received insulin glar- SPSS version 18.0 software (SPSS Inc., Chicago, gine therapy. Three patients in each group IL, USA) with a significance level of a = 0.05. A refused to undergo continuous glucose moni- previous study demonstrated that insulin glar- toring again after 16 weeks of intervention. gine treatment results in a MAGE change from Therefore, 39 patients completed the study baseline at week 36 of - 0.3 ± 1.3 [31], while (Fig. 1). The baseline characteristics of the exenatide used twice daily results in an patients in the two treatment groups, including approximate MAGE change from baseline at age and diabetes duration, were well matched week 16 of - 2.91 ± 3.15 [32]. Using SMBG data (Table 1). [33] from a head-to-head study between Diabetes Ther (2018) 9:1253–1267 1257 measured at 30 min (P = 0.009) using a stan- dard meal tolerance test was found to be sig- nificantly decreased with insulin glargine treatment (Table 2). The exenatide treatment group exhibited larger decreases in FBG (- 0.17 mmol/L; 95% CI - 1.64 to 1.30 mmol/ L) and HbA1c (- 0.16%; 95% CI - 0.67% to 0.35%) than the insulin glargine group, although there were no statistical differences between the two groups (Table 2). Following both treatments, increases in 1/HOMA-IR (exe- natide: from 0.35 ± 0.05 to 0.54 ± 0.08 lIU/ mL, mmol/L, P = 0.035; insulin glargine: from Fig. 1 Flow chart of study participants. Of the 52 0.28 ± 0.03 to 0.47 ± 0.06 lIU/mL, mmol/L, participants recruited for the present study, 3 were P = 0.017), ISI (exenatide: from 4.43 ± 0.71 to unwilling to participate and 4 did not meet the inclusion 7.13 ± 1.06 lIU/mL, mmol/L, P = 0.035), dis- criteria position index 30 (exenatide: from 41.94 ± 5.30 to 68.50 ± 11.06 mmol/L, Biomedical Parameters mmol/L, P = 0.008; insulin glargine: from 36.40 ± 4.40 to 60.39 ± 6.24 mmol/L, mmol/L, After intervention, body weight (D = - 3.51 kg, P = 0.001), and disposition index 120 (exe- P \ 0.001), BMI (D = - 1.22 kg/m , P \ 0.001), natide: from 61.31 ± 9.64 to and waist circumference (D = - 3.03 cm, 97.12 ± 17.30 mmol/L, mmol/L, P = 0.020; P = 0.012) significantly decreased in the exe- insulin glargine: from 52.71 ± 6.87 to natide treatment group but did not change in 88.33 ± 10.21 mmol/L, mmol/L, P = 0.006) the insulin glargine group. Larger reductions in were observed (Table 2); however, there were no body weight (- 3.96 kg; 95% CI -5.65 to statistical differences between the two groups. - 2.27 kg; P \ 0.001) and BMI (- 1.38 kg/m ; 95% CI - 1.99 to - 0.77 kg/m ; P \ 0.001) Continuous Glucose Monitoring occurred in the exenatide group compared with the insulin glargine group. Although TC Overall, blood glucose fluctuations significantly (P = 0.019) and LDL (P = 0.010) levels were improved after the exenatide and insulin glar- significantly reduced after exenatide interven- gine treatments (Fig. 2). There were similarly tion, and ALT (P = 0.027) levels were signifi- significant reductions in the MBG (exenatide cantly decreased after the insulin glargine D = - 1.25 ± 0.29 mmol/L, P \ 0.001; insulin treatment, there were no significant differences glargine D = - 1.83 ± 0.66 mmol/L, P = 0.007), between the two groups (Table 1). continuous overlapping net glycemic action (CONGA) (exenatide D = - 1.13 ± 0.30 mmol/ Glycemic Control and Insulin Function L, P \ 0.001; insulin glargine D = - 2.09 ± 0.70 mmol/L, P = 0.015), and MAGE (exenatide After 16 weeks of intervention, HbA1c was sig- D = - 1.20 ± 0.54 mmol/L, P = 0.001; insulin nificantly reduced by both exenatide treatment glargine D = - 1.47 ± 0.52 mmol/L, P = 0.001) and by insulin glargine treatment (exenatide values after the exenatide and insulin glargine 8.01 ± 0.21 vs 6.83 ± 0.25%, D = - 1.18%, interventions. There appeared to be smaller P \ 0.001; insulin glargine 8.35 ± 0.24 vs decreases in MBG (0.36 mmol/L; 95% CI - 1.00 7.14 ± 0.16%, D = - 1.21%, P \ 0.001; to 1.72 mmol/L), CONGA (0.45 mmol/L; 95% CI Table 2). Both exenatide and insulin glargine - 0.94 