Journal of Endocrinological Investigation https://doi.org/10.1007/s40618-018-0991-0 ORIGINAL ARTICLE Sitagliptin vs. pioglitazone as add‑ on treatments in patients with uncontrolled type 2 diabetes on the maximal dose of metformin plus sulfonylurea P. Khaloo1 · S. Asadi Komeleh1 · H. Alemi1 · M. A. Mansournia2 · A. Mohammadi1 · A. Yadegar1 · M. Afarideh1 · S. Esteghamati1 · M. Nakhjavani1 · A. Esteghamati1 Received: 23 June 2018 / Accepted: 27 November 2018 © Italian Society of Endocrinology (SIE) 2018 Abstract Aims To compare the efficacy of sitagliptin versus pioglitazone as add-on drugs in patients with poorly controlled diabetes with metformin and sulfonylureas. Methods This is a randomized, open-label, parallel assignment clinical trial. Patients who had inadequate glycemic control [7% (53 mmol/mol) ≤ A1C < 11% (97 mmol/mol)] despite a minimum 6-month period of active treatment with metformin 2000 mg/day plus gliclazide 240 mg/day were enrolled in the study. HbA1C, fasting blood glucose (FBG), fasting plasma lipid parameters [total cholesterol (TC0, low-density lipoprotein cholesterol (LDL-C), triglycerides (TG) and high-density lipoprotein cholesterol (HDL-C)], systolic and diastolic blood pressure (SBP, DBP), weight, waist circumference, and body mass index were measured at baseline and after 17, 34, and 52 weeks of treatment. Generalized estimating equation analysis was done to compare treatment groups for continuous efficacy parameters. Results No significant difference in HbA1C reduction was observed between the treatment groups during the study course. (P = 0.149, adjusted P = 0.434; coefficient − 0.11 ± 0.08). The FBG (P = 0.032; coefficient 7.44 ± 3.48), HDL-C (P = 0.001; coefficient − 2.69 ± 0.83), TG (P = 0.027; coefficient 12.63 ± 5.71) and SBP (P < 0.001; coefficient 5.43 ± 1.26) changes from baseline, and weight gain were greater in the pioglitazone group. The mean changes in LDL-C and TC from baseline to week 52 were greater in the sitagliptin group (P = 0.034; coefficient − 7.40 ± 3.50, P = 0.013; coefficient − 7.16 ± 2.88, respectively). Conclusion Sitagliptin and pioglitazone were equally effective in improvement of HbA1C. There were some differences in terms of lipid indices, weight gain, and SBP. The current study confirmed that both sitagliptin and pioglitazone are effective treatment options and the decision should be made for each individual based on the baseline characteristics. Keywords Sitagliptin · Pioglitazone · Efficacy · Type2 diabetes Introduction The number of people with diabetes is rising as a result of population growth, aging, urbanization, increased prevalence of obesity, and physical inactivity. It is estimated that the number of patients with diabetes may rise from A. Esteghamati [email protected] 1 Endocrinology and Metabolism Research Center (EMRC), Vali-Asr Hospital, School of Medicine, Tehran University of Medical Sciences, P.O. Box: 13145-784, Tehran, Iran 2 Department of Epidemiology and Biostatistics, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran 424 million in 2017 to 629 million in 2045 [1]. Metformin is the first line of treatment, while gliclazide is the most common conventional second line used in controlling type 2 diabetes mellitus (T2DM) [2]. Sitagliptin, approved by FDA in 2006, is the first dipeptidyl peptidase-4 (DPP-4) inhibitor which acts by preventing inactivation of incretin hormones glucagon-like peptide-1 (GLP-1) and glucose-dependent insulin tropic polypeptide (GIP). Recently, sitagliptin has been proposed as an alternative for uncontrolled diabetes in patients who failed to respond to either metformin or gliclazide. It has been shown that sitagliptin helps in improving the glycemic control and lipid profile in addition to reducing the blood pressure and body weight simultaneously [3–7]. 