Efficacy and safety of sodium-glucose cotransporter 2 inhibitors as add-on to metformin and sulfonylurea treatment for the management of type 2 diabetes: a meta-analysis

Abstract

This study evaluates the efficacy and safety of sodium-glucose cotransporter 2 (SGLT2) inhibitors as add-on to metformin and sulfonylurea treatment for type 2 diabetes management. The literature search was conducted in electronic databases and meta-analyses of mean differences in the changes from baseline in selected disease endpoints (efficacy endpoints) or odds ratios (for safety endpoints) were performed to compare outcomes between SGLT2 inhibitor- and placebo-/comparator-treatments. Seven studies (5,143 patients; age 56.75 years [95% CI: 56.19, 57.37]; body mass index 29.53 kg/m² [28.23, 30.83]; and 51.87% [50.46, 53.57] males) were included. Compared to placebo, SGLT2 inhibitors significantly (p < 0.00001) reduced glycated hemoglobin (HbA1c; –0.79% [95% CI: –0.90, –0.68]), fasting plasma glucose (FPG; –1.73 mmol/L [–1.86, –1.60]) and body weight (–1.85 kg [–2.11, –1.59]) after 52–78 weeks of treatment. There were no significant differences in reduction of either HbA1c, FPG or body weight between 18–24 weeks and after 52–76 weeks of treatment. Treatment with SGLT2 inhibitors as add-on to metformin and sulfonylurea was also associated with significant reductions in blood pressure and triglycerides and increase in high-density lipoprotein-cholesterol. Incidence of hypoglycemia was significantly higher, but incidence of hyperglycemia was significantly lower in SGLT2 inhibitor group. Overall, drug-related adverse events were more common in SGLT2 group mainly due to higher incidence of genital tract infections.

TYPE 2 DIABETES MELLITUS (T2DM) is characterized by chronic hyperglycemia frequently associated with dyslipidemia, hypertension, and vascular complications and is a major risk factor for cardiovascular vascular disease [1]. Substantial morbidity and mortality is linked to T2DM [2]. According to International Diabetes Federation, 415 million adults have T2DM and by 2040 the prevalence will rise to 642 million [3]. The prevalence of T2DM is not restricted to later age only rather it is increasing even in young groups. Early-onset T2DM is a more aggressive phenotype with more rapid decline in β-cell function associated with comorbid conditions [4, 5].

Developing countries may face a significant public health crisis from T2DM because it is projected that between 2010 and 2030 there can be a 70% increase in the prevalence of T2DM there as compared to 20% increase for developed countries [6]. Causal factors being the population growth, increase in life expectancy, urbanization, lack of physical inactivity and obesity [7].

For the management of T2DM, primary advice is to adapt a healthy lifestyle and dietary pattern. Metformin, which is an efficacious drug because of its glycemic control, insulin sensitizing and body weight effects, is the first line pharmacotherapy for T2DM [8]. However, metformin cannot provide effect for very long periods which necessitates use of additional drugs. Sulfonylurea drugs (SU) are often added upon metformin’s inadequacy. Glycemic control improves with the addition of SU to metformin, but the efficacy is not persistent in the longer run [9]. Among the management options for T2DM, one strategy recommended by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD) is addition of sodium-glucose cotransporter 2 (SGLT2) inhibitors to ongoing metformin or SU plus metformin, if glycemic goals are not met [10].

Selective and reversible inhibition of SGLT2 lowers blood glucose levels in an insulin-independent fashion and is also found to be favorable for hypertension and body weight control. Chemically, SGLT2 is a high-capacity and low-affinity protein located abundantly in the proximal renal tubules that reabsorb 80–90% of glucose from the glomerular filtrate [11]. Canagliflozin, dapagliflozin, empagliflozin, ipragliflozin, luseogliflozin and tofogliflozin are SGLT2 inhibitor drugs which are well-studied for their efficacy, safety, bioavailability, pharmacokinetic and pharmacodynamic characteristics [12, 13].