to 1.84 mmol/L), and MAGE (0.20 mmol/ significantly reduced FBG (P = 0.002, L; 95% CI - 1.00 to 1.41 mmol/L) in the exe- P = 0.007, respectively); the blood glucose level natide group when compared to the insulin 1258 Diabetes Ther (2018) 9:1253–1267 Table 1 Subject characteristics at baseline and after treatment Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin glargine Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) Mean (95% CI) Mean – SE Mean – SE Mean – SE Mean – SE Number (n)19 – – 20 – – – – – – Sex (male/female) 12/7 – – 10/10 – – – – – – Age 46.74 ± 2.31 – – 49.45 ± 2.17 – – – – – – Diabetic duration 6.37 ± 0.99 – – 4.35 ± 0.68 – – – – – – Weight (kg) 80.36 ± 2.35 76.85 ± 2.36 \ 0.001 74.75 ± 1.95 75.20 ± 1.98 0.263 - 3.51 0.45 - 3.96 \ 0.001 (- 5.09 to - 1.93) (- 0.37 to 1.27) (- 5.65 to - 2.27) BMI (kg/m ) 28.03 ± 0.50 26.81 ± 0.51 \ 0.001 27.08 ± 0.52 27.24 ± 0.55 0.258 - 1.22 0.17 - 1.38 \ 0.001 (- 1.78 to - 0.65) (- 0.13 to 0.47) (- 1.99 to - 0.77) Waist circumference 96.74 ± 1.50 93.71 ± 1.59 0.012 93.70 ± 1.52 93.30 ± 1.48 0.366 - 3.03 - 0.40 - 2.62 0.061 (cm) (- 5.30 to - 0.75) (- 1.30 to 0.50) (- 5.03 to - 0.22) ALT (U/L) 31.79 ± 4.33 31.56 ± 3.96 0.949 36.46 ± 4.82 28.83 ± 3.81 0.027 - 0.23 - 7.63 7.39 0.183 (- 7.75 to 7.28) (- 14.27 to - 0.99) (- 2.27 to 17.05) AST (U/L) 22.28 ± 2.13 21.97 ± 2.18 0.855 24.78 ± 2.15 22.39 ± 1.92 0.103 - 0.32 - 2.39 2.07 0.522 (- 3.92 to 3.29) (- 5.31 to 0.53) (- 2.37 to 6.51) TG (mmol/L) 1.75 ± 0.17 1.61 ± 0.23 0.286 1.70 ± 0.17 1.49 ± 0.19 0.295 - 0.15 - 0.21 0.07 0.744 (- 0.42 to 0.13) (- 0.63 to 0.20) (- 0.42 to 0.55) TC (mmol/L) 4.55 ± 0.31 4.04 ± 0.28 0.019 4.45 ± 0.23 4.30 ± 0.15 0.557 - 0.51 - 0.14 - 0.37 0.183 (- 0.93 to - 0.10) (- 0.64 to 0.35) (- 1.00 to 0.25) HDL (mmol/L) 1.04 ± 0.05 1.00 ± 0.06 0.324 1.03 ± 0.05 1.11 ± 0.07 0.147 - 0.04 0.09 - 0.12 0.083 (- 0.11 to 0.04) (- 0.03 to 0.21) (- 0.26 to 0.01) LDL (mmol/L) 2.42 ± 0.21 2.07 ± 0.20 0.010 2.36 ± 0.17 2.28 ± 0.11 0.617 - 0.35 - 0.09 - 0.26 0.153 (- 0.60 to - 0.10) (- 0.44 to 0.27) (- 0.69 to 0.16) P values of\ 0.05 were considered to indicate significant differences between the groups BMI body mass index, ALT alanine aminotransferase, AST aspartate aminotransferase, TG triglyceride, TC total cholesterol, LDL low-density lipoprotein cholesterol, HDL high-density lipoprotein cholesterol Diabetes Ther (2018) 9:1253–1267 1259 Table 2 Glucose control, insulin sensitivity, and b-cell function at baseline and after treatment Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) glargine Mean – SE Mean – SE Mean – SE Mean – SE Mean (95% CI) FBG (mmol/L) 8.65 ± 0.53 6.88 ± 0.40 0.002 8.83 ± 0.48 7.23 ± 0.39 0.007 - 1.77 - 1.60 - 0.17 0.580 (- 2.83 to - 0.72) (- 2.70 to - 0.50) (- 1.64 to 1.30) 30 min glucose 11.03 ± 0.74 9.06 ± 0.87 0.061 11.90 ± 0.68 10.08 ± 0.59 0.009 - 1.97 - 1.82 - 0.15 0.497 (mmol/L) (- 4.05 to 0.10) (- 3.13 to - 0.52) (- 2.48 to 2.19) 60 min glucose 14.27 ± 0.99 11.82 ± 1.03 0.080 14.70 ± 0.67 12.18 ± 0.66 0.091 - 2.45 - 2.52 0.08 0.844 (mmol/L) (- 5.23 to 0.33) (- 3.92 to - 1.12) (- 2.82 to 2.97) 120 min glucose 13.31 ± 0.88 11.71 ± 0.75 0.106 13.99 ± 0.90 12.77 ± 0.86 0.204 - 1.59 - 1.22 - 0.37 0.457 (mmol/L) (- 3.56 to 0.38) (- 3.17 to 0.74) (- 3.04 to 2.29) Fasting insulin 10.92 ± 1.85 10.03 ± 1.75 0.701 10.48 ± 1.02 9.62 ± 1.88 0.671 - 0.89 - 0.86 - 0.03 0.903 (uIU/mL) (- 5.71 to 3.93) (- 5.07 to 3.35) (- 6.22 to 6.15) 30 min insulin 19.69 ± 3.03 17.57 ± 3.96 0.523 18.47 ± 2.12 22.48 ± 4.83 0.388 - 2.12 4.01 - 6.13 0.298 (uIU/mL) (- 8.99 to 4.74) (- 5.57 to 13.59) (- 17.37 to 5.10) 60 min insulin 30.26 ± 5.48 33.86 ± 8.82 0.646 29.92 ± 3.53 33.99 ± 6.26 0.525 3.60 4.06 - 0.46 0.970 (uIU/mL) (- 12.64 to 19.84) (- 9.19 to 17.32) (- 20.78 to 19.85) 120 min insulin 35.65 ± 5.61 39.33 ± 8.87 0.549 38.38 ± 6.05 44.93 ± 8.61 0.436 3.67 6.54 - 2.87 0.756 (uIU/mL) (- 9.00 to 16.35) (- 10.81 to 23.90) (- 23.36 to 17.62) 1/HOMA-IR 0.35 ± 0.05 0.54 ± 0.08 0.035 0.28 ± 0.03 0.47 ± 0.06 0.017 0.19 