Background Image
Table of Contents Table of Contents
Previous Page  23 / 48 Next Page
Information
Show Menu
Previous Page 23 / 48 Next Page
Page Background

SA JOURNAL OF DIABETES & VASCULAR DISEASE

REVIEW

VOLUME 12 NUMBER 2 • NOVEMBER 2015

65

above 90, with heterogeneities between the studies (

Q

2

= 8.7,

I

2

= 77%,

p

< 0.1). The pooled OR was 0.61 (95% CI: 0.31–1.21,

p

> 0.05, Fig. 6A) in studies whose sample size was 90 or less,

without heterogeneities between the studies (

Q

2

= 2.39,

I

2

= 16%,

p

> 0.1).

The pooled OR was 0.45 (95% CI = 0.27–0.77,

p

< 0.01, Fig. 6B)

in studies whose subjects were European, without heterogeneities

between the studies (

Q

2

= 3.71,

I

2

= 46%,

p

> 0.1). The pooled OR

was 0.37 (95% CI: 0.11–1.27,

p

> 0.05, Fig. 6B) in studies whose

subjects were American and Asian, with heterogeneities between

the studies (

Q

2

= 15.55, I2 = 87%,

p

< 0.1).

The pooled OR was 0.28 (95% CI: 0.19–0.42,

p

< 0.01, Fig. 6C)

in studies whose study method was RCT, without heterogeneities

between the studies (

Q

2

= 2.4, I2 = 0%,

p

> 0.1).

The pooled OR was 0.87 (95% CI: 0.61–1.24,

p

> 0.05, Fig.

6C) in studies whose method of study was non-RCT, without

heterogeneities between the studies (

Q

2

= 0.92, I2 = 0%,

p

> 0.1).

By removing one study at a time, a sensitivity analysis was

performed and the model was rerun to determine the effect on

each estimate. It showed that the above meta-analysis estimates

did not change significantly after removal of each study, implying

that these results were statistically reliable.

SES

BMS

Odds ratio

Odds ratio

Study or subgroup

Events

Total

Events

Total

Weight (%)

M–H, random, (95% CI)

M–H, random, (95% CI)

Aoki J,

et al.

13

112

8

118

22.2

1.81 (0.72–4.54)

Baumgart D,

et al.

5

94

3

96

12.1

1.29 (0.34–4.97)

Chan C,

et al.

0

54

2

29

10.3

0.10 (0.00–2.18)

Daemen J,

et al.

15

206

16

252

43.0

1.16 (0.56–2.40)

Jimenez-Quevedo P,

et al.

1

80

2

80

6.4

0.49 (0.04–5.56)

Maresta A,

et al.

3

68

2

70

6.1

1.57 (0.25–9.70)

Total (95% CI) 614 645

100.0

1.19 (0.74–1.92)

Total events

37

34

Heterogeneity: Chi

2

= 3.88, df = 5 (

p

= 0.57);

I

2

= 0%

Test for overall effect: Z = 0.72 (

p

= 0.47)

Discussion

A growing number of studies has shown the efficacy and safety

of SES versus BMS for treating CAD patients with diabetes,

9,29

but

the outcome has been controversial. In this analysis, we retrieved

six studies, which included 1 259 CAD subjects with diabetes, and

performed a meta-analysis. It showed that the SES group had a

significant reduction in major adverse cardiac events, as well as

target-lesion revascularisations, compared with the BMS group.

There was no significant difference for myocardial infarction or

mortality.

These results are consistent with a recent study that suggested

a significant reduction in target-vessel revascularisations with SES,

but with similar mortality rates.

9

Unlike this study, in which the

incidence of myocardial infarction was higher, our analysis showed

no difference for myocardial infarctions between the groups.

Another recent study conducted in Europeans confirmed

the efficacy of SES compared with BMS, along with comparable

mortality rates and myocardial infarctions,

11

which further proved

the validity of our analysis. The efficacy and safety of SES have been

receiving more and more supportive reports.

30-33

The uniqueness of

our analysis and findings is that it proved the efficacy and safety of

SES in CAD patients with diabetes.

Figure 5

. Forest plots of studies with mortality events in the SES group versus the BMS group.

A

SES

BMS

Odds ratio

Odds ratio

Study or subgroup

Events

Total

Events

Total

Weight (%)

M–H, random, (95% CI)

M–H, random, (95% CI)

6.2.2 Both groups’ sample size > 90

Aoki J,

et al.

27

112

37

118

18.5

0.71 (0.39–1.24)

Baumgart D,

et al.

15

94

38

96

17.2

0.29 (0.15–0.58)

Daemen J,

et al.

44

206

54

252

20.1

1.00 (0.64–1.56)

Subtotal (95% CI) 412 466

55.8

0.61 (0.31–1.20)

Total events

86

129

Heterogeneity: Tau

2

= 0.28, Chi

2

= 8.70, df = 2 (

p

= 0.01);

I

2

= 77%

Test for overall effect: Z = 1.42 (

p

= 0.15)

6.2.3 Both groups’ sample size ≤ 90

Chan C,

et al.

8

54

12

29

12.8

0.25 (0.09–0.71)

Jimenez-Quevedo P,

et al.

8

80

31

80

15.0

0.18 (0.07–0.41)

Maresta A,

et al.

15

68

28

70

16.4

0.42 (0.20–0.90)

Subtotal (95% CI) 202 179

44.2

0.28 (0.16–0.48)

Total events

31

71

Heterogeneity: Tau

2

= 0.04, Chi

2

= 2.39, df = 2 (

p

= 0.30);

I

2

= 16%

Test for overall effect: Z = 4.60 (p < 0.00001)

Total (95% CI) 614 645

100.0

0.42 (0.24–0.74)

Total events

117

200

Heterogeneity: Tau

2

= 0.36, Chi

2

= 20.14, df = 5 (

p

= 0.001); I

2

= 75%

Test for overall effect: Z = 3.00 (

p

= 0.003)

Test for subgroup differences: Chi

2

= 3.12, df = 1 (

p

= 0.08);

I

2

= 68.0%

Figure 6. A:

Forest plots of sample size subgroups