13 Journal of Endocrinological Investigation Pioglitazone is a thiazolidinedione approved by FDA in 1999 that has been recommended as an alternative in patients with diabetes treated with a combination of metformin and sulfonylurea (SU) but having poor glycemic control. Pioglitazone improves blood glucose control by decreasing target tissues’ insulin resistance through activating peroxisome proliferator-activated receptor-gamma (PPRA-gamma) [8]. Pioglitazone could also improve abnormal lipid profile by acting on PPRA-alpha in T2DM. Notably, changes in liver function tests, blood pressure, and body weight have been also reported [9–11]. However, many patients still could not reach glycemic targets with metformin and gliclazide and require other agents in combination to control blood glucose level [12]. Numerous clinical trials have been done to show the efficacy and safety of pioglitazone and sitagliptin as monotherapies or in combination with other antidiabetic drugs as dual therapies [13–15]. However, few trials have compared triple therapies consisting each of these two oral agents with metformin and gliclazide so far. A 24-week study has shown that pioglitazone and sitagliptin achieved similar improvements in overall glycemic control in patients inadequately controlled with metformin and an SU, but there were some differences in FBG, lipids, and body weight changes [16]. The durability of these drugs is also of great importance. Both sitagliptin and pioglitazone have been reported to have long-term effects on glycemic control and improve durability when given in combination with other oral agents [17–19]. Therefore, in the current study, we investigated the improvement in glycemic control variables, lipid profile indices, blood pressure, and weight control after adding sitagliptin or pioglitazone to conventional oral therapy (metformin + gliclazide) to compare the efficacy and safety of these two drug regimens through the 52 weeks of follow-up. Materials and methods Study population Men and women with type 2 diabetes mellitus based on the American Diabetes Association (ADA) criteria who had inadequate glycemic control were enrolled in this study. Their treatment regimen included a minimum 6 months period of active treatment with the maximum dose of metformin, 2000 mg/day (500 mg QID) [20], plus the advised maximum dose of gliclazide, 240 mg/day (80 mg TDS) [21]. Inadequate glycemic control was defined as baseline level of 7% (53 mmol/mol) ≤ A1C < 11% (97 mmol/mol). Eligible patients were aged 25–70 years. The following exclusion criteria were applied: cardiovascular disease (including myocardial infarction, unstable angina, history of revascularization procedure 13 or cerebrovascular accident) or uncontrolled hypertension, estimated glomerular filtration rate (eGFR) < 60 ml/ min/1.73 m2 (calculated using the CKD–EPI equation), treatment with corticosteroids or other drugs interfering with glucose metabolism, any history of malignant disease, active infectious disease or history of infectious disease in the last 6 months, and a documented diagnosis of interstitial or obstructive lung disease. Study design This is a randomized, open-label, parallel assignment clinical trial conducted in the diabetes clinic of Vali-Asr Hospital, affiliated to Tehran University of Medical Sciences in Tehran, Iran. The study was performed between February 2015 and April 2017. Eligible patients were randomly divided into two 1.1 ratio treatment groups by the use of randomizations software. One took sitagliptin 100 mg daily, and the other took pioglitazone 30 mg daily both in combination with metformin 500 mg QID and gliclazide 80 mg TDS for 52 weeks. Patients were instructed to continue the lifestyle they had prior to entry to the trial through these 52 weeks. Participants were withdrawn from the study if they met any severe adverse effects including hypoglycemia, heart failure, or hepatic failure. Prior to enrollment, written informed consents were taken from all participants. The study was conducted in accordance with the Declaration of Helsinki, and the Ethics Committee of the Tehran University of Medical Sciences approved the study protocol. This trial was registered with ClinicalTrials.gov (NCT03125694). Study end point The primary outcome of this study was a change in the mean glycated hemoglobin (A1C) levels from baseline by sitagliptin vs. pioglitazone during the 52 weeks of treatment. HbA1C, fasting blood glucose, fasting plasma lipid parameters [total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG) and