1. Search results and baseline characteristics of patients
Study selection procedure was detailedly proscribed in Fig1. Total 641 studies were found after excluding the duplicate records. 519 studies were excluded after reading titles and abstracts,. 111 of the remained 122 studies were excluded after reading full text. Eventually 11 studies were identified in our meta-analysis, which involves 4 RCTs and 7 observation studies. 5 of 11 studies were conducted in Europe, 3 were in China and 3were in Korea. Given to the ethnic difference, 5 of all are west European while 6 are east Asian. Baseline characteristics of all studies were available except the IMPACTOR-CTO trial. The baseline characteristics of IRCTO trial were not analysed by PSM method, while clinical outcomes were analysed by PSM method. Our analysis includes total 5664 CTO patients, of whom 2904 patients received PCI therapy and 2760 patients received OMT therapy. The baseline characteristics of included studies were showed in Table1.
Table 1
Baseline characteristics in PCI vs OMT arms
Study ID
|
Years
|
Ethicity
|
Number of Population
|
Follow up
(Months)
|
Age(years)
|
Male
|
Hypertension
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
Choi SY
|
2017
|
Asian
|
158
|
158
|
60M
|
64±9
|
64±11
|
117 (74.1)
|
113 (71.5)
|
107 (67.7)
|
111(70.3)
|
Lee SW
|
2018
|
Asian
|
417
|
398
|
36M
|
62.2 ±10.2
|
62.9±9.9
|
344(83.3)
|
319(81.6)
|
262(63.4)
|
238(60.9)
|
Yang JH
|
2016
|
Asian
|
533
|
533
|
45.8M
|
63.5±10.3
|
64.5±11.4
|
425 (79.7)
|
419 (78.6)
|
348 (65.3)
|
350 (65.7)
|
Gao L
|
2018
|
Asian
|
80
|
80
|
47.2M
|
64.76 ± 9.58
|
64.55 ± 11.24
|
56(70)
|
58(72.5)
|
54(67.5)
|
55(68.8)
|
Gao L*
|
2019
|
Asian
|
286
|
286
|
43.2M
|
63.3±9.6
|
63.4±10.5
|
207 (72.4)
|
217 (75.9)
|
199 (69.6)
|
188 (65.7)
|
Yan YF
|
2019
|
Asian
|
118
|
118
|
45M
|
62.00 (54.50-67.00)
|
58.00 (51.00-68.75)
|
90 (76.3)
|
93 (78.8)
|
84 (71.2)
|
79 (66.9)
|
Choo EH
|
2018
|
Asian
|
264
|
264
|
26.4M
|
61.5 ± 09.8
|
61.5 ±10.5
|
199 (75.4)
|
201 (76.1)
|
156 (59.1)
|
159 (60.2)
|
Obedinskiy
|
2018
|
West-European
|
39
|
33
|
1M
|
N
|
N
|
N
|
N
|
N
|
N
|
Werner GS
|
2018
|
West-European
|
259
|
137
|
12M
|
65.2±9.7
|
64.7±9.9
|
215 (83.0)
|
118 (86.1)
|
189 (73.0)
|
98 (71.5)
|
Ladwiniec A
|
2015
|
West-European
|
294
|
294
|
60M
|
64.3±10.0
|
63.9±10.2
|
220 (74.8)
|
220 (74.8)
|
159 (54.1)
|
157 (53.4)
|
Mashayekhi K
|
2018
|
West-European
|
101
|
104
|
12M
|
65 (57–72)
|
68 (61–74)
|
91 (90.1)
|
90 (86.5)
|
81 (80.2)
|
93 (89.4)
|
Study ID
|
Diabetes Mellitus
|
Dyslipidemia
|
Smoking
|
Previous MI
|
Previous PCI
|
Previous CABG
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
Choi SY
|
78 (49.4)
|
72 (45.6)
|
51 (32.3)
|
49 (31.0)
|
79 (50.0)
|
94(59.5)
|
N
|
N
|
N
|
N
|
N
|
N
|
Lee SW
|
132 (32.0)
|
134(34.3)
|
249(60.3)
|
217(55.5)
|
125(30.3)
|
102(26.1)
|
45 (10.9)
|
34(8.7)
|
64(15.5)
|
75(19.2)
|
4(1.0)
|
5(1.3)
|
Yang JH
|
252 (47.3)
|
242 (45.4)
|
143 (26.8)
|
139 (26.1)
|
168 (31.5)
|
156 (29.3)
|
