ArticlePDF Available

Effect of rosiglitazone on progression of coronary atherosclerosis in patients with type 2 diabetes mellitus and coronary artery disease: the assessment on the prevention of progression by rosiglitazone on atherosclerosis in diabetes patients with cardiov

Authors:

Figures

Content may be subject to copyright.
Coronary Heart Disease
Effect of Rosiglitazone on Progression of Coronary
Atherosclerosis in Patients With Type 2 Diabetes Mellitus
and Coronary Artery Disease
The Assessment on the Prevention of Progression by Rosiglitazone on
Atherosclerosis in Diabetes Patients With Cardiovascular History Trial
Hertzel C. Gerstein, MD, MSc; Robert E. Ratner, MD; Christopher P. Cannon, MD;
Patrick W. Serruys, MD, PhD; He´ctor M. García-García, MD, MSc; Gerrit-Anne van Es, PhD;
Nikheel S. Kolatkar, MD, MPH; Barbara G. Kravitz, MS; Diane M. Miller, PhD; Chun Huang, PhD;
Peter J. Fitzgerald, MD, PhD; Richard W. Nesto, MD; and the APPROACH Study Group
Background—Rosiglitazone has several properties that may affect progression of atherosclerosis. The Assessment on the
Prevention of Progression by Rosiglitazone on Atherosclerosis in Diabetes Patients With Cardiovascular History
(APPROACH) study was undertaken to determine the effect of the thiazolidinedione rosiglitazone on coronary
atherosclerosis as assessed by intravascular ultrasound compared with the sulfonylurea glipizide.
Methods and Results—This was a randomized, double-blind, controlled 18-month study in 672 patients aged 30 to 80
years with established type 2 diabetes mellitus treated by lifestyle, 1 oral agent, or submaximal doses of 2 oral agents
who had at least 1 atherosclerotic plaque with 10% to 50% luminal narrowing in a coronary artery that had not
undergone intervention during a clinically indicated coronary angiography or percutaneous coronary intervention. The
primary outcome was change in percent atheroma volume in the longest and least angulated epicardial coronary artery
that had not undergone intervention. Secondary outcomes included change in normalized total atheroma volume and
change in total atheroma volume in the most diseased baseline 10-mm segment. Rosiglitazone did not significantly
reduce the primary outcome of percent atheroma volume compared with glipizide (0.64%; 95% confidence interval,
1.46 to 0.17; P0.12). The secondary outcome of normalized total atheroma volume was significantly reduced by
rosiglitazone compared with glipizide (5.1 mm
3
; 95% confidence interval, 10.0 to 0.3; P0.04); however, no
significant difference between groups was observed for the change in total atheroma volume within the most diseased
baseline 10-mm segment (1.7 mm
3
; 95% confidence interval, 3.9 to 0.5; P0.13).
Conclusions—Rosiglitazone did not significantly decrease the primary end point of progression of coronary atherosclerosis
more than glipizide in patients with type 2 diabetes mellitus and coronary atherosclerosis.
Clinical Trial Registration http://www.clinicaltrials.gov. Unique Identifier: NCT00116831.
(Circulation. 2010;121:1176-1187.)
Key Words: atherosclerosis ultrasonography, intravascular rosiglitazone type 2 diabetes mellitus
Type 2 diabetes mellitus is a strong independent risk factor
for cardiovascular diseases,
1
and epidemiological studies
have shown that this risk is progressively related to the degree
of hyperglycemia as measured by the hemoglobin A
1C
.
2
Whereas recent trials of more intensive versus less intensive
glucose lowering with a menu of drugs did not detect a clear
cardiovascular benefit,
3,4
the observation of reduced myocar-
dial infarction and all-cause mortality in metformin-treated
obese subjects during active therapy in the United Kingdom
Prospective Diabetes study
5
and in all participants after an
additional 8.5 years of passive follow-up
6
suggests that
specific glucose-lowering drugs may have cardiovascular
benefits. Coronary atherosclerosis forms the substrate for
vulnerable plaque, which is hypothesized to precede clinical
coronary events. Coronary atherosclerosis can be directly
assessed with the use of intravascular ultrasonography
Received June 11, 2009; accepted December 7, 2009.
From the McMaster University and Hamilton Health Sciences, Hamilton, Ontario, Canada (H.C.G.); MedStar Research Institute, Washington, DC
(R.E.R.); Brigham and Women’s Hospital, Boston, Mass (C.P.C.); Erasmus Medical Center, Rotterdam, the Netherlands (P.W.S.); Cardialysis,
Rotterdam, the Netherlands (H.M.G.-G., G.v.E.); GlaxoSmithKline Research and Development, King of Prussia, Pa (N.S.K., B.G.K., D.M.M., C.H.);
Stanford University Medical Center, Palo Alto, Calif (P.J.F.); and Lahey Clinic, Burlington, Mass (R.W.N.).
A full list of the participants in the APPROACH trial is provided in Acknowledgments.
Correspondence to Dr H.C. Gerstein, McMaster University and Hamilton Health Sciences, Department of Medicine, Room 3V38, 1200 Main St W,
Hamilton, Ontario L8N 3Z5, Canada. E-mail gerstein@mcmaster.ca
© 2010 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.109.881003
1176
Downloaded from http://ahajournals.org by on September 4, 2019
(IVUS), and progression of IVUS-determined atherosclerosis
has been correlated with an increased risk of coronary
events.
7,8
Furthermore, reductions in IVUS-detected plaque
volume have been demonstrated in response to antihyperten-
sive
9
and lipid-lowering therapies
10–13
that also reduce the
incidence of coronary events.
Editorial see p 1165
Clinical Perspective on p 1187
Thiazolidinediones have effects on cardiovascular risk
factors, including insulin sensitivity,
14,15
inflammatory bi-
omarkers,
16
endothelial function,
17
coagulability,
16,18,19
plaque instability,
16
and blood pressure,
20
that may slow the
progression of coronary atherosclerosis. Some controlled
trials of both thiazolidinediones (rosiglitazone and pioglita-
zone) in patients with type 2 diabetes mellitus have suggested
a favorable effect of these agents on carotid atherosclerosis
21
and in-stent restenosis
22
; however, the applicability of these
data to native vessel coronary disease is uncertain. Moreover,
a recent large randomized trial of 1 thiazolidinedione (pio-
glitazone) demonstrated that it reduced progression of coro-
nary atherosclerosis in arteries that had not undergone inter-
vention more than glimepiride.
23
To date, comparable data
with the use of rosiglitazone have not been available.
The Assessment on the Prevention of Progression by
Rosiglitazone on Atherosclerosis in Diabetes Patients With
Cardiovascular History (APPROACH) trial was designed to
compare the effect of rosiglitazone and glipizide, agents that
reduce glucose through different mechanisms, on progression
of coronary atherosclerosis. Importantly, the trial was de-
signed to provide comparable glycemic control between
treatment arms and to evaluate the treatment effect on a
background of optimized contemporary therapy for second-
ary prevention of coronary disease including statins, anti-
platelet agents, and antihypertensive medications.
Methods
Study Design and Eligibility Criteria
A detailed description of the APPROACH trial has been published
previously.
24
APPROACH was a prospective, multicenter, double-
blind, randomized, active-controlled trial (Figure 1) of 672 patients
from 92 centers in 19 countries, who were aged 30 to 80 years with
established type 2 diabetes mellitus and who had clinically indicated
coronary angiography or percutaneous coronary intervention (PCI)
between February 2005 and January 2007. Patients were included if
they had at least 1 atherosclerotic plaque with 10% to 50% luminal
narrowing in a coronary artery that had not undergone intervention
and if their diabetes mellitus was treated with either lifestyle
approaches alone (with a hemoglobin A
1C
7% and 10%) or with
oral agents comprising 1 oral agent at any dose or 2 oral agents, in
which case each was prescribed at 50% of its maximal dose (with
a hemoglobin A
1C
6.5% and 8.5%). Exclusion criteria were as
follows: ST-segment elevation myocardial infarction in the prior 30
days; coronary artery bypass graft surgery; severe valvular heart
disease; left ventricular ejection fraction 40%; any heart failure
(New York Heart Association class I to IV); uncontrolled hyperten-
sion (systolic blood pressure 170 mm Hg or diastolic blood
pressure 100 mm Hg); renal insufficiency (serum creatinine 1.5
mg/dL for men or 1.4 mg/dL for women); and active liver disease.
Participant safety was monitored by an Independent Data Monitoring
Committee who periodically reviewed rates of clinical outcomes
according to unmasked therapy. The study protocol and consent
forms were reviewed by the institutional review board at each site,
and all patients provided written informed consent. Study design,
implementation, and analysis were performed under the supervision
of the Steering Committee, which was composed of 7 members from
Figure 1. Disposition of APPROACH
participants.
Gerstein et al Rosiglitazone and Coronary Atherosclerosis 1177
Downloaded from http://ahajournals.org by on September 4, 2019
external academic institutions and 2 from the sponsor. Data analysis
was performed according to a prespecified plan that was developed
with the approval of the Steering Committee.
Management of Glycemia and Follow-Up
Patients were randomized in a 1:1 ratio to receive masked rosiglita-
zone (4 mg/d) or glipizide (5 mg/d) in 1 pill. At the time of
randomization, the doses of other oral antidiabetic drugs were
reduced by 50% and were discontinued during a visit 1 month later.
At that time and after 2 and 3 months, the dose of masked study drug
was increased if tolerated and if the mean daily glucose level
calculated from the patient’s logbook of capillary tests in the 3 days
before the visit was 126 mg/dL (7.0 mmol/L). If 1 titration was
required, 2 pills per day were given. Titration doses of rosiglitazone
at the first, second, and third titration were 4 mg/d (1 pill; unchanged
dose), 8 mg/d (as 2 pills with active drug in the morning and placebo
in the evening), and 4 mg BID, respectively; glipizide dosing at these
visits were 10 mg/d (1 pill), 10 mg in the morning and 5 mg in the
evening (as 2 pills), and this same dose as the third titration,
respectively. Open-label metformin (maximal total daily dose, 2550
mg) and then once-daily basal insulin or both were added after the
first 3 months if needed to maintain a hemoglobin A
1C
7% with the
use of a glycemic titration algorithm designed to provide comparable
glycemic control between treatment groups. Nonstudy drugs were
reduced before study drugs in the event of hypoglycemia requiring
dose reductions. Unless informed consent was formally withdrawn,
all patients were followed until 18 months from randomization, and
clinical status was ascertained regardless of whether they continued
to take study medication.
IVUS Examination and Image Analysis
The longest and least angulated epicardial coronary artery that had
not undergone intervention was selected for IVUS examination.
Angiographers were instructed to choose a plaque within which there
was at least 20% stenosis by visual assessment. They were also
instructed to ensure that the region of interest was flanked by 2
anatomic landmarks that could be easily identified at follow-up (ie,
side branches). After intracoronary administration of nitroglycerin,
an ultrasound catheter (2.5F Atlantis SR Pro Imaging 40 MHz)
connected to a Galaxy G2 digital imaging console (Boston Scientific,
Natick, Mass) was advanced into the target vessel. The imaging
transducer was positioned just distal to an identifiable side branch,
and then motorized pullback of the transducer was performed at
0.5 mm/s. Follow-up IVUS examination was performed at study
completion in all patients providing informed consent (irrespective
of whether they continued to take study medication) with imaging of
the same coronary artery segment identified at the baseline exami-
nation. If a participant required cardiac catheterization for a clinical
indication between 9 and 18 months, follow-up IVUS examination
could be performed at that time instead of at study completion.
IVUS Outcomes
Core laboratory personnel (Cardialysis, Rotterdam, the Netherlands)
who were blinded to treatment assignment analyzed all IVUS images
using validated software (Curad, version 3.1, Wijk bij Duurstede, the
Netherlands), which facilitates detection of luminal and external
elastic membrane (EEM) boundaries in reconstructed longitudinal
planes. To obtain a smooth appearance of the vessel wall structures
in the longitudinal views, the Intelligate image-based gating method
was applied.
25,26
The primary IVUS outcome is the change in percent
atheroma volume (PAV),
9,27
calculated as PAV[(EEM
CSA
LUMEN
CSA
)/EEM
CSA
]100, where EEM
CSA
is the external elas-
tic membrane cross-sectional area (CSA) and LUMEN
CSA
is the
luminal cross-sectional area. A secondary IVUS outcome is the
change in normalized total atheroma volume (TAV
N
), calculated as
the product of the mean atheroma area and the median segment
length in the entire population, as follows:
TAV
N
[(EEM
CSA
LUMEN
CSA
)/N]overall median
segment length.
