Evidence Table 4Key Question 1: Studies of rosiglitazone and pioglitazone

Study Characteristics
Author, Year
Trial Name (if app.)
Duration
Country
Funding
Quality
Inclusion and Exclusion CriteriaOverall Sample Size
Interventions
Group Sizes
Baseline Population Characteristics
Mean Age, years
Race/Ethnicity
% Female
Background MedicationsIntermediate Outcomes
HbA1c
Health and Utilization Outcomes
Microvascular Disease
Macrovascular Disease
Lower Extremity Ulcers
All-Cause Mortality
Quality of Life
Hospitalization
Medical Visits (diabetes)
Other
Head-to-head studies
Brackenridge, 2009
Poor
data not abstracted because of poor quality rating
Chogtu, 2009
12 weeks
India
Funding NR
Poor
Inclusion: both genders; age 30–70; T2DM; perscribed glimeperide and required an add-on therapy for poor glycemic control, normotensive, not on antihypertensive or hypolipidaemic drugs

Exclusion: NR
N=63 patients randomized

G1: (pioglitazone, titrated dose + glimeperide 2mg/day) n=28

G2: (rosiglitazone, titrated dose + glimeperide 2mg/day) n=28
NRGlimepirideHbA1c: NR, states that decreases in A1c not significant between groups P >0.05NR
Berneis, 2008
12 weeks
Switzerland
Government funding
Poor
Inclusion: T2DM ≥ 6 months; HbA1c 6.5–9.0; maximum of two oral antidiabetic agents

Exclusion: reated with insulin or glitazones, New York Heart Association stage class III/IV congestive heart failure, active neoplasia, unstable cardiovascular disease, severly impaired liver or kidney function
N=9 patients randomized

NA; all one group. Patients randomized to either receive pioglitazone (30mg/day × 4 weeks followed by 45mg/day × 8 weeks) or rosiglitazone (4mg/day × 4 weeks followed by 8mg/day × 8 weeks). Then crossed-over to other group after washout
Identical groups (cross-over).

Age: 61
Race: NR
Female: 44.4%
NRHbA1c: mean change after 12 weeks of treatment
G1: −0.54

G2: −0.59

NS vs. baseline

NS, G1 vs. G2
NR
Beysen, 2008
20 weeks
US
Funding NR
Fair
Inclusion: T2D; HbA1c > 7.5 or fasting glucose > 180mg/dl; not controlled with metformin alone or metformin in combination with sulfonylurea; hypertriglyceridemia (150–400mg/dl)

Exclusion: pregnancy; ALT>1.5 times normal upper limit; creatinine > 1.4mg/dl; congestive heart failure; history of coronary artery, pulmonary or neurological disesae; treatment with insulin; treatment with statin or fibric acid derivative within 2 months of study
N=12 patients randomized

G1: (rosiglitazone 15–30mg/day × 4 weeks, 45mg/day × 16 weeks) n=6

G2: (pioglitazone 4mg/day × 4 weeks, 8mg/day × 16 weeks) n=6
G1:
Age: 56
Race: NR
Female: 66.7%

G2:
Age: 53
Race: NR
Female: 50%
NRHbA1c: mean change after 20 weeks of treatment:
G1: −1.3 (SD 0.8)

G2: −1.1 (SD 0.6)

P <0.05, G1/G2 vs. baseline

NS, G1 vs. G2
NR
Vijay, 2009
16 weeks
India
UGC, India
Fair
Inclusion: HbA1c > 8%; cardiovascular risk factors; age 30–70; BMI <36; stable body weight for 3 months prior to study

Exclusion: Hepatic or other preexisting chronic disease; any smoking in 6 months prior to study; previous use of insulin or thiazonlidinediones; history of stroke; patients taking glucocorticoids or other drugs that affect glucose metabolism, lipid lowering drugs, alcohol, or psychoactive substances.
N=50

G1: (pioglitazone 30–45mg/day)
n=20

G2: (rosiglitazone 4–8mg/day)
n=20

G3: (controls (sulfonylureas/other secretagogues))
n=10
G1:
Age: 48.1
Race/Ethnicity: NR Female: NR

G2:
Age: 47.75
Race/Ethinicity: NR Female: NR

G3:
Age: 49.7
Race/Ethnicity: NR Female: NR

Study reports that overall the male/female ratio was 3:2
NRG1: −1.27 (SD NR)
P =0.00, G1 vs. baseline

G2: −1.26 (SD NR)
P =0.00, G2 vs. baseline

G3: −0.94 (SD NR)
P =0.00, G3 vs. baseline

NR, G1 vs. G2 vs. G3
NR
Oz, 2008
12 weeks
Turkey
Funding NR
Fair
Inclusion: Newly diagnosed T2DM (<6 months)