In a recent network meta-analyses of several drugs used for the management of T2DM after inadequacy of metformin and SU, SGLT2 inhibitors were found as effective as many other antidiabetic drugs in controlling HbA1c levels but with regards to the incidence of hypoglycemia, weight loss and blood pressure, SGLT2 inhibitors were better than other drugs [14, 15]. Similar results were found in a meta-analysis of randomized controlled trials (RCTs) in which SGLT2 inhibitors were used in combination with metformin against placebo-controlled trials [16]. To our knowledge, there is no systematic review of trials which investigated the efficacy and safety of SGLT2 inhibitors as add-on to metformin and SU treatment. Objective of the present study was to conduct a comprehensive literature search for the identification of trials which recruited T2DM patients to investigate the efficacy and safety of SGLT2 inhibitors as add-on to metformin and SU treatment and to perform meta-analyses of important efficacy and safety endpoints that could display the overall effectiveness of this therapeutic combination.

Methods

This study was performed by following the Cochrane Collaboration’s guidelines for conducting systematic reviews and meta-analyses and is reported in accordance with the PRISMA statement.

Inclusion and exclusion criteria

Inclusion criteria were: Study—a) recruiting adult T2DM patients to evaluate the efficacy and safety of a SGLT2 inhibitor as add-on to metformin and SU treatment by comparing it with either placebo or a suitable non-SGLT2 comparator drug controlled group; and b) reported at least one indicator of disease condition of interest (study endpoints). Exclusion criteria were: relevant studies—a) of less than 16 weeks duration; b) examined the efficacy of SGLT2 inhibitors as add-on to metformin and SU in a single arm trial; c) compared SGLT2 inhibitor monotherapy with metformin and/or SU either alone or in combination with other antidiabetic drugs; and d) compared SGLT2 inhibitors in combination with other non-SGLT2 drugs with any other combination or monotherapy as add-on to metformin and/or SU.

Literature search

Literature search was conducted in Embase, Google Scholar, Medline/PubMed, Scopus, Ovid SP, and Web of Science databases. Medical subject headings (MeSH) and keywords used in different combinations were: sodium-glucose cotransporter-2 inhibitor, SGLT2 inhibitor, dapagliflozin (DAPA), canagliflozin (CANA), ipragliflozin (IPRA), empagliflozin (EMPA), tofogliflozin (TOFO), luseogliflozin (LUSEO), metformin, sulfonyl­urea, type 2 diabetes, randomized trial, efficacy, safety, tolerability, adverse effects, add-on treatment, glycosylated hemoglobin (HbA1c), fasting plasma glucose (FPG), blood pressure, body weight, body mass index (BMI), postprandial glucose (PPG), cholesterol, high density lipoprotein cholesterol (HDL-chol), low density lipoprotein cholesterol (LDL-chol), triglyceride, and estimated glomerular filtration rate (eGFR). The cross references of important articles were also searched. Search encompassed research articles published before October 2017 in English language.

Primary and secondary endpoints

Participants of the included studies were T2DM patients having inadequate control on disease with metformin and SU therapy. Primary outcome measures of interest were the changes from baseline in percent HbA1c, FPG) levels, and body weight. Secondary endpoints were the changes from baseline in systolic and diastolic blood pressure (SBP and DBP), and the blood levels of HDL-chol, LDL-chol, and triglycerides, and eGFR. Safety endpoints were the incidence of hypoglycemia and hyperglycemia, incidence of genital and urinary tract infections, and incidence of ketoacidosis during treatment period.

Quality assessment and data synthesis

Quality assessment of the RCTs included in this meta-analysis was carried out by using the Cochrane Collaboration’s Tool for Quality Assessment of Randomized Controlled Trials. Data pertaining to the participants’ demographic and clinical characteristics, interventions, trial eligibility criteria, outcome measures, and outcomes were extracted from the selected research articles by adapting a standardized procedure. Changes from baseline in the selected endpoints were either extracted raw from the respective research articles if provided, or calculated from baseline and week 18, 24, 52 and 76 of treatment duration values reported as outcomes of the treatments.

Statistical analysis

Data and Analyses module of RevMan software (version 5.3; Cochrane Collaboration) was used for the meta-analyses of weighted mean differences (WMD) between SGLT2 inhibitors and control groups in changing efficacy endpoints or weighted odds ratios (OR) for safety endpoints where overall effect size or subgroup size was an inverse variance weighted average of the individual effect sizes. Between-studies inconsistency (heterogeneity) was tested by I² statistics. Subgroup analyses were performed with regards to control group (placebo vs. DPP4 inhibitors). Subgroup analyses were also performed to examine the effect of BMI (over 30 vs. under 30) in the changes from baseline in % HbA1c, FPG, and body weight between SGLT2 inhibitor- and placebo-treated groups.