high-density lipoprotein cholesterol (HDL-C)], systolic and diastolic blood pressure (SBP and DBP), weight, waist circumference (WC), hip circumference (HC),and body mass index (BMI) were measured at baseline and after 17, 34, and 52 weeks of treatment. Secondary efficacy end points were changes in HbA1C of patients who achieve the HbA1C < 7% (53 mmol/ mol), HbA1C < 6.5% (48 mmol/mol), and changes in other previously mentioned measurements at week 17, 34, and 52. Statistical analysis Intent-to-treat (ITT) analysis was done using all randomized participants who received at least one dose of study treatment, had baseline measurements of HbA1C, Journal of Endocrinological Investigation and at least one follow-up visit after initiation of treatment. Mean ± standard deviation (SD) for continuous and frequencies (%) for categorical variables were used to show the baseline characteristics of participants. Regression models with generalized estimating equation (GEE) analysis were done to compare treatment groups for continuous efficacy parameters, taking into account the correlation between repeated variable measurements of the same subjects. Treatments as indicator variables were the main predictors of the model. In the model, the variable “time” was defined as months from the baseline visit. In our study, univariable regression analysis was per- formed for each potential risk factor including age, sex, duration of diabetes, weight, waist circumference, BMI, SBP, DBP, HbA1C, FBG, and lipid profile. Later on, covariates with a P value < 0.2 in the initial univariable analysis were selected to enter the multivariable model. Age, sex, weight, duration of diabetes, time, HbA1C, and baseline values were entered in the final model as covariates. The interaction between time and treatment was examined in the multivariate model. Considering that there was no significant interaction, we did not enter their product term in regression models with the generalized estimating equation. Missing data were han- dled using the lastobservation-carried-forward method. We used Huber-white cluster robust standard error for calculating 95% confidence interval. The proportion of patients achieving HbA1C ≤ 7 (53) or ≤ 6.5% (48 mmol/ mol) was compared among groups using logistic regres- sion analysis. Statistical analyses were performed using STATA version 12; P values ≤ 0.05 were considered sta- tistically significant. Results Baseline characteristics A total of 250 participants were recruited. 125 patients were randomized in each treatment group. One hundred and ten patients with the mean age of 62.7 in the pioglitazone group and 112 patients with the mean age of 60.8 in the sitagliptin group completed the 52 weeks of treatment including 128 females and 94 males. The factors leading to early study discontinuation were weight gain (nine in the pioglitazone group), edema (six in the pioglitazone group), GI upset (five in the sitagliptin group),and cost consideration (8 in the sitagliptin group) (Fig. 1). Percentage of females and duration of diabetes were significantly higher in patients randomly assigned to the pioglitazone group at baseline. Additionally, SBP and weight were higher in the sitagliptin group. Randomized and treated (N=250) Sitagliptin 100mg (N=125) Pioglitazone 30mg (N=125) Discontinued study (N=13) GI upset 5 Cost consideration 8 Discontinued study (N=15) Weight gain 9 Edema 6 Completed 52 weeks (N=112) Completed 52 weeks (N=110) Fig. 1 Study population flowgram Table 1 Baseline characteristics and demographics of randomized patients Pioglitazone Sitagliptin P value Sex Female Male Age DM Duration Weight BMI WC HC SBP DBP FBG TC HDL-C LDL-C TG 71 (55.5%) 39 (41.5%) 62.7 ± 8.2 14.3 ± 6.9 74.2 ± 13.5 29 ± 4.8 107.2 ± 85.9 104.3 ± 9.3 129.1 ± 14.1 79.4 ± 6.1 180.2 ± 47.5 167 ± 32.8 43.2 ± 10.6 90.4 ± 25.6 169 ± 70 57 (44.5%) 55 (58.5%) 60.8 ± 8.1 11.3 ± 6.2 78.3 ± 12.5 29.7 ± 4.3 101.6 ± 10 106.5 ± 9.9 135.7 ± 16.2 78.8 ± 8.2 171.4 ± 44.1 158.1 ± 34.6 42.3 ± 9 87.9 ± 26.8 150.2 ± 75.3 0.04 N/S 0.001 0.019 N/S N/S N/S 0.001 N/S N/S N/S N/S N/S N/S HbA1C 9 ± 1.2 8.8 ± 1.1 N/S DM diabetes, SBP systolic blood pressure, DBP diastolic blood pressure, BMI body mass index, WC waist circumferences, HC hip circumference, FPG fasting plasma glucose, TC total cholesterol, TG triglyceride, HDL-C high-density lipoprotein cholesterol, LDL-C lowdensity lipoprotein cholesterol There was no significant difference in other baseline measurements and clinical characteristics of the two treatment groups (Table 1). 