Nineteen studies enrolling 12,953 patients (8,649 undergoing TF-TAVR and 4,304 undergoing TA-TAVR) met the inclusion criteria and were included for the final meta-analysis [5, 14–31]. The search and selection process are shown in Fig. 1. The main characteristics of the included studies are shown in Table 1. Of the 19 studies, 1 was RCT, 9 were prospective observational studies and 9 were retrospective observational studies. The Study quality assessment is summarized in Table 2. Publication bias and heterogeneity for each outcome are listed in Table 3.
Table 1. Study characteristics
Lead Author
|
Mohammed Al-Hijji[25]
|
Takahide Arai[26]
|
Edward Koifman[27]
|
Takashi Murashita[20]
|
Martine Gilard[29]
|
Vinod H Thourani[23]
|
Fausto Biancari[21]
|
Publication Year
|
2019
|
2016
|
2016
|
2016
|
2016
|
2016
|
2015
|
Journal
|
Catheter Cardio Inte
|
JACC-Cardiovasc Inte
|
Cardiovasc Revasc Med
|
Ann Thorac Surg
|
JACC
|
Lancet
|
Am J Cardiol
|
Study design
|
retrospective
|
prospective
|
retrospective
|
retrospective
|
prospective
|
prospective
|
prospective
|
Study period
|
2012-2016
|
2011-2014
|
2007-2014
|
2008-2015
|
2010-2012
|
2014.2-2014.9
|
2010-2012
|
Procedure
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
|
Cohort number
|
115
|
115
|
467
|
42
|
516
|
132
|
351
|
216
|
3064
|
735
|
948
|
126
|
199
|
199
|
|
Age, years
|
82.5±7.7
|
82.8±7.8
|
83.8±7.1
|
81.3±7.7
|
83±8
|
84±7
|
79.6±9.7
|
82.0±7.5
|
83.2±7.0
|
81.7±7.5
|
82.1±6.57
|
80.7±6.69
|
81.5±6.2
|
81.2±6.6
|
|
Male sex
|
60(52.2%)
|
63(55.3%)
|
234(50%)
|
30(71%)
|
264(51%)
|
58(44%)
|
211 (60.1%)
|
123 (56.9%)
|
1448(47.3%)
|
428(58.2%)
|
577(60.9%)
|
85(67.5%)
|
111 (55.8%)
|
104 (52.3%)
|
|
STS score,%
|
10.0±5.2
|
10.6±4.7
|
6.2±3.9
|
7.1±4.2
|
8.7±4.5
|
10.4±4.6
|
8.8±6.5
|
9.4±5.4
|
N/A
|
N/A
|
5.3±1.29
|
5.6±1.28
|
14.9±11.8
|
15.0±10.6
|
|
EuroScore, %
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
21.2±14.0
|
23.9±14.8
|
N/A
|
N/A
|
8.1±7.1
|
8.4±7.3
|
|
Diabetes mellitus
|
44(38.3%)
|
46(40.4%)
|
103(22%)
|
11(26%)
|
171(35%)
|
38(30%)
|
141 (40.2%)
|
84 (38.9%)
|
753(24.7%)
|
192(26.6%)
|
N/A
|
N/A
|
52 (26.1%)
|
50 (25.1%)
|
|
Chronic renal failure
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
6 (1.7%)
|
5 (2.3%)
|
79(2.6%)
|
18(2.5%)
|
N/A
|
N/A
|
5 (2.5%)
|
5 (2.5%)
|
|
COPD
|
77(67%)
|
70(61.4%)
|
59(13%)
|
8(19%)
|
164(33%)
|
47(37%)
|
236 (67.2%)
|
126 (58.3%)
|
740(24.3%)
|
158(21.7%)
|
270 (28.5%)
|
51(40.5%)
|
44 (22.1%)
|
50 (25.1%)
|
|
Atrial fibrillation
|
52(45.2%)
|
48(42.1%)
|
132(28%)
|
18(43%)
|
212(43%)
|
52(41%)
|
N/A
|
N/A
|
823(27.6%)
|
160(22.1%)
|
342(36.1%)
|
43/(34.1%)
|
N/A
|
N/A
|
|
Previous stroke
|
10(8.7%)
|
14(12.3%)
|
13(3%)
|
2(5%)
|