This calculation adjusts for differing segment lengths across patients,
thereby providing equal weighting of each patient in the calculation
of atheroma volume. An additional secondary IVUS outcome is the
change in TAV in the most diseased baseline 10-mm segment. This
was calculated as the follow-upbaseline difference in the TAV
within the 10-mm contiguous segment with the greatest atheroma
volume at baseline. Intraobserver variability was assessed with the
use of IVUS recordings from 20 randomly selected patients. Baseline
and follow-up IVUS examinations were each analyzed twice by the
same analyst who did not know whether the baseline or follow-up
IVUS recording was being read. The mean (SD) differences were
0.09 (0.18) mm
2
for vessel area and 0.02 (0.23) mm
2
for lumen
area. To evaluate variability between IVUS analysis methods, the
same patients were analyzed twice longitudinally and twice cross
sectionally.
10
The mean (SD) differences (in mm
2
) were 0.10
(0.36) mm
2
for vessel area and 0.001 (0.46) mm
2
for lumen area.
Finally, to assess variability between core laboratories, the same
patients were analyzed twice cross sectionally at different core
laboratories (Cardialysis, Rotterdam, the Netherlands, and MedStar
Research Institute, Division of Cardiology, Washington, DC). The
mean (SD) differences (in mm
2
) were 0.53 (0.37) mm
2
for vessel
area and 0.07 (0.45) mm
2
for lumen area.
Clinical Cardiovascular Outcomes
Investigators submitted end point forms for any event that could
potentially represent a myocardial ischemic event or heart failure. An
independent end point committee blinded to treatment assignment
prospectively adjudicated these cardiovascular events, which in-
cluded cardiovascular and noncardiovascular death, nonfatal myo-
cardial infarction and stroke, coronary revascularization, hospitaliza-
tion for recurrent myocardial ischemia, and heart failure.
Statistical Methods
Continuous variables are expressed as mean and SD or median and
interquartile range if nonnormally distributed, with categorical vari-
ables reported as percentage. IVUS outcomes were analyzed with the
use of ANCOVA with terms for treatment group, baseline value,
geographic region, gender, entry cardiac procedure (angiography or
PCI), and prior oral antidiabetic medication. The change in PAV for
each allocated group was calculated by estimating the model-
adjusted mean change in PAV across the cohort. All Pvalues are
2-sided and not adjusted for multiple testing, with Pvalues 0.05
considered significant. A worst-rank sensitivity analysis was per-
formed as described previously to assess the potential influence on
the primary end point of randomized patients who did not complete
the follow-up IVUS examination because of a cardiovascular
event.
28
For this analysis, all patients with evaluable baseline and
follow-up IVUS examinations were assigned a rank value based on
their change in PAV (ordered from least to greatest). Patients without
a follow-up IVUS as noted above were assigned a rank value that
was worse than that of the patient with the greatest increase in PAV.
The rank value for these patients was first assigned on the basis of a
prespecified hierarchy determined by the Steering Committee ac-
counting for both clinical severity and relationship of the event to
coronary atherosclerosis, as follows (lowest to highest rank): con-
gestive heart failure, hospitalization for recurrent myocardial ische-
mia, coronary revascularization, noncardiovascular death, nonfatal
stroke, nonfatal myocardial infarction, and cardiovascular death.
Within each event category, the final rank was determined by the
time to event, with earlier events assigned a worse (higher) rank. The
worst-rank analysis compared the distribution of rank values be-
tween treatment groups with the use of a univariate Wilcoxon-Mann-
Whitney test.
Analysis of the primary IVUS outcome in prespecified subgroups,
including region, angiography versus PCI, prior oral agent use (drug
naive, sulfonylurea, metformin, dual therapy), age (60 versus 60
years), gender, systolic blood pressure (130 versus 130 mm Hg),
statin use, body mass index, diabetes mellitus duration, baseline
high-sensitivity C-reactive protein, hemoglobin A
1C
, high-density
1178 Circulation March 16, 2010
Downloaded from http://ahajournals.org by on September 4, 2019
lipoprotein (HDL), low-density lipoprotein (LDL), triglycerides, and
PAV (less than or equal to the median versus more than the median
value), was performed with the use of ANCOVA with a test for
treatment-by-subgroup interaction. The effect of treatment allocation
on time to first occurrence of the various cardiovascular outcomes
was estimated with a Cox proportional hazards model that included
terms for treatment group, region, angiography versus PCI, and prior
oral agent use.
Sample size calculations determined that 206 patients per group
with evaluable baseline and follow-up IVUS examinations were
required to provide 90% power with a 2-sided
of 0.05 to detect a
treatment difference between the groups of 1.6%, with the assump-
tion of a 5.0% SD for the primary IVUS outcome. These assumptions
were based on prior IVUS studies of lipid-lowering therapies in
patients with diabetes mellitus.
10
Given prior noncompletion rates
among patients with type 2 diabetes mellitus in contemporary IVUS
studies ranging from 25% to 35%,
9,10,23,27
a total sample size of 634
randomized patients was required under the worst-case assumption
of a 35% noncompletion rate. All analyses were performed with the
use of SAS version 9.1 (SAS Institute, Cary, NC). Finally, to ensure
the accuracy of the analyses, data for the primary and secondary end
points were provided to an external biostatistician (Dr Todd A.
Table 1. Baseline Characteristics of Randomized Patients
Glipizide
(n339)
Rosiglitazone
(n333) P
Age, mean (SD), y 60.2 (9.0) 61.8 (8.4) 0.03
Male, n (%) 223 (65.8) 233 (70.0) 0.25
Current smoker, n (%) 57 (16.8) 55 (16.5) 1.00
Duration of diabetes mellitus, median (IQR), y 4.6 (1.7–8.9) 5.0 (2.2–7.9) 0.90
Hypertension, n (%) 272 (80.2) 266 (79.9) 0.92
Dyslipidemia, n (%) 227 (67.0) 232 (69.7) 0.46
Prior myocardial infarction, n (%) 82 (24.2) 81 (24.3) 1.00
Presenting condition, n (%) 130 (38.3) 128 (38.4) 1.00
Acute coronary syndrome
Elective procedure 209 (61.7) 205 (61.6)
Baseline procedure, n (%) 171 (50.4) 166 (49.8) 0.94
Coronary angiography
PCI 168 (49.6) 167 (50.2)
Medication use, n (%) 279 (82.3) 280 (84.1) 0.61
Aspirin
Other antiplatelet 195 (57.5) 196 (58.9) 0.75
-blocker 223 (65.8) 241 (72.4) 0.07
ACE inhibitor or ARB 238 (70.2) 237 (71.2) 0.80
Nitrates 137 (40.4) 125 (37.5) 0.48
Statin 262 (77.3) 248 (74.5) 0.42
Fibrate or other lipid-lowering agent 24 (7.1%) 34 (10.2%) 0.17
Weight, mean (SD), kg 83.8 (18.5) 82.0 (19.1) 0.22
Body mass index, mean (SD), kg/m
2
29.8 (5.3) 29.3 (5.5) 0.14
Blood pressure, mean (SD), mm Hg 131.0 (15.1) 127.9 (16.1) 0.004
Systolic
Diastolic 76.3 (10.0) 75.2 (10.2) 0.16
Serum creatinine, mean (SD), mg/dL 0.98 (0.22) 1.02 (0.25) 0.02
Hemoglobin A
1C
, mean (SD), % 7.2 (0.9) 7.1 (0.8) 0.08
BNP,* median (IQR), pg/mL 25 (12–53) 25 (11–58) 0.98
Fasting insulin,* median (IQR),
U/mL 13.0 (8.6–18.1) 13.0 (8.6–20.7) 0.49
LDL cholesterol, mean (SD), mg/dL 91.2 (35.5) 89.6 (35.9) 0.63
HDL cholesterol, mean (SD), mg/dL 42.7 (10.7) 42.4 (11.1) 0.77
Triglycerides, median (IQR), mg/dL 159.3 (122.1–204.9) 162.0 (122.1–211.5) 0.93
hsCRP, median (IQR), mg/L 5.4 (2.5–11.0) 4.9 (2.2–11.3) 0.54
MMP-9, median (IQR),
g/L 86.9 (43.8–195.1) 88.6 (44.1–221.6) 0.44
IQR indicates interquartile range; ACE, angiotensin-converting enzyme; ARB, angiotensin II receptor blocker; hsCRP,
high-sensitivity C-reactive protein; and MMP-9, matrix metalloproteinase-9. SI conversion factors: to convert fasting
plasma glucose to mmol/L, multiply by 0.0555; LDL and HDL cholesterol values to mmol/L, by 0.0259; triglyceride
values to mmol/L, by 0.0113; serum creatinine to
mol/L, by 88.4; fasting insulin to pmol/L, by 6.945; and c-reactive
protein to nmol/L, by 9.524.
*Performed in a subset of patients (B-natriuretic peptide BNP,n464; fasting insulin, n435.
Gerstein et al Rosiglitazone and Coronary Atherosclerosis 1179
Downloaded from http://ahajournals.org by on September 4, 2019
MacKenzie, Dartmouth Medical School), who independently re-
peated and confirmed all of the findings reported herein. All authors
had access to study data and take responsibility for the integrity of
the data and the accuracy of the data analysis. All authors have read
and agree to the manuscript as written.
Results
Participants
Of 1147 people who were screened, 672 (68% men) of mean
(SD) age 61 (9) years with median diabetes mellitus duration
of 4.8 years and mean hemoglobin A
1C
of 7.2% (0.9) were
randomized to either glipizide (n339) or rosiglitazone
(n333) from 92 sites in 19 countries (Figure 1). Baseline
characteristics by treatment group are noted in Table 1; by
chance, patients allocated to rosiglitazone versus glipizide
were slightly older and had a slightly higher serum creatinine
and lower systolic blood pressure. Because randomization
was stratified by the cardiac procedure, 50% of enrolled
patients had diagnostic coronary angiography, and 50% had a
PCI. A total of 38% presented with acute coronary syndrome,
and 76% were on statins. A total of 229 of 339 patients
(67.5%) allocated to glipizide and 233 of 333 (70%) allocated
to rosiglitazone had an evaluable baseline and follow-up
IVUS. Compared with participants who did not have 2
evaluable IVUS examinations, those who did had a slightly
lower diastolic blood pressure (P0.04), were more likely to
be from South America (P0.001), were less likely to be on
2 oral antidiabetic agents (P0.004), and were more likely to
have had a stent inserted (P0.006). An assessment of vital
status at the 18-month final visit was available in 317 patients
(93.5%) allocated to glipizide and 316 patients (94.9%)
allocated to rosiglitazone.
Patients were followed for a median of 18.6 months
(interquartile range, 18.2 to 18.9) and a mean (SD) of 16 (6)
months; patients allocated to glipizide were adherent (took
80% and 120% of their study medications) at 90.7% of
visits, and those allocated to rosiglitazone were adherent at
92.7% of visits. Adverse effects that were either of interest on
the basis of prior studies, that occurred in 5% of partici-
pants in either group, or that significantly differed in fre-
quency between groups are noted in Table 2. Compared with
patients in the glipizide group, those allocated to rosiglitazone
had less hypoglycemia and more anemia. No between-group
difference was noted in the rate of cardiovascular events (all
adjudicated) that occurred infrequently during the trial (Table
3); 5 of the cardiovascular events in the rosiglitazone group (1
revascularization, 2 nonfatal myocardial infarctions, 1 nonfa-
tal stroke, and 1 cardiovascular death) occurred within 5 days
of the baseline cardiac catheterization and were classified as
procedure related.
The mean (SD) dose of study drug for patients who had a
baseline and follow-up IVUS was 12.3 (4.3) mg for glipizide
and 6.8 (1.8) mg for rosiglitazone. Of these patients, 220
(96.1%) allocated to glipizide and 220 (94.4%) allocated to
rosiglitazone had their follow-up IVUS done 17 months
after the baseline IVUS. Figure 2 illustrates the change from
baseline in hemoglobin A
1C
, blood pressure, lipids, and
weight for each group, and Table 4 lists the postrandomiza-
tion mean or median values for these and other variables, as
well the final use of concomitant medications by treatment
group. During the first 3 months of therapy, when rosiglita-
zone or glipizide was substituted for other oral agents,
hemoglobin A
1C
levels were higher on rosiglitazone. Subse-
quently, hemoglobin A
1C
levels were the same, and the final
hemoglobin A
1C
levels did not differ between groups; how-
ever, these fluctuations led to a slightly higher average
postrandomization hemoglobin A
1C
value on rosiglitazone
compared with glipizide (7.0% versus 6.9%, respectively;
P0.01). In addition, compared with patients on glipizide,
those on rosiglitazone had no significant difference in weight
or systolic blood pressure; a significantly lower postrandom-
ization diastolic blood pressure, high-sensitivity C-reactive
protein, and matrix metalloproteinase-9; and a significantly
higher postrandomization LDL cholesterol and HDL
cholesterol.