Results

Literature search

Seven studies fulfilled the eligibility criteria and were included in this meta-analysis [17-23] (Fig. 1). These studies recruited 3,321 patients to treat with a SGLT2 inhibitor by comparing the outcomes with 1,318 placebo-treated patients and 504 comparator (dipeptidyl dipeptidase 4 (DPP4) inhibitors) treated patients as add-on to metformin and SU treatment.

Fig. 1

A flowchart of study screening and selection process.

Meta-analysis results

Important characteristics of the included studies are presented in Table 1. Age, percentage of males and BMI of the SGLT2 inhibitor treated patients were 56.85 years [95% confidence interval (CI): 56.24, 57.45], 52.12% [50.46, 53.77], and 29.72 kg/m² [28.39, 31.06] and those of control patients were 56.69 years [56.11, 57.26], 51.55% [49.91, 53.20] and 29.36 kg/m² [28.06, 30.65], respectively. Quality of the included RCTs was high when these studies were subjected to the Cochrane Collaboration’s Tool for Quality Assessment of Randomized Controlled Trials (Table 2).

Table 1 Characteristics of the included studies

Study	TD	n		Drug & dose (mg)		Age (years)		Males (%)		White (%)		Black (%)		Asian (%)
		SGLTI	COMP	SGLTI	COMP	SGLTI	COMP	SGLTI	COMP	SGLTI	COMP	SGLTI	COMP	SGLTI	COMP
Cha 2017	24	60	124	DAPA	LINA/GLEMI	52.6 ± 6.5	53.4 ± 7.1	58.3	50					100	100
Haring 2013	24	225	225	EMPA 10	PBO	57 ± 9.2	56.9 ± 9.2	50.22	50	39.56	39	1.33	1	57.33	56
Haring 2013	24	216	225	EMPA 25	PBO	57.4 ± 9.3	56.9 ± 9.2	53	50	39	39	1	1	58	56
Haring 2015	52	225	225	EMPA 10	PBO	57 ± 9.2	56.9 ± 9.2	50.22	50	39.56	39	1.33	1	57.33	56
Haring 2015	52	216	225	EMPA 25	PBO	57.4 ± 9.2	56.9 ± 9.2	53	50	39	39	1	1	58	56
Haring 2015	76	225	225	EMPA 10	PBO	57 ± 9.2	56.9 ± 9.2	50.22	50	39.56	39	1.33	1	57.33	56
Haring 2015	76	216	225	EMPA 25	PBO	57.4 ± 9.2	56.9 ± 9.2	53	50	39	39	1	1	58	56
Ji 2015	18	223	226	CANA 100	PBO	56.5 ± 8.3	55.8 ± 9.4	55.60	55.31					100	100
Ji 2015	18	227	226	CANA 300	PBO	56.4 ± 9.2	55.8 ± 9.4	49.78	55.31					100	100
Matthaei 2015	24	108	108	DAPA 10	PBO	61.1 ± 9.7	60.9 ± 9.2	42.59	55.56	96.23	94.44
Wilding 2013	24	157	156	CANA 100	PBO	57.4 ± 10.5	56.8 ± 8.3	48.41	48.72	84.08	82.05	3.18	6.41	1.27	1.28
Wilding 2013	24	156	156	CANA 300	PBO	56.1 ± 8.9	56.8 ± 8.3	55.77	48.72	81.41	82.05	7.05	6.41	0	1.28
Wilding 2013	52	157	156	CANA 100	PBO	57.4 ± 10.5	56.8 ± 8.3	48.41	48.72	84.08	82.05	3.18	6.41	1.27	1.28
Wilding 2013	52	156	156	CANA 300	PBO	56.1 ± 8.9	56.8 ± 8.3	55.77	48.72	81.41	82.05	7.05	6.41	0	1.28
Schrenthaner 2013	24	377	378	CANA 100	SITA	56.6 ± 9.3	56.7 ± 9.3	54.91	56.88	64.98	63.49	11.40	11.90	17.77	17.19
Schrenthaner 2013	52	377	378	CANA 300	SITA	56.6 ± 9.3	56.7 ± 9.3	54.91	56.88	64.98	63.49	11.40	11.90	17.77	17.19

Abbreviations: CANA, canagliflozin; COMP, comparator; DAPA, dapagliflozin; EMPA, empagliflozin; IPRA, ipragliflozin; PBO, placebo; SGLTI, SGLT2 inhibitor; SITA, sitagliptin; TD, trial duration (weeks).