146 (27.4)
|
144 (27.0)
|
142 (26.6)
|
142 (26.6)
|
N
|
N
|
Gao L
|
24(30)
|
22(27.5)
|
42(52.5)
|
41(51.2)
|
32(40)
|
35(43.8)
|
21(26.3)
|
20(25.0)
|
5(6.3)
|
10(12.5)
|
N
|
N
|
Gao L*
|
107 (37.4)
|
103 (36.0)
|
213 (74.5)
|
210 (73.4)
|
112 (39.2)
|
111 (38.8)
|
111 (38.8)
|
127 (44.4)
|
35 (12.2)
|
40 (14.0)
|
N
|
N
|
Yan YF
|
118(100%)
|
118(100%)
|
59 (50.0)
|
61 (51.7)
|
68 (57.6)
|
66 (55.9)
|
66 (55.9)
|
67 (56.8)
|
50 (42.4)
|
57 (48.3)
|
N
|
N
|
Choo EH
|
117 (44.3)
|
121 (45.8)
|
N
|
N
|
71 (26.9)
|
64 (24.2)
|
N
|
N
|
N
|
N
|
N
|
N
|
Obedinskiy
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
Werner GS
|
85(32.8)
|
40(29.2)
|
210 (81.1)
|
111 (81.0)
|
190 (73.4)
|
92 (67.2)
|
59 (22.8)
|
25 (18.3)
|
145 (56.0)
|
71 (51.8)
|
34 (13.1)
|
10 (7.3)
|
Ladwiniec A
|
62 (21.1)
|
53 (18.0)
|
146 (49.7)
|
135 (45.9)
|
56 (19.1)
|
62 (21.1)
|
151 (51.4)
|
143 (48.6)
|
18 (6.1)
|
24 (8.2)
|
N
|
N
|
Mashayekhi K
|
32 (31.6)
|
31 (29.8)
|
N
|
N
|
23 (22.8)
|
21 (20.2)
|
39 (38.6)
|
38 (36.5)
|
28 (27.7)
|
33 (31.7)
|
12 (11.9)
|
14 (13.5)
|
Study ID
|
CKD
|
Stroke
|
LVEF(%)
|
LAD
|
LCX
|
RCA
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
PCI
|
OMT
|
Choi SY
|
14 (8.9)
|
13 (8.2)
|
21 (13.3)
|
17 (10.8)
|
50±12
|
50±12
|
52 (32.9)
|
49 (31)
|
45 (28.5)
|
43 (27.2)
|
73 (46.2)
|
78 (49.4)
|
Lee SW
|
6(1.5)
|
5 (1.3)
|
29 (7.0)
|
31 (7.9)
|
57.3±9.8
|
57.6±9.1
|
185(44.8)
|
163 (41.7)
|
42(10.2)
|
42 (10.7)
|
186(45.0)
|
186 (47.6)
|
Yang JH
|
46 (8.6)
|
47 (8.8)
|
47 (8.8)
|
45 (8.4)
|
55.6±11.9
|
55.7±11.7
|
152 (28.5)
|
146 (27.4)
|
170 (31.9)
|
174 (32.6)
|
288 (54.0)
|
293 (55.0)
|
Gao L
|
12(15.0)
|
12(15.0)
|
7(8.8)
|
8(10.0)
|
53.70 ± 8.29
|
53.63± 7.21
|
30(36.1)
|
26(31.3)
|
19(22.9)
|
22(26.5 )
|
34(41.0)
|
35 (42.2)
|
Gao L*
|
19 (6.6)
|
21 (7.3)
|
N
|
N
|
54.6±7.8
|
53.6±8.2
|
112 (39.2)
|
103 (36.0)
|
65 (22.7)
|
60 (21.0)
|
132 (46.2)
|
146 (51.0)
|
Yan YF
|
4 (3.4)
|
4 (3.4)
|
7 (5.9)
|
6 (5.1)
|
62.00 (57.00-68.00)
|
61.50 (53.10-67.00)
|
33 (28.0)
|
36 (30.5)
|
31 (26.3)
|
33 (28.0)
|
54 (45.8)
|
49 (41.5)
|
Choo EH
|
26 (9.8)
|
26 (9.8)
|
25 (9.5)
|
24 (9.1)
|
54.8± 11.3
|
53.6 ± 12.3
|
84 (31.8)
|
80 (30.3)
|
66 (25.0)
|
72 (27.3)
|
124 (47.0)
|
118 (44.7)
|
Obedinskiy
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
N
|
Werner GS
|
N
|
N
|
N
|
N
|
54.5±10.8
|
55.7±10.8
|
66 (25.5)
|
38 (27.0)
|
28(10.8)
|
22 (15.6)
|
165 (63.7)
|
81 (57.4)
|
Ladwiniec A
|
59 (20.1)
|
64 (21.8)
|
17 (5.8)
|
20 (6.8)
|
213 (72.5)
|
220 (74.8)
|
N
|
N
|
N
|
N
|
N
|
N
|
Mashayekhi K
|
N
|
N
|
5 (5.0)
|
9 (8.6)
|
N
|
N
|
23 (22.8)
|
17 (16.3)
|
20 (19.8)
|
16 (15.4)
|
58 (57.4)
|
71 (68.3)
|
Data are expressed as mean ± SD or absolute numbers (%).
MI: myocardial infarction; PCI:percutaneous coronary intervention; OMT: optimal medical therapy; CABG: coronary artery bypass grafting; CKD: chronic kidney disease; LVEF: left ventricular ejection fraction; LAD: left anterior descending artery; LCX: left circumflex artery; RCA: right coronary artery; N: null.