Effect on IVUS End Points
Table 5 notes the effect of allocated therapy on the primary
and secondary IVUS end points. During the course of the
study, PAV (the primary outcome) did not significantly
change from baseline in patients allocated to glipizide
(0.43%; 95% confidence interval [CI], 0.22 to 1.08;
P0.19) or in patients allocated to rosiglitazone (0.21%;
95% CI, 0.86 to 0.44; P0.53). Moreover, rosiglitazone
did not significantly reduce PAV compared with glipizide
(0.64%; 95% CI, 1.46 to 0.17; P0.12). Similar findings
were noted (1) when the data were analyzed after adjustment
for baseline differences in age, creatinine, and systolic blood
Table 2. Major Adverse Events in All Randomized Patients
Glipizide
(n339)
Rosiglitazone
(n333) P
Events requiring change or stop
in study medication
Peripheral edema 1 (1) 2 (1) 0.62
Hypoglycemia 12 (4) 0 (0) 0.0004
Peripheral edema 24 (7) 29 (9) 0.48
Severe hypoglycemia (requiring
external assistance)
3(1) 0 (0) 0.25
Hypoglycemia 96 (28) 27 (8) 0.0001
Bone fracture 2 (1) 6 (2) 0.17
Hemoglobin decrease 3
g/dL from baseline
10 (3) 26 (8) 0.006
Angina pectoris 35 (10) 31 (9) 0.70
Chest pain 17 (5) 12 (4) 0.45
Cough 22 (6) 13 (4) 0.17
Diarrhea 17 (5) 12 (4) 0.45
Dizziness 18 (5) 18 (5) 1.00
Fatigue 14 (4) 18 (5) 0.47
Headache 23 (7) 15 (5) 0.24
Hypertension 22 (6) 14 (4) 0.23
Values are number of patients with an event (%). Adverse events requiring
changes in study drug, reported in 5% of patients in either group,
significantly differed between groups or were of interest on the basis of other
studies are listed.
1180 Circulation March 16, 2010
Downloaded from http://ahajournals.org by on September 4, 2019
pressure (0.60%; 95% CI, 1.43 to 0.23; P0.15) and (2)
when the data were analyzed with a worst-rank analysis
comprising 243 glipizide (72%) and 247 rosiglitazone pa-
tients (74%), in which the 28 patients (14 per group) who did
not have 2 evaluable IVUS examinations due to a cardiovas-
cular event were assigned a rank on the basis of their event as
described above (P0.20). The main secondary end point of
TAV
N
did not significantly change in patients allocated to
glipizide (1.2 mm
3
; 95% CI, 2.7 to 5.1; P0.54). However,
it significantly decreased by 3.9 mm
3
(95% CI, 7.8 to
0.02; P0.049) in patients allocated to rosiglitazone, and,
compared with glipizide, rosiglitazone significantly reduced
TAV
N
by 5.1 mm
3
(95% CI, 10.0 to 0.3; P0.04).
Atheroma volume within the most diseased baseline 10-mm
segment was the other main secondary IVUS end point and
decreased from baseline by 3.6 mm
3
(95% CI, 5.3 to 1.8;
P0.0001) in patients allocated to glipizide and by 5.3 mm
3
(95% CI, 7.0 to 3.5; P0.0001) in patients allocated to
rosiglitazone; the effect of rosiglitazone did not significantly
differ from that of glipizide (1.7 mm
3
; 95% CI, 3.9 to 0.5;
P0.13).
When analyzed according to prespecified subgroups (Fig-
ure 3), an interaction between treatment allocation and
diabetes mellitus duration was noted (P0.005) such that
rosiglitazone reduced the PAV more than glipizide in patients
with diabetes mellitus duration longer than the median
duration of 4.9 years (ie, a 1.8% decrease versus a 0.5%
increase in those with a shorter diabetes mellitus duration).
Discussion
In this 1.5-year trial, rosiglitazone did not reduce progression
of coronary artery plaque as measured by the primary IVUS
end point of PAV compared with glipizide. The prespecified
subgroup analyses suggesting an effect of rosiglitazone on
PAV in participants with the longest diabetes mellitus dura-
tion raise the possibility of some antiatherosclerotic effect in
this subgroup but should be viewed as hypothesis generating
because of the many subgroups tested. Similarly, the obser-
vation that rosiglitazone significantly reduced the secondary
outcome of TAV
N
does not prove that rosiglitazone reduces
Table 3. Adjudicated Clinical Cardiovascular Outcomes in All
Randomized Patients
Glipizide
(n339)
Rosiglitazone
(n333) P*
Composite of all-cause death, nonfatal
myocardial infarction, nonfatal stroke,
coronary revascularization, or
hospitalization for myocardial ischemia
38 (11.2) 39 (11.7) 0.58
Composite of cardiovascular death,
nonfatal myocardial infarction, or
nonfatal stroke
10 (2.9) 14 (4.2) 0.31
All-cause death 7 (2.1) 8 (2.4) 0.72
Cardiovascular death 3 (0.9) 4 (1.2) 0.50
Myocardial infarction
Nonfatal 6 (1.8) 7 (2.1) 0.71
Fatal 1 (0.3) 1 (0.3) 0.89
Stroke 0.13
Nonfatal 1 (0.3) 5 (1.5)
Fatal 0 (0) 0 (0)
Coronary revascularization 27 (8.0) 26 (7.8) 0.82
Hospitalization for myocardial ischemia 7 (2.1) 11 (3.3) 0.25
Congestive heart failure 3 (0.9) 8 (2.4) 0.14
Values are number of patients with an event (%). Noncardiovascular causes
of death for the glipizide participants include cancer (n2), septic shock
(n1), and spinal fracture with cord compression (n1) and for the rosiglita-
zone participants include cancer (n3) and chronic obstructive lung disease
(n1).
*All Pvalues calculated with the use of the Cox proportional hazards model
with time to first event.
Figure 2. Effect of rosiglitazone vs glipiz-
ide on mean change in hemoglobin A
1C
,
blood pressure, weight, HDL cholesterol
(HDL-c), and LDL cholesterol (LDL-c) level
and on median change in triglyceride level
in patients with 2 evaluable IVUS mea-
sures. Pvalues refer to differences of the
change between groups. Gray bars depict
the change for the rosiglitazone group,
and black bars depict the change for the
glipizide group.
Gerstein et al Rosiglitazone and Coronary Atherosclerosis 1181
Downloaded from http://ahajournals.org by on September 4, 2019
atherosclerotic plaque progression more than glipizide in
light of the absence of benefit on the primary IVUS outcome.
Although not significant, the APPROACH findings are
qualitatively similar to those of a similar trial in which
another thiazolidinedione (pioglitazone) was compared with
glimepiride.
23
In that trial, pioglitazone significantly reduced
the primary outcome of PAV by 0.89% compared with
glimepiride during 18 months of therapy (P0.002). The
Table 4. Postrandomization Values and Medication Use at Final Visit
Patients With Evaluable IVUS
Measurement
Glipizide
(n229)
Rosiglitazone
(n233) P
Continuous variables
Mean hemoglobin A
1C
(95% CI), % 6.9 (6.8–7.0) 7.0 (7.0–7.1) 0.01
Mean blood pressure (95% CI), mm Hg
Systolic 131.1 (129.6–132.6) 130.7 (129.3–132.2) 0.71
Diastolic 76.7 (75.8–77.7) 75.4 (74.5–76.4) 0.03
Mean LDL (95% CI), mg/dL 84.9 (80.7–89.1) 95.3 (91.0–99.6) 0.0001
Mean HDL (95% CI), mg/dL 45.5 (44.3–46.7) 48.7 (47.5–49.9) 0.0001
Median triglycerides (95% CI), mg/dL 156.2 (150.4–164.6) 146.5 (138.9–151.3) 0.14
Median hsCRP (95% CI), mg/L 1.9 (1.7–2.1) 0.9 (0.8–1.0) 0.0001
Median MMP-9 (95% CI), ug/L 56.8 (54.2–61.0) 47.9 (44.4–50.8) 0.0001
Mean weight (95% CI), kg 83.6 (83.1–84.1) 84.0 (83.5–84.5) 0.22
Final visit medication use, No. (%)
Aspirin 190 (83.0) 199 (85.4) 0.52
Other antiplatelet 93 (40.6) 82 (35.2) 0.25
-blocker 152 (66.4) 158 (67.8) 0.77
ACE inhibitor or ARB 166 (72.5) 176 (75.5) 0.46
Nitrates 78 (34.1) 76 (32.6) 0.77
Statin 179 (78.2) 190 (81.5) 0.42
Fibrate or other lipid-lowering agent 27 (11.8) 35 (15.0) 0.34
Metformin 153 (66.8) 152 (65.2) 0.77
Any insulin 21 (9.2) 14 (6.0) 0.22
Pvalues pertain to between-group differences in variables after randomization and in medication use at the final
visit. hsCRP indicates high-sensitivity C-reactive protein; MMP-9, matrix metalloproteinase-9; ACE, angiotensin-
converting enzyme; and ARB, angiotensin II receptor blocker.
Table 5. IVUS End Points
Mean Value of IVUS
Measurement (SD)
Glipizide Rosiglitazone
Treatment Difference
(95% CI)Baseline Follow-Up Change* (95% CI) Baseline Follow-Up Change* (95% CI)
Mean (SD) PAV† 40.6 (11.0) 41.0 (11.2) 0.43 (0.22, 1.08) 40.4 (11.8) 40.2 (11.4) 0.21 (0.86, 0.44) 0.64 (1.46, 0.17)§
Mean (SD) TAV
N
,mm
3
232.8 (115.2) 233.2 (116.5) 1.2 (2.68, 5.08) 226.1 (100.6) 221.6 (100.7) 3.9 (7.82, 0.02)5.12 (9.98, 0.26)¶
Mean (SD) atheroma
volume in the most
diseased 10-mm
segment, mm
3
75.6 (32.6) 72.2 (33.3) 3.6 (5.31, 1.80)# 71.0 (30.0) 66.0 (30.7) 5.3 (7.04, 3.51)# 1.7 (3.93, 0.49)
Mean (SD) total vessel
volume, mm
3
609.4 (311.8) 603.1 (304.3) 4.6 (11.40, 2.27) 555.1 (298.0) 547.2 (298.2) 8.1 (14.9, 1.32)** 3.6 (12.15, 5.02)
Mean (SD) total lumen
volume, mm
3
359.7 (195.7) 353.5 (192.2) 4.9 (11.88, 2.05) 332.7 (192.4) 328.7 (191.9) 4.6 (11.52, 2.34) 0.3 (8.40, 9.05)
*Change from baseline for each allocated group was estimated with the use of ANCOVA with terms for treatment group, baseline value, geographic region, gender,
entry cardiac procedure (angiography or PCI), and prior oral antidiabetic medication.
†Primary IVUS outcome.
‡Key IVUS secondary outcomes.
§P0.12.
P0.049.
P0.04.
#P0.0001.
**P0.02.
Similar Pvalues were obtained for the primary and key secondary outcomes when reanalyzed with a nonparametric test on rank-transformed data. The treatment
difference in PAV between the 21 glipizide and 28 rosiglitazone participants who both had a clinical outcome and a final evaluable IVUS was 0.24% (95% CI, 3.66,
3.18; P0.89).
1182 Circulation March 16, 2010
Downloaded from http://ahajournals.org by on September 4, 2019
absence of a significant effect of rosiglitazone in this study
may be due to the fact that the PAV in the glipizide control
group only increased modestly with time. These observations
and reports of similar beneficial effects of these 2 drugs on
carotid intima-media thickness
21,29
and the need for revascu-
larization after PCI
30
suggest that they have similar effects on
atherosclerosis. Whether they have similar or different effects
on cardiovascular outcomes remains uncertain, especially in
light of recent meta-analyses relative to the cardiovascular
safety of rosiglitazone.
31–37
The effect of these drugs on
cardiovascular outcomes can only be assessed by directly
comparing them with each other and with placebo in a large
head-to-head cardiovascular outcome trial. Such a trial (Thia-
zolidinedione Intervention With Vitamin D Evaluation
[TIDE]) is now under way (http://www.clinical trials.gov;
Unique Identifier NCT00879970).
Strengths of this study are the large sample size, the use of
masked therapies, high adherence rates, achievement of
similar levels of cardiovascular risk factors in both groups,
and careful validated measurements of IVUS indices. These
findings are limited by the fact that 32.5% of patients
allocated to glipizide and 30% of patients allocated to
rosiglitazone did not have 2 evaluable IVUS measurements.