0.19 0.00 0.718 (lIU/mL, (0.02 to 0.37) (0.04 to 0.33) (- 0.22 to 0.22) mmol/L) ISI (lIU/mL, 4.43 ± 0.71 7.13 ± 1.06 0.035 3.81 ± 0.54 5.29 ± 0.70 0.059 2.70 1.48 1.22 0.218 mmol/L) (0.22 to 5.19) (- 0.07 to 3.03) (- 1.62 to 4.06) HOMA-B 55.55 ± 11.55 120.79 ± 48.44 0.153 54.17 ± 10.90 84.09 ± 30.50 0.361 65.24 29.91 35.33 0.536 (IU/mol) (- 26.68 to 157.17) (- 37.74 to 97.57) (- 76.88 to 147.54) InsAUC30/ 12.09 ± 1.90 15.29 ± 3.32 0.181 11.20 ± 1.64 14.77 ± 3.20 0.237 3.20 3.58 - 0.37 0.898 GluAUC30 (- 1.65 to 8.05) (- 2.613 to 9.76) (- 7.86 to 7.11) (IU/mol) 1260 Diabetes Ther (2018) 9:1253–1267 Table 2 continued Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) glargine Mean – SE Mean – SE Mean – SE Mean – SE Mean (95% CI) InsAUC120/ 17.39 ± 3.17 22.03 ± 5.78 0.158 16.43 ± 2.73 21.94 ± 4.63 0.227 4.64 5.51 - 0.87 0.836 GluAUC120 (- 2.02 to 11.29) (- 3.88 to 14.90) (- 11.64 to 9.89) (IU/mol) DI 30 (mmol/L, 41.94 ± 5.30 68.50 ± 11.06 0.008 36.40 ± 4.40 60.39 ± 6.24 0.001 26.57 23.99 2.57 0.872 mmol/L) (8.20 to 44.93) (12.71 to 35.27) (- 18.36 to 23.51) DI 120 (mmol/L, 61.31 ± 9.64 97.12 ± 17.30 0.020 52.71 ± 6.87 88.33 ± 10.21 0.006 35.81 35.62 0.19 0.949 mmol/L) (6.65 to 64.97) (12.24 to 59.00) (- 35.79 to 36.17) HbA1c (%) 8.01 ± 0.21 6.83 ± 0.25 \ 0.001 8.35 ± 0.24 7.14 ± 0.16 \ 0.001 - 1.21 - 1.05 - 0.16 0.537 (- 1.57 to - 0.85) (- 1.41 to - 0.69) (- 0.67 to 0.35) InsAUC30/GluAUC30 was calculated as the total insulin area under the curve divided by the total glucose area under the curve during the first 30 min of the 75-g oral glucose tolerance test (OGTT); InsAUC120/GluAUC120 was calculated as the total insulin area under the curve divided by the total glucose area under the curve during the 120 min of the OGTT P values of \ 0.05 were considered to indicate significant differences between the groups FBG fasting blood glucose, 1/HOMA-IR 1/homeostasis model assessment of insulin resistance, ISIM Matsuda insulin sensitivity index, HOMA-b basal homeostasis model assessment of insulin secretion, DI30 disposition index 30, DI120 disposition index 120, HbA1c hemoglobin A1c Diabetes Ther (2018) 9:1253–1267 1261 Fig. 2 Mean interstitial glucose values at baseline and after treatment with exenatide (a) and insulin glargine (b) glargine group, although there were no statisti- DISCUSSION cally significant differences. After exenatide intervention, there were significant reductions This study compared the effects of treatment in the standard deviation of the mean glucose with the short-term GLP-1 receptor agonist (D = 0.70 ± 0.24 mmol/L, P = 0.001) and largest exenatide and insulin glargine treatment on GV amplitude of glycemic excursions in overweight and obese patients with T2DM (D = - 2.32 ± 1.06 mmol/L, P = 0.004), while and poor glycemic control who were receiving no changes were observed after the insulin glar- metformin monotherapy. Importantly, our gine intervention (Table 3). results using a CGMS revealed that similar Both interventions significantly decreased improvements in GV, HBA1c, and b-cell func- the percentage of the time that TBG [ tion were achieved with exenatide treatment 10.0 mmol/L [exenatide D = - 14.79 ± 4.97%, and insulin glargine treatment, although P = 0.008; insulin glargine D = - 25.11 ± greater body weight and BMI reductions were 10.16%, P = 0.024 (Fig. 3b)] and the highest achieved with the exenatide treatment during blood glucose level [exenatide D = - 2.70 ± the 16-week intervention. After treatment, waist 0.72 mmol/L, P = 0.001; insulin glargine circumference was significantly decreased in the D = - 2.49 ± 1.16 mmol/L, P = 0.045 (Fig. 3e)]. exenatide group, but there was no statistically The percentage of the time that TBG was significant difference between the two groups. [ 7.8 mmol/L was significantly decreased in the Recent studies have suggested that glycemic insulin glargine treatment group (P = 0.030) fluctuation is a useful indicator of the effec- but not in the exenatide group (Fig. 3a). How- tiveness