Baseline Clinical and Echocardiographic Characteristics
Baseline characteristics were significantly different across the 4 age groups in several aspects (Table 1). Patients > 80 years of age more often had coronary artery disease and anemia, and were more often symptomatic at baseline and had higher STS score. Patients < 65 years of age more often had chronic kidney disease including dialysis. Regarding the etiologies of AS, bicuspid AS accounted for 35% of patients undergoing surgical AVR in patients < 65 years of age, while degenerative AS was most frequently diagnosed in patients > 80 years of age. With increasing age, there was an incremental decrease in left ventricular end diastole/systole diameter and AVA, while there was no significant difference in the AVA index. Acute HF hospitalization at the index echocardiography was more often seen in patients > 80 years of age (Table 1).
AVR procedural characteristics
Of 1163 patients undergoing surgical AVR, bioprosthetic valves were more frequently used than mechanical valves: 899 (77%) patients (< 65 years of age: 59 [32%] patients; 65–74 years of age: 310 [79%] patients; 75–80 years of age: 313 [91%] patients; >80 years of age: 217 [91%] patients). Of 239 patients > 80 years of age, relatively small mechanical valves were needed in 20 patients (8%) due to small aortic annulus despite predominant use of bioprosthetic valves (Table 2). Concomitant procedures were performed in 545 (47%) patients. Concomitant mitral valve surgery and ascending aortic replacement were more frequently performed in patients < 65 years of age, while concomitant coronary artery bypass grafting was less frequently performed in these patients (Table 2).
Table 2
Procedural Characteristics of Surgical AVR
|
< 65 years of age
|
65–74 years of age
|
75–80 years of age
|
> 80 years of age
|
P value
|
N = 185
|
N = 394
|
N = 345
|
N = 239
|
|
Types of prosthetic valve
|
|
|
|
|
|
Bioprosthetic valve
|
59 (32%)
|
310 (79%)
|
313 (91%)
|
217 (91%)
|
< 0.001
|
Mechanical valve
|
124 (67%)
|
81 (21%)
|
31 (9%)
|
20 (8%)
|
Unknown
|
2 (1%)
|
3 (1%)
|
1 (0.3%)
|
2 (1%)
|
Concomitant procedures
|
92 (50%)
|
195 (49%)
|
166 (48%)
|
92 (38%)
|
0.01
|
AVR with CABG
|
29 (16%)
|
97 (25%)
|
100 (29%)
|
62 (26%)
|
0.007
|
AVR with mitral valve surgery
|
30 (16%)
|
54 (14%)
|
40 (12%)
|
19 (8%)
|
0.03
|
with mitral valve plasty/annuloplasty
|
13
|
20
|
22
|
14
|
|
with mitral valve replacement
|
17
|
34
|
18
|
5
|
|
AVR with tricuspid valve surgery
|
17 (9%)
|
37 (9%)
|
23 (7%)
|
17 (7%)
|
0.74
|
with tricuspid valvuloplasty/ annuloplasty
|
17
|
36
|
22
|
16
|
|
with tricuspid valve replacement
|
0
|
1
|
1
|
1
|
|
AVR with ascending aortic replacement
|
32 (17%)
|
42 (11%)
|
26 (8%)
|
5 (2%)
|
< 0.001
|
AVR with annular dilatation
|
2 (1%)
|
3 (1%)
|
1 (0.3%)
|
