Methods

The study was conducted on data from the cohort of patients prospectively enrolled in the FRANCE2 registry, composed of 33 French centers and the Monaco center, having performed TAVI procedures from January 2010 to December 2011. The method has previously been described. Included patients had symptomatic severe aortic stenosis and were eligible for TAVI. The study included the 2 available valves (Edwards and CoreValve systems) together with pre-, peri- and postprocedural clinical and echocardiographic data (1 month, 6 months, 1 year, and 2 years). Within this population, we studied patients in whom the use of 2 valves was reported during the same procedure (second immediate valve). The initial clinical and echocardiographic data were analyzed as well as the periprocedural data and postprocedural evolution. Echocardiographic data (aortic area, mean aortic valve gradient, left ventricular function, and aortic regurgitation) were assessed using transthoracic echocardiography. For each of these patients, the indication for implantation of the second prosthesis was specified. The study end points were descriptive. The primary end point was to investigate whether the use of a second valve was a predictor of all-cause mortality. The secondary end points were clinical and echocardiographic data, the reason for the use of a second valve, cardiovascular mortality, and morbidity (cardiovascular, neurological, bleeding, and valve related). In addition, these data were compared with those from the remaining FRANCE2 registry population.

Statistical analysis was performed using Stata version 12 (StataCorp LP, College Station, Texas). Tests were 2 sided, with a type I error rate set at α = 0.05. Baseline characteristics are presented as mean ± SD or as median (interquartile range) for each group (deceased and alive, 1 and 2 valves) for continuous data and as numbers of patients and associated percentages for categorical parameters. Comparisons of baseline characteristics between groups were analyzed using chi-square or Fisher’s exact tests for categorical variables and Student’s t tests or Mann-Whitney tests for quantitative variables, with normality verified by the Shapiro-Wilk test and homoscedasticity by the Fisher-Snedecor test. Overall survival was defined as the interval from intervention (implantation) to death, regardless of the cause of death. Overall survival curves and estimates were constructed using the Kaplan-Meier method. The log-rank test was used in a univariate analysis to test the prognostic value of patient characteristics. Cox proportional-hazards regression was used to investigate prognostic factors in multivariate situation by backward and forward stepwise analysis of the factors considered significant in univariate analysis (entered into the model at p ≤0.15) and according to clinically relevant parameters. The proportional-hazards hypotheses were verified using Schoenfeld’s test and plotting residuals. The interactions between possible predictive factors were also tested. Results are expressed as hazard ratios (HRs) and 95% confidence intervals (CIs).

Results

A total of 3,919 consecutive patients treated with TAVI were studied, 91 of whom (2.3%) were reported to have received second valves during the same procedure. Five of these patients were excluded from the analysis because the first device had not reached the aortic valve (desterilization of the system or obstruction in the approach route), while 2 others were excluded because they died before receiving the second valve (because of annular rupture). Of the 86 remaining patients (2.1% of the TAVI population), 9 other patients were excluded because the cause that led to the implantation of a second valve was not specified (implantation failure or incident during installation without further details); these 9 cases included 4 Edwards valves and 5 CoreValve prostheses. Finally, there were 3 solely described because of rare cause for the second valve (2 annular ruptures and 1 leaflet dysfunction; these were observed only with the Edwards valves). The analysis was therefore conducted on the remaining 72 patients (1.8% of the TAVI population) who required the use of a second valve during the TAVI procedure.

The clinical and echocardiographic preprocedural data of the population receiving 2 valves were comparable with those of the remaining FRANCE2 registry population ( Table 1 ). Thus, the 2 groups exhibited comparable clinical and technical situations. There were therefore no predictive clinical or echocardiographic factors indicating the need for a second valve.