of blood glucose control and the pre- ever, there were no statistically significant dif- vention of diabetes complications [1, 2, 10]. ferences between the two groups. Less frequent monitoring of blood glucose levels using SMBG may lead to inaccurate GV assessment and could falsely suggest a lack of or Correlative Analysis a decrease in glycemic fluctuation [3, 5, 6]. A CGMS automatically records blood glucose levels every 5 min, achieving up to 288 mea- For all subjects, there were significant positive surements per day, which provide a better correlations between DHbA1c and the reduc- measure of the extent of glycemic fluctuation tions in MBG, CONGA, and Max BG (0 \ r \ 1; and overall trends [7, 8]. Therefore, we used a P \ 0.05; Table 4). CGMS in this study to monitor blood glucose. 1262 Diabetes Ther (2018) 9:1253–1267 Table 3 Glycemic variability at baseline and after treatment Exenatide Insulin glargine Change in value Estimated difference in P value between groups Before After P Before After P Exenatide Insulin glargine Mean (95% CI) treatment treatment treatment treatment Mean (95% CI) Mean (95% CI) Mean – SE Mean – SE Mean – SE Mean – SE MBG (mmol/L) 9.77 ± 0.40 8.32 ± 0.44 \ 0.001 10.85 ± 0.54 9.04 ± 0.54 0.007 - 1.45 - 1.82 0.36 0.824 (- 2.06 to - 0.85) (- 3.01 to - 0.57) (- 1.00 to 1.72) SDBG (mmol/L) 1.87 ± 0.15 1.19 ± 0.13 0.001 2.02 ± 0.14 1.71 ± 0.16 0.187 - 0.68 - 0.31 - 0.37 0.022 (- 1.06 to - 0.31) (- 0.79 to 0.17) (- 0.96 to 0.21) MODD (mmol/L) 1.75 ± 0.21 1.52 ± 0.17 0.215 2.01 ± 0.18 1.82 ± 0.13 0.248 - 0.24 - 0.19 - 0.05 0.294 (- 0.63 to 0.16) (- 0.53 to 0.15) (- 0.53 to 0.44) MAGE (mmol/L) 4.51 ± 0.32 2.94 ± 0.33 0.001 5.62 ± 0.46 3.84 ± 0.39 0.001 - 1.57 - 1.78 0.20 0.322 (- 2.42 to - 0.73) (- 2.70 to - 0.86) (- 1.00 to 1.41) LAGE (mmol/L) 7.82 ± 0.59 5.18 ± 0.53 0.004 8.60 ± 0.63 6.97 ± 0.87 0.166 - 2.64 - 1.63 - 1.00 0.083 (- 4.29 to - 0.99) (- 4.01 to 0.75) (- 3.77 to 1.77) CONGA (mmol/L) 8.82 ± 0.39 7.53 ± 0.40 \ 0.001 9.55 ± 0.49 7.80 ± 0.47 0.015 - 1.30 - 1.75 0.45 0.973 (- 1.91 to - 0.68) (- 3.12 to - 0.38) (- 0.94 to 1.84) P values of\ 0.05 were considered to indicate statistically significant differences between the groups MBG mean blood glucose level, SDBG standard deviation of the mean glucose value, MODD absolute mean of daily differences, MAGE mean amplitude of glycemic excursions, LAGE largest amplitude of glycemic excursions, CONGA continuous overlapping net glycemic action Diabetes Ther (2018) 9:1253–1267 1263 Fig. 3 Effects of exenatide treatment and glargine treatment on the percentage of the time that the blood glucose value (TBG) was[7.8 mmol/L (a),[10.0 mmol/L (b), or[11.1 mmol/L (c) as well as the lowest (d) and the highest (e) blood glucose (BG) values. *P \ 0.05, 16 weeks vs baseline patients with suboptimally controlled diabetes Table 4 Correlations between glycemic variability and using metformin or a sulfonylurea [18, 19]. glycated hemoglobin However, the average daily risk range as well as DHBA1c the low and high blood glucose indices, rP indicative of the fluctuation in blood glucose levels, were based on SMBG and not on the DMBG (mmol/L) 0.498 0.001 more accurate CGMS [18, 19]. We found that DCONGA (mmol/L) 0.512 0.002 FBG and HbA1c were significantly reduced after exenatide and insulin glargine intervention. DMax BG (mmol/L) 0.439 0.006 Compared with baseline levels, glycemic fluc- P \ 0.05 was considered to indicate statistical significance tuation results calculated by the CGMS indi- MBG mean blood glucose level, CONGA continuous cated that both exenatide treatment and insulin overlapping net glycemic action, HbA1c hemoglobin A1c glargine treatment significantly decreased MBG, CONGA, MAGE, and the highest blood glucose value, as well as the percentage of time that the blood glucose was [ 10.0 mmol/L through the Glycemic fluctuation following the use of a day. Following the standard meal tolerance test, GLP-1 receptor agonist was compared with the insulin glargine significantly