2 (1%)
|
0.69
|
AVR with Maze
|
16 (9%)
|
24 (6%)
|
16 (5%)
|
12 (5%)
|
0.20
|
Valve size
|
|
|
|
|
|
Bioprosthetic valve
|
|
|
|
|
-
|
19 mm
|
13/59 (22%)
|
81/310 (26%)
|
146/313 (47%)
|
108/217 (50%)
|
|
21 mm
|
25/59 (42%)
|
129/310 (42%)
|
110/313 (35%)
|
76/217 (35%)
|
|
23 mm
|
13/59 (22%)
|
66/310 (21%)
|
37/313 (12%)
|
24/217 (11%)
|
|
25 mm
|
6/59 (10%)
|
25/310 (8%)
|
17/313 (5%)
|
3/217 (1%)
|
|
27 mm
|
0
|
4/310 (1%)
|
1/313 (0.3%)
|
0
|
|
29 mm
|
1/59 (2%)
|
1/310 (0.3%)
|
0
|
0
|
|
Unknown
|
1/59 (2%)
|
4/310 (1%)
|
2/313 (1%)
|
6/217 (3%)
|
|
Mechanical valve
|
|
|
|
|
-
|
16, 17, 18mm
|
14/124 (11%)
|
20/81 (25%)
|
11/31 (35%)
|
19/20 (95%)
|
|
19 mm
|
36/124 (29%)
|
25/81 (31%)
|
13/31 (42%)
|
1/20 (5%)
|
|
20 mm
|
3/124 (2%)
|
2/81 (2%)
|
1/31 (3%)
|
0
|
|
21 mm
|
34/124 (27%)
|
19/81 (23%)
|
3/31 (10%)
|
0
|
|
22 mm
|
8/124 (6%)
|
0
|
1/31 (3%)
|
0
|
|
23 mm
|
23/124 (19%)
|
11/81 (14%)
|
1/31 (3%)
|
0
|
|
25 mm
|
2/124 (2%)
|
1/81 (1%)
|
0
|
0
|
|
27 mm
|
0
|
2/81 (2%)
|
1/31 (3%)
|
0
|
|
29 mm
|
2/124 (2%)
|
0
|
0
|
0
|
|
Unknown
|
2/124 (2%)
|
1/81 (1%)
|
0
|
0
|
|
Name of the prosthetic valves
|
|
|
|
|
|
Bioprosthetic valve
|
|
|
|
|
-
|
Carpentier Edwards Perimaunt/Magna/Magna Ease
|
41/59 (69%)
|
225/310 (73%)
|
230/313 (73%)
|
167/217 (77%)
|
|
Mosaic
|
9/59 (15%)
|
56/310 (18%)
|
55/313 (18%)
|
34/217 (16%)
|
|
Freestyle/Prima plus
|
7/59 (12%)
|
16/310 (5%)
|
19/313 (6%)
|
2/217 (1%)
|
|
Epic
|
2/59 (3%)
|
9/310 (3%)
|
8/313 (3%)
|
11/217 (5%)
|
|
Unknown
|
0
|
4/310 (1%)
|
1/313 (0.3%)
|
3/217 (1%)
|
|
Mechanical valve
|
|
|
|
|
-
|
St. Jude Medical
|
56 (45%)
|
44 (54%)
|
20 (65%)
|
15 (75%)
|
|
ATS
|
32 (26%)
|
17 (21%)
|
6 (19%)
|
5 (25%)
|
|
CarboMedics
|
17 (14%)
|
13 (16%)
|
3 (10%)
|
0
|
|
On-X
|
18 (15%)
|
7 (9%)
|
2 (6%)
|
0
|
|
Sorin Bicarbon LNFJ
|
1 (1%)
|
0
|
0
|
0
|
|
Abbreviations were same as in Table 1. |
Clinical Outcomes
The mean follow-up period in the entire cohort was 3.9 years with 93% follow-up rate at 2-year (maximum 11.5 years). The 30-day mortality after SAVR with or without concomitant procedure in the entire cohort was 1.8%. There was no significant difference in the 30-day mortality rate across the age groups (< 65 years of age: 1.1%, 65 to 74 years of age: 1.0%, 75 to 80 years of age: 2.9%, and > 80 years of age: 2.1%, P = 0.22). SAVR with concomitant procedure was associated with a higher 30-day mortality rate compared with isolated SAVR (2.9% versus 0.8%, P = 0.007). When stratified by dialysis, SAVR with concomitant procedure was significantly associated with higher 30-day mortality compared with isolated SAVR in dialysis patients, but not in non-dialysis patients (30-day mortality: 11.7% versus 1.9%, P = 0.04 in dialysis patients; 1.5% versus 0.7%, P = 0.22 in non-dialysis patients) (Supplementary Table 1).