Exclusion: Impaired hepatic function or renal function; Serious cardiovascular disease including heart failure, history of myocardial infarction or stroke; Pregnant or breastfeeding women; Severe anemia
N=35

G1: (pioglitazone 30mg/day)
n=14

G2: (rosiglitazone 4mg/day)
n=11

G3: (placebo + medical nutrition therapy)
n=10
Age: 55.2
Race/Ethnicity: NR Female: 49%

NR by individual groups
NRG1: −1.22 (SD NR)
P =0.003, G1 vs. baseline

G2: −0.8 (SD NR)
P =0.019, G2 vs. baseline

G3: 0.06 (SD NR)
NS, G3 vs. baseline

NR, G1/G2 vs. G3
NR
OZ Gul, 2010
12 weeks
Turkey
Funding NR
Fair
Inclusion: Newly diagnosed type 2 diabetes mellitus (<6 months) naïve to prior antidiabetic therapy

Exclusion: Taking statins, ACE inhibitors, ARBs; Acute complications with need for insulin therapy; Impaired hepatic function or renal function; Severe anemia; Serious cardiovascular disease including heart failure, history of myocardial infarction or stroke; Pregnant or breastfeeding women.
N=60

G1: (pioglitazone 30mg/day)
n=19

G2: (rosiglitazone 4mg/day)
n=20

G3: (placebo + medical nutrition therapy)
n=21
Age: 56.4
Race/Ethnicity: Turkish 100%
Female: 42%

NR by individual groups
NoneHbA1c: mean change at 12 weeks:
G1: −1.1 (SD NR)
P <0.001, G1 vs. baseline

G2: −1.1 (SD NR)
P =0.003, G2 vs. baseline

G3: −0.1 (SD NR)
NS, G3 vs. baseline

NR, G1/G2 vs. G3
NR
Active-control studies
Erdem, 2008
12 weeks
Turkey
Gulthane School of Medicine
Poor
Inclusion: age between 30 and 70 years, body mass index (BMI) less than 35 kg/m2, no other illnesses including liver failure, renal failure, heart failure or other chronic disease as determined by history, physical examination, and screening tests

Exclusion: NR
N=53 patients randomized

G1: (pioglitazone 15mg/day, titrated up to 45 mg in 15mg increments if mean serum glc >110 mg/dL)
n=21

G2: (Metformin, 1000mg/day; up to 2000mg if mean serum glc >110mg/dL)
n=23
G1: Age: 54.9
Race: NR
Female: 62%

G2: Age: 55.1
Race: NR
Female: 52%
NRChange in HbA1c at 12 weeks:
G1: −.74, calculated
G2: −0.59, calculated.

G1 vs baseline: P =0.01
G2 vs baseline: P =0.02
NR
DeFronzo, 2010
20 weeks
USA
Eli Lilly
Fair
Inclusion: age 18 –75 years, BMI 25–40 kg/m2, stable body weight for at least 6 months prior to screening, A1C 6.8–10.0%, stable dose of metformin for at least 6 weeks prior to screening and no treatment with any other antidiabetic medication, and absence of islet cell autoantibodies.

Exclusion: NR
N = 137

G1: Exenatide 10mcg
N = 45

G2: Exenatide 10mcg + Rosiglitazone 4mcg
N = 47

G3: Rosglitazone 4mg
N = 45
Baseline characteristics not reported for each arm. For entire study population:
Mean age 56 yrs
61% white
49% female
MetforminChange in HbA1c
G1: −0.9 (0.1)
G2: −1.3 (0.1)
G3: −1.0 (0.1)
G1 vs. G2 P = 0.016
G1 vs. G3 P = 0.720
G3 vs. G2 P = 0.039

Change in weight
G1: −2.8 (0.5)
G2: −1.2 (0.5)
G3: +1.5 (0.5)
G1 vs. G2 P = 0.038
G1 vs. G3 P < 0.001
G3 vs. G2 P < 0.001
NR
Gerstein, 2010
APPROACH
18 months
Multinational
GlaxoSmithKline
Fair
Inclusion: 30–80y; established T2DM; clinically indicated coronary angiography or percutaneous coronary intervention; ≥ atherosclerotic plaque with 10%–50% luminal narrowing in a coronary artery that had not undergone intervention and if their DM was treated with either lifestyle approaches alone or with oral agents.