Table 2 Risk of bias assessment in the included studies

	Other bias	Selective reporting	Incomplete outcome data	Blinding of outcome assessment	Blinding of participants/personnel	Allocation concealment	Random sequence generator
Cha et al., 2017	L	L	L	H	H	H	H
Haering et al., 2013/2015	L	L	L	L	L	L	L
Haring et al., 2015	L	L	L	L	L	L	L
Ji et al., 2015	L	L	L	L	L	L	L
Matthaei et al., 2013	L	L	L	L	L	L	L
Schrenthaner et al., 2013	L	L	L	L	L	L	L
Wilding et al., 2013	L	L	L	L	L	L	L

Legends: H: high risk; L: low risk; U: unclear risk

In these studies, canagliflozin was used in 3 studies, and empagliflozin in 2 and dapagliflozin in 2 studies. Five studies compared a SGLT2 inhibitor against a placebo-controlled group whereas two studies used a DPP4 inhibitor drug as a comparator. Five studies investigated 2 doses of SGLT2 inhibitors.

As add-on to metformin and SU treatment, SGLT2 inhibitors significantly reduced percent HbA1c after 18–24 weeks (–0.64% [–0.71, –0.58]; p < 0.00001) as well as after 52–78 weeks (–0.79% [–0.90, –0.68]; p < 0.00001) in comparison with placebo. There was no significant difference in reduction of HbA1c after 18–24 weeks and after 52–76 weeks of treatment (Fig. 2). Although, SGLT2 inhibitors treatment also reduced HbA1c significantly more than DPP4 inhibitors (–0.28% [–0.36, –0.19]; p < 0.00001), but decrease in HbA1c was significantly more (p < 0.00001) against placebo treated than in DPP4 inhibitors-treated patients.

Fig. 2

Forest graph showing the mean differences in the changes in percent HbA1c between SGLT2 inhibitors and comparators.

Compared to placebo-controlled patients, the SGLT2 inhibitors significantly reduced FPG after 18–24 weeks (–1.58 mmol/L [–1.75, –1.42]; p < 0.00001) as well as after 52–78 weeks (–1.73 mmol/L [–1.86, –1.60]; p < 0.00001) (Fig. 3). There was no significant difference in the reduction of FPG after 18–24 weeks and after 52–76 weeks of treatment. Although, SGLT2 inhibitors treatment also reduced FPG significantly more than DPP4 inhibitors (–1.14 mmol/L [–1.52, –0.77]; p < 0.00001), but decrease in FPG was significantly more (p < 0.00001) against placebo than in DPP4 inhibitors treated patients.

Fig. 3

Forest graph showing the mean differences in the changes in percent fasting plasma glucose between SGLT2 inhibitors and comparators.

The SGLT2 inhibitors also significantly reduced body weight after 18–24 weeks (–1.75 kg [–1.99, –1.51]; p < 0.00001) as well as after 52–78 weeks (–1.85 kg [–2.11, –1.59]; p < 0.00001) in comparison with placebo treated group. There was no significant difference in reduction of body weight after 18–24 weeks and after 52–76 weeks of treatment. Not only the SGLT2 inhibitors treatment reduced body weight significantly more than DPP4 inhibitors (–2.23 kg [–2.56, –1.90]; p < 0.00001), but also the decrease in body weight was significantly more (p < 0.00001) against DPP4 inhibitors than in placebo treated patients (Fig. 4).

Fig. 4

Forest graph showing the mean differences in the changes in body weight between SGLT2 inhibitors and comparators.

No significant differences were found between placebo-controlled subgroup (under 30 BMI vs. over 30 BMI) in mean differences in the changes in % HbA1c, FPG, and body weight.

The SGLT2 inhibitor treatment was also associated with significantly reduced SBP, DBP, and blood triglyc­erides (Table 3). Whereas, SGLT2 inhibitors increased HDL-chol levels, there was no significant change in LDL-chol levels (Table 3). These outcomes did not differ with respect to placebo- and DPP4 inhibitors-treated subgroups.

Table 3 Outcomes of secondary efficacy endpoints showing mean differences between SGLT2 inhibitors-treated and placebo-treated groups

Endpoint	Mean difference [95% CI]	Significance
Systolic blood pressure (mm Hg)	–2.85 [–3.51, –2.19]	p < 0.00001
Diastolic blood pressure (mm Hg)	–0.76 [–1.19, –0.34]	p < 0.0001
HDL-chol (%)	5.98 [4.64, 7.32]	p < 0.00001
LDL-chol (%)	1.79 [–1.05, 4.63]	p = 0.22
Triglycerides (%)	–5.25 [–9.12, –1.38]	p = 0.008
eGFR (mL/min/1.73 m2)	0.11 [–0.79, 1.01]	p = 0.81

Abbreviations: chol, cholesterol; CI, confidence interval; eGFR, estimated glomerular filtration rate; H/LDL, high/low density lipoprotein.

