In recent years, transcatheter aortic valve implantation (TAVI) has been widely recognized as a safe and effective alternative to surgical treatment for patients with severe inoperable aortic stenosis (AS) or at high surgical risk.1,2 At present, TAVI is indicated only for patients with severe symptomatic AS who are assessed by a multidisciplinary team and considered inoperable (indication I, level of evidence B) or who are operable but at high surgical risk (indication IIa, level of evidence B).3

Various studies show that the TAVI procedure yields good short- and mid-term results4–7; however, there is a paucity of long-term data on the durability and clinical outcomes of percutaneous valves.8 Recently, the first series was published with long-term results on the use of the Edwards-SAPIEN valve.9–12 In the case of the self-expanding CoreValve prosthesis, very little evidence is available from long-term follow-up because only a few studies have been published, with follow-up periods of 3 to 5 years.13–16

The main aim of this study was to evaluate the long-term mortality, clinical outcomes, and durability of the self-expanding CoreValve prosthesis at 3 Spanish hospitals with initial results that represent early experience with this treatment in Spain.17

A total of 108 patients were included. The baseline characteristics for the population have been previously published.17 The mean age at the time of implantation was 78.6 ± 6.7 (range, 50-92) years; 49 (45.4%) patients were men, and the mean logistic EuroSCORE was 16% ± 13.9% (2.3%-86.9%). The success rate for the procedure was 98.1%. The median follow-up for all patients was 6.1 years (2232 days). For living patients at the end of follow-up, the median follow-up was 7.48 years (2731 days) with a maximum and minimum follow-up of 8.5 and 7.02 years (3086 and 2563 days).

Primary Study Endpoint: Long-term Survival

Survival at the end of years 1 to 6 was 84.3% (92.6% after the hospitalization period), 77.8%, 72.2%, 66.7%, 58.3%, and 52.8%. The survival estimated by the Kaplan-Meier method for all-cause and cardiac mortality was 0.32 and 0.79, respectively (Figure 1 and Figure 2). The median survival time estimated by the same method was 2207 days (95% confidence interval, 1856-2557).

Figure 1.

Survival estimated by the Kaplan-Meier method for all-cause mortality.

(0.1MB).

Figure 2.

Survival estimated by the Kaplan-Meier method for cardiac mortality.

(0.08MB).

Secondary EndpointsTime Trend in Mortality and its Causes

The time trend in mortality during follow-up and its causes are listed in Table 1. A total of 71 (65.7%) patients died during the follow-up period: 18 (25.3%) due to cardiac cause, 5 (7%) during hospitalization, and 13 (18.3%) during subsequent follow-up. Among the 3 patients who died from cardiac arrest, 1 had a pacemaker (the other 2 had no known conduction abnormalities), and 2 had ventricular dysfunction. Among the 9 deaths due to heart failure after hospital discharge, 6 patients had ventricular dysfunction and 4 had grade III aortic regurgitation.

Table 1.

Evolution and Causes of Mortality During Follow-up

	Hospitalization	First year (except hospitalization)	Second year	Third year	Fourth year	Fifth year	Sixth year	Seventh year	Eighth year
Total, n, yearly %	8 (7.4)	9 (8.3)	7 (7.7)	6 (7.1)	6 (7.7)	9 (12.5)	6 (9.5)	15 (26.3)	5 (11.9)
Noncardiac, n	3	7	4	6	6	6	3	13	5
Causes	Respiratory failure (2) Vascular sepsis	Respiratory failure (3) SBP Pancreatitis Cancer Stroke	Sepsis (osteomyelitis) MDS Multiorgan failure Renal failure	Pulmonary sepsis Sepsis (pyelonephritis) Sepsis after hip surgery Multiorgan failure Renal failure (2)	Breast cancer (2) Stomach cancer Multiorgan failure Respiratory failure Pulmonary sepsis	Pulmonary sepsis (2) Sepsis after hip surgery Multiorgan failure Respiratory failure SAH	Multiorgan failure (2) Stomach cancer	Multiorgan failure (6) Abdominal sepsis (2) Pulmonary sepsis (2) Renal failure Stomach cancer Lung cancer	Multiorgan failure (4) Pulmonary sepsis
Cardiac, n	5	2	3	0	0	3	3	2	0
Causes	Heart failure (3) Ventricular perforation (2)	Heart failure Sudden cardiac death	Heart failure (2) Sudden cardiac death			Heart failure (2) AMI	Heart failure (2) Sudden cardiac death	Heart failure (2)

AMI, acute myocardial infarction; MDS, myelodysplastic syndrome; SAH, subarachnoid hemorrhage; SBP, spontaneous bacterial peritonitis.