Exclusion: ST-segment elevation myocardial infarction in past 30 days; coronary artery bypass graft surgery; severe valvular heart disease; left ventricular efection fraction <40%; any heart failure NY Heart Association class I–IV; systolic blood pressure >170 mmHG or diastolic blood pressure > 100 mm Hg; serum creatinine ≥ 1.5 mg/dL for men; serum creatinine ≥ 1.4 mg/dL for women; active liver disease
N = 672

G1: Glipizide (10–15mg/d)
N = 339

G2: Rosiglitazone (4–8mg/d)
N = 333
G1: Age 60.2, Race NR, Female 34.2%

G2: Age 61.8, Race NR, Female 30.0%

Age, G1 vs. G2, p = 0.03
Metformin max 2550 mg/d and once-daily basal insulin or both if needed to maintain a HbA1c of ≤ 7%Change in HbA1c
G1: −0.2 SD NR
G2: −0.3 SD NR
G1 vs. G2, P = 0.44
Composite of all-cause death, nonfatal myocardial infarction, nonfatal stroke, coronary revascularization, or hospitalization for myocardial ischemia, No. of patients (%):
G1: 38 (11.2)
G2: 39 (11.7)
G1 vs. G2, P = 0.58

Composite of cardiovascualr death, nonfatal myocardial infarction, or nonfatal stroke, No. of patients (%)
G1: 10 (2.9)
G2: 14 (4.2)
G1 vs. G2, P = 0.31

Cardiovascular Death, No. of patients (%)
G1: 3 (0.9)
G2: 4 (1.2)
G1 vs. G2, P =0.50

Myocardial Infarction, No. of patients (%) - Nonfatal
G1: 6 (1.8)
G2: 7 (2.1)
G1 vs. G2, P =0.71

Myocardial Infarction, No. of patients (%) - Fatal
G1: 1 (0.3)
G2: 1 (0.3)
G1 vs. G2, P =0.89

Stroke, No. of patients (%) - Nonfatal
G1: 1 (0.3)
G2: 5 (1.5)
G1 vs. G2,P = 0.13

Stroke, No. of patients (%) - Fatal
G1: 0 (0)
G2: 0 (0)
G1 vs. G2, P = NR

Coronary Revascularization No. of patients (%) -
G1: 27 (8.0)
G2: 26 (7.8)
G1 vs. G2, P = 0.82

All cause mortality:
No. of patients (%)
G1: 7 (2.1)
G2: 8 (2.4)
G1 vs. G2, P =0.72

Hospitalization for myocardial ischemia, No. of patients (%)
G1: 7 (2.1)
G2: 11 (3.3)
G1 vs. G2, p = 0.25

Congestive heart failure, No. of patients (%)
G1: 3 (0.9)
G2: 8 (2.4)
G1 vs G2 P = 0.14
Kadoglou, 2010
14 weeks
Greece
European Social Fund and National Resources - PYTHAGORAS II & Alexander S Onassis Public Benefit Foundation
Fair
Inclusion: 50–70 y; T2DM; treated with metformin (850 mg/d) alone for ≥ 4 mo; HbA1c > 6.5%; BMI > 25 kg/m2

Exclusion: Creatinine > 2mg/dL; Alanine amino transferase > 3 times higher than the upper normal limit; congestive heart failure (NY Heart Association II-IV); Prior TZD treamtent; >5% change in body weight for up to 4 mo prior study initiation.
N = 100

G1: Rosiglitazone(8 mg/d) + Metformin (850 mg/d)
N= 50
analyzed = 49

G2: Metformin (titration from 850 mg/d – 2550 mg/d)
N = 50
analyzed = 48
G1: Age 62, Race NR, Female 74%
G2: Age 62.7, Race NR, Female 67%
Metformin 850 mg/dChange in HbA1c (calculated change from baseline)
G1: −0.87 SD NR, P <0.001
G2: −0.54 SD NR, P = 0.014
G1 vs. G2, P =0.291
NR
Kato, 2009
12 weeks
Japan
NR
Fair
Inclusion: Recent diagnosis of T2DM associated with metabolic syndrome; Abdominal ultrasound determining fatty liver; no history of treamtne with oral antihyperglycemic drugs, antihyperlipidemic drugs, or antihypertensive drugs.