13 Journal of Endocrinological Investigation Fig. 2 HBA1C levels over time up to 52 weeks of treatment The mean changes in LDL-C and TC from baseline to week 52 were greater in the sitagliptin group (P = 0.034; coefficient − 7.40 ± 3.50, P = 0.013; coefficient − 7.16 ± 2.88, respectively). On the other hand, the mean changes observed during 52 weeks in HDL-C and TG were significantly higher in the pioglitazone group (P = 0.001; coefficient − 2.69 ± 0.83, P = 0.027; coefficient 12.63 ± 5.71, respectively). At week 52, a decrease in body weight and HC with sitagliptin and an increase with pioglitazone were observed which led to a significant difference between the two treatment groups (P < 0.001; coefficient − 2.44 ± 0.34, P < 0.001; coefficient − 1.68 ± 0.45, respectively). SBP increased during the study course in both sitagliptin (3 ± 15.4) and pioglitazone (2.4 ± 14.6) groups, but the changes in the sitagliptin group were significantly greater (P < 0.001; coefficient 5.43 ± 1.26). DBP changes were not different between the two treatment groups (P = 0.161). No subjects showed any clinically significant adverse events. Changes in all study end points during the follow-ups are shown in Table 2. Discussion Fig. 3 FBG changes over time up to 52 weeks of treatment Efficacy The mean HbA1C changes from baseline to week 52 were − 1.9 ± 1.2% for the pioglitazone group and − 1.8 ± 1% for the sitagliptin group, but no significant difference in HbA1C reduction was observed between the treatment groups during the study course (P = 0.149, adjusted P = 0.434; coefficient − 0.11 ± 0.08). HbA1C values over time up to 52 weeks are shown in Fig. 2. After 52 weeks of treatment, the percentage of participants reaching the HbA1C goal of < 7% (53 mmol/mol) was significantly higher in the pioglitazone group (57.3%) compared to the sitagliptin group (42.7%) (Adjusted P = 0.048; IRR 1.31 ± 0.18). However, no difference was seen between the treatment groups (P = 0.139) by changing the target definition to HbA1C < 6.5% (48 mmol/ mol). Both pioglitazone and sitagliptin decreased FBG during the study course (− 45 ± 52, − 35 ± 46.4 mg/dL, respectively), but the FBG changes from baseline were greater in the pioglitazone group (P = 0.032; coefficient 7.44 ± 3.48). Figure 3 shows the FBG values over time up to week 52. 13 While metformin is widely recognized as the first line of diabetes treatment, physicians should choose the second and third line of antidiabetic agents according to the patients’ baseline characteristics and glycemic control, suggested by international and national guidelines. There is no consensus about how to intensify treatment in patients with poor glycemic control [22]. Here, we compared the efficacy of sitagliptin and pioglitazone as add-on drugs in triple therapy with metformin and SU. We found that although adding sitagliptin or pioglitazone both result in better glycemic control, there was no significant difference in HbA1C improvement between these two drug regimens after 52 weeks of treatment. However, pioglitazone was more potent in reaching HbA1C target goal [HbA1C < 7% (53 mmol/mol)]. Our results also showed a greater change in FBG with pioglitazone despite the similar effect on HbA1C. Significant reductions in HbA1C, FBG, and postprandial plasma glucose have been shown in previous studies using pioglitazone and sitagliptin alone or in combination with other treatments in patients with T2DM. The observed effects of pioglitazone and sitagliptin on HbA1c in the current study are similar to those reported in previous clinical studies [23]. In Kutoh et al.’s study, significant changes of HbA1c (from 10.05 to 8.01%) were observed in drug-naïve Japanese patients with diabetes [24]. Pioglitazone was indicated to be as effective as metformin elsewhere [25]. Pioglitazone and sitagliptin