Clinical Status

At the end of follow-up, 31.6% of patients were in NYHA I and 50.9% in NYHA II. Data on functional class in the first 6 years of follow-up are shown in Figure 3.

Figure 3.

Time trend in New York Heart Association functional class during follow-up. NYHA, New York Heart Association functional class.

(0.23MB).

Valve Function

The echocardiographic time trends during follow-up (maximum and mean gradients, effective valve area, and left ventricular function) are shown in Table 2 and are compared according to follow-up year in Figure 4. The rates of aortic regurgitation > grade II observed by echocardiography in years 1 to 6 of follow-up were 2.2%, 2.5%, 2.6%, 2.7%, 3.2%, and 7.0% (Figure 5). Upon completion of follow-up and according to the VARC-2 criteria, 6 (5.5% of total) patients had valve dysfunction. There were no cases of severe prosthetic dysfunction requiring reoperation. In the first year of follow-up, 3 patients had moderate valve stenosis with mean gradients of 22, 22, and 27mmHg, respectively. Of those 3 patients, 1 died in the second year due to septic cause and the other 2 were alive at the end of follow-up, with no increase in the gradients. In the second, third, and fourth years, another 3 patients had mean gradients of 24, 22, and 23mmHg, with oscillations < 2mmHg up to the end of follow-up.

Table 2.

Time Trend in Echocardiographic Parameters During Follow-up

Follow-up, y	1	2	3	4	5	6
Maximum gradient, mmHg	18.6 ± 8.4	19.9 ± 8.7	20.6 ± 8.7	21.1 ± 10.3	24.3 ± 17	21.1 ± 8.4
Mean gradient, mmHg	9.7 ± 4.7	10.5 ± 5.4	11.1 ± 5.6	11.2 ± 5.7	13.9 ± 13	11.9 ± 5.5
Aortic valve area, cm2	1.6 ± 0.4	1.5 ± 0.3	1.43 ± 0.3	1.4 ± 0.32	1.4 ± 0.4	1.33 ± 0.24
LVEF, %	63 ± 9.6	61.9 ± 8.8	60.6 ± 9.7	61.4 ± 9.1	61.5 ± 8.9	58.8 ± 10.5
Patients, n, %*	80 (88)	73 (87)	62 (79)	58 (81)	49 (78)	43 (77)

LVEF, left ventricular ejection fraction.

*

Number of patients who underwent echocardiography each year and percentage of total living patients in each period.

Figure 4.

Time trend in echocardiographic data during follow-up. A: Time trend in maximum gradient. B: Time trend in mean gradient. C: Time trend in effective valve area. D: Time trend in left ventricular function.

(0.25MB).

Figure 5.

Time trend in the grade of aortic regurgitation during follow-up.

(0.22MB).

The results of our series of patients who received a self-expanding transcatheter aortic valve, with a very long-term follow-up (median of 2232 days/6.1 years) show that mortality at the end of the sixth year of follow-up was high (47.2%), particularly due to noncardiac causes, and that valve durability was good (5.5% prosthetic dysfunction with no need for reoperation). Our data are taken from the multicenter cohort with the longest, protocol-based follow-up studied to date in Spain, as it includes the first patients treated with the self-expanding valve by this procedure.17 Complete clinical and echocardiographic parameters and the causes of death each year are listed.