Exclusion: Diabetic retinopathy, nephropathy, or neuropathy whose condition was unstable or underwent sudden progression; Aspartate aminotransferase or alanin aminotransferase > 1.5 times the upper limit of normal level; serum creatinine > 133 μmol/L; anemia; myocardial infar4ction; angina pectoris; congestive heart failure; history of cerebrovascular disease.
N = 50

G1:Pioglitazone (15mg/d)
N = 25

G2: Metformin (500 mg/d)
N = 25
G1: Age 51.4, Race NR, Female 52%
G2: Age 58.6, Race NR, Female 44%
All patients received diet therapy and exercise therapy.
Parameters: total energy intake within 1200–1800kcal, fat ration of caloric intake to < 25–30% and to do ≥ 150 min of exercise per wk.
Change in HbA1c

Change from baseline [%]
G1: −1.05 (P <0.01 within group)

G2: −0.83 (P <0.01 within group)

P = NS between groups
NR
Papathanassiou, 2009
6 months
Greece
Funding NR
Fair
Inclusion: T2DM treated only with metformin for 6 months prior to study; HbA1c > 6.5%; normal liver enzymes and renal function

Exclusion: History of coronary artery, cerebrovascular, or peripheral vascular disease; chronic heart failure; liver or renal disease; anemia; thyroid dysfunction; and the new onset of any medications within the previous 8 weeks.
N=28

G1: (glimepiride 4mg/day)
n=14

G2: (pioglitazone 30mg/day)
n=14
G1:
Age: 63.6
Race/Ethnicity: NR Female: 78.6%

G2:
Age: 62.8
Race/Ethnicity: NR Female: 78.6%
MetforminG1: −0.56 (SD 0.57)

G2: −0.60 (SD 0.85)

P =0.398, G1 vs. G2
NR
Perez, 2009
24 weeks
Multinational
Takeda
Fair
Inclusion: 18 y; TWDM; baseline HbA1c ≥ 7.5% but ≤ 10.0%; treatment -naïve; BMI ≤ 45 kg/m2; received counseling on lifestyle modification for T2DM including diet and exercise

Exclusion: Type 1 diabetes; NY Heart Association Class II or IV heart failure; history of myocardial infarction, cerebrovascular accident, percutaneous coronary intervention, coronary arter bypass graft, transient ischemic attach within 6 mo; serum creatinine level males ≥ 1.5 mg/dL; serum creatinine level females ≥ 1.4 mg/dL; triglyceride level >500mg/dL; ALT level > 2.5 times upper limit of normal; active liver disease; jaundice; discontinuation from TZD or metformin therapy due to lack of efficacy; clinical or laboratory signs of intolerance of TZD or metformin; pregnant; intent to become pregnant; lactating during the study period.
N = 600

G1: Pioglitazone (15mg) + Metformin (850mg) bid:
N = 201

G2: Pioglitazone (15mg) bid
M = 189

G3: Metformin (850mg) bid
N = 210
Overall: Age 54.1, American Indian 32%, Asian 2.2%, Black 6.5%, White 89.0%, Multiracial 29.7%, Hispanic/Latino 25.5%, Non-hispanic/non-Latino 20.7% Female 57.7%

G1: Age 54.7, American Indian 31.3%, Asian 1.5%, Black 6.0%, White 91.5%, Multiracial 30.3%, Hispanic/Latino 24.4%, Non-hispanic/non-Latino 20.9% Female 55.2%

G2: Age 54.0, American Indian 32.8%, Asian 2.6%, Black 6.9%, White 87.3%, Multiracial 29.6%, Hispanic/Latino 25.9%, Non-hispanic/non-Latino 19.0% Female 65.1%

G3: Age 53.7, American Indian 31.9%, Asian 2.4%, Black 6.7%, White 88.1%, Multiracial 29.0%, Hispanic/Latino 26.2%, Non-hispanic/non-Latino 21.9% Female 53.3%
None% Change in HbA1c
G1: −1.83% SD NR
G2: −0.96% SD NR
G3: −0.99% SD NR
G1 vs. G2, P < 0.0001
G1 vs. G3, P < 0.0001
Coronary Artery Disease (No. of patients)
G1: 0
G2: 2
G3: 0

Myocardial Infarction(No. of patients)
G1: 0
G2: 1
G3: 0

Anterior bundle Branch Block (No. of patients)
G1: 0
G2: 1
G3: 0

Myocardial Ischemia (No. of patients)
G1: 0
G2: 0
G3: 1
Petrica, 2009
Poor
data not abstracted because of poor quality rating
Rigby, 2010
16 weeks
Multinational
Daiichi Sankyo
Fair
Inclusion: Male and female; 18–80; T2DM diagnosis; HbA1c 7.0% – 10.0%; Taking a dose of metformin 1500–2550 mg/d; LDl cholesterol ≥ 60 mg/dL; Triglycerides < 500 mg/dL

Exclusion: LDL Cholesterol < 60 mg/dL; Triglycerides ≥ 500 mg/dL; BMI > 40kg/m2; History of type 1 DM; Ketoacidosis; Insulin therapy > 2mo; dysphagia; swallowing disorders; intestinal motility disorders; pancreatitis; AIDS/HIV; drug or alcohol abuse w/in 2 yrs; Allergic/toxic response to colesevelam; Current treatment with TZD, colesevelam, or FDCP including metformin; Pulmonary, hepatic, gastrointestinal, uncontrolled endocrine/metabolic, hematologic/oncologic, neruologic, or psychiatric disease; Acute coronary syndrome, coronary intervention, congestive heart failure, transient ischemic attack within 3 mo; hospitalization within 14 days; participation in a weight-loss program or intensive exercise program
N = 169