Moreover, the observation that patients without 2 evaluable
IVUS measurements were less likely to have had a stent
inserted at baseline and were on fewer antidiabetic agents
than those who had the 2 IVUS measurements suggests that the
patients who did have the 2 IVUS measurements were those
with more advanced atherosclerosis at baseline. However, the
findings of the worst-rank analysis, which includes 490 of the
672 randomized participants (73%) who received at least 1
dose of blinded medication and which was similar to the
Figure 3. Effect of rosiglitazone vs glipizide on the primary outcome of change in PAV according to predefined subgroups in patients
with 2 evaluable IVUS measures. Pvalues reflect the test for an interaction between the subgroups and allocation to rosiglitazone vs
glipizide. Median values are as follows: body mass index (BMI), 28.9 kg/m
2
; diabetes mellitus duration, 4.9 years; hemoglobin A
1C
,
7.1%; HDL, 42.5 mg/dL; LDL, 87.0 mg/dL; triglycerides, 165.0 mg/dL; high-sensitivity C-reactive protein (hsCRP), 5.2 mg/L; PAV,
41.3%. AS indicates Asia; EU, Europe, NA, North America, SA, South America; OAD, oral antidiabetic drug; SU, sulfonylurea; and BP,
blood pressure.
Gerstein et al Rosiglitazone and Coronary Atherosclerosis 1183
Downloaded from http://ahajournals.org by on September 4, 2019
analysis of patients with the 2 evaluable IVUS measurements,
suggest that it is unlikely that data from these individuals
would have substantively altered the findings. These findings
are also limited by the observation that rosiglitazone participants
had a higher hemoglobin A
1C
level and that glipizide-treated
patients had a lower LDL level after randomization despite
blinded adjustment of medications designed to achieve similar
levels of hemoglobin A
1C
and LDL.
In summary, this study did not prove the hypothesis that
rosiglitazone has a greater antiatherosclerotic effect than
glipizide in patients with type 2 diabetes mellitus. However,
reduced plaque progression in the presence of more advanced
diabetes mellitus and the reduced secondary IVUS outcome
suggest that there may be a benefit in some subgroups.
Ongoing analyses will evaluate which factors relate to
changes in atherosclerosis as measured by IVUS and will
explore the effects of rosiglitazone on atherosclerosis and
plaque in more detail.
Acknowledgments
Steering Committee: R. Ratner, P. Fitzgerald (cochairs), C. Cannon,
H. Gerstein, R. Nesto, P. Serruys, G.-A. van Es, A. Zalewski.
Statistical Data and Analysis Center: Axio Research, Seattle, Wash.
L. Shemanski, A. Slee. End Point Adjudication Committee: M.
Savage (chair), H. Weitz, I. Ahmed. Independent Data Monitoring
Committee: G. Mintz, J. Lachin (cochairs), M. Abrahamson, P.
Carson, P. Jones. APPROACH Study Management Team: N. Kolat-
kar, B. Kravitz, A. Wolstenholme, K. Saarinen, R. Fowler, Janet
Hoffman, D. Steele-Norwood, R. Russell, S. Young, and former
team members Paul Aftring, Hubert Chou, Steve McMorn, Courtney
Kirsch, Bonnie Louridas, Teresa Olivieria, and Debra Mattioli.
APPROACH Statistics and Data Management Team: D. Miller, C.
Huang, C. Nguyen, L. Jahnke. Investigators—Argentina: Liliana
Grinfeld, Daniel Nul, Graciela Gili, Sandra Silveiro, Silvina Ramos,
Sonia Hermida, Hugo Torres, Estevan Zuanich, Pablo Severino,
Ramiro Costello, Silvio Severino, Pablo Czerniuk, Liliana Grinfeld,
Carlos Rojas Matas, Marcelo Halac, Alejandro Fernandez, Jose´
Gabay, Daniel Berrocal, Jorge Bluguerman, Daniel Mauro, Rodolfo
Zazzali, Oscar Carlevaro, Jorge Alvarinas, Julia Blanco, Aristo´bulo
Balestrini, Arnoldo Girotti, Mario Principato, Susana Righi, Norma
Del Hoyo, Pedro Tesone, Marcelo Mule, Jorge Belardi, Lucio
Padilla, Mariano Albertal, Fernando Cura, Carlos Dall’asta, Guill-
ermo Bortman, Pablo Czerniuk, Martin Gabriel Nicolini, Jorge
Novas, Jorge Pinzetta Martinez, Marcos Litvak, Sebastia´n Schanz
Estrada, Carlos Busso, Alejandro Cueto, Ma´ximo Senesi, Francisco
Gadaleta, Graciela Pedrale, Sandra Geraci, Carlos Wechsler, Edith
Guerra, Elsa Ramirez de Martini, Susana Llois, Hugo Londero,
Eugenio Trucco, Leandro Martinez Riera, Jose Pozzi, Silvana Solis,
Jorge Waitman, María Cabrera, Francisco Paoletti, Daniel Mercado,
Horacio Jure, Maria Pepe, Carolina Nadaya, Gabriela Morra, Jose
Tibaldi, Cecilia Luquez, Alberto Villamil, Daniel Bekier, Judith
Zilberman, Fabian Contreras, Paulo Reisin, Santiago Bruzone, Ad-
riana Osorio, Sergio Kuznicki, Natalia Souto, Daniel Rico´n, Sebas-
tian Rizzone, Maria Jimena Soutelo, Gabriel Sumay, Marcelo Ma-
suelli, Sergio Chekherdenian, Leo´n Litwak, Ventura Simonovich,
Lina Capurro, Ricardo Rey, Daniel Seinhart, Laura Brescacin, Pedro
Forcada, Oscar Montan˜a, Christian Ponce, Alberto Marani, Jose
Spolidoro, Virginia Visco, Claudia Mahfoud, Silvia Orio, Carlos
Damico, Gustavo Costa, Simon Salzberg, Gustavo Alfredo Leiva,
Miguel Oscar Payaslian, Ricardo Basile, Karina Crotto, Patricia
Gibelman, Sonia Sassone, Fernando Mario Krinsky, Pablo Perez
Balin˜o, Jorge Said Nisi, Felipe Martinez, Oscar Alfredo Allall,
Sergio Rafael Tamiozzo, Rodolfo Sarjanovich, Viviana Arias, Mar-
cela Giacometti, Marcela Giacometti, Victorino Jose´ Fuentealba,
Moises Azize, Jimena María Martinez, Pablo Nicolas Guzman,
Gabriel Sumay, Hugo Luquez. Brazil: Fa´bio de Brito, Jr, Marco
Perin, Breno Almeida, Joa˜o Sa´, Alexandre Abizaid, Luiz Tanajura,
Luiz Mattos, Amanda Sousa, Roselei Graebin, Jose´ Sousa, Galo
Andrade, Rodolfo Staico, Fausto Feres, A
´urea Chaves, Denise
Franco, Carlos Eduardo Soares, Felipe Maia, Ricardo Costa, Jose´
Ju´nior, Henry Zatz, Moyse´s Lima Filho, Geraldo Figueiredo, Jose´
Antonio Marin Neto, Ingrid Dick de Paula, Maria Cristina Freitas,
Raul dos Santos, Expedito Silva, Pedro Horta, Eulo´gio Filho, Marcio
Miname, Pedro Neto, Luiz Ce´sar, Ana Paula Chacra, Wilson Filho,
Carolina Chrisman. Canada: Jeff Pang, Harinda Wijeysundera, John
Graham, Eric Cohen, Sam Radhakrishnan, Dennis Ko, Manu Prab-
hakar, Ram Vijayaraghavan, Salim Naqvi, Ivy Fettes, Jaffer Syed,
Alexander Dick, Mina Madan, Asim Cheema, Bradley Strauss, Jason
Burstein, Thierry Charron, Tej Sheth, Michael Kutryk, Warren
Cantor, Robert Chisholm, Ruth McManus, Kumar Sridhar, David
Almond, Patrick Teefy, Madhu Natarajan, Dominic Raco, James
Velianou, Tej Sheth, David Crosby, Shamir Mehta, Michael Rokoss,
Nicholas Valettas, William Harper, Ally Prebtani. Czech Republic:
Josef Stasek, Irma Dresslerova, Pavel Polansky, Josef Bis, Dusan
Cernohorsky, Miroslav Brtko, Jan Vojacek, Jaroslav Dusek. France:
Bernard Vialettes, Catherine Mattei, Jean Louis Bonnet, Herve´Le
Breton, Anne-Marie Leguerrier, Guillaume Lecoq, Dominique Boul-
mier, Marc Bedossa, Catherine Petit, Guillaume Charpentier, David
Pesenti-Rossi, Christophe Caussin. Germany: Gabriele von Berg-
mann, Guenter Boehm, Iris Donati-Hirsch, Klaus Busch, Horst-
Michael Frick, Susanne Hoeltz, Eckart Lohr, Ludwig Merker, I.
Kreuzer, Michael Mikolaiczik, Adolf Trautmann, Raimund Erbel,
Sebastian Alexander Philipp, Igor Kordish, Dirk Boese, Stefan Sack,
Achim Gutersohn, Karl-Eugen Hauptmann, Herbert Dasch, B.
Gestrich, Anne-Marie Ley, Gerrit Schneider, Bruno Sontag, Bernd
Liesenfeld, Helmut Kuecherer, Stefan Hardt, Celine Johanssen, G.
Korosoglou, Alexander Hansen, Raffi Bekeredjian, Andreas
Muegge, Harald Klein, Antje Wagner, Tusgut Brodherr, Michael
Lindstaedt, Alfried Germing, Eckard Rau, Christian Stute, Johannes
Brachmann, Harald Rittger, Klaus-Juergen Gutleben, Guido
Ritscher, Kai Koegler, Hasan Alawi, Ilka Simon-Wagner, Georg
Plassmann, Dirk Weber, Martin List. Hong Kong: John Sanderson,
Cheuk-Man Yu, Pui Wai Lee, Juliana Chan, Eugene Wu, Norman
Chan, Wing Bun Chan, Alice Kong, Leo Kum, Edmond Wong,
Wilson Chan, Linda Lam, Wai-Hong Chen, Stephen Lee, Wing-Sun
Chow, Annette Tso, Pui-Yin Lee, Chu-Pak Lau, Raymond Miu,
Ryan Ko, Chi Sang Hung, Man Wo Tsang, Tai Pang Ip, Shing Chung
Siu, Wai-Hong Chen, Michael Lee, Chiu Lai Fu, Steven LI, Duncan
Ho, Cheung Hei Choi, Sau Cheung Tiu, Kam Tim Chan. India: Ajit
Mullasari, Arpandev Bhattacharyya, Rajiv Passey, Balram Bhargava,
Anil Dhall, Naveen Aggrawal, Sachin Khandelwal, M.K. Garg, N.K.
Mahesh, Naveen Aggrawal, Anupam Goel, Ashok Seth, Dhiraj
kapoor, Rajesh Saxena, Praveen Kumar Chandra. Italy: Giorgio
Morocutti, Laura Tonutti, Guglielmo Bernardi, Claudio Noacco,
Leonardo Spedicato, Patrizia Presbitero, Stefano Genovese, Silvia
Rebuzzini, Dennis Zavalloni Parenti. Korea: Hyo-Soo Kim, HyunJae
Kang, Bon-Kwon Koo, Young Min Cho, Sang Ho Jo, Hae Sung
Kim, Soo Lim, Hyun-Jae Kang, Jin Wook Chung, Kyong Soo Park,
Eun-Jung Lee, You Jin Lee, Mi-Yeon Kang, Cheol-Whan Lee, Cheol-
Whan Lee, Young Hak Kim, Jung-Min Koh, DongHoon Choi, Bong-
Soo Cha, Yang Soo Jang, Se-Eun Park, SeungJea Tahk, Byoung Joo
Choi, Joon-Han Shin, Soo-Jin Kang, So-Yeon Choi, Hong-Seok Lim,
Myoung Ho Yoon, Hae-Jin Kim, Dae-Jung Kim, Kwan-Woo Lee.
Latvia: Valdis Pirags, Andrejs Erglis, Irina Kutajeva, Agnis Zvaig-
zne, Karlis Trusinskis, Indulis Kumsars, Sanda Jegere, Dace Son-
dore, Inga Narbute, Sandra Steina, Inga Balcere, Rota Ritenberga.