are also considered to be effective in patients with poor glycemic control using metformin alone Journal of Endocrinological Investigation Table 2 Changes of clinical data from baseline during treatment with sitagliptin or pioglitazone Week 17 Pioglitazone Week 34 Sitagliptin Pioglitazone Week 52 Sitagliptin Pioglitazone Weight (kg)* 0.4 ± 1.1 − 0.1 ± 0.8 0.7 ± 1.3 − 0.2 ± 0.9 0.9 ± 1.5 BMI (kg/m2) 1 ± 2.8 − 0.4 ± 2 1.9 ± 3.3 − 0.6 ± 2.4 2.3 ± 3.8 WC (cm) − 7.6 ± 85.7 − 0.3 ± 3.5 − 6.8 ± 85.8 − 0.4 ± 4.2 − 6.6 ± 85.8 HC (cm)* 1.1 ± 3 − 0.5 ± 2.9 1.6 ± 4.9 − 1 ± 3.5 2 ± 5.3 FBG (mg/dL)* − 42 ± 47.7 − 30.3 ± 45.3 − 41.4 ± 52.8 − 31 ± 43.8 − 45 ± 52 HbA1C (%) − 1.3 ± 1.1 − 1.1 ± 0.9 − 1.6 ± 1.2 − 1.5 ± 1 − 1.9 ± 1.2 TG (mg/dL)* − 28.1 ± 56.6 − 4.1 ± 56.5 − 21.1 ± 62.9 4.2 ± 64.3 − 22.9 ± 58.6 TC (mg/dL)* − 1.7 ± 27.9 − 7.3 ± 29.4 − 3.1 ± 33.2 − 4.7 ± 31.1 − 3.4 ± 33.7 HDL-C (mg/dL)* 2.4 ± 7 − 0.7 ± 7.8 2.7 ± 7.9 − 0.2 ± 8.2 3.2 ± 10.1 LDL-C (mg/dL)* 8 ± 93.1 − 5.6 ± 23.3 − 0.6 ± 28.4 − 3.2 ± 26.6 − 0.9 ± 29.1 SBP (mmHg)* − 2.1 ± 12 1.5 ± 13.3 0.6 ± 12.7 1.6 ± 15.7 2.4 ± 14.6 DBP (mmHg) − 1.7 ± 6.3 − 1.8 ± 7.9 − 0.3 ± 7.4 − 1.4 ± 9.1 − 0.6 ± 7.8 Sitagliptin − 0.5 ± 1.1 − 1.2 ± 2.8 − 0.1 ± 4.3 − 0.7 ± 3.1 − 35 ± 46.4 − 1.8 ± 1 − 1.7 ± 65.3 − 6.9 ± 34.6 1 ± 9.6 − 6 ± 27.5 3 ± 15.4 − 0.6 ± 8.9 DM diabetes, SBP systolic blood pressure, DBP diastolic blood pressure, BMI body mass index, WC waist circumferences, HC hip circumference, FPG fasting plasma glucose, TC total cholesterol, TG triglyceride, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol *Refers to significant difference between two treatment groups during the study course or metformin in combination with SU [6, 26]. Nevertheless, there are not adequate data to decide which one is the preferred alternative to intensify the metformin and SU dual therapy. In concordance with our results, Sung-Chen et al.’s study showed a similar HbA1C reduction using pioglitazone and sitagliptin in patients with T2DM inadequately controlled with metformin and an SU. They also showed a greater FBG change in the pioglitazone group which was in favor of our results [16]. The desirable effect of pioglitazone and sitagliptin on lipid profile is well established. Sitagliptin is known to be effective on all lipid components [24, 25], while the effect of pioglitazone is mainly limited to TG and HDL-C [23, 27, 28]. Moreover, a recent meta-analysis indicated that sitagliptin is also more beneficial in the improvement of serum TG and HDL-C levels, especially when used in combination with other treatments [29]. Our results extended the present data and showed a positive effect on TG, HDL-C, LDL-C, and TC for both sitagliptin and pioglitazone when added to dual therapy with metformin and SU. Furthermore, we observed that improvements in HDL-C and TG were significantly greater with pioglitazone, while changes in TC and LDL-C were in favor of sitagliptin. In Sung-Chen et al.’s study, the only prominent improvement observed in the pioglitazone group was on TG and HDL-C with no significant difference in other lipid components [16]. It can be deduced that serum levels of lipid parameters are important factors for choosing the third line of treatment. In agreement with the established effect of pioglitazone on body weight [24, 30], we found a significant increase in HC and body weight with pioglitazone during the study course. On the contrary to the ADA/EASD statement [31] on the neutral effect of sitagliptin on weight gain, a significant decrease in body weight and HC was observed in our study. These results led to a significant difference in body weight and HC between the two groups of treatment in favor of sitagliptin. Weight gain resulted in early discontinuation of pioglitazone in nine patients and peripheral edema led to drug withdrawal in six patients in the pioglitazone group, whereas discontinuation in the sitagliptin group was mainly because of cost consideration. Although both drugs were well tolerated during the study and no adverse events were reported, taking all the observations into account, sitagliptin was a better choice according to the side effects in our study. Cardiovascular safety is another