decreased blood GV observed after the introduction of basal glucose levels at 30 min; glucose levels at 60 and insulin. Results showed that although the exe- 120 min showed a downward trend after exe- natide and glargine treatments were both asso- natide intervention and insulin glargine inter- ciated with similar significant improvements in vention. The decrease in MAGE with insulin HbA1c from baseline, exenatide treatment glargine treatment might due to improved resulted in greater improvements in GV than b-cell function and insulin resistance associated insulin glargine treatment for previously with improved glycemic control. uncontrolled glucose levels when combined Other long-term GLP-1 receptor agonists with metformin and sulfonylurea therapy. that only require once-daily or once a week Improvements in GV were also observed in 1264 Diabetes Ther (2018) 9:1253–1267 injection have become widely used or are insulin glargine facilitate glycemic control and beginning to find wider use in China. The long- reduce GV, thus aiding diabetes management. term GLP-1 receptor agonists liraglutide and The main AEs reported in the exenatide- semaglutide have shown beneficial effects on treated group were gastrointestinal intolerance/ CVD and mortality-related CVD in T2DM appetite suppression (17 subjects), nausea patients, although there would appear to be no (which relieved over time; 3 subjects), and difference in the effects of once-weekly exe- abdominal distension (4 subjects). No patients natide, the other long-term GLP-1 receptor in the group treated with insulin glargine suf- agonist, or lixisenatide, a short-term GLP-1 fered gastrointestinal intolerance, hypo- receptor agonist [34–36]. HbA1c and FBG glycemia, or other AEs. No hypoglycemia events showed greater reductions with liraglutide occurred in these patients, which may be treatment rather than exenatide treatment because prandial insulin and drugs stimulating [23, 24], and significantly greater improvements insulin secretion were not used. No participants in HbA1c were seen with liraglutide and withdrew from the study due to these AEs, and semaglutide than with insulin glargine [20–22]. no SAEs occurred in the present study. Whether the beneficial effects of liraglutide and There are a number of limitations of our semaglutide for CVD were the result of greater study. Only 39 patients were enrolled, and this decreases in GV requires further study. small sample size may have had a negative Regarding insulin resistance and b-cell func- impact on the statistical analysis. It might affect tion, the 1/HOMA-IR, disposition index 30, and the comparison of waist circumference between disposition index 120 improved with either the two intervention groups, for greater reduc- exenatide treatment or insulin glargine treat- tion of waist circumference in exenatide group ment. Only the ISI improved to a greater might be confirmed after enlarging the sample extent with exenatide treatment than with size. The results of the present study could be insulin glargine treatment. These results indi- further confirmed by expanding the sample cate that b-cell function and insulin resistance size, using a blinded design, testing long-term improved following a 16-week course of exe- GLP-1 receptor analogists, and potentially natide or insulin glargine treatment. developing a multicenter research study. In Overweight and obese patients in the coop- addition, the study period was 16 weeks, which erative meta-analysis group of the China Obe- may not have been sufficient to assess addi- sity Task Force were enrolled in our study [25]. tional benefits relating to the GV. Since a fixed The results showed that exenatide treatment dosage of exenatide was given in the exenatide resulted in weight loss while insulin glargine group after 4 weeks and the dosage of the treatment did not. Standardized weight is an insulin glargine was only titrated during the important part of integrated T2DM manage- first four weeks, rising