Mexico: Guillermo Gonzalez-Galvez, Emilia Susana Pelayo-Orozco,
Blanca Leticia Sa´nchez-Michel, Armando García-Castillo, Fernando
Lavalle-Gonza´lez, Violeta Imelda Gallardo-Montejano, Arturo
Abundes-Velasco. The Netherlands: H.R. Michels, R.S.W. Entjes, B.
Bravenboer, K. Krasznai, P.W. Serruys, Martin van der Ent, A.H.
Bootsma, P.C. Smits, L.J.D.M. Schelfhout. Poland: Henryk
Wysocki, Agnieszka Rutkowska, Dagmara Przymuszala, Mieczys-
law Dziarmaga, Janusz Tarchalski, Agnieszka Trawczynska, Jan
Peruga, Ewa Mazurek, Malgorzata Sporna, Aleksander Jachniewicz,
Adam Witkowski, Lukasz Kalinczuk, Jerzy Pregowski, Bozena
Norwa-Otto, Andrzej Ciszewski, Zbigniew Chmielak, Witold
1184 Circulation March 16, 2010
Downloaded from http://ahajournals.org by on September 4, 2019
Ruzyllo, Pawel Buszman, Agata Gruszka, Leszek Stachaczyk, Mar-
cin Debinski, Stanislaw Trznadel, Aleksander Zurakowski, Krzysz-
tof Milewski. Russian Federation: Ludmila Nedosugova, Bagrat
Alekyan, Simon Matskeplishvili, Yuriy Buziashvili, Inna Klefor-
tova, Samson Madoyan, Igor Shuvaev, Dzhamshed Kamardinov,
Marat Aripov, Nino Tsereteli, Elvina Tugeeva, Natalya Trubitsina,
Leonid Kokov, Andrey Ardashev, Svetlana Voloshko, Andrey Sha-
varov, Dmitriy Mangutov, Elena Potekhina, Vasiliy Chestukhin,
Anna Sibyakina, Alexander Pokatilov, Alexey Mironkov. Spain:
Carlos Macaya, Luis Azcona Varela, M
a
Luz Capote Toledo,
Fernando Alfonso Manterola, Manuel Sabate´, Alfonso Sua´rez, A
´ngel
Díaz, Javier Goicolea, Jose´ Ramo´ n Domínguez Puente, Raymundo
Ocaranza Sa´nchez, Juan Francisco Oteo, Alberto Simal, Arturo
García Touchard, Víctor Castro, Amadeo Betriu, Mo´nica Masotti,
Rafael Ruiz, Sergio Rueda, Enric Esmatjes, Vicente Mainar, Juan
Miguel Ruiz Nodar, Javier Pineda, Jose´ Valencia, Pascual Bordes,
O
´scar Moreno, Antonio Pico, Alberto Berenguer, Enric Domingo,
Josep Guindo, Adriana Soto, Jorge Mesa, Eladio Jose´ Losada
Grande, Inocencio Anivarro, Juan A
´ngel, Edgardo Kaplinsky, Ima-
nol Otaegui, Bruno García del Blanco, Nuria Batalla, Antonio
Espolita, In˜igo Lozano, Ce´sar Morís, Pablo Avanzas, Juan Ronda´n
Murillo, Concepcio´n Sua´rez Sua´rez, Elías Delgado, Angel Cequier,
Joan Antoni Gomez Hospital, Francesc Jara, Jaume Maristany,
Marcelo Jime´nez, Eduard Montanya, Emili Iraculis, Jose´ María
Herna´ndez, María Jose´ Pico´ n, Juan Alonso Briales, Manuel Jime´nez
Navarro, Araceli Frutos García, Pilar Carrillo, Vicente Bertomeu,
Francisco Jose´ Pomares Go´ mez, Jose´ Ramo´ n Domínguez Escribano,
Ramo´n Lo´pez Palop, Fina Mauri, Andreu Nubiola Calonge, Vicente
Valle Tudela, Gerardo Aguilar, Oriol Rodríguez Leor, Eduardo
Ferna´ndez Nofredias. Sweden: Bertil Wennerblom, Thomas Hedner,
Inga-Britt Kling, Eva Dahl, Per Albertsson, Lars Grip, Asa Dahl,
Staffan Svensson, Stanko Skrtic, Farshad Tahmasebiepour, Tage
Nilsson, Karin Astrom Olsson, Marie Escar, Thomas Hedner, Bjorn
Wahlstrand, Go¨ran Matejka, Per Tornvall, Kerstin Hoglund, Camilla
Hage, Eva Arnqvist-Gustafsson. Thailand: Srun Kuanprasert, Am-
pica Mangklabruks, Supawan Buranapin, Kaset Chimplee, Natapong
Kosachunhanun, Padiporn Vasinanukorn, Sarat Sunthornyothin,
Surapun Sitthisook, Jarkarpun Chaipromprasit, Sompongse Suwan-
walaikorn. United States: Steven Bailey, Bogdana Ioana Balas,
Barbara Walz, Ralph DeFronzo, Devang Patel, Marc Feldman,
Robert Chilton, Umesh Arora, Michael Ball, Catherine Bain, Greg-
ory Eisner, Matthew French, Edward Fry, John Fry, Jack Hall, James
Hermiller, Zachary Hodes, Elaine Moen, Charles Orr, Kirk Parr,
Thomas Peters, Ali Rizvi, Donald Rothbaum, Charles Taylor, James
VanTassel, Joseph Barrera, Daniel Kulick, Arthur Loussararian,
Mim Mulford, Venkatesh Nadar, Stanley Lewin, Renu Joshi, Robert
Chadband, Harold Settle, Michael Simpson, Michael Rowland,
Robert Foster, Van Reeder, Andrew Kronenberg, Andrew Brian,
Brian Snoddy, Rachel Wilkinson, Arthur Chernoff, Catherine Gar-
rison, Dennis Morris, Catherine Anastasopoulou, Shahriar Yazdan-
far, Robert Chow, Robert Ferrera, Robert Schott, Linda Morrow,
Sailesh Shah, Burritt Haag, Kamal Shoukri, Michael LeMay, Mar-
garet Sharron, Ayesha Parvoz, Vasantha Reddy, Vilma Carlone,
Marc Schweiger, Ibitoro Osakwe, Dominica Kelley, Gregory Guigli-
ano, Satyendra Giri, Gopal Gododia, Shashin Desai, Nikhita Dhruv,
Rajesh Desai, Theodore Heierman, Nancy Parker, Lucia Lilien,
Laura Bryan, Henry Lui, Keith Atkins, Eie Hage-Korbin, Michelle
Magee, Robert Ratner, Asha Thomas, Angela Silverman, Wayman
Cheatham, Gabriel Uwaifo, Debra Wells, Ron Waksman, Jennifer
Holst, Kenneth Kent, John Laird, Augusto Pichard, Lowell Satler,
John Sharretts, William Suddath, Marc Mitchell, Scott Eberly, Allen
Geltzer, Thomas Amidon, Christopher Kozlowski, James Leggett,
Joseph Doucette, Neal Perlmutter, Jorge Calles-Escandon, Talal
Baki, William Little, Robert Applegate, Michael Kutcher, Sanjay
Gandhi, Renato Santos, George Dangas, Mark Apfelbaum, Michael
Collins, Daniel Donovan, Edward Kreps, Martin Leon, Roxana
Mehran, Jeffrey Moses, Issam Moussa, Leroy Rabbani, Gregg Stone,
Hal Wasserman, Judah Weinberger, Giora Weisz, Charles Whit-
comb, Jason Rogers, Thomas Aoki, Todd Brown, Ann Munson, Jon
Resar, Julie Miller, Alan Heldman, Kathleen Citro, Mary Ellen
Ehlers, Jan Houghton, William Georgitis, Christopher Lang, David
Kinnard, Steven Resnick, Albert Tseng, William Herzog, Michael
Silverman, George Groman, Keith Friedman, Alexander Chud-
novsky, Gebreye Rufael, Marie Gregory, Stephen Hippler, Craig
Kurtz, Alan Chu, Robert Crawford, Ronald Rabjohns, David Best,
Darrel Gumm, John Rashid, Barry Clemson, S. Craig Kurtz, Paul
Schmidt, Stephen Schneider, Alpesh Patel, Jai Agarwal, Sheri
Funderburk, Mark Moshiyakhov, Lilliana Cohen, Abel Moreyra,
Gretchen Perilli, Anthony Lombardo, Shelley Greenhaus, Lisa Mo-
tivalli, Tudor Vaganescu, William Schafranek, Pirouz Parang, Ajay
Chander, Anthony Messina, Syed Hussassian, Chirag Shah, Louis
Amorosa, Rita Louard, Ray Matthews, Thomas Shook, Steven
Burstein, Guy Mayeda, David Shavelle, Mariola Ficinski, Jane
Reusch, Irene Schauer, Sarah Bull, John Messenger, Cecilia Wang,
Rocio Pereira, Ivan Casserly, Stacey Mitchell, Kylie Carson, Barbara
Watson, Paul Rosenblit, Reynaldo Mulingtapang, John Sullebarger,
Robyn Aydelott, Nancy Grove, Huy Khuu, Beth Major, Manuel
Sainz de la Pena, John Sullebarger, Fadi Martar, Marc Taylor,
Anthony Morrison, Joel Strom, Reynaldo Mulingtapang, Tywaun
Tillman, MaryAnn Yarborough, Helen Taylor, Cynthia Williams,
Benjamin Rosin, Laryenth Lancaster, Veronica Piziak, David Rob-
ertson, Charles Wilmer, Charles Wickliffe, Kenneth Taylor, Spencer
King, III, William Jacobs, Mina Jacob, John Hurst, Jr, Harold
Carlson, Charles Brown, III, William Blincoe, William Ballard,
Scott Anderson, John Tayek, David Shavelle, Matthew Budoff, Gul
Khawar, Edward Kosinski, Jeffrey Lins, Jay Zdunek, Alexander
Kalenak, Tammy Outly, Richard Begg, Minn Mach, Daniel Eisen-
berg, Roger On, Joseph Lee, Eric Lee, A. Farfel, Khakloun Sroujieh,
Augusta Silva, Bharat Shah, Boris Larreta, Sako Chen, Douglas
Schreck, David Sato, Barry Rosen, William Gifford, Shawna
Weisler, Tony DeMartini, Norman Soler, Linda L McCall, Romesh
Kahdori, Gabor Matos, Frank Aguirre, Vincent Zuck, Wilfred Lam,
Raymond Kacich, Ziad Issa, John Gill, Mansura Ghani, Chandhiran
Rangaswamy, Nasar Nallamothu, Brian Miller, James Leibsohn, Beth
Lalande, Catherine Zyniecki, John Patterson, Michael Sollenberger,
David Bohle, Jaime Zapata, Davidson Givens, Charles Harris, John
McCabe, Usman Khawaja, Gary Reynaldo, Gregory Ledger, Thomas
Rossiter, Anthony Giglio, Kelvin VanOsdol, David Cochran, Donald
Myears, Robert Merritt, John Hawkins, Osvaldo Brusco, Srikanth
Damaraju, John Pappas, Greg Arsenis, R. Kiesz, Thomas Rossiter,
Alexander Shepherd, Krishnaswami Vijayaraghavan, W. Scott Rob-
ertson, Fuad Ibrahim, Alan Tenaglia, Mehrdad Saririan, Bradley
Oswood, Leslie Barnett-Avery.
Sources of Funding
This study was supported by GlaxoSmithKline. The sponsor partic-
ipated in the design of the study and provided logistical support
during the trial. Monitoring of the study was performed by the
sponsor, which also maintained the trial database. Statistical analyses
of the primary and secondary end points were independently per-
formed by both the sponsor and an independent academic statisti-
cian, and the results were cross-checked. Manuscript development
was led by the corresponding author in collaboration with the other
authors. Authors employed by the sponsor reviewed the manuscript
and suggested changes, but the final decision on content was retained
by the academic authors.
Disclosures
Dr Gerstein has received honoraria from AstraZeneca, Boehringer
Ingelheim, GlaxoSmithKline, Lilly, Merck, Novo Nordisk, and
Sanofi-Aventis and grants from GlaxoSmithKline, King, Merck,
Novo Nordisk, and Sanofi-Aventis. Dr Ratner has received grants
from AstraZeneca, Bayhill Therapeutics, Boehringer Ingelheim,
GlaxoSmithKline, Merck, Novo Nordisk, Pfizer, Takeda, and Ver-
alight; served as an advisor to Amylin, AstraZeneca, Eli Lilly,
GlaxoSmithKline, Lifescan, Novo Nordisk, Roche, Sanofi-Aventis,
Sirtris, Takeda, and Tethys; and holds stock ($10 000 value) in
Abbott, Johnson and Johnson, and Merck. Dr Cannon has received
grants from Accumetrics, AstraZeneca, Bristol-Myers Squibb/Sanofi
Partnership, GlaxoSmithKline, Merck, and the Merck/Schering
Gerstein et al Rosiglitazone and Coronary Atherosclerosis 1185
Downloaded from http://ahajournals.org by on September 4, 2019
Plough Partnership and served as advisor to and holds an equity
position in Automedics Medical Systems. Drs Serruys, van Es,
García-García, and Fitzgerald received honoraria from GlaxoSmith-
Kline. Dr Nesto has served on speakers’ bureaus for GlaxoSmithK-
line and Takeda and served as an advisor to GlaxoSmithKline. Drs
Kolatkar, Miller, and Huang and B.G. Kravitz are employed by and
own stock in GlaxoSmithKline.