noticeable element in diabetes therapy, considering the high risk of cardiovascular events among these patients. Presently, the effect of sitagliptin on blood pressure remains controversial. Some studies indicate an improvement in the blood pressure [32, 33], while others report that sitagliptin either increases the blood pressure [34] or does not affect it [35]. In contrast, literature mostly indicates favorable effects on BP for pioglitazone [36, 37]. In our study, SBP increased in both pioglitazone and sitagliptin groups and the change was greater with pioglitazone. Our results showed no difference regarding DBP between the two groups. Our results confirmed the context-dependent effect of sitagliptin on BP, which could be due to the complex modification in the levels of biologically active peptides that would accompany the inhibition of DPP4 [38]. The observed discrepancy for pioglitazone can be due to racial differences, since it is reported that different ethnicities in different regions or even same geographical regions have different outcomes and drug responses to the same endocrine disease [39, 40]. 13 Journal of Endocrinological Investigation Glycemic durability of sitagliptin and pioglitazone as monotherapies or in combination with other drugs was reported in 4–5 years follow-up studies which all indicated the long-term effect of these two drugs [18, 19]. Ching-Jung et al. showed that in the initial 6 months of sitagliptin therapy, the HbA1C concentration was reduced significantly, but the HbA1C levels fluctuated mildly, followed by a 0.1% increase from 6 to 48 months of therapy [41]. Another trial stated that sitagliptin was effective in patients with suboptimal glycemic control with metformin, SU or dual therapy, maximum between 36 and 48 weeks [42]. Besides, it was found that pioglitazone could result in improved glycemic durability when given in combina- tion with other oral glucose-lowering drugs compared with oral glucose-lowering treatment alone [17]. Considering the drugs’ durability and the fact that present data on the comparison of pioglitazone and sitagliptin efficacy in triple therapy is limited to one 28 weeks length study [16], we decided to follow patients for 52 weeks. We found that HbA1C and FBG decreased in both groups after 4 months and the glucose-lowering effect was maintained through all the follow-ups up to 52 weeks. We did not observe any time–treatment interaction for both sitagliptin and pioglitazone, and no difference in glycemic durability between the two groups was seen. There are several limitations in our study. We did not have a control group, so the changes observed in glycemic or lipid indices in each group from baseline can be due to the placebo effect. However, the main question of the study is to compare pioglitazone and sitagliptin in a triple therapy which is not affected by this limitation. Additionally, we did not have any data about the patients’ physical activity and diet. These two factors could both modify the evaluated variables. Hence, to minimize the potential effects of these factors, patients were given a diet plan, and they were asked to continue the lifestyle they had prior to entry to the trial through the 52 weeks of the study. Moreover, there were slight differences in the baseline characteristics of the two treatment groups, which might have affected the obtained results. All the potential confounders including the whole baseline demographic and measurements were considered in our analysis to minimize the possible effect of basal differences of the two cohorts studied. In conclusion, sitagliptin and pioglitazone were equally effective in improvement of HbA1C. Pioglitazone was superior in reducing the FBG compared to sitagliptin. There were also some differences regarding body weight and SBP in favor of sitagliptin. In the case of lipid control indices, TG and HDL-C were better controlled with pioglitazone, while TC and LDL-C had greater improvement with sitagliptin. The current study confirmed that both sitagliptin and pioglitazone are effective treatment options in patients treated with metformin and SU who require more intensive therapy. 