FBG levels in some ment because weight loss is associated with insulin-glargine-treated patients during the fol- improvements in b-cell function and insulin lowing 12 weeks may have affected glucose action, it makes long-term glycemic control control in the insulin glargine group. management easier, and it may favorably affect other comorbid diseases such as dyslipidemia, CONCLUSIONS hypertension, and hyperuricemia [15]. An increasing waist circumference contributes to Our study showed that comparable improve- central obesity and is an independent risk factor ments in GV, HBA1c, and b-cell function can be for T2DM, dyslipidemia, hypertension, and achieved in overweight and obese T2DM CVD [26, 37]. In the present study, waist cir- patients with inadequate glycemic control who cumference only decreased in the exenatide- were receiving metformin monotherapy by treated group. combining that treatment with either exenatide An important finding from our study was a or insulin glargine. Compared to insulin glar- positive association between GV and HbA1c gine, exenatide had a superior impact on body level, which may indicate that exenatide and Diabetes Ther (2018) 9:1253–1267 1265 weight and BMI. Waist circumference was sig- Compliance with Ethics Guidelines. All nificantly reduced with exenatide treatment, procedures performed in studies involving but there was no statistically significant differ- human participants were in accordance with ence from the insulin glargine treatment. the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed ACKNOWLEDGEMENTS consent was obtained from all participants included in the study. We thank the participants of the study. Data Availability. The datasets analyzed for Funding. This work was supported by grants the current analysis are available from the cor- from the Investigator Initiated-Sponsored Clin- responding author on reasonable request. ical Research from AstraZeneca Investment (China) Co., Ltd. (ISSEXEN0034), the National Open Access. This article is distributed Natural Science Foundation of China under the terms of the Creative Commons (81570736, 81570737), Medical and Health Attribution-NonCommercial 4.0 International Research Projects of Nanjing Health Bureau in License (http://creativecommons.org/licenses/ Jiangsu Province of China (YKK14055), Project by-nc/4.0/), which permits any non- of Standardized Diagnosis and Treatment of Key commercial use, distribution, and reproduction Diseases in Jiangsu Province of China in any medium, provided you give appropriate (2015604), China Diabetes Young Scientific credit to the original author(s) and the source, Talent Research Project (2017-N-05), and Nan- provide a link to the Creative Commons license, jing University Central University Basic Scien- and indicate if changes were made. tific Research (14380296). Drum Tower Hospital (affiliated with Nanjing University Medical School, China) provided sponsorship for article processing charges. Our study sponsor did not REFERENCES fund the journal’s article processing charges. All authors had full access to all of the data in this 1. Siegelaar SE, Holleman F, Hoekstra JB, DeVries JH. study and take complete responsibility for the Glucose variability; does it matter? Endocr Rev. 2010;31(2):171–82. integrity of the data and accuracy of the data analysis. 2. Jin YP, Su XF, Yin GP, et al. Blood glucose fluctua- tions in hemodialysis patients with end stage dia- Editorial Assistance. Editage, a brand of betic nephropathy. J Diabetes Complicat. 2015;29(3):395–9. Cactus Communications, offered professional English language editing. 3. Lee MK, Lee KH, Yoo SH, Park CY. Impact of initial active engagement in self-monitoring with a tele- Authorship. All the named authors meet the monitoring device on glycemic control among International Committee of Medical Journal patients with type 2 diabetes. Sci Rep. 2017;7(1):3866. Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of 4. Tildesley HD, Mazanderani AB, Ross SA. Effect of the work as a whole, and have given their Internet therapeutic intervention on A1C levels in approval for this version to be published. patients with type 2 diabetes treated with insulin. Diabetes Care. 2010;33(8):1738–40. Disclosures. Ting-Ting Yin, Yan Bi, Ping Li, 5. Clar C, Barnard K, Cummins E, Royle P, Waugh N, Shan-Mei Shen, Xiao-Lu Xiong, Li-Jun Gao, Can Aberdeen Health Technology Assessment Group. Jiang, Yan Wang, Wen-Huan Feng, and Da- Self-monitoring of blood glucose in type 2 diabetes: Long Zhu have nothing to disclose. systematic review. Health Technol Assess. 2010;14(12):1–140. 1266 Diabetes Ther (2018) 9:1253–1267 6. Malanda UL, Bot SD, Nijpels G. Self-monitoring of 17. Buse JB, Bergenstal RM, Glass LC, et al. Use of twice- blood glucose in noninsulin-using type 2 diabetic daily exenatide in basal insulin-treated patients patients: it is time to face the evidence. Diabetes with type 2 diabetes: a randomized, controlled trial. Care. 2013;36(1):176–8. Ann Intern Med. 2011;154(2):103–12. 7. Levy JC, Davies MJ, Holman RR, Group TS. Con- 18. Barnett AH, Burger J, Johns D, et al. Tolerability and tinuous glucose monitoring detected hypogly- efficacy of exenatide and titrated insulin glargine in caemia in the Treating to Target in Type 2 Diabetes adult patients with type 2 diabetes previously Trial (4-T). Diabetes Res Clin Pract. 2017;131:161–8. uncontrolled with metformin or a sulfonylurea: a multinational, randomized, open-label, two-period, 8. Wei Q, Sun Z, Yang Y, Yu H, Ding H, Wang S. Effect crossover noninferiority trial. Clin Ther. of a CGMS and SMBG on maternal and neonatal 2007;29(11):2333–48. outcomes in gestational diabetes mellitus: a ran- domized controlled trial. Sci Rep. 2016;6:19920. 19. McCall AL, Cox DJ, Brodows R, Crean J, Johns D, Kovatchev B. Reduced daily risk of glycemic vari- 9. Brownlee M. The pathobiology of diabetic compli- ability: comparison of exenatide with insulin glar- gine. Diabetes Technol Therap. 2009;11(6):339–44. cations: a unifying mechanism. Diabetes. 2005;54(6):1615–25. 20. Aroda VR, Bain SC, Cariou B, et al. Efficacy and 10. Di Flaviani A, Picconi F, Di Stefano P, et al. Impact safety of once-weekly semaglutide versus once-daily of glycemic and blood pressure variability on sur- insulin glargine as add-on to metformin (with or rogate measures of cardiovascular outcomes in type without sulfonylureas) in insulin-naive patients 2 diabetic patients. Diabetes Care. 2011;34(7): with type 2 diabetes (SUSTAIN 4): a randomised, 1605–9. open-label, parallel-group, multicentre, multina- tional, phase 3a trial. Lancet Diabetes Endocrinol. 11. Murata GH, Duckworth WC, Shah JH, Wendel CS, 2017;5(5):355–66. Hoffman RM. Sources of glucose variability in insulin-treated type 2 diabetes: the Diabetes Out- 21. Russell-Jones D, Vaag A, Schmitz O, et al. Liraglu- comes in Veterans Study (DOVES). Clin Endocrinol. tide vs insulin glargine and placebo in combination 2004;60(4):451–6. with metformin and sulfonylurea therapy in type 2 diabetes mellitus (LEAD-5 met ? SU): a randomised 12. Figueira FR, Umpierre D, Casali KR, et al. Aerobic controlled trial. Diabetologia. 2009;52(10): and combined exercise sessions reduce glucose 2046–55. variability in type 2 diabetes: crossover randomized trial. PLoS One. 2013;8(3):e57733. 22. Diamant M, Van Gaal L, Stranks S, et al. Once weekly exenatide compared with insulin glargine 13. Cohen O, Korner A, Chlup R, et al. Improved gly- titrated to target in patients with type 2 diabetes cemic control through continuous glucose sensor- (DURATION-3): an open-label randomised trial. augmented insulin pump therapy: prospective Lancet. 2010;375(9733):2234–43. results from a community and academic practice patient registry. J Diabetes Sci Technol. 2009;3(4): 23. Twigg SM, Daja MM, O’Leary BA, Adena MA. Once- 804–11. daily liraglutide (1.2 mg) compared with twice- daily exenatide (10 mug) in the treatment of type 2 14. Peterson K, Zapletalova J, Kudlova P, et al. Benefits diabetes patients: an indirect treatment comparison of three-month continuous glucose monitoring for meta-analysis. J Diabetes. 