The crude 5- and 10-year incidences of all-cause death in the entire cohort were 22.0% and 42.4% respectively (Supplementary Fig. 1). The crude cumulative incidence of all-cause death was significantly lower in patients < 65 years of age than in the other age groups (Fig. 2). The lower adjusted mortality risk of patients < 65 years of age relative to those 75–80 years of age (reference) was significant (HR: 0.46; 95% CI: 0.26–0.79; P = 0.005), while the mortality risks of 65–74 years of age and > 80 years of age relative to 75–80 years of age were not significant (HR: 0.86; 95% CI: 0.60–1.25; P = 0.43, and HR: 1.26; 95% CI 0.84–1.89; P = 0.26, respectively) (Table 3).
Table 3
Crude and Adjusted Risks for Clinical Events According to the Age Categories at Baseline.
|
< 65 years of age versus 75–80 years of age
|
65 to 74 years of age versus 75–80 years of age
|
> 80 years of age versus 75–80 years of age
|
Unadjusted HR (95% CI)
|
P value
|
Adjusted
HR (95% CI)
|
P value
|
Unadjusted HR (95% CI)
|
P value
|
Adjusted
HR (95% CI)
|
P value
|
Unadjusted HR (95% CI)
|
P value
|
Adjusted
HR (95% CI)
|
P value
|
All-cause death
|
0.53 (0.32–0.85)
|
0.007
|
0.46 (0.26–0.79)
|
0.005
|
0.96 (0.69–1.34)
|
0.82
|
0.86 (0.60–1.25)
|
0.43
|
1.15 (0.79–1.68)
|
0.46
|
1.26 (0.84–1.89)
|
0.26
|
CI = confidence interval; HR = hazard ratio. |
The observed mortality rates after SAVR in patients < 65, 65–74, and 75–80 years of age were apparently higher than the expected mortality rates in the age- and sex-matched general population, while the mortality rate in patients > 80 years of age was comparable to the expected mortality rate in the age- and sex-matched general population (Fig. 3). A trend toward loss in life expectancy in patients after SAVR was similarly observed both in men and women (Supplementary Fig. 2).
When stratified by the STS score, there were 637 (55%) patients at low risk (STS score < 3%), 415 (36%) patients at intermediate risk (STS score 3–8%), and 111 (10%) patients at high risk (STS score > 8%). The observed mortality rates in patients with low surgical risk (STS score < 3%) were comparable to the expected mortality rates in the age- and sex-matched general population across all the age groups, while patients with high surgical risk (STS score > 8%) had much higher mortality rate compared with the age- and sex-matched general population regardless of the age groups (Fig. 3). Among patients with intermediate surgical risk (STS score 3–8%), the observed mortality rates in patients < 65, and 65–74 years of age were higher than the expected mortality rates in the age- and sex-matched general population, while the observed mortality rates in patients 75–80, and > 80 years of age were comparable to the expected mortality rates in the age- and sex-matched general population (Fig. 4).
There was no significant difference in postoperative mortality between patients with bioprosthetic valve and those with mechanical valve in the entire cohort (Supplementary Fig. 4). When stratified by age, patients with bioprosthetic valves had better survival than those with mechanical valves in patients ≥ 65 years of age, while there was no significant difference in survival between bioprosthetic and mechanical valves in patients < 65 years of age (Supplementary Fig. 3). Among 978 patients ≥ 65 years of age in whom bioprosthetic valves are recommended in the current guidelines, the favorable effect of bioprosthetic valves on survival compared to mechanical valves was observed after adjusting the confounders (adjusted HR: 0.61; 95%CI: 0.41–0.92, P = 0.02). During the follow-up period, 10 patients with bioprosthetic valve and 2 patients with mechanical valve underwent redo-SAVR. The causes of bioprosthetic valve failure were as follows; endocarditis (n = 5), structural valve deterioration (n = 5), while the causes of mechanical valve failure were endocarditis (n = 1) and hemolytic anemia (n = 1). The 30-day mortality rate after reoperation was 8.3% (1/12).