13 This decision should be made separately for each individual based on the baseline characteristics. Acknowledgements The authors wish to thank the patients for their participation and kind cooperation. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Data availability All data generated or analyzed during this study are included in this submitted article. Compliance with ethical standards Conflict of interest The authors declare that they have no competing interest. Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the insti- tutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent Informed consent was obtained from all individual participants included in the study. References 1. Ogurtsova K et al (2017) IDF diabetes atlas: global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Res Clin Pract 128:40–50 2. 2 Diabetes. 2013 Sep. https://www.ncbi.nlm.nih.gov/books/ NBK361996/, C.R.P.i.B.I.O.O.C.A.f.D.a.T.i.H.-O.S.-a.T.-L.T.i.T 3. Ahren B (2010) Use of DPP-4 inhibitors in type 2 diabetes: focus on sitagliptin. Diabetes Metab Syndr Obes 3:31–41 4. Drucker DJ, Nauck MA (2006) The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes. Lancet 368(9548):1696–1705 5. Derosa G et al (2014) Sitagliptin added to previously taken antidiabetic agents on insulin resistance and lipid profile: a 2-year study evaluation. Fundam Clin Pharmacol 28(2):221–229 6. Ba J et al (2017) Randomized trial assessing the safety and efficacy of sitagliptin in Chinese patients with type 2 diabetes mellitus inadequately controlled on sulfonylurea alone or combined with metformin. J Diabetes 9(7):667–676 7. Harashima SI et al (2012) Sitagliptin add-on to low dosage sulphonylureas: efficacy and safety of combination therapy on glycaemic control and insulin secretion capacity in type 2 diabetes. Int J Clin Pract 66(5):465–476 8. Smith U (2001) Pioglitazone: mechanism of action. Int J Clin Pract Suppl 121:13–18 9. Al-Azzam SI et al (2012) Effects of pioglitazone add-on to gliclazide and metformin on glycemic control in patients with type 2 diabetes. Endocr Res 37(1):7–11 10. Filipova E et al (2017) Effects of pioglitazone therapy on blood parameters, weight and BMI: a meta-analysis. Diabetol Metab Syndr 9(1):90 11. Hirose H et al (2002) Effects of pioglitazone on metabolic parameters, body fat distribution, and serum adiponectin levels in Japanese male patients with type 2 diabetes. Metab Clin Exp 51(3):314–317 Journal of Endocrinological Investigation 12. Turner RC et al (1999) Glycemic control with diet, sulfonylurea, metformin, or insulin in patients with type 2 diabetes mellitus: progressive requirement for multiple therapies (UKPDS 49). JAMA 281(21):2005–2012 13. Sharma M et al (2017) Effectiveness of sitagliptin compared to sulfonylureas for type 2 diabetes mellitus inadequately controlled on metformin: a systematic review and meta-analysis. BMJ Open 7(10):e017260 14. Nauck M et al (2014) Efficacy and safety of dulaglutide versus sitagliptin after 52 weeks in type 2 diabetes in a randomized controlled trial (AWARD-5). Diabetes Care 37(8):2149–2158 15. Herz M et al (2003) A randomized, double-blind, placebocontrolled, clinical trial of the effects of pioglitazone on glycemic control and dyslipidemia in oral antihyperglycemic medication-naive patients with type 2 diabetes mellitus. Clin Ther 25(4):1074–1095 16. Liu SC et al (2013) Efficacy and safety of adding pioglitazone or sitagliptin to patients with type 2 diabetes insufficiently controlled with metformin and a sulfonylurea. Endocr Pract 19(6):980–988 17. Stringer F et al (2015) Evaluation of the long-term durability and glycemic control of fasting plasma glucose and glycosylated hemoglobin for pioglitazone in Japanese patients with type 2 diabetes. Diabetes Technol Ther 17(3):215–223 18. Ku EJ et al (2015) Four-Year durability of initial combination therapy with sitagliptin and metformin in patients with type 2 diabetes in clinical practice; COSMIC study. PLoS one 10(6):e0129477 19. Derosa G, D’Angelo A, Maffioli P (2015) Sitagliptin in type 2 diabetes mellitus: efficacy after 5 years of therapy. Pharmacol Res 100:127–134 20. International Diabetes Federation. Recommendations for managing type 2 diabetes in primary