In safety meta-analysis, incidence of hypoglycemia was significantly higher in SGLT2 inhibitors treated patients whereas the incidence of hyperglycemia was significantly higher in the placebo-treated group. Overall, drug-related adverse events were more common in SGLT2 group especially the genital tract infections (Table 4). These outcomes did not differ with respect to placebo-treated and DPP4 inhibitors-treated subgroups.

Table 4 Outcomes of the safety meta-analysis (odds ratios [95% confidence interval]) between SGLT2 inhibitors-treated and placebo-treated groups in the incidence of adverse events (AEs).

Safety endpoint	n	Overall	Low dose	High dose
Any AE/s	2,695	1.02 [0.88, 1.17]; p = 0.82	0.98 [0.80, 1.19]; p = 0.82	1.07 [0.86, 1.32]; p = 0.56
Drug-related AEs	2,695	1.61 [1.37, 1.89]; p < 0.0001	1.63 [1.30, 2.05]; p < 0.0001	1.58 [1.25, 2.00]; p < 0.00001
Serious AEs	2,695	0.68 [0.51, 0.90]; p = 0.007	0.76 [0.52, 1.11]; p = 0.16	0.58 [0.38, 0.89]; p = 0.01
AEs causing discontinuation	2,695	1.09 [0.78, 1.52]; p = 0.62	0.95 [0.59, 1.52]; p = 0.82	1.25 [0.78, 1.99]; p = 0.35
Male genital tract infections	2,695	2.73 [1.29, 5.75]; p = 0.009	3.04 [1.10, 8.38]; p = 0.03	2.38 [0.79, 7.20]; p = 0.12
Female genital tract infections	2,695	4.71 [2.63, 8.44]; p < 0.00001	6.59 [2.57, 16.93]; p < 0.0001	3.63 [1.71, 7.70]; p = 0.00008
Urinary tract infections	2,680	1.31 [1.02, 1.67]; p = 0.03	1.37 [0.98, 1.91]; p = 0.06	1.24 [0.87, 1.77]; p = 0.24
Hypoglycemia	2,244	1.75 [1.43, 2.15]; p < 0.00001	1.88 [1.42, 2.50]; p < 0.001	1.62 [1.20, 2.187]; p < 0.001
Hyperglycemia	1,284	0.30 [0.22, 0.42]; p < 0.00001	0.30 [0.19, 0.47]; p < 0.00001	0.31 [0.20, 0.48]; p < 0.00001

Discussion

This meta-analysis finds that SGLT2 inhibitors as add-on to metformin and SU treatment are significantly more efficacious in controlling T2DM disease indices including HbA1c, FPG, body weight, SBP/DBP, HDL-chol and triglycerides than placebo or DPP4 inhibitors. However, drug-related adverse events were found to be significantly more in SGLT2 inhibitor treated group than in comparators mainly due to the higher incidence of genital tract infections in SGLT2 inhibitor group. Moreover, incidence of hypoglycemia was significantly higher, but the incidence of hyperglycemia was significantly lower in SGLT2 inhibitor-treated patients.

For the management of T2DM, when lifestyle and dietary interventions remain inadequate to control dis‍ease, metformin is the first line drug [8]. However, dis‍ease progression usually makes metformin use in­adequate. Addition of a SU drug to metformin is reported to reduce HbA1c by up to 0.8% after 1 year of treatment [24]. A meta-analysis of 6 studies (1,364 patients) revealed that decrease in HbA1c from baseline was 0.9% but the odds of experiencing a hypoglycemic event was significantly higher in SU treated than in comparator treated patients [25].