60(13%)
|
29(24%)
|
83 (23.6%)
|
71 (32.9%)
|
286(9.4%)
|
84(11.7%)
|
81(8.5%)
|
16(12.7%)
|
4 (2.0%)
|
7 (3.5%)
|
|
Previous infectious endocarditis
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
|
Previous valve surgery
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
90 (25.6%)
|
44 (20.4%)
|
50(1.6%)
|
12(1.7%)
|
51(5.4%)
|
4(3.2%)
|
N/A
|
N/A
|
|
Previous myocardial infarction
|
N/A
|
N/A
|
15(3%)
|
2(5%)
|
N/A
|
N/A
|
100 (28.5%)
|
80 (37.0%)
|
439(14.4%)
|
169(23.4%)
|
133 (14.0%)
|
39 (31.0%)
|
54 (27.1%)
|
53 (26.6%)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Data are n (%), or mean±SD; TF-TAVR= transfemoral transcatheter aortic valve replacement; TA-TAVR= transapical transcatheter aortic valve replacement; COPD =chronic obstructive pulmonary disease; STS=Society of Thoracic Surgeons
Table 1. Study characteristics (Continued)
Lead Author
|
Eugene H. Blackstone[24]
|
Gerhard Schymik[28]
|
Martyn Thomas[5]
|
Craig R. Smith[22]
|
Johan M. Bosmans[15]
|
See Hooi Ewe[16]
|
Publication Year
|
2015
|
2015
|
2011
|
2011
|
2011
|
2011
|
Journal
|
Circulation
|
Circ-Cardiovasc Int
|
Circulation
|
NEJM
|
Inter Cardiov Th
|
Ann Thorac Surg
|
Study design
|
prospective
|
prospective
|
retrospective
|
RCT
|
retrospective
|
retrospective
|
Study period
|
2017-2012
|
2008-2012
|
2007-2009
|
2007-2009
|
-2010
|
N/A
|
Procedure
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
Cohort number
|
501
|
501
|
354
|
354
|
463
|
575
|
492
|
207
|
99
|
88
|
45
|
59
|
Age, years
|
218 (44%)
|
85±6.3
|
81.7±5.0
|
81.8±5.9
|
81.7±6.7
|
80.7±7.0
|
84.4±6.7
|
83.2±6.5
|
84±5
|
82±6
|
82.2±7.1
|
79.4±8.3
|
Male sex
|
283(56%)
|
272(54%)
|
164(46.3%)
|
161(45.4%)
|
208(44.9%)
|
254(45.2%)
|
284 (57.8%)
|
115 (55.8%)
|
N/A
|
N/A
|
21(46.7%)
|
31(52.5%)
|
STS score,%
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
11.7±3.3
|
11.8±3.5
|
N/A
|
N/A
|
8.5±3.8
|
8.9±3.5
|
EuroScore,%
|
N/A
|
N/A
|
23.5±16.3
|
23.0±15.6
|
25.8±14.4
|
29.1±16.2
|
29.1±16.1
|
29.8±15.9
|
29±15
|
33±17
|
20.1±11.7
|
22.6±11.9
|
Diabetes mellitus
|
184(37%)
|
185(375)
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
10(10%)
|
19(18%)
|
13(28.9%)
|
16(27.1%)
|
Chronic renal failure
|
96(195)
|
92(18)
|
29(8.2%)
|
26(7.3%)
|
118(25.5%)
|
187(32.5%)
|
46(9.5%)
|
16(7.9)
|
N/A
|
N/A
|
10(22.2%)
|
13(22%)
|
COPD
|
221(44%)
|
214(43%)
|
46(13%)
|
47(13.3%)
|
114(24.6%)
|
172(29.9%)
|
211(42.9%)
|
91(44.0%)
|
N/A
|
N/A
|
11(24.4%)
|
17(28.8%)
|
Atrial fibrillation
|
109(22%)
|
100(20%)
|
N/A
|
N/A
|
N/A
|
N/A
|
106(38.7%)
|
47(50.5)
|
N/A
|
N/A
|
8(17.8%)
|
14(23.7%)
|
Previous stroke
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
116(25.4%)
|
66(35.7%)
|
N/A
|
N/A
|
2(4.4%)
|
10(17%)