References
1. Huxley R, Barzi F, Woodward M. Excess risk of fatal coronary heart
disease associated with diabetes in men and women: meta-analysis of 37
prospective cohort studies. BMJ. 2006;332:73–78.
2. Selvin E, Marinopoulos S, Berkenblit G, Rami T, Brancati FL, Powe NR,
Golden SH. Meta-analysis: glycosylated hemoglobin and cardiovascular
disease in diabetes mellitus. Ann Intern Med. 2004;141:421– 431.
3. ADVANCE Collaborative Group, Patel A, MacMahon S, Chalmers J,
Neal B, Billot L, Woodward M, Marre M, Cooper M, Glasziou P,
Grobbee D, Hamet P, Harrap S, Heller S, Liu L, Mancia G, Mogensen
CE, Pan C, Poulter N, Rodgers A, Williams B, Bompoint S, de Galan BE,
Joshi R, Travert F. Intensive blood glucose control and vascular outcomes
in patients with type 2 diabetes. N Engl J Med. 2008;358:2560 –2572.
4. Action to Control Cardiovascular Risk in Diabetes Study Group, Gerstein
HC, Miller ME, Byington RP, Goff DC Jr, Bigger JT, Buse JB, Cushman
WC, Genuth S, Ismail-Beigi F, Grimm RH Jr, Probstfield JL, Simons-
Morton DG, Friedewald WT. Effects of intensive glucose lowering in
type 2 diabetes. N Engl J Med. 2008;358:2545–2559.
5. UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive
blood glucose control with metformin on complications in overweight
patients with type 2 diabetes (UKPDS 34). Lancet. 1998;352:854 – 865.
6. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-Year
follow-up of intensive glucose control in type 2 diabetes. N Engl J Med.
2008;359:1577–1589.
7. Ricciardi MJ, Meyers S, Choi K, Pang JL, Goodreau L, Davidson CJ.
Angiographically silent left main disease detected by intravascular ultra-
sound: a marker for future adverse cardiac events. Am Heart J. 2003;146:
507–512.
8. von Birgelen C, Hartmann M, Mintz GS, van Houwelingen KG, Dep-
permann N, Schmermund A, Bo¨ se D, Eggebrecht H, Neumann T, Go¨ ssl
M, Wieneke H, Erbel R. Relationship between cardiovascular risk as
predicted by established risk scores versus plaque progression as
measured by serial intravascular ultrasound in left main coronary arteries.
Circulation. 2004;110:1579 –1585.
9. Nissen SE, Tuzcu EM, Libby P, Thompson PD, Ghali M, Garza D,
Berman L, Shi H, Buebendorf E, Topol EJ; CAMELOT Investigators.
Effect of antihypertensive agents on cardiovascular events in patients
with coronary disease and normal blood pressure: the CAMELOT study:
a randomized controlled trial. JAMA. 2004;292:2217–2225.
10. Nissen SE, Tuzcu EM, Schoenhagen P, Brown BG, Ganz P, Vogel RA,
Crowe T, Howard G, Cooper CJ, Brodie B, Grines CL, DeMaria AN;
REVERSAL Investigators. Effect of intensive compared with moderate
lipid-lowering therapy on progression of coronary atherosclerosis: a ran-
domized controlled trial. JAMA. 2004;291:1071–1080.
11. Okazaki S, Yokoyama T, Miyauchi K, Shimada K, Kurata T, Sato H,
Daida H. Early statin treatment in patients with acute coronary syn-
drome: demonstration of the beneficial effect on atherosclerotic
lesions by serial volumetric intravascular ultrasound analysis during
half a year after coronary event: the ESTABLISH Study. Circulation.
2004;110:1061–1068.
12. Schartl M, Bocksch W, Koschyk DH, Voelker W, Karsch KR, Kreuzer
J, Hausmann D, Beckmann S, Gross M. Use of intravascular
ultrasound to compare effects of different strategies of lipid-lowering
therapy on plaque volume and composition in patients with coronary
artery disease. Circulation. 2001;104:387–392.
13. Takagi T, Yoshida K, Akasaka T, Hozumi T, Morioka S, Yoshikawa
J. Intravascular ultrasound analysis of reduction in progression of
coronary narrowing by treatment with pravastatin. Am J Cardiol. 1997;
79:1673–1676.
14. Tiikkainen M, Hakkinen AM, Korsheninnikova E, Nyman T, Makimattila
S, Yki-Jarvinen H. Effects of rosiglitazone and metformin on liver fat
content, hepatic insulin resistance, insulin clearance, and gene expression
in adipose tissue in patients with type 2 diabetes. Diabetes. 2004;53:
2169 –2176.
15. Kahn SE, Haffner SM, Heise MA, Herman WH, Holman RR, Jones NP,
Kravitz BG, Lachin JM, O’Neill MC, Zinman B, Viberti G; ADOPT
Study Group. Glycemic durability of rosiglitazone, metformin, or gly-
buride monotherapy. N Engl J Med. 2006;355:2427–2443.
16. Haffner SM, Greenberg AS, Weston WM, Chen H, Williams K, Freed
MI. Effect of rosiglitazone treatment on nontraditional markers of car-
diovascular disease in patients with type 2 diabetes mellitus. Circulation.
2002;106:679 – 684.
17. Natali A, Baldeweg S, Toschi E, Capaldo B, Barbaro D, Gastaldelli A,
Yudkin JS, Ferrannini E. Vascular effects of improving metabolic control
with metformin or rosiglitazone in type 2 diabetes. Diabetes Care. 2004;
27:1349 –1357.
18. Derosa G, Gaddi AV, Piccinni MN, Ciccarelli L, Salvadeo S, Peros E,
Ghelfi M, Ferrari I, Cicero AF. Antithrombotic effects of rosiglitazone-
metformin versus glimepiride-metformin combination therapy in patients
with type 2 diabetes mellitus and metabolic syndrome. Pharmacotherapy.
2005;25:637– 645.
19. Kelly AS, Thelen AM, Kaiser DR, Gonzalez-Campoy JM, Bank AJ.
Rosiglitazone improves endothelial function and inflammation but not
asymmetric dimethylarginine or oxidative stress in patients with type 2
diabetes mellitus. Vasc Med. 2007;12:311–318.
20. Qayyum R, Adomaityte J. A meta-analysis of the effect of thiazo-
lidinediones on blood pressure. J Clin Hypertens. 2006;8:19 –28.
21. Mazzone T, Meyer PM, Feinstein SB, Davidson MH, Kondos GT,
D’Agostino RB Sr, Perez A, Provost JC, Haffner SM. Effect of piogli-
tazone compared with glimepiride on carotid intima-media thickness in
type 2 diabetes: a randomized trial. JAMA. 2006;296:2572–2581.
22. Rosmarakis ES, Falagas ME. Effect of thiazolidinedione therapy on
restenosis after coronary stent implantation: a meta-analysis of ran-
domized controlled trials. Am Heart J. 2007;154:144 –150.
23. Nissen SE, Nicholls SJ, Wolski K, Nesto R, Kupfer S, Perez A, Jure H,
De Larochellie` re R, Staniloae CS, Mavromatis K, Saw J, Hu B, Lincoff
AM, Tuzcu EM; PERISCOPE Investigators. Comparison of pioglitazone
vs glimepiride on progression of coronary atherosclerosis in patients with
type 2 diabetes: the PERISCOPE randomized controlled trial. JAMA.
2008;299:1561–1573.
24. Ratner RE, Cannon CP, Gerstein HC, Nesto RW, Serruys PW, Van Es
GA, Kolatkar NS, Kravitz BG, Zalewski A, Fitzgerald PJ; APPROACH
Study Group. Assessment on the Prevention of Progression by Rosigli-
tazone on Atherosclerosis in Diabetes Patients With Cardiovascular
History (APPROACH): study design and baseline characteristics. Am
Heart J. 2008;156:1074 –1079.
25. De Winter SA, Hamers R, Degertekin M, Tanabe K, Lemos PA, Serruys
PW, Roelandt JR, Bruining N. Retrospective image-based gating of
intracoronary ultrasound images for improved quantitative analysis: the
Intelligate method. Catheter Cardiovasc Interv. 2004;61:84 –94.
26. Rodriguez-Granillo GA, Vos J, Bruining N, García-García HM, de
Winter S, Ligthart JM, Deckers JW, Bertrand M, Simoons ML, Ferrari R,
Fox KM, Remme W, De Feyter PJ; Investigators of the EUROPA Study.
Long-term effect of perindopril on coronary atherosclerosis pro-
gression (from the Perindopril’s Prospective Effect on Coronary Ath-
erosclerosis by Angiography and Intravascular Ultrasound Evaluation
[PERSPECTIVE] study). Am J Cardiol. 2007;100:159 –163.
27. Nissen SE, Nicholls SJ, Sipahi I, Libby P, Raichlen JS, Ballantyne CM,
Davignon J, Erbel R, Fruchart JC, Tardif JC, Schoenhagen P, Crowe T,
Cain V, Wolski K, Goormastic M, Tuzcu EM; ASTEROID Investigators.
Effect of very high-intensity statin therapy on regression of coronary
atherosclerosis: the ASTEROID trial. JAMA. 2006;295:1556 –1565.
28. Lachin JM. Worst-rank score analysis with informatively missing obser-
vations in clinical trials. Control Clin Trials. 1999;20:408 – 422.
29. Stocker DJ, Taylor AJ, Langley RW, Jezior MR, Vigersky RA. A ran-
domized trial of the effects of rosiglitazone and metformin on inflam-
mation and subclinical atherosclerosis in patients with type 2 diabetes. Am
Heart J. 2007;153:445– 446.
30. Riche DM, Valderrama R, Henyan NN. Thiazolidinediones and risk of
repeat target vessel revascularization following percutaneous coronary
intervention: a meta-analysis. Diabetes Care. 2007;30:384 –388.
31. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial
infarction and death from cardiovascular causes. N Engl J Med. 2007;
356:2457–2471.
32. Singh S, Loke YK, Furberg CD. Long-term risk of cardiovascular events
with rosiglitazone: a meta-analysis. JAMA. 2007;298:1189 –1195.
33. Shuster JJ, Jones LS, Salmon DA. Fixed vs random effects metaanalysis
in rare event studies: the rosiglitazone link with myocardial infarction and
cardiac death. Stat Med. 2007;26:4375– 4385.
1186 Circulation March 16, 2010
Downloaded from http://ahajournals.org by on September 4, 2019
34. Lago RM, Singh PP, Nesto RW. Congestive heart failure and cardiovas-
cular death in patients with prediabetes and type 2 diabetes given thia-
zolidinediones: a meta-analysis of randomised clinical trials. Lancet.
2007;370:1129 –1136.
35. US Food and Drug Administration. FDA briefing document. Joint
meeting of the Endocrinologic and Metabolic Drugs Advisory Com-
mittee and the Drug Safety and Risk Management Advisory Com-
mittee, July 30, 2007. Available at: http://www.fda.gov/ohrms/dockets/
ac/07/briefing/2007-4308b1-02-fda-backgrounder.pdf. Accessed May
21, 2009.
36. Friedrich JO, Beyene J, Adhikari NKJ. Rosiglitazone: can meta-analysis
accurately estimate excess cardiovascular risk given the available data?
Re-analysis of randomized trials using various methodologic approaches.
BMC Res Notes. 2009;2:5.
37. Mannucci E, Monami M, Di Bari M, Lamanna C, Gori F, Gensini GF,
Marchionni N. Cardiac safety profile of rosiglitazone: a comprehensive
meta-analysis of randomized clinical trials. Int J Cardiol. Mar 26, 2009.
DOI: 10.1016/j.ijcard.2009.01.064. Available at: http://www.
internationaljournalofcardiology.com/article/S0167-5273%2809%2900133-
8/fulltext/. Accessed September 12, 2009.