Efficacy of SGLT2 inhibitors in improving glycemic control, blood pressure and body weight is also reported in other meta-analytical reviews and is now well-defined [26-28]. Safety concerns, especially with regards to higher incidence of genital tract infections, are also well-pursued in literature and generally glucosauria is considered as the main causal factor; although, immune weakness may also play a role in the etiology [29, 30]. Serious safety issues such as bone fractures, pyelonephritis, urosepsis, and ketoacidosis, are rarely reported with SGLT2 inhibitors. Rather, a robust improvement in cardiovascular outcomes in T2DM patients treated with SGLT2 inhibitor, empagliflozin, supports the use of this therapeutic regimen. Therefore, overall this class of oral anti-diabetic medication is a valuable addition to available treatment options for T2DM that can be a prominent therapy in future [28]. A simulation study conducted in Canada revealed that canagliflozin treatment was associated with better health outcomes and lower costs than sitagliptin as a third-line therapy added-on to metformin and sulfonylurea in T2DM patients [31].

Sulfonylurea use is found to be associated with significantly higher incidence of hypoglycemia than with metformin [32]. Incidence and severity of hypoglycemia is reported to be associated with increased patient anxiety about hypoglycemia and lower health-related quality of life in T2DM patients under metformin and a SU treatment [33]. In comparison with SU, a SGLT2 inhibitor (dapagliflozin) is reported to cause significantly lower incidence of hypoglycemia (7% vs. 29%) [34]. Other RCTs have also found that in combination with metformin, SGLT2 treatment is associated with significantly lower incidence of hypoglycemia in comparison with DPP-4 inhibitors [35, 36] and incretin-based therapies [37, 38]. However, in the present study, the incidence of hypoglycemia was significantly higher in SGLT2 inhibitor group than in the placebo group. Whether there can be any synergistic relationship between SGLT2 inhibitors and SU in this regard needs to be further researched.

Taken together, the SGLT2 inhibitors have insulin-independent mechanisms of action whereas the efficacy of SU drugs depends on adequate insulin-secretory capacity of pancreas. Generally, the SGLT2 inhibitors lowers the risk of hypoglycemia, whereas, SU drugs increase hypoglycemia risk. Using both these regimens in combination may need to adjust SU dosage to mitigate the risk of hypoglycemia. Moreover, increasing concerns about of the safety of SU drugs in the long-run also needs due considerations. Thus, the identification of patients who can be potentially benefitted with such a combination of drugs keeping in view the efficacy as well as safety should be an important consideration of a patient-centered approach to T2DM management.

Conclusion

As add-on to metformin and sulfonylurea, SGLT2 inhibitors are found significantly more efficacious than placebo and DPP4 inhibitors in improving HbA1c, FPG, and body weight. Systolic and diastolic blood pressures, HDL-cholesterol and triglyceride levels also significantly improved with SGLT2 inhibitor treatment in comparison with placebo controlled patients. Incidence of genital tract infections and hypoglycemic events was significantly higher in SGLT2 inhibitor treated group whereas hyperglycemic events were significantly higher in control group.

Acknowledgements

None.

Authors Contributions

J.L., Y.H.S. and X.G.W. conceptualized and designed the study; Y.P.G. and C.L.L. extracted and analyzed data, and drafted the manuscript; Y.H.L. provided technical support in the analyses and writing. All authors agree to be accountable for all aspects of the work.

Funding

No fund was received for this research.

Conflicts of Interest

There is no conflict of interests.

Ethical Privacy Statement

No approval was required for ethical privacy because the study does not involve direct contact with patients or their personal information.