G1: Rosiglitazone (4mg/d)
N = 56

G2: Sitagliptin (100 mg/d)
N = 56

G3: Colesevelam (3.75 g/d)
N = 57

G1: Age 54.7; White 28.6%, Black 3.6%, Asian 0%, Hispanic 67.9%, Multiple 0%, Other 0%; Female 58.9%

G2: Age 54.8; White 23.2%, Black 1.8%, Asian 0%, Hispanic 73.2%, Multiple 0%, Other 1.8%; Female 64.3%
Metformin (1500–2550 mg/d)Change in HbA1c
Least-squares mean change from baseline (95% CI):
G1: −0.6% (−0.83 to −0.32), P < 0.001

G2: −0.4% (−0.64 to −0.13), P = 0.009
NR
Tolman, 4039
Poor
data not abstracted because of poor quality rating
Tsuchiya, 2009
Poor
data not abstracted because of poor quality rating
Van der Meer, 2009
PIRAMID
24 weeks
The Netherlands Eli Lilly, Takeda
Good
Inclusion: Men with uncomplicated T2DM; ages 45–65; HbA1c 6.5–8.5; BMI 25 to 32; Blood pressure lower than 150/85

Exclusion: Any clinically significant disorder; particularly any history of cardiovascular or liver disease or diabetes-related complications; any prior use of thiazolidinediones or insulin.
N=78

G1: (pioglitazone 30mg/day)
n=39

G2: (metformin 2000mg/day)
n=39
G1:
Age: 56.8
Race/Ethnicity: NR Female: 0%

G2:
Age: 56.4
Race/Ethinicity: NR Female: 0%
Glimepiride monotherapy, titrated during the 10-week run-in periodG1: −0.6 (SD NR)
P <0.001, G1 vs. baseline

G2: −0.7 (SD NR)
P <0.001, G2 vs. baseline

P =0.146, G1 vs. G2
NR
Schernthaner, 2004
Quarter Study
12 months
Multinational
Funding NR
Good
Inclusion: age 35–75; T2DM; inadequately treated with diet alone, HbA1c between 7.5% and 11% with stable or worsening glycemic control for at least 3 months

Exclusion: prior use of glucose-lowering pharmacotherapy; specific contraindications to either drug
N=1199 randomized

G1: (pioglitazone 30–45mg/day + placebo)
n=597

G2: (metformin up to 850mg–2550mg/day + placebo)
n=597
G1:
Age: 57
Race: NR
Female: 47.4%

G2:
Age: 56
Race: NR
Female: 42.2%
NoneG1: −1.41 (SD 0.04)

G2: −1.50 (SD 0.04)

90% CI (−0.01, 0.19), difference between G1 vs. G2
Microvascular disease:
Albumin:Cr ratio
G1: −19%
G2: −1%
P =0.002, G1 vs. G2

All-cause Mortality
G1: n=3
G2: n=2
Kusaka, 2008
4 months
Japan
Funding NR
Fair
Inclusion: T2DM; inadequate glucose control

Exclusion: cardiovascular disease; apparent liver or kidney disease; severe diabetic complications
N=35 patients randomized

G1: (metformin 750mg/day)
n=17

G2: (pioglitazone 15–30mg/day)
n=16
G1:
Age: 60
Race: NR
Female: 41.2%

G2:
Age: 64
Race: NR
Female: 43.8%
Patients stayed on sulfonylurea if on them (82% and 75%, respectively)HbA1cmean change at 4 months:
G1: −1.0 (SD NR)

G2: −1.1 (SD NR)

P <0.0005, G1&G2 vs. baseline

significance NR for G1 vs. G2
NR
Nissen, 2008
PERISCOPE
18 months
Multinational
Takeda
Pharmaceuticals
Fair
Inclusion: age 35–85; HbA1c 6.0–9.0 (if taking glucose-lowering meds) and 6.5–10.0 (if not); one angiographic stenosis at least 20% narrowing; a “target vessel” for ultrasound was required to have less than 50% obstruction throughout a 40mm or longer segment

Exclusion: T1DM; 3 or more antidiabetic meds; received any TZD within 12 weeks; serum creatinine > 2.0mg/dL; triglycerides > 500mg/dl; blood pressure >160/100 despite therapy; active liver disease; left main coronary artery stenosis more than 50%;
N=547 patients randomized