|
Previous infectious endocarditis
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
Previous valve surgery
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
Previous myocardial infarction
|
136(27%)
|
140(28%)
|
46(13%)
|
47(13.3%)
|
N/A
|
N/A
|
128(26.4%)
|
67(33.2%)
|
N/A
|
N/A
|
10(22.2%)
|
14(23.7%)
|
Data are n (%), or mean±SD; TF-TAVR= transfemoral transcatheter aortic valve replacement; TA-TAVR= transapical transcatheter aortic valve replacement; COPD =chronic obstructive pulmonary disease; STS=Society of Thoracic Surgeons
Table 1. Study characteristics (Continued)
Lead Author
|
Peter Wenaweser[18]
|
Rafal Dworakowski[19]
|
Josep Rodés-Cabau[30]
|
Martyn Thomas[14]
|
Helene Eltchaninoff[17]
|
Nawwar Al-Attar[31]
|
Publication Year
|
2011
|
2011
|
2010
|
2010
|
2010
|
2009
|
Journal
|
Am Heart J
|
Am Heart J
|
JACC
|
Circulation
|
European Heart Journal
|
Ann Thorac Surg
|
Study design
|
prospective
|
prospective
|
retrospective
|
retrospective
|
retrospective
|
prospective
|
Study period
|
N/A
|
2007-2009
|
2005-2009
|
2007-2009
|
2009.2-2009.6
|
2006-2008
|
Procedure
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
TP-TAVR
|
TA-TAVR
|
Cohort number
|
130
|
27
|
67
|
84
|
162
|
177
|
463
|
575
|
161
|
71
|
35
|
15
|
Age, years
|
82.9±5.0
|
83.9±4.0
|
83±0.8
|
82.2±0.8
|
83±8
|
80±8
|
81.7±6.7
|
80.7±7.0
|
82.3±7.3
|
82.1±7.3
|
83±6
|
83±10
|
Male sex
|
50(23%)
|
9(33%)
|
43(51%)
|
39(58%)
|
91(56.1%)
|
61(34.5%)
|
208(44.9%)
|
254(45.2%)
|
86(53%)
|
46(64.7%)
|
18(51.4%)
|
9(60%)
|
STS score,%
|
N/A
|
N/A
|
N/A
|
N/A
|
9.0±5.8
|
10.5±6.9
|
N/A
|
N/A
|
18.9±12.8
|
18.4±12.1
|
15±6
|
19±9
|
EuroScore,%
|
N/A
|
N/A
|
19.4±1.1
|
23.4±1.5
|
N/A
|
N/A
|
25.8±14.4
|
29.1±16.2
|
25.6±11.4
|
26.8±11.6
|
26±14
|
30±12
|
Diabetes mellitus
|
27(20.8%)
|
8(29.6%)
|
18(26.9%)
|
17(20.2%)
|
37(22.8%)
|
42(23.7%)
|
N/A
|
N/A
|
46(28.5%)
|
18(25.3%)
|
6(17%)
|
4(27%)
|
Chronic renal failure
|
N/A
|
N/A
|
28(41.8%)
|
52(61.9%)
|
7(4.3%)
|
3(1.7%)
|
118(25.5%)
|
187(32.5%)
|
N/A
|
N/A
|
9(26%)
|
8(53%)
|
COPD
|
N/A
|
N/A
|
15(22.4%)
|
26(31%)
|
45(27.8%)
|
55(31.1%)
|
114(24.6%)
|
172(29.9%)
|
N/A
|
N/A
|
10(29%)
|
4(27%)
|
Atrial fibrillation
|
37(28.5%)
|
6(22.2%)
|
N/A
|
N/A
|
66(40.7%)
|
49(27.7%)
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
Previous stroke
|
N/A
|
N/A
|
N/A
|
N/A
|
27(16.7%)
|
50(28.2%)
|
N/A
|
N/A
|
16(9.9%)
|
6(8.4%)
|
4(11%)
|
3(20%)
|
Previous infectious endocarditis
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
Previous valve surgery
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
N/A
|
Previous myocardial infarction
|
24(18.5%)
|
4(14.8%)
|
N/A
|
N/A
|
82(50.6%)
|
91(51.4%)
|
N/A
|
N/A
|
42(26%)
|
10(14%)
|
4(11%)