Periprocedural data showed that the procedural characteristics (approach routes, anesthesia, performance of transesophageal echocardiography, arterial closure) were similar between the 2 groups ( Table 2 ). In contrast, in the population with 2 valves, the first valve used was an Edwards valve in 44% and a CoreValve prosthesis in 56%, as opposed to 67% and 33%, respectively, of the population with 1 valve (p <0.001). The second valve was always from the same manufacturer as the first. The sizes of the first valve in the group that received 2 valves were 23 mm (14 patients), 26 mm (16 patients), and 29 mm (2 patients) for the Edwards valve and 26 mm (15 patients), 29 mm (20 patients), and 31 mm (5 patients) for the CoreValve. In 64 patients (88.9%), the second valve was of the same size as the first, whereas in the remaining 8 cases (11.1%), the size of the second valve differed. In 2 cases, it was the Edwards 23-mm valve that embolised (1 in the aortic arch and 1 in the abdominal aorta) and was replaced by Edwards 26-mm valves. Finally, the 6 other replacements were due to malpositioning, 3 of which resulted in larger second valves (2 23-mm Edwards valves replaced by 26-mm valves and 1 29-mm CoreValve replaced by a 31-mm valve), while the 3 others (CoreValve devices) were replaced by smaller second valves (1 31-mm replaced with a 29-mm valve and 2 29-mm valves replaced by a 26-mm valve).

Postprocedural data found that from before to after the procedure, a favorable evolution of clinical and echocardiographic data was observed in the 2 groups. Thus, improvements in aortic valve area, aortic valve gradient, and New York Heart Association class was observed. Although their persisting improvement over time was known for the TAVI procedures in general, they were also found in patients receiving 2 valves ( Figure 1 ). The aortic surface increased from 0.68 cm ² before the procedure to 1.80 cm ² afterward and to 1.90 cm ² at 2 years. The mean gradient decreased from 46.9 to 11.8 mm Hg after the procedure and 10 mm Hg at 2 years. With regard to the ejection fraction, an improvement from before to after the procedure was found in the group with 1 valve (53.3 ± 14.2% vs 55.6 ± 12.6%, p <0.001) relative to the group with 2 valves (53.7 ± 13.1% vs 53.5 ± 11.9%, p = 0.53). Comparison of the postprocedural data between the group with 1 valve and the group with 2 valves revealed no differences in aortic valve area (1.81 ± 0.5 vs 1.80 ± 0.58 cm ² , p = 0.82), mean gradient (10.5 ± 5.3 vs 11.8 ± 9.6 mm Hg, p = 0.29), New York Heart Association classes III and IV (322 [11.1%] vs 5 [10.4%], p = 0.88), aortic regurgitation grade >2 (471 [14.4%] vs 12 [21.8%], p = 0.12), and the left ventricular ejection fraction (55.6 ± 12.6% vs 53.5 ± 11.9%, p = 0.23).

Gradient (millimeters of mercury) and valve area (square centimeters) in patients receiving 1 or 2 valves before the procedure and during follow-up. (Top chart) Gradient. (Bottom chart) Area. Blue denotes 1 valve; red denotes 2 valves. M0 = preprocedure; T1 = 1 month; T2 = 6 months; T3 = 1 year; T4 = 2 years.

The 2 main reasons for the implantation of a second valve during the same procedure were device malpositioning in 52 patients (72%) and embolization of the prosthesis in the vascular system in 20 patients (28%). Device malpositioning was responsible for major leaks, hence the need for implantation of the second valve. Of the 52 instances of prosthetic malpositioning, 16 valves (30.8%) were too high, 18 (34.6%) were too low, including 6 that ultimately ended up in the left ventricle, and 18 (34.6%) were incorrectly positioned with significant leakage, with no further details. There were as many instances of malpositioning with the Edwards valve (26 [50%]) as with the CoreValve prosthesis (26 [50%]). Of the 20 prostheses that embolized, 16 (80%) embolized into the ascending aorta, 1 (5%) into the aortic arch, 2 (10%) into the abdominal aorta, and 1 (5%) into the iliac artery. There was a greater number of embolizations with CoreValve prostheses (14 of 20 [70%]) than with Edwards valves (6 of 20 [30%]).