CLINICAL PERSPECTIVE
The thiazolidinedione class of drugs has many favorable metabolic and vascular anatomic effects in people with type 2
diabetes mellitus. Whereas the Assessment on the Prevention of Progression by Rosiglitazone on Atherosclerosis in
Diabetes Patients With Cardiovascular History (APPROACH) trial did not show a clear reduction in coronary
atherosclerosis versus glipizide, its results are consistent with other studies of rosiglitazone and pioglitazone that did
suggest reduced carotid and coronary atherosclerosis. Moreover, the large randomized outcomes trials of the thiazo-
lidinediones that have been completed to date are consistent with the hypothesis that (1) pioglitazone may reduce
cardiovascular outcomes compared with placebo and (2) the effect of rosiglitazone on cardiovascular outcomes is similar
to that of metformin and to that of sulfonylureas. With the exception of fluid retention and pulmonary edema, these trial
findings support the importance of clearly testing the cardiovascular effects of both of these drugs within 1 trial. The
Thiazolidinedione Intervention With Vitamin D Evaluation (TIDE) trial is a large placebo-controlled trial of 16 000
participants that is currently assessing the cardiovascular effects of both thiazolidinediones versus placebo when added to
current therapy. It will also clearly evaluate whether either of the 2 thiazolidinediones differ with respect to cardiovascular
outcomes and will clearly determine whether either or both of these drugs prevents, promotes, or has a neutral effect on
serious cardiovascular outcomes.
Gerstein et al Rosiglitazone and Coronary Atherosclerosis 1187
Downloaded from http://ahajournals.org by on September 4, 2019
... The result also has been reported in SYMPHONY substudy [51]. Although these studies highlighted the central role of insulin resistance in progression of atherothrombotic events in DM patients, these outcomes have not been established in the later published APPROACH trial [52]. In this study, there was no significant difference between the group treated with thiazolinedione rosiglitazone and the group treated with glipizide, an insulin secretagogue, in patients with type-2 DM and CAD [52]. ...
... Although these studies highlighted the central role of insulin resistance in progression of atherothrombotic events in DM patients, these outcomes have not been established in the later published APPROACH trial [52]. In this study, there was no significant difference between the group treated with thiazolinedione rosiglitazone and the group treated with glipizide, an insulin secretagogue, in patients with type-2 DM and CAD [52]. ...
Article
Full-text available
Patients with Diabetes Mellitus (DM) have accelerated atherosclerosis, which is the main essential factor contributing to the high risk of atherothrombotic events in these patients. Atherothrombotic complications are the principal cause of morbidity and mortality in patients with DM. Both atherosclerosis and the increased risk of thrombotic vascular events may result from dyslipidaemia, endothelial dysfunction, platelet hyperreactivity, an impaired fibrinolytic balance, and abnormal blood flow. Platelets of DM patients are characterised by dysregulation of several signalling pathways causing increased adhesion, activation and aggregation. Platelet function of patients with DM is complicated by several mechanisms, such as hyperglycaemia, insulin deficiency and resistance, associated metabolic conditions, and cellular abnormalities. The present manuscript purposes to provide a review on the up-to-date status of data on platelet abnormalities that characterise patients with DM.
... 23 Studies have shown that the administration of statins and thiazolidinedione (pioglitazone and rosiglitazone) causes a significant regression of the coronary atherosclerotic plaques and they can be used in the secondary prevention of cardiovascular disease. [24][25][26] A wide angle between the LM and LAD could predict in-stent restenosis after stenting of the proximal LAD, as reported by Konishi et al. 27 The effect of the bifurcation angle on 12-month outcome after stenting of the LM-LAD was investigated by Amemiya et al. 28 The angle was evaluated by three-dimensional quantitative coronary angiography analysis. They reported a significant relationship between the pre-procedure bifurcation angle and adverse cardiac events. ...
Article
Aim: The aim of this study was to evaluate the association between the left main coronary artery (LM) bifurcation angle and the severity of the proximal left anterior descending coronary artery (LAD) stenosis. Methods: Two hundred patients with suspected coronary artery disease who had coronary angiography were included in this observational study. The severity of coronary artery stenosis was analysed using quantitative coronary angiography software (QCA analysis). The LM-LAD and LAD-left circumflex artery (LCX) angles were measured using software (IC MEASURE) in two-dimensional axial images. Results: The patients were divided into two groups. The first group included 100 patients with significant proximal LAD stenosis (≥ 50%) and the second, those with LAD stenosis < 50% (100 patients). Patients with significant proximal LAD stenosis were older and had a higher frequency of diabetes mellitus, and higher serum creatinine and low-density lipoprotein levels than those with non-significant LAD stenosis. The LM-LAD and LAD-LCX angles in patients with significant proximal LAD stenosis were wider than in patients with non-significant LAD stenosis (p < 0.001). The cut-off value of 42° of the LM-LAD angle had a sensitivity of 73% and specificity of 70% to predict significant proximal LAD stenosis. The cut-off value of 68° of the LAD-LCX angle had a sensitivity of 68% and specificity of 62% to predict significant proximal LAD disease. In a multivariate logistic regression analysis, LM-LAD and LAD-LCX angles were independent factors for the development of significant proximal LAD stenosis. Conclusions: Wider LM-LAD and LAD-LCX angles were associated with the severity of proximal LAD disease. Preventative measures and close follow up are needed in such cases to improve their cardiovascular outcome.
... In the Pioglitazone Effect on Regression of Intravascular Sonographic Coronary Obstruction Prospective Evaluation (PERISCOPE) study, 110 pioglitazone was superior to glimepiride in inhibiting plaque progression (PAV: −0.16% vs. +0.73%, P = 0.02), whereas the Assessment on the Prevention of Progression by Rosiglitazone on Atherosclerosis in Diabetes Patients With Cardiovascular History (APPROACH) study 111 showed no difference between the effect of rosiglitazone and glipizide on PAV (see Supplementary material online, Table S1). ...
Article
Intravascular imaging has been often used over the recent years to examine the efficacy of emerging therapies targeting plaque evolution. Serial intravascular ultrasound, optical coherence tomography, or near infrared spectroscopy-intravascular ultrasound studies have allowed us to evaluate the effects of different therapies on plaque burden and morphology, providing unique mechanistic insights about the mode of action of these treatments. Plaque burden reduction, a decrease in necrotic core component or macrophages accumulation – that have been associated with inflammation - and an increase in fibrous cap thickness over fibroatheromas have been used as surrogate endpoints to assess the value of several drugs in inhibiting plaque evolution and improving clinical outcomes. However, some reports have demonstrated weak associations between the effects of novel treatments on coronary atheroma and composition and their prognostic implications. This review examines the value of invasive imaging in assessing pharmacotherapies targeting atherosclerosis. It summarizes the findings of serial intravascular imaging studies assessing the effects of different drugs on atheroma burden and morphology and compares them with the results of large-scale trials evaluating their impact on clinical outcome. Furthermore, it highlights the limited efficacy of established intravascular imaging surrogate endpoints in predicting the prognostic value of these pharmacotherapies and introduces alternative imaging endpoints based on multimodality/hybrid intravascular imaging that may enable more accurate assessment of the athero-protective and prognostic effects of emerging therapies.
... Sulphonylureas were associated with a significantly increased risk of cardiovascular mortality. [103][104][105][106][107][108][109] Adapted with permission from Phung et al. 101 ...
Article
Full-text available
Despite the importance of individualised strategies for patients with type 2 diabetes mellitus (T2DM) and the availability of alternative treatments, including glucagon‐like peptide‐1 receptor agonists (GLP‐1 RAs), sulphonylureas are still widely used in practice. Clinical evidence shows that GLP‐1 RAs may provide better and more durable glycaemic control than sulphonylureas, with lower risk of hypoglycaemia. Other reported benefits of GLP‐1 RAs include weight loss rather than weight gain (as observed with sulphonylureas), blood pressure reductions and improvements in lipid profiles. In general, the main adverse events with GLP‐1 RAs are gastrointestinal in nature. The respective modes of action of GLP‐1 RAs and sulphonylureas contribute to differences in durability of glycaemic control (related to effects on beta‐cells) and effects on body weight. Moreover, the glucose‐dependent mode of action of GLP‐1 RAs, which favours a low incidence of hypoglycaemia, contrasts with the glucose‐independent mode of action of sulphonylureas. Evidence from cardiovascular outcomes trials indicates a consistent finding of cardiovascular safety across the GLP‐1 RAs and suggests a class benefit for the long‐acting GLP‐1 RAs in reducing three‐point major adverse cardiovascular events, cardiovascular mortality and all‐cause mortality. In contrast, potential concerns relating to an increased incidence of adverse cardiovascular events with sulphonylureas have yet to be fully resolved. Recent updates to management guidelines recommend that treatment selection for patients with T2DM should consider clinical trial evidence of cardiovascular safety. Available evidence suggests that this selection should give preference to GLP‐1 RAs over sulphonylureas, especially for patients at high cardiovascular risk. This article is protected by copyright. All rights reserved.
Article
Full-text available
Objective To evaluate sodium-glucose cotransporter-2 (SGLT-2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists in patients with type 2 diabetes at varying cardiovascular and renal risk. Design Network meta-analysis. Data sources Medline, Embase, and Cochrane CENTRAL up to 11 August 2020. Eligibility criteria for selecting studies Randomised controlled trials comparing SGLT-2 inhibitors or GLP-1 receptor agonists with placebo, standard care, or other glucose lowering treatment in adults with type 2 diabetes with follow up of 24 weeks or longer. Studies were screened independently by two reviewers for eligibility, extracted data, and assessed risk of bias. Main outcome measures Frequentist random effects network meta-analysis was carried out and GRADE (grading of recommendations assessment, development, and evaluation) used to assess evidence certainty. Results included estimated absolute effects of treatment per 1000 patients treated for five years for patients at very low risk (no cardiovascular risk factors), low risk (three or more cardiovascular risk factors), moderate risk (cardiovascular disease), high risk (chronic kidney disease), and very high risk (cardiovascular disease and kidney disease). A guideline panel provided oversight of the systematic review. Results 764 trials including 421 346 patients proved eligible. All results refer to the addition of SGLT-2 inhibitors and GLP-1 receptor agonists to existing diabetes treatment. Both classes of drugs lowered all cause mortality, cardiovascular mortality, non-fatal myocardial infarction, and kidney failure (high certainty evidence). Notable differences were found between the two agents: SGLT-2 inhibitors reduced mortality and admission to hospital for heart failure more than GLP-1 receptor agonists, and GLP-1 receptor agonists reduced non-fatal stroke more than SGLT-2 inhibitors (which appeared to have no effect). SGLT-2 inhibitors caused genital infection (high certainty), whereas GLP-1 receptor agonists might cause severe gastrointestinal events (low certainty). Low certainty evidence suggested that SGLT-2 inhibitors and GLP-1 receptor agonists might lower body weight. Little or no evidence was found for the effect of SGLT-2 inhibitors or GLP-1 receptor agonists on limb amputation, blindness, eye disease, neuropathic pain, or health related quality of life. The absolute benefits of these drugs vary substantially across patients from low to very high risk of cardiovascular and renal outcomes (eg, SGLT-2 inhibitors resulted in 5 to 48 fewer deaths in 1000 patients over five years; see interactive decision support tool ( https://magicevidence.org/match-it/200820dist/#!/ ) for all outcomes. Conclusions In patients with type 2 diabetes, SGLT-2 inhibitors and GLP-1 receptor agonists reduced cardiovascular and renal outcomes, with notable differences in benefits and harms. Absolute benefits are determined by individual risk profiles of patients, with clear implications for clinical practice, as reflected in the BMJ Rapid Recommendations directly informed by this systematic review. Systematic review registration PROSPERO CRD42019153180.
Article
Full-text available
Purpose of Review Thiazolidinediones (TZDs) are the only pharmacologic agents that specifically treat insulin resistance. The beneficial effects of TZDs on the cardiovascular risk factors associated with insulin resistance have been well documented. TZD use has been limited because of concern about safety issues and side effects. Recent Findings Recent studies indicate that cardiovascular toxicity with rosiglitazone and increase in bladder cancer with pioglitazone are no longer significant issues. There are new data which show that pioglitazone treatment reduces myocardial infarctions and ischemic strokes. New data concerning TZD-mediated edema, congestive heart failure, and bone fractures improves the clinician’s ability to select patients that will have minimal significant side effects. Summary Thiazolidinediones are now generic and less costly than pharmaceutical company–promoted therapies. Better understanding of the side effects coupled with clear benefits on the components of the insulin resistance syndrome should promote TZD use in treating patients with type 2 diabetes.