References

• 1   American Diabetes  Association (2012) Diagnosis and classification of diabetes mellitus. Diabetes Care 35: S64–S71.
• 2   Deshpande  AD,  Harris-Hayes  M,  Schootman  M (2008) Epidemiology of diabetes and diabetes-related complications. Phys Ther 88: 1254–1264.
• 3  International Diabetes Federation (2017) IDF Diabetes Atlas. Brussels, Belgium. http://www.idf.org/diabetesatlas
• 4  TODAY Study Group,  Zeitler  P,  Hirst  K,  Pyle  L,  Linder  B, et al. (2012) A clinical trial to maintain glycemic control in youth with type 2 diabetes. N Engl J Med 366: 2247–2256.
• 5  Song SH, Hardisty CA (2009) Early onset type 2 diabetes mellitus: a harbinger for complications in later years—clinical observation from a secondary care cohort. QJM 102: 799–806.
• 6   Shaw  JE,  Sicree  RA,  Zimmet  PZ (2010) Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract 87: 4–14.
• 7   Hu  FB (2011) Globalization of diabetes: the role of diet, lifestyle, and genes. Diabetes Care 34: 1249–1257.
• 8   Scheen  AJ,  Paquot  N (2013) Metformin revisited: a critical review of the benefit-risk balance in at-risk patients with type 2 diabetes. Diabetes Metab 39: 179–190.
• 9   Cook  MN,  Girman  CJ,  Stein  PP,  Alexander  CM,  Holman  RR (2005) Glycemic control continues to deteriorate after sulfonylureas are added to metformin among patients with type 2 diabetes. Diabetes Care 28: 995–1000.
• 10   Inzucchi  SE,  Bergenstal  RM,  Buse  JB,  Diamant  M,  Ferrannini  E, et al. (2015) Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care 38: 140–149.
• 11   Musso  G,  Gambino  R,  Cassader  M,  Pagano  G (2012) A novel approach to control hyperglycemia in type 2 diabetes: sodium glucose co-transport (SGLT) inhibitors: systematic review and meta-analysis of randomized trials. Ann Med 44: 375–393.
• 12   Heerspink  HJ,  Perkins  BA,  Fitchett  DH,  Husain  M,  Cherney  DZ (2016) Sodium glucose cotransporter 2 inhibitors in the treatment of diabetes mellitus: cardiovascular and kidney effects, potential mechanisms, and clinical applications. Circulation 134: 752–772.
• 13   Carlson  CJ,  Santamarina  ML (2016) Update review of the safety of sodium-glucose cotransporter 2 inhibitors for the treatment of patients with type 2 diabetes mellitus. Expert Opin Drug Saf 15: 1401–1412.
• 14   Mearns  ES,  Saulsberry  WJ,  White  CM,  Kohn  CG,  Lemieux  S, et al. (2015) Efficacy and safety of anti­hyperglycaemic drug regimens added to metformin and sulphonylurea therapy in type 2 diabetes: a network meta-‍analysis. Diabetes Med 32: 1530–1540.
• 15   Lozano-Ortega  G,  Goring  S,  Bennett  HA,  Bergenheim  K,  Sternhufvud  C, et al. (2016) Meta-analysis of treatments for type 2 diabetes mellitus following failure with metformin plus sulfonylurea. Curr Med Res Opin 32: 807–816.
• 16   Zhang  Q,  Dou  J,  Lu  J (2014) Combinational therapy with metformin and sodium-glucose cotransporter inhibitors in management of type 2 diabetes: systematic review and meta-analyses. Diabetes Res Clin Pract 105: 313–321.
• 17   Cha  SA,  Park  YM,  Yun  JS,  Lim  TS,  Song  KH, et al. (2017) A comparison of effects of DPP-4 inhibitor and SGLT2 inhibitor on lipid profile in patients with type 2 diabetes. Lipids Health Dis 16: 58.
• 18   Haring  HU,  Merker  L,  Seewaldt-Becker  E,  Weimer  M,  Meinicke  T, et al. (2013) Empagliflozin as add-on to metformin plus sulfonylurea in patients with type 2 diabetes: a 24-week, randomized, double-blind, placebo-controlled trial. Diabetes Care 36: 3396–3404.
• 19   Haering  HU,  Merker  L,  Christiansen  AV,  Roux  F,  Salsali  A, et al. (2015) Empagliflozin as add-on to metformin plus sulphonylurea in patients with type 2 diabetes. Diabetes Res Clin Pract 110: 82–90.
• 20   Ji  L,  Han  P,  Liu  Y,  Yang  G,  Dieu Van  NK, et al. (2015) Canagliflozin in Asian patients with type 2 diabetes on metformin alone or metformin in combination with sulphonylurea. Diabetes Obes Metab 17: 23–31.
• 21   Matthaei  S,  Bowering  K,  Rohwedder  K,  Sugg  J,  Parikh  S, et al. (2015) Durability and tolerability of dapagliflozin over 52 weeks as add-on to metformin and sulphonylurea in type 2 diabetes. Diabetes Obes Metab 17: 1075–1084.