Patients requiring second valves had lower 2-year survival than those with only 1 valve (51.7% vs 62.3%, p <0.001). The 1-month and 1-year survival rates between patients with 2 and 1 valve were, respectively, 79.1% versus 92.8% (p <0.001) and 62.0% versus 78.1% (p <0.001). Thus, at 2 years, 778 patients (20.3%) with 1 valve and 26 patients (36.1%) with 2 valves were deceased ( Figure 2 ). In contrast, there were no differences in terms of mortality as a function of the first valve used in the overall study population. The 2-year survival rate for valve malpositioning and embolization in patients with 2 valves was lower than in patients with 1 valve (p = 0.019 and p = 0.0013, respectively; Figure 3 ). In univariate analysis, the predictive criteria of all-cause mortality after TAVI, other than the use of a second periprocedural valve (HR 2.23, 95% CI 1.47 to 3.38, p <0.001), were numerous and are listed in Table 3 . In multivariate analysis, the fact of having received 2 valves stood out as an independent predictor of all-cause mortality (HR 2.32, 95% CI 1.50 to 3.60, p <0.001). Among other factors, the 2 most significant were previous renal dialysis and New York Heart Association class III or IV. Other factors are listed in Table 4 . Cardiovascular mortality was higher in the group with 2 valves, with 13 deaths among 72 patients compared with 342 deaths among 3,835 patients (8.9%) with single valves (18.1% vs 8.9%, p = 0.008). The 1-month, 1-year, and 2-year survival rates between patients with 2 and 1 valve were, respectively, 85.7% versus 96.3% (p <0.001), 81% versus 91.5% (p <0.001), and 81% versus 87% (p <0.001). The need for a second valve was an independent predictor of cardiovascular mortality (HR 2.64, 95% CI 1.35 to 5.15, p = 0.005). Periprocedural mortality was also higher for patients with 2 valves (8 patients [11.1%] vs 98 patients [2.6%], p <0.001). The rates of vascular, neurologic, and bleeding complications were identical between the 2 groups. Instances of postprocedural third-degree atrioventricular block were more numerous in the group with 2 valves (42.11% vs 14.73%), although the number of new pacemakers did not differ between the 1-valve and 2-valve groups. In the overall study population, the number of postprocedural pacemakers was higher in patients with CoreValve devices (36.8% vs 21.7%, p <0.001).

Survival of patients with 1 or 2 valves. Continuous line denotes 1 valve; dashed line denotes 2 valves.

Survival of patients with malpositioning, embolization, or 1 valve. Continuous line denotes 1 valve; dashed line denotes patients with malpositioning; dotted line denotes patients with embolization.