G1 (glimepiride titrated): 273 randomized, 181 included in primary analysis

G2 (pioglitazone titrated): 274 randomized, 179 included in primary analysis
G1:
Age: 59.7

Race: White 80.6%, Black 9.9%, Asian 5.9%, Native American 3.7%;

Female: 34.1%

G2:
Age: 60.0

Race: White 83.3%, Black
11.1%, Asian 4.4%, Native American 1.1%

Female: 31.1%
Patients stayed on baseline therapy (unless a TZD or sulfonylurea)HbA1c mean change (95% CI at 18 months):

G1: −0.36 (−0.48, −0.24)

G2: −0.55 (−0.68, −0.42)

P =0.03, G1 vs. G2
All-cause Mortality
G1: n=2
G2: n=3
Marre, 2009
LEAD-1SU
26 weeks
Multinational
Novo Nordisk
Fair
Inclusion: T2DM treated with oral glucose-lowering agents (OGLAs) for ≥ 3 months; 18–80 years of age; HbA1c 7.0–11.0% (previous OGLA monotherapy) or 7.0–10.0% (previous OGLA combination therapy); body mass index (BMI) ≤ 45.0 kg/m 2.

Exclusion: Insulin within 3 months, impaired liver or renal function, uncontrolled hypertension (≥ 180/100 mmHg), cancer or used any drugs apart from OGLAs likely to affect glucose concentrations
N=1041

G1: (liraglutide 0.6 mg)
n=233

G2: (liraglutide 1.2 mg)
n=228

G3: (liraglutide 1.8 mg)
n=234

G4: (placebo) n=114

G5: (rosiglitizone)
n=232
Age: 56

Race/Ethnicity: NR

% Female:
G1: 46
G2: 55
G3: 47
G4: 53
G5: 53
glimepiride (2–4 mg)HbA1c

G1: −0.6% vs. placebo −0.8% (−1.1, − 0.6) P <0.0001

G2: −1.1% vs. placebo −1.3% (1.5, − 1.1) P <0.0001

G3: −1.1% vs. placebo −1.4% (1.6, − 1.1) P <0.0001

G4: +0.2%

G5: −0.4% vs. placebo −0.7% (−0.9, − 0.4) P <0.0001
NR
Bao, 2009
48 weeks
China
Government funding
Poor
Inclusion: newly diagnosed T2D; no previous treatment with hypoglycemic agents or lipid drugs, HbA1c > 6.5 during wash out period

Exclusion: NR
N=82 patients randomized

G1 (repaglinide 1.5– 6mg/day): n=35

G2 (metformin 0.75–1.5mg/day): n=22

G3: (rosiglitazone 4–8mg/day): n=25
NRNRHbA1c: mean change at 48 weeks:
G1: −2.15 (SD NR)

G2: −1.92 (SD NR)

G3: −2.47 (SD NR)

P <0.01, G1/G2/G3 vs. baseline

NS, G1 vs. G2 vs. G3
NR
Pop-Busui, 2009
6 months
US
GlaxoSmithKline, Eli Lilly, Research Foundations
Fair
Inclusion: Subjects with T2DM without known coronary artery disease; HbA1c between 6% and 9%; treatment with diet/exercise or sulfonylurea therapy or insulin < 20 U/d; If previously on metformin, 4- wk washout period prior to study.

Exclusion: NR
N = 27

G1: (rosiglitazone 8mg/day)
n=14

G2: (glyburide 10mg/day)
n=13
Age: 49.5
Race/Ethnicity: NR Female: 48%

NR by individual groups
NRG1: − 0.5 (SD NR)

G2: −0.9 (SD NR)

NS, G1 vs. G2
NR
Turkmen Kemal, 2007
Poor
data not abstracted because of poor quality rating
von Bibra, 2008
16 weeks (32-week cross-over)
Germany
Funding NR
Fair
Inclusion: T2DM of relative short duration; taking metformin monotherapy; age 35–75; BMI 25–35; HbA1c 6.5–9%; no major complications of macrovascular disease; normal left ventricular function by 2-dimensional echocardiography; blood pressure normal or <140/90 if treated; cholesterol <250 mg/dL; triglyceride <250mg/dL; no microvascular complications and no albuminuria

Exclusion: Atrial fibrillation; ischemic heart disease; severe left ventricular hypertrophy; history or signs of heart failure; hepatic, or renal insufficiency
N=12

G1: (rosiglitazone 8mg/day)
n=12

G2: (glimpiride 3mg/day)
n=12
Age: 59 y
Race/Ethnicity: NR Female: 33.3%
Metformin and other previous medications continuedHbA1c:
G1: −0.4 (SD NR)
P =0.208, G1 vs. baseline