Article
Full-text available
Background The last international consensus on the management of type 2 diabetes (T2D) recommends SGLT-2 inhibitors or GLP-1 agonists for patients with clinical cardiovascular (CV) disease; metformin remains the first-line glucose lowering medication. Last studies suggested beneficial effects of SGLT-2 inhibitors or GLP-1 agonists compared to DPP-4 inhibitors, in secondary CV prevention. Recently, a potential benefit of SGLT-2 inhibitors in primary CV prevention also has been suggested. However, no comparison of all the new and the old hypoglycemic drugs is available on CV outcomes. We aimed to compare the effects of old and new hypoglycemic drugs in T2D, on major adverse cardiovascular events (MACE) and mortality. Methods and findings We conducted a systematic review and network meta-analysis of clinical trials. Randomized trials, blinded or not, assessing contemporary hypoglycemic drugs on mortality or MACE in patients with T2D, were searched for in Medline, the Cochrane Central Register of Controlled Trials (CENTRAL), and ClinicalTrials.gov. References screening and data extraction were done by multiple observers. Each drug was analyzed according to its therapeutic class. A random Bayesian network meta-analysis model was used. The primary outcomes were overall mortality, cardiovascular mortality, and MACE. Severe adverse events and severe hypoglycemia were also recorded. 175,966 patients in 34 trials from 1970 to 2018 were included. No trials evaluating glinides or alpha glucosidase inhibitors were found. 17 trials included a majority of patients with previous cardiovascular history, 16 trials a majority of patients without. Compared to control, SGLT-2 inhibitors were associated with a decreased risk of overall mortality (OR = 0.84 [95% CrI: 0.74; 0.95]), SGLT-2 inhibitors and GLP-1 agonists with a decreased risk of MACE (OR = 0.89 [95% CrI: 0.81; 0.98] and OR = 0.88 [95% CrI: 0.81; 0.95], respectively). Compared to DPP-4 inhibitors, SGLT-2 inhibitors were associated with a decreased risk of overall mortality (OR = 0.82 [95% CrI: 0.69; 0.98]), GLP-1 agonists with a decreased risk of MACE (OR = 0.88 [95% CrI: 0.79; 0.99]). Insulin was also associated with an increased risk of MACE compared to GLP-1 agonists (OR = 1.19 [95% CrI: 1.01; 1.42]). Insulin and sulfonylureas were associated with an increased risk of severe hypoglycemia. In the trials including a majority of patients without previous CV history, the comparisons of SGLT-2 inhibitors, metformin and control did not showed significant differences on primary outcomes. We limited our analysis at the therapeutic class level. Conclusions SGLT-2 inhibitors and GLP-1 agonists have the most beneficial effects, especially in T2D patients with previous CV diseases. Direct comparisons of SGLT-2 inhibitors, GLP-1 agonists and metformin are needed, notably in primary CV prevention. Trial registration PROSPERO CRD42016043823.
Article
Full-text available
Purpose of Review Intravascular ultrasound (IVUS) is an essential tool in the cardiac cath lab, providing tomographic slices series of intra-luminal and vascular wall images with high quality and spatial resolution, that complement angiographic luminal information. It has been successfully used in research for the understanding of atherosclerotic process, and it was the first clinical method that delineated diverse patterns of atherosclerosis. IVUS also contributed to the advancement in the knowledge of restenosis after percutaneous coronary interventions (PCI). Recent Findings Routine clinical of IVUS is hindered in most Latin American countries due to lack of reimbursement. In comparison with the rate of IVUS utilization around 5 % in the USA and of 5 % in Europe, the rate is less than 1 % in Latin America. These low rates occur despite the fact that the European, American, and Brazilian guidelines agreed regarding IVUS utilization in specific clinical settings: assessment of non-left main intermediate lesions as evaluated by angiography, assessment of angiographically uncertain severity of left main coronary artery disease, and as a guidance to optimization of coronary stent implantation. Summary In this review, we emphasize the importance of IVUS as a research tool in selected patients with coronary heart disease, especially those treated with PCI, particularly in diabetic patients, and also discuss some data that support its routine clinical use.
Article
Previous studies suggested that dyslipidemia was potentially associated with anti-diabetic medications of sulfonylureas (SUs). The results were, however, inconsistent. We conducted a meta-analysis of randomized controlled trials (RCTs) to assess the effects of SUs on the level of lipids in patients with type 2 diabetes mellitus (T2DM). We searched PubMed, EMBASE and CENTRAL databases for RCTs that addressed the effects of second- and/or third-generation SUs on T2DM patients and reported lipids profiles with study duration of at least 12 weeks. Two reviewers independently screened literature, collected data, and assessed methodological quality of included studies. The meta-analysis was performed by using the RevMan5.1 software. A total of 59 RCTs were included, of which 52 were included for final meta-analysis. The results suggested that SUs statistically increased the levels of FFA (SMD = 0.24, 95%CI 0.06 to 0.42) and TG (MD = 0.06, 95%CI 0.02 to 0.10), but decreased HDL-C (MD = -0.07, 95%CI -0.11 to -0.04) and LDL-C (MD = -0.11, 95%CI -0.17 to -0.04); but the SUs had no effect on TC (MD = 0.01, 95%CI -0.05 to 0.08), ApoA1 (MD = 0.01, 95%CI -0.03 to 0.04), and Apo B (MD = -0.01, 95%CI -0.05 to 0.03). When compared to metformin, SUs could increase TC and LDL-C; compared to glinides, SUs increased TC and lowered HDL-C; compared to thiazolidinediones, SUs reduced TC, LDL-C, HDL-C, and increase TG. SUs have a small effect on lipids, although they may statistically increase the level of FFA and TG, and decrease LDL-C and HDL-C. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
Article
Full-text available
glucose levels represents one of the principal treatment goals of diabetes therapy, it has been difficult to demonstrate a fa-vorable effect of improved glycemic control on the macrovascular compli-cations of this disease. 1,2 No antidia-betic regimen has demonstrated the ability to reduce the progression of coronary atherosclerosis. Accord-ingly, there is little evidence to sup-port a preference of one class of glucose-lowering medication over any other as a means to reduce atherosclerotic dis-ease burden. 3 Sulfonylureas have been available for decades, lower blood glu-cose by acting as insulin secreta-gogues, and represent one of the most commonly-used classes of antidia-betic therapy. Thiazolidinediones (TZDs) are a relatively new class of an-tidiabetic agents that reduce glucose pri-For editorial comment see p 1603.
Article
Full-text available
The effect of antihypertensive drugs on cardiovascular events in patients with coronary artery disease (CAD) and normal blood pressure remains uncertain. To compare the effects of amlodipine or enalapril vs placebo on cardiovascular events in patients with CAD. Double-blind, randomized, multicenter, 24-month trial (enrollment April 1999-April 2002) comparing amlodipine or enalapril with placebo in 1991 patients with angiographically documented CAD (>20% stenosis by coronary angiography) and diastolic blood pressure <100 mm Hg. A substudy of 274 patients measured atherosclerosis progression by intravascular ultrasound (IVUS). Patients were randomized to receive amlodipine, 10 mg; enalapril, 20 mg; or placebo. IVUS was performed at baseline and study completion. The primary efficacy parameter was incidence of cardiovascular events for amlodipine vs placebo. Other outcomes included comparisons of amlodipine vs enalapril and enalapril vs placebo. Events included cardiovascular death, nonfatal myocardial infarction, resuscitated cardiac arrest, coronary revascularization, hospitalization for angina pectoris, hospitalization for congestive heart failure, fatal or nonfatal stroke or transient ischemic attack, and new diagnosis of peripheral vascular disease. The IVUS end point was change in percent atheroma volume. Baseline blood pressure averaged 129/78 mm Hg for all patients; it increased by 0.7/0.6 mm Hg in the placebo group and decreased by 4.8/2.5 mm Hg and 4.9/2.4 mm Hg in the amlodipine and enalapril groups, respectively (P<.001 for both vs placebo). Cardiovascular events occurred in 151 (23.1%) placebo-treated patients, in 110 (16.6%) amlodipine-treated patients (hazard ratio [HR], 0.69; 95% CI, 0.54-0.88 [P = .003]), and in 136 (20.2%) enalapril-treated patients (HR, 0.85; 95% CI, 0.67-1.07 [P = .16]. Primary end point comparison for enalapril vs amlodipine was not significant (HR, 0.81; 95% CI, 0.63-1.04 [P = .10]). The IVUS substudy showed a trend toward less progression of atherosclerosis in the amlodipine group vs placebo (P = .12), with significantly less progression in the subgroup with systolic blood pressures greater than the mean (P = .02). Compared with baseline, IVUS showed progression in the placebo group (P<.001), a trend toward progression in the enalapril group (P = .08), and no progression in the amlodipine group (P = .31). For the amlodipine group, correlation between blood pressure reduction and progression was r = 0.19, P = .07. Administration of amlodipine to patients with CAD and normal blood pressure resulted in reduced adverse cardiovascular events. Directionally similar, but smaller and nonsignificant, treatment effects were observed with enalapril. For amlodipine, IVUS showed evidence of slowing of atherosclerosis progression.
Article
Background Epidemiologic studies have shown a relationship between glycated hemoglobin levels and cardiovascular events in patients with type 2 diabetes. We investigated whether intensive therapy to target normal glycated hemoglobin levels would reduce cardiovascular events in patients with type 2 diabetes who had either established cardiovascular disease or additional cardiovascular risk factors. Methods In this randomized study, 10,251 patients (mean age, 62.2 years) with a median gly- cated hemoglobin level of 8.1% were assigned to receive intensive therapy (targeting a glycated hemoglobin level below 6.0%) or standard therapy (targeting a level from 7.0 to 7.9%). Of these patients, 38% were women, and 35% had had a previous car- diovascular event. The primary outcome was a composite of nonfatal myocardial infarction, nonfatal stroke, or death from cardiovascular causes. The finding of higher mortality in the intensive-therapy group led to a discontinuation of intensive therapy after a mean of 3.5 years of follow-up. Results At 1 year, stable median glycated hemoglobin levels of 6.4% and 7.5% were achieved in the intensive-therapy group and the standard-therapy group, respectively. During follow-up, the primary outcome occurred in 352 patients in the intensive-therapy group, as compared with 371 in the standard-therapy group (hazard ratio, 0.90; 95% confidence interval (CI), 0.78 to 1.04; P = 0.16). At the same time, 257 patients in the intensive-therapy group died, as compared with 203 patients in the standard- therapy group (hazard ratio, 1.22; 95% CI, 1.01 to 1.46; P = 0.04). Hypoglycemia requiring assistance and weight gain of more than 10 kg were more frequent in the intensive-therapy group (P
Article
The prevalence of Type 2 diabetes is increasing rapidly and is associated with significant cardiovascular adverse effects. The authors tested the hypothesis that, in patients with type II diabetes, intensive glucose control as evidenced by lowering the glycated hemoglobin level to to ≤6.5% would result in lower rates of major microvascular and macrovascular events. There were 11,140 patients with Type 2 diabetes who were randomly assigned to undergo standard glucose control or intensive glucose control. Intensive glucose control was achieved with the use of gliclazide (modified release) and other drugs as required to reach a glycated hemoglobin ≤6.5%. Primary end points were composites of major microvascular events (new or worsening nephropathy or retinopathy) or macrovascular events (death from cardiovascular causes, non-fatal myocardial infarction, or non-fatal stroke). End points were assessed both jointly and separately. Yes Yes
Article
Some meta-analyses of randomized clinical trials suggested that rosiglitazone could be associated with increased risk for myocardial infarction (MI). Available meta-analyses, based on studies sponsored by GlaxoSmithKline (GSK), failed to include all trials performed with rosiglitazone. Aim of this analysis is the assessment of the cardiovascular risk with rosiglitazone, using a comprehensive data set. Results of 164 trials with duration >4 weeks were retrieved from http://ctr.gsk.co.uk/welcome.asp and from Medline, while unpublished studies were identified from www.clinicaltrials.gov. A total of 164 trials (42,922 and 45,483 patient years for rosiglitazone and comparators, respectively) were included in the analysis. The OR for all-cause and cardiovascular mortality with rosiglitazone was 0.93[0.76;1.14] and 0.94[0.68;1.29], respectively; rosiglitazone-associated risk for nonfatal MI and heart failure was 1.14[0.90;1.45] and 1.69[1.21;2.36], respectively. The risk of heart failure was higher (2.20[1.28;3.78]) when rosiglitazone was administered as add-on therapy to insulin. Figures for rosiglitazone-associated risk for myocardial infarction could be lower than those previously reported on the basis of a smaller number of clinical trials. No increase of all-cause or cardiovascular mortality were observed with rosiglitazone. Conversely, treatment with rosiglitazone is associated with a relevant increase in the risk of heart failure, particularly in insulin-treated patients.