2016;8(6):866–76. persons with diabetes using insulin pumps and sensors. Biomed Pap Med Fac Univ Palacky Olo- 24. Meier JJ, Rosenstock J, Hincelin-Mery A, et al. mouc Czechoslovakia. 2009;153(1):47–51. Contrasting effects of lixisenatide and liraglutide on postprandial glycemic control, gastric emptying, 15. Garber AJ, Abrahamson MJ, Barzilay JI, et al. Con- and safety parameters in patients with type 2 dia- sensus Statement by the American Association of betes on optimized insulin glargine with or without Clinical Endocrinologists and American College of metformin: a randomized, open-label trial. Diabetes Endocrinology on the Comprehensive Type 2 Dia- Care. 2015;38(7):1263–73. betes Management Algorithm—2017 Executive Summary. Endocr Pract. 2017;23(2):207–38. 25. Zhou BF, Cooperative Meta-Analysis Group of the Working Group on Obesity in China. Predictive values of body mass index and waist circumference 16. Eng C, Kramer CK, Zinman B, Retnakaran R. Glu- cagon-like peptide-1 receptor agonist and basal for risk factors of certain related diseases in Chinese insulin combination treatment for the manage- adults—study on optimal cut-off points of body ment of type 2 diabetes: a systematic review and mass index and waist circumference in Chinese meta-analysis. Lancet. 2014;384(9961):2228–34. adults. Biomed Environ Sci. 2002;15(1):83–96. Diabetes Ther (2018) 9:1253–1267 1267 26. Murphy KG. Editorial overview: endocrine and and beta-cell function in early type 2 diabetes. Acta metabolic diseases: waistline weapons: new thera- Diabetol. 2013;50(4):587–95. peutic avenues for the treatment of obesity and metabolic disease. Curr Opin Pharmacol. 32. Irace C, Fiorentino R, Carallo C, Scavelli F, Gnasso 2015;25:iv–vi. A. Exenatide improves glycemic variability assessed by continuous glucose monitoring in subjects with 27. Li C, Yang H, Tong G, et al. Correlations between type 2 diabetes. Diabetes Technol Therap. A1c, fasting glucose, 2 h postload glucose, and beta- 2011;13(12):1261–3. cell function in the Chinese population. Acta Dia- betol. 2014;51(4):601–8. 33. Heine RJ, Van Gaal LF, Johns D, et al. Exenatide versus insulin glargine in patients with subopti- 28. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, mally controlled type 2 diabetes: a randomized Treacher DF, Turner RC. Homeostasis model trial. Ann Intern Med. 2005;143(8):559–69. assessment: insulin resistance and beta-cell func- tion from fasting plasma glucose and insulin con- 34. Marso SP, Daniels GH, Brown-Frandsen K, et al. centrations in man. Diabetologia. 1985;28(7): Liraglutide and cardiovascular outcomes in type 2 412–9. diabetes. N Engl J Med. 2016;375(4):311–22. 29. Matsuda M, DeFronzo RA. Insulin sensitivity indi- 35. Holman RR, Bethel MA, Mentz RJ, et al. Effects of ces obtained from oral glucose tolerance testing: once-weekly exenatide on cardiovascular outcomes comparison with the euglycemic insulin clamp. in type 2 diabetes. N Engl J Med. 2017;377(13): Diabetes Care. 1999;22(9):1462–70. 1228–39. 30. Stancakova A, Javorsky M, Kuulasmaa T, Haffner 36. Pfeffer MA, Claggett B, Diaz R, et al. Lixisenatide in SM, Kuusisto J, Laakso M. Changes in insulin sen- patients with type 2 diabetes and acute coronary sitivity and insulin release in relation to glycemia syndrome. N Engl J Med. 2015;373(23):2247–57. and glucose tolerance in 6,414 Finnish men. Dia- betes. 2009;58(5):1212–21. 37. Cui R, Qi Z, Zhou L, Li Z, Li Q, Zhang J. Evaluation of serum lipid profile, body mass index, and waist- 31. Pistrosch F, Kohler C, Schaper F, Landgraf W, Forst line in Chinese patients with type 2 diabetes mel- T, Hanefeld M. Effects of insulin glargine versus litus. Clin Interv Aging. 2016;11:445–52. metformin on glycemic variability, microvascular

Journal

Diabetes TherapySpringer Journals

Published: May 9, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off