G2: −0.2 (SD NR)
P =0.196, G2 vs. baseline

NR, G1 vs. G2
Iliadis, 2007
18 weeks
Greece
Funding NR
Poor
Inclusion: recently diagnosed T2DM (<3years); not on any anti-diabetic medication; fasting hyperglycemia after 1 month of intensive dietary intervention

Exclusion: renal and liver impairment
48 patients randomized, 41 patients included in analysis

G1 (diet alone):
n=12

G2 (diet + rosiglitazone 8mg/day): n=14

G3 (diet + metformin 1700mg/day): n=15
G1:
Age: 58.0
Race: NR
Female: NR

G2:
Age: 56.3
Race: NR
Female: NR

G3:
Age: 57.8
Race: NR
Female: NR
noneHbA1C
mean change at 18 weeks:
G1: −0.6 (SD 1.8)

G2: −1.0 (SD 0.7)

G3: −1.7 (SD 1.1)

NS, G1 vs. baseline

P <0.01, G2 vs. baseline

P <0.001, G3 vs. baseline

NR, G1 vs. G2 vs. G3
Home, 2009
RECORD
7 year study, mean follow up time 5.5 years
Multinational
GlaxoSmithKline
Fair
Inclusion: age between 40–75; BMI > 25; being on maximum tolerated doses of metformin or a sulfonylurea monotherapy

Exclusion: hospitalizations for a major cardiovascular event in prior 3 months; planned cardiovascular intervention; presence, history or treatment for heart failure
N=4458 randomized

G1 (addition of rosiglitazone)
n=2,220

G1a (rosiglitazone + metformin)
n=1,117

G1b (rosiglitazone + sulfornylurea)
n=1,103

G2 (metformin + sulfonylurea)
n=2,227

G2a (background metformin)
n=1,105

G2b (background sulfonylurea)
n=1,122
G1a: Age: 57.0; White 98.9%, Other 1.1%; Female 46.2%

G1b: Age: 59.8; White 99.3%, Other 0.7%; Female 51.0%

G2a: Age: 57.2; White 98.4%; Other 1.6%; Female 47.1%

G2b: Age: 59.7; White 99.1%; Other 0.9%; Female 49.4%
all patients stayed on their metformin or sulfonylurea that they used as monotherapyHbA1c mean change at 5 years:
G1a: −0.28 (SD 0.03)
G2a: 0.01 (SD 0.04)

P <0.0001, G1a vs. G2a

G1b: −0.44 (SD 0.03)
G2b: −0.18 (0.04)

P <0.0001, G1b vs. G2b
Cardiovascular death or hospitalization:
G1: n=321
G2: n=323
HR = 0.99 (0.85–1.16)

CV death:
G1: n=60
G2: n=71
HR 0.84 (0.59–1.18)

Myocardial infarction:
G1: n=64
G2: n=56
HR 1.14 (0.80–1.63)

Stroke:
G1: n=46
G2: n=63
HR: 0.72 (0.49–1.06)

All-cause Mortality
G1: n=136
G2: n=157
HR 0.86 (0.68–1.08)
Kiyici, 2009
12 months
Turkey
Funding NR
Fair
Inclusion: age 30–65; baseline HbA1c<8; BMI < 40

Exclusion: usage of any medications for T2DM before study; presence of cardiovascular, gastrointestinal, hepatic, renal, rhematologic, neoplastic, infectious or other endocrine diseases (except hyperlipidemia), micro or macrovascular complications of diabetes, previous history of substance abuse
N=50 randomized

G1 (medical nutrition therapy):
n=15

G2 (metformin + medical nutrition therapy):
n=16

G3 (rosiglitazone + medical nutrition therapy):
n=19
G1: Age: 52.1; Race NR; Female NR

G2: Age: 52.4; Race NR; Female NR

G3: Age 50.7; Race NR; Female NR
noneHbA1c mean change at 12 months:
G1: +0.1 (SD NR)
G2: −0.3 (SD NR)
G3: −0.7 (SD NR)

P =0.014, G2/G3 vs. G1
NR
Scott, 2008
18 weeks
Multinational
Merck
Fair
Inclusion: age 18–75; taking metformin monotherapy >1500mg/day for at least 10 weeks prior to screening; HbA1c 7–11%

Exclusion: T1DM; insulin use within 8 weeks of the screening visit; contraindications for use of TZDs or metformin; impaired renal function, ALT or AST levels more than 2-fold the upper limit of normal, fasting glucose values >270mg/dl
N=273 randomized

G1 (placebo): n=92

G2 (sitagliptin 100mg/day): n=94

G3 (rosiglitazone 8mg/day): n=87
G1: Age: 55.3; White 61%, Asian 39%, Other 0%; Female 41%