• 22   Schernthaner  G,  Gross  JL,  Rosenstock  J,  Guarisco  M,  Fu  M, et al. (2013) Canagliflozin compared with sitagliptin for patients with type 2 diabetes who do not have adequate glycemic control with metformin plus sulfonylurea: a 52-week randomized trial. Diabetes Care 36: 2508–2515.
• 23   Wilding  JPH,  Charpentier  G,  Hollander  P,  Gonzalez-Galvez  G,  Mathieu  C, et al. (2013) Efficacy and safety of canagliflozin in patients with type 2 diabetes mellitus inadequately controlled with metformin and sulphonyl­urea: a randomised trial. Int J Clin Pract 67: 1267–1282.
• 24   Stargardt  T,  Yin  DD,  Alexander  CM (2009) Treatment choice and effectiveness of adding sulphonylurea or glitazones to metformin for the treatment of type 2 diabetes mellitus. Diabetes Obes Metab 11: 491–497.
• 25   Belsey  J,  Krishnarajah  G (2008) Glycaemic control and adverse events in patients with type 2 diabetes treated with metformin + sulphonylurea: a meta-analysis. Diabetes Obes Metab 10: 1–7.
• 26   Liu  XY,  Zhang  N,  Chen  R,  Zhao  JG,  Yu  P (2015) Efficacy and safety of sodium-glucose cotransporter 2 inhibitors in type 2 diabetes: a meta-analysis of randomized controlled trials for 1 to 2 years. J Diabetes Complications 29: 1295–1303.
• 27   Reed  JW (2016) Impact of sodium-glucose cotransporter 2 inhibitors on blood pressure. Vasc Health Risk Manag 12: 393–405.
• 28   Carlson  CJ,  Santamarina  ML (2016) Update review of the safety of sodium-glucose cotransporter 2 inhibitors for the treatment of patients with type 2 diabetes mellitus. Expert Opin Drug Saf 15: 1401–1412.
• 29   Geerlings  S,  Fonseca  V,  Castro-Diaz  D,  List  J,  Parikh  S (2014) Genital and urinary tract infections in diabetes: impact of pharmacologically-induced glucosuria. Diabetes Res Clin Pract 103: 373–381.
• 30   Arakaki  RF (2016) Sodium-glucose cotransporter-2 inhibitors and genital and urinary tract infections in type 2 diabetes. Postgrad Med 128: 409–417.
• 31   Sabapathy  S,  Neslusan  C,  Yoong  K,  Teschemaker  A,  Johansen  P, et al. (2016) Cost-effectiveness of canagliflozin versus sitagliptin when added to metformin and sulfonylurea in type 2 diabetes in Canada. J Popul Ther Clin Pharmacol 23: e151–e168.
• 32   van Dalem  J,  Brouwers  MC,  Stehouwer  CD,  Krings  A,  Leufkens  HG, et al. (2016) Risk of hypoglycaemia in users of sulphonylureas compared with metformin in relation to renal function and sulphonylurea metabolite group: population based cohort study. BMJ 354: i3625.
• 33   Vexiau  P,  Mavros  P,  Krishnarajah  G,  Lyu  R,  Yin  D (2008) Hypoglycaemia in patients with type 2 diabetes treated with a combination of metformin and sulphonylurea therapy in France. Diabetes Obes Metab 10: 16–24.
• 34   Wan Seman  WJ,  Kori  N,  Rajoo  S,  Othman  H,  Mohd Noor  N, et al. (2016) Switching from sulphonylurea to a sodium-glucose cotransporter2 inhibitor in the fasting month of Ramadan is associated with a reduction in hypoglycaemia. Diabetes Obes Metab 18: 628–632.
• 35   Ferrannini  E,  Berk  A,  Hantel  S,  Pinnetti  S,  Hach  T, et al. (2013) Long-term safety and efficacy of empagliflozin, sitagliptin, and metformin: an active-controlled, parallel-group, randomized, 78-week open-label extension study in patients with type 2 diabetes. Diabetes Care 36: 4015–4021.
• 36   DeFronzo  RA,  Lewin  A,  Patel  S,  Liu  D,  Kaste  R, et al. (2015) Combination of empagliflozin and linagliptin as second-line therapy in subjects with type 2 diabetes inadequately controlled on metformin. Diabetes Care 38: 384–393.
• 37   Nauck  MA,  Del Prato  S,  Meier  JJ,  Durán-García  S,  Rohwedder  K, et al. (2011) Dapagliflozin versus glipizide as add-on therapy in patients with type 2 diabetes who have inadequate glycemic control with metformin: a randomized, 52-week, double-blind, active-controlled noninferiority trial. Diabetes Care 34: 2015–2022.
• 38   Ridderstrale  M,  Andersen  KR,  Zeller  C,  Kim  G,  Woerle  HJ, et al. (2014) Comparison of empagliflozin and gli­mepiride as add-on to metformin in patients with type 2 diabetes: a 104-week randomised, active-controlled, double-blind, phase 3 trial. Lancet Diabetes Endocrinol 2: 691–700.