|
7(47%)
|
Data are n (%), or mean±SD; TF-TAVR= transfemoral transcatheter aortic valve replacement; TA-TAVR= transapical transcatheter aortic valve replacement; COPD =chronic obstructive pulmonary disease; STS=Society of Thoracic Surgeons
Table 2. Publication bias analysis
Study (RCT)
|
Random sequence generation
|
Allocation concealment
|
Blinding of participants and personnel
|
Blinding of outcome assessment
|
Incomplete outcome data
|
Selective reporting
|
Other bias
|
|
Craig R. Smith, 2011
|
Low
|
Unclear
|
Low
|
Low
|
Low
|
Low
|
Low
|
|
Study (Observational)
|
Bias due to confounding
|
Bias in selection of participants into the study
|
Bias in measurement of interventions
|
Bias due to departures from intended interventions
|
Bias due to missing data
|
Bias in measurement of outcomes
|
Bias in selection of reported result
|
Overall bias
|
Mohammed Al-Hijji, 2019
|
Serious
|
Serious
|
Low
|
Low
|
Moderate
|
Serious
|
Low
|
Serious
|
Takahide Arai, 2016
|
Serious
|
Low
|
Low
|
Low
|
Moderate
|
Moderate
|
Low
|
Moderate
|
Edward Koifman, 2016
|
Serious
|
Moderate
|
Low
|
Low
|
Moderate
|
Moderate
|
Low
|
Moderate
|
Takashi Murashita, 2016
|
Serious
|
Low
|
Low
|
Low
|
Low
|
Moderate
|
Low
|
Moderate
|
Martine Gilard, 2016
|
Serious
|
Low
|
Low
|
Low
|
Low
|
Serious
|
Low
|
Moderate
|
Vinod H Thourani, 2016
|
Serious
|
Low
|
Low
|
Low
|
Moderate
|
Serious
|
Low
|
Moderate
|
Fausto Biancari, 2015
|
Serious
|
Moderate
|
Moderate
|
Low
|
Moderate
|
Serious
|
Low
|
Serious
|
Eugene H. Blackstone, 2015
|
Serious
|
Low
|
Low
|
Low
|
Moderate
|
Moderate
|
Low
|
Moderate
|
Gerhard Schymik, 2015
|
Serious
|
Low
|
Low
|
Low
|
Serious
|
Moderate
|
Low
|
Moderate
|
Martyn Thomas, 2011
|
Serious
|
Low
|
Moderate
|
Moderate
|
Serious
|
Moderate
|
Low
|
Serious
|
Johan M. Bosmansa, 2011
|
Serious
|
Low
|
Moderate
|
Low
|
Serious
|
Serious
|
Low
|
Serious
|
See Hooi Ewe, 2011
|
Serious
|
Moderate
|
Low
|
Low
|
Low
|
Serious
|
Low
|
Moderate
|
Peter Wenaweser, 2011
|
Serious
|
Low
|
Moderate
|
Serious
|
Moderate
|
Serious
|
Low
|
Serious
|
Rafal Dworakowski, 2011
|
Serious
|
Moderate
|
Moderate
|
Low
|
Serious
|
Moderate
|
Low
|
Moderate
|
Josep Rodés-Cabau, 2010
|
Serious
|
Low
|
Low
|
Low
|
Low
|
Moderate
|
Low
|
Moderate
|
Martyn Thomas, 2010
|
Serious
|
Low
|
Low
|
Low
|
Serious
|
Moderate
|
Low
|
Moderate
|
Helene Eltchaninoff, 2010
|
Serious
|
Low
|
Low
|
Low
|
Serious
|
Moderate
|
Low
|
Moderate
|
Nawwar Al-Attar, 2009
|
Serious
|
Serious
|
Moderate
|
Low
|
Moderate
|
Moderate
|
Low
|
Serious
|
Table 3. Test of heterogeneity and publication bias for each outcome
Outcomes
|
Chi-Square
|
df
|
P value
|
I square
|
Heterogeneity
|
Publication bias
|
In-hospital mortality
|