G2: Age: 55.2; White 61%, Asian 38%, Other 1%; Female 45%

G3: Age: 54.8; White 59%, Asian 38%, Other 3%; Female 37%
metforminHbA1c mean change from baseline (95% CI):
G1: −0.22 (−0.36, −0.08)

G2: −0.73 (−0.87, −0.60)
G2 vs. G1: −0.51 (−0.70, −0.32)

G3: −0.79 (−0.92, −0.65)
G3 vs. G1: −0.57 (−0.76, −0.37)
G3 vs. G2: −0.06 (−0.25, 0.14)
NR
Hamann, 2008
52 weeks
Multinational
Funding NR
Fair
Inclusion: BMI ≥ 25 T2DM; HbA1c 7–10; received metfromin for at least 8 weeks prior to screening

Exclusion: used any oral diabetic drug other than metformin in last 12 weeks; insulin at any time other than pregnancy or emergency; history of metabolic acidosis; edema requiring treatment; anemia; renal or hepatic disease; known congestive heart failure; unstable or severe angina; history of myocardial infarction; angioplasty; coronary artery bypass graft; stroke within 3 months; left ventricular dysfunction within 6 monhts; fasting C-peptide ≤ 0.5nmol/L; systolic blood pressure > 170; diastolic > 100
818 entered run-in, 596 randomized

G1: (rosiglitazone 4mg/day + metformin 2g/day) n=294

G2: (sulfonylurea (glibenclamide 5mg/day or glicazide 80mg/day + metformin 2g/day)
n=288

All medications uptitrated
G1: Age 58.5; 94% white, 6% other; Female 47%

G2: Age 59.3; 95% white, 5% other; Female 48%
MetforminHbA1c mean change at 52 weeks:
G1: −0.78 (SD 0.06)
G2: −0.86 (SD 0.06)

treatment difference: 0.09 (−0.08, 0.25), NS
All-cause Mortality
G1: n=2
G2: n=2
Nauck, 2009
LEAD-1 / LEAD-2
26 weeks
Multinational
Funding NR
Fair
(subset of LEAD1 and LEAD2 studies)
Inclusion: age 18–80; T2DM treated with monotherapy and had HbA1c between 7–11% on oral antidiabetic monotherapy for 3 months or between 7 – 10% on combination oral antidiabetic therapy for 3 months.

Exclusion: patients on combination therapy (although these were in original trial); insulin use within 3 months; impaired liver or renal function; blood pressure ≥ 180/100; cancer; use of drugs besides antidiabetic drugs likely to affect glucose
N=386 patients from 2 other RCTs

G1 (liglutinide 1.8mg + glimepiride): n=63

G2 (placebo + glimepiride): n=37

G3 (rosiglitazone + glimepiride): n=73

G4 (liraglutide 1.8mg + metformin): n=83

G5 (placebo + metformin): n=41

G6 (glimepiride + metformin): n=89
G1: Age 55.6; White 68.3%, Black 3.2%, Asian 28.6%, Other 0%; Female 47.6%

G2: Age 55.8; White 75.7%, Black 0%, Asian 24.3%, Other 0%; Female 45.9%

G3: Age: 57.1; White 68.5%, Black 2.7%, Asian 27.7%, Other 1.4%; Female 53.4%

G4: Age 55.3; White 86.7, Black 4.8%, Asian 8.4%, Other 0%; Female 44.6%

G5: Age 54.2; White 82.9%, Black 4.9%, Asian 12.2%, Other 0%; Female 41.5%

G6: Age 56.4; White 87.6%, Black 2.2%, Asian 10.1%, Other 0%; Female 39.3%
In LEAD-1 (Arms 1–3), background glimepiride

In LEAD-2 (Arms 4–6), background metformin
mean change at 26 weeks:
G1: −1.4 (SD NR)
G2: −0.3 (SD NR)
G3: −0.8 (SD NR)

P <0.0001, G1 vs. G2
P <0.0001, G1 vs. G3

G4: −1.3 (SD NR)
G5: −0.4 (SD NR)
G6: −1.2 (SD NR)

P <0.0001, G4 vs. G5
Non-inferior, G4 vs. G6
NR

From: Evidence Tables

Cover of Drug Class Review: Newer Diabetes Medications, TZDs, and Combinations
Drug Class Review: Newer Diabetes Medications, TZDs, and Combinations: Final Original Report [Internet].
Jonas D, Van Scoyoc E, Gerrald K, et al.
Portland (OR): Oregon Health & Science University; 2011 Feb.
Copyright © 2011, Oregon Health & Science University.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.