At present, there is little information on long-term follow-up of TAVI. The currently published series can be analyzed according to the type of valve used, Edwards-SAPIEN and CoreValve, as these valves were used initially and, therefore, have longer follow-up periods. Most notably, the Edwards-SAPIEN valve was studied in a series by Toggweiler et al.9 with 88 patients, which reported a 5-year survival of 35% with yearly mortality rates of 17%, 26%, 47%, 58%, and 65%. Unbenhaun et al.11 reported a total mortality of 61.4% at 5 years, with yearly rates of 18.4%, 33.1%, 42.8%, 51.8%, and 61.4% and a final survival of 39.6%. Mortality data for arm B was also recently published by the PARTNER study,10 which observed yearly mortality rates of 30.7%, 43%, 53.9%, 64.1%, and 71.8%. A study by Escárcega et al.19 found that the survival of the transfemoral group was < 50% at 5 years. In our setting, experience has recently been published on follow-up at a single hospital in Spain12 with yearly mortality rates of 20.3%, 29.1%, 41.1%, 50.5%, and 68% and a final survival of 32%.

Regarding experience with TAVI using the CoreValve self-expanding valve, Barbanti et al.16 recently published their experience in a 5-year follow-up study of a cohort of 353 patients, reporting yearly mortality rates of 21%, 29%, 38%, 48%, and 55% and cardiovascular mortality rates of 10%, 14%, 19%, 23%, and 28%. In a study by Gulino et al.,15 the yearly survival of a cohort of 125 patients was 83.2%, 76.8%, 73.6%, and 66.3%. Codner et al.20 published data on a mixed series with 360 patients, 71% of whom were treated with CoreValve and 26% with SAPIEN, with 3- and 5-year mortality rates of 71.6% and 56.4%, respectively.

Consequently, it can be assumed that total 5-year mortality is between the rate of 71.8% reported by PARTNER B10 and the rate of 55% for the series studied by Barbanti et al.16 Our data are consistent with those studies, particularly with the series by Barbanti et al.,16 and underscore the fact that the highest mortality rate is seen in the first year of follow-up, a fact essentially explained by higher mortality rates in the first month due to procedure-related complications. Following hospital discharge, the main cause of mortality is noncardiac, due to the high rate of comorbidities in these patients. Infectious diseases, mainly as a complication of previous conditions, and respiratory failure are very common causes of mortality, followed by the development of other diseases clearly related to aging, such as heart failure, multiorgan failure, renal impairment, and cancer.

A very important data point of our sample is the proven durability of the valve, which is seen in the rate of valve dysfunction found in our study (5.5%). A comparison of this rate with data reported by other series is difficult: first, because there is little information on follow-up periods of similar length with the same valve, and second, due to the heterogeneity of the definition used to diagnose valve dysfunction. For instance, various series published with the Edwards-SAPIEN valve have reported that moderate-to-severe aortic regurgitation appeared in 3.8% of cases at 5 years.20 Barbanet et al.16 considered the sum of severe degeneration plus increased mean gradient above 20mmHg at 5 years and found a rate of prosthetic dysfunction of 4.2%, a figure comparable to ours, as well as the rate of 4.5% recently reported by Del Trigo et al.21 A series by Salinas et al.12 reported even higher (15.3%) valve dysfunction figures. In our series, there were no cases of prosthetic valve dysfunction requiring valve replacement, similar to the studies conducted by D’Onofrio et al.22 (at 3 years) and Salinas et al.12 and below the levels reported by Unbehaun et al.11 (3%), Toggweiler et al.9 (1.1%), Kapadia et al.10 (1.1%), Rodés-Cabau et al.8 at 4 years (0.6%), and Barbanti et al.16 (0.6%). In this regard, although additional studies are needed to understand this aspect more clearly, the valve durability observed in our series is consistent with the largest series published to date with the same valve.

In view of the results and limitations of our series, it can be concluded that long-term survival with TAVI is acceptable, as shown by the fact that more than half the patients survive beyond the sixth year. Although the main cause of mortality is cardiovascular in the first year, other causes more commonly seen with aging, such as infections and cancer, are more prevalent with longer follow-up. Patients also show an acceptable long-term functional status. In our series, although there was a slight, progressive increase in valve gradients, valve functionality remained steady over time. Further studies with even longer follow-up are needed to confirm the functional durability of TAVI and to investigate potential factors related to the procedure.

J.H. Alonso-Briales, J.M. Hernández-García, and C. Morís are CoreValve proctors.