care, 2017. http://www.idf.org/ managing-type2-diabetes. Accessed 20 Aug 2018 21. Landman GW et al (2014) Safety and efficacy of gliclazide as treatment for type 2 diabetes: a systematic review and meta-analysis of randomized trials. PLoS One 9(2):e82880 22. Aschner P, Beck-Nielsen H, Bennett P, Boulton A, Colagiuri R, Colagiuri S, Franz M, Gadsby R, Gagliardino JJ, Home P, McGill M, Manley S, Marshall S, Mbanya JC, Neil A, Ramachandran A, Ramaiya K, Roglic G, Schaper N, Siminerio L, Sinclair A, Snoek F, Van Crombrugge P, Vespasiani G, Viswanathan V, Sim K (2014) Global guideline for type 2 diabetes. Diabetes Res Clin Pract 104(1):1–52 23. Ikeda H, Sugiyama Y (2001) Pharmacological effects of a thiazolidinedione derivative, pioglitazone. Nihon Rinsho 59(11):2191–2194 24. Kutoh E (2010) Differential effects of pioglitazone on metabolic parameters in newly diagnosed, drug-naive Japanese patients with type 2 diabetes with or without metabolic syndrome. Endocr Res 35(3):118–127 25. Erem C et al (2014) Comparison of effects of gliclazide, metformin and pioglitazone monotherapies on glycemic control and cardiovascular risk factors in patients with newly diagnosed uncontrolled type 2 diabetes mellitus. Exp Clin Endocrinol Diabetes 122(5):295–302 26. Moses RG et al (2016) A randomized clinical trial evaluating the safety and efficacy of sitagliptin added to the combination of sulfonylurea and metformin in patients with type 2 diabetes mellitus and inadequate glycemic control. J Diabetes 8(5):701–711 27. Baba S (2001) Pioglitazone: a review of Japanese clinical studies. Curr Med Res Opin 17(3):166–189 28. Lu ZH et al (2011) A randomized, double blind, placebo-controlled, parallel and multicenter study to evaluate the safety and efficacy of pioglitazone with sulphonylurea in type 2 diabetic patients. Zhonghua Nei Ke Za Zhi 50(10):826–830 29. Fan M, Li Y, Zhang S (2016) Effects of sitagliptin on lipid Profiles in patients with type 2 diabetes mellitus: a meta-analysis of randomized clinical trials. Med (Baltim) 95(2):e2386 30. Aghamohammadzadeh N et al (2015) The effect of pioglitazone on weight, lipid profile and liver enzymes in type 2 diabetic patients. Ther Adv Endocrinol Metab 6(2):56–60 31. Inzucchi SE et al (2012) Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American diabetes association (ADA) and the European association for the study of diabetes (EASD). Diabetes Care 35(6):1364–1379 32. Yuasa S et al (2017) Primary care-based investigation of the effect of sitagliptin on blood pressure in hypertensive patients with type 2 diabetes. J Clin Med Res 9(3):188–192 33. Hussain M et al (2016) Effect of sitagliptin on glycemic control, body weight, blood pressure and serum lipid profile in type 2 diabetic hyperlipidemic patients. J Ayub Med Coll Abbottabad 28(2):369–372 34. Jackson EK, Dubinion JH, Mi Z (2008) Effects of dipeptidyl peptidase iv inhibition on arterial blood pressure. Clin Exp Pharmacol Physiol 35(1):29–34 35. Zhang X, Zhao Q (2016) Effects of dipeptidyl peptidase-4 inhibitors on blood pressure in patients with type 2 diabetes: a systematic review and meta-analysis. J Hypertens 34(2):167–175 36. Giles TD, Sander GE (2007) Effects of thiazolidinediones on blood pressure. Curr Hypertens Rep 9(4):332–337 37. Takahashi A, Kushiro T (2008) Pioglitazone effects on blood pressure in patients with metabolic syndrome. Nihon Rinsho 66(8):1591–1595 38. Jackson EK et al (2015) Effect of dipeptidyl peptidase 4 inhibition on arterial blood pressure is context dependent. Hypertension 65(1):238–249 39. Chong SA et al (2009) Depressive symptoms and diabetes mellitus in an Asian multiracial population. Asian J Psychiatr 2(2):66–70 40. Golden SH et al (2012) Health disparities in endocrine disorders: biological, clinical, and nonclinical factors–an Endocrine Society scientific statement. J Clin Endocrinol Metab 97(9):E1579–E1639 41. Hsieh CJ, Shen FC (2014) The durability of sitagliptin in elderly patients with type 2 diabetes. Clin Interv Aging 9:1905–1911 42. Mamza J et al (2015) Comparative efficacy of adding sitagliptin to metformin, sulfonylurea or dual therapy: a propensity scoreweighted cohort study. Diabetes Ther 6(2):213–226 13