CKD in Ladwiniec A et al study is CKD≥3 stage, LVEF defined in Ladwiniec A et al study is ≥50%
|
Quality assessment of 4 RCTs were low-to-moderate quality, while, 7 observation studies were moderate quality. Detailed description was showed in supplement 2.
2. Meta-Analysis
2.1 Comparison of PCI vs. OMT in MACE
There were 10 studies reporting the MACE event. Pooled analysis showed that patients with CTO who received PCI had no significant difference in risk of MACE compared with OMT (RR 0.83, 95% CI 0.64-1.06, P=0.14). However, there was a discrepancy in ethnicity. Sub-analysis by ethnicity indicated that PCI is associated with lower incidence of MACE compared with OMT in west European (RR 0.48, 95% CI 0.28-0.84, P=0.01), but not in east Asian (RR 0.94, 95% CI 0.77-1.16, P=0.58). Shown in Fig2.
2.2 Comparison of PCI vs. OMT in cardiac death
There were 9 studies reporting the cardiac death event. Pooled analysis indicated that compared with OMT, PCI had a lower risk of cardiac death in patients with CTO (RR 0.59, 95%CI 0.45-0.77, P<0.001). Result of sub-analysis by ethnicity was in consistence with the main result. However, there was more reduction risk of cardiac death in west European compared with east Asian (east Asian: RR0.67, 95% CI 0.49-0.90, P=0.009; west European: RR0.40, 95% CI 0.22-0.71, P=0.002). Shown in Fig3.
2.3 Comparison of PCI vs. OMT in myocardial infarction
There were 10 studies reporting outcomes of myocardial infarction. When pooled analysis of all studies, we found that there was no significant difference between PCI and OMT in terms of myocardial infarction (RR 0.98, 95%CI 0.78-1.24, P=0.86). Result of sub-analysis according to ethnicity conformed with main results (east Asian: RR 1.13,95% CI 0.84-1.51, P=0.42; west European: RR 0.75,95% CI 0.51-1.12, P=0.16). Shown in Fig4.
2.4 Comparison of PCI vs. OMT in repeated revascularization
Outcomes of revascularization were reported in 10 studies. Pooled analysis illustrated that there was no significant difference between PCI and OMT regarding repeated revascularization (RR 0.95, 95%CI 0.71-1.25, P=0.71). Sub-analysis showed that whatever west European or east Asian, risk of revascularization is similar between PCI and OMT (east Asian: RR 1.08, 95% CI 0.81-1.45, P=0.59; west European: RR 0.58, 95% CI 0.30-1.13, P=0.11). Shown in Fig5
2.5 Comparison of PCI vs. OMT in stroke
There were 4 out of 11 studies report outcomes of stroke. Pooled analysis demonstrated that there were no significant difference between PCI and OMT in terms of stroke (RR 0.62, 95%CI 0.29- 1.34, P=0.22). Result of sub-analysis by ethnicity is similar to main results (east Asian: RR 0.64, 95% CI 0.26-1.56, P=0.33; west European: RR 0. 55, 95% CI 0.11-2.69, P=0.46). Shown in Fig6.
3. Heterogeneity, sensitivity analysis and Publication bias
There was significant heterogeneity in MACE and repeat revascularization with I2 was 64% and 55% correspondingly. We conducted a sensitivity analysis by removing one study a time to locate the origin of heterogeneity. We found that pooled RR of MACE was dramatically changed after removing Gao (2018) study (RR0.78, 95%CI 0.62-0.98, P=0.04). AS for repeat revascularization, there was no significant change in the overall RR. For significant heterogeneity existed in MACE, we performed a subgroup analysis by ethnicity. And we found that significant heterogeneity still existed in Asian (I2 of MACE and Revascularization was 54% and 55% separately in east Asian), but no significant heterogeneity in west European (I2 of MACE and Revascularization was 34% and 42% separately in west European).
There no publication bias measured by Egger’s value and funnel plots in our analysis. Detailed has been shown in Table2 and Figure7.
Table2. Test of association, Heterogeneity analysis and Publication bias
Outcome
|
N
|
Test of association
|
Heterogeneity analysis
|
Publication bias
|
RR (95% CI)
|
P value
|
Model
|
Q value
|
P value
|
I2(%)
|
Egger’s P
|
MACE
|
10
|
0.83(0.64-1.06)
|
0.14
|
R
|
25.11
|
0.003
|
64%
|
0.952
|
Cardiac Death
|
9
|
0.59 (0.45-0.77)
|
0.0001
|
F
|
8.05
|
0.43
|
1%
|
0.918
|
Myocardial Infarction
|
10
|
0.98 (0.78- 1.24)
|
0.86
|
F
|
14.1
|
0.12
|
36%
|
0.761
|
Repeat Revascularization
|
10
|
0.95 (0.71- 1.25)
|
0.71
|
R
|
19.85
|
0.02
|
55%
|
0.109
|
Stroke
|
4
|
0.62 (0.29- 1.34)
|
0.22
|
F
|
0.74
|
0.86
|
0%
|
0.73
|
R: random effect model; F: fixed effect model
|