32.05
|
17
|
0.01
|
47%
|
low
|
none
|
1-year mortality
|
1.84
|
4
|
0.77
|
0%
|
insignificant
|
none
|
Mid-term mortality
|
17.08
|
3
|
0.0007
|
82%
|
high
|
none
|
Major vascular complication
|
36.06
|
10
|
<0.00001
|
72%
|
moderate
|
none
|
Pacemaker implantation
|
22.66
|
15
|
0.09
|
34%
|
low
|
none
|
Major bleeding
|
31.07
|
8
|
0.0001
|
74%
|
moderate
|
none
|
Acute kidney injury
|
69.19
|
11
|
<0.00001
|
84%
|
high
|
none
|
Length of hospital stay
|
10.86
|
5
|
0.05
|
54%
|
moderate
|
none
|
Stroke
|
13.36
|
9
|
0.15
|
33%
|
low
|
none
|
Mortality
Postoperative in-hospital mortality was reported in 18 studies. One RCT and 17 observational studies with 11,915 patients were included. In the pooled analysis, in-hospital mortality was significantly lower with TF-TAVR compared with TA-TAVR (OR = 0.63, 95% CI 0.54–0.74, P < 0.001, Fig. 2A).Postoperative 1-year mortality was reported in 5 studies. One RCT and 4 observational studies with 2,313 patients were included. In the pooled analysis, 1-year mortality was still significantly lower with TF-TAVR compared with TA-TAVR (OR = 0.53, 95% CI 0.41–0.69, P < 0.001, Fig. 2B). Postoperative mid-term mortality was reported in 4 observational studies with 5,907 patients. The pooled analysis did not demonstrate a statistically significant difference in the risk of mid-term mortality when comparing TF-TAVR versus TA-TAVR (OR = 0.68, 95% CI 0.46–1.01, P = 0.06, Fig. 2C).
Morbidity and Other Complications
Results for the other outcomes are summarized in Fig. 3. The pooled analysis of 11 studies (5,456 patients) demonstrated a higher risk of major vascular complication with TF-TAVR compared with TA-TAVR (OR = 3.21, 95% CI 1.90–5.42, P < 0.001, Fig. 3A). Meanwhile, in the pooled analysis of 16 studies (n = 7,837), there was a significantly higher incidence of pacemaker implantation in TF-TAVR when compared with TA-TAVR (OR = 1.47, 95% CI 1.22–1.76, P < 0.001, Fig. 3B).
On the other hand, pooled analyses revealed that TF-TAVR was associated with lower risk for major bleeding (9 studies, 5,174 patients, OR = 0.52, 95% CI 0.32–0.85, P = 0.009, Fig. 3C) and acute kidney injury (12 studies, 5,622 patients, OR = 0.40, 95% CI 0.21–0.76, P = 0.005, Fig. 3D), while TF-TAVR has a shorter length of hospital stay (6 studies, 6,890 patients, mean difference=-2.92 days, 95% CI -3.71 to -2.14, P < 0.001, Fig. 3E). Pooled analysis of 10 studies (5,716 patients) demonstrated no statistically significant difference in the risk of stroke among patients assigned to TF-TAVR versus TA-TAVR (OR = 0.80 95% CI 0.59–1.09, P = 0.160, Fig. 3F).Funnel plots for each outcome are shown in Supplementary Figures. No significant publication biases were detected.