Table 3

Univariate analysis, predictors of all-cause mortality

	ALIVE	DEAD	HR	p
	(n=3103)	(n=804)
Age (yrs)	82.7 ± 7.3	83.3 ± 6.8	1.01 [1.00 – 1.02]	0.09
Male sex	1524 (49.1%)	454 (56.5%)	1.35 [1.16 – 1.56]	<0.001
NYHA
I – II	810 (26.3%)	146 (18.3%)
III – IV	2274 (73.7%)	652 (81.7%)	1.53 [1.26 – 1.85]	<0.001
Coronary artery disease	1457 (47.4%)	398 (49.7%)	1.10 [0.95 – 1.27]	0.21
Previous CABG	555 (18.0%)	136 (17.0%)	0.86 [0.71 – 1.05]	0.14
Previous valvuloplastie	490 (15.9%)	165 (20.6%	1.27 [1.06 – 1.53]	0.01
Peripheral vascular disease	590 (19.2%)	199 (24.8%)	1.34 [1.13 – 1.59]	<0.001
Aortic abdominal aneurysm	126 (4.1%)	52 (6.5%)	1.51 [1.12 – 2.02]	0.006
Chronic obstructive pulmonary disease	727 (23.6%)	223 (27.8%)	1.24 [1.06 – 1.47]	0.01
Renal dialysis, n (%)	62 (2.0%)	36 (4.5%)	2.13 [1.50 – 3.02]	<0.001
TIA-CVA, n (%)	298 (9.7%)	84 (10.5%)	1.15 [0.90 – 1.46]	0.26
Previous surgical aortic-valve replacement	51 (1.7%)	14 (1.7%)	1.08 [0.61 – 1.91]	0.80
Hypertrophic cardiomyopathy	22 (0.7%)	7 (0.9%)	1.12 [0.46 – 2.69]	0.81
Smoker	96 (3.1%)	31 (3.9%)	1.26 [0.87 – 1.82]	0.23
HT	2141 (69.6%)	537 (67.0%)	0.92 [0.79 – 1.08]	0.30
Diabetes mellitus	778 (25.3%)	214 (26.7%)	1.10 [0.93 – 1.29]	0.27
Dyslipidemia	1494 (48.6%)	363 (45.3%)	0.84 [0.73 – 0.98]	0.02
Vitamin K antagonist	720 (23.4%)	221 (27.6%)	1.23 [1.05 – 1.45]	0.01
EUROSCORE > 20	1288 (42.7%)	451 (58.1%)	1.69 [1.45 – 1.97]	<0.001
EURO 1	682 (22.1%)	206 (25.7%)	1.21 [1.02 – 1.43]	0.03
EURO 2	818 (26.5%)	267 (33.3%)	1.31 [1.12 – 1.53]	<0.001
EURO 3	236 (7.6%)	69 (8.6%)	1.13 [0.87 – 1.47]	0.37
EURO 4	638 (20.7%)	159 (19.8%)	0.89 [0.74 – 1.07]	0.21
EURO 5	210 (6.8%)	124 (15.4%)	2.22 [1.81 – 2.71]	<0.001
EURO 6	9 (0.3%)	1 (0.1%)	0.75 [0.10 – 5.30]	0.77
EURO 7	101 (3.3%)	61 (7.6%)	2.34 [1.77 – 3.09]	<0.001
EURO 8	95 (3.1%)	21 (2.6%)	0.95 [0.61 – 1.49]	0.83
EURO 9	931 (30.1%)	292 (36.4%)	1.29 [1.11 – 1.50]	0.001
EURO 10	207 (6.7%)	75 (9.3%)	1.47 [1.15 – 1.89]	0.002
EURO 11	37 (1.2%)	10 (1.2%)	1.32 [0.71 – 2.46]	0.39
EURO 12	748 (24.2%)	256 (31.9%)	1.41 [1.20 – 1.65]	<0.001
EURO 13	64 (2.1%)	27 (3.4%)	1.94 [1.29 – 2.92]	0.001
EURO 14	2193 (71.0%)	567 (70.7%)	0.88 [0.75 – 1.03]	0.11
EURO 15	20 (0.6%)	12 (1.5%)	1.86 [1.02 – 3.37]	0.04
EURO 16	1 (0.03%)	2 (0.25%)	11.34 [2.82 – 45.57]	<0.001
STS score >10	1314 (47.2%)	405 (55.9%)	1.39 [1.19 – 1.63]	<0.001
CHARLSON	2.5 ± 1.9	3.0 ± 2.9	1.09 [1.06 – 1.11]	<0.001
Permanent pacemaker	421 (13.6%)	129 (16.2%)	1.21 [0.99 – 1.47]	0.06
Aortic annulus diameters (mm)	22.1 ± 2.2	22.3 ± 2.2	1.06 [1.02 – 1.10]	0.001
Mean aortic-valve gradient (mmHg)	48.9 ± 16.6	45.5 ± 16.3	0.99 [0.98 – 0.99]	<0.001
Aortic-valve area (cm 2 )	0.7 ± 0.2	0.7 ± 0.2	1.32 [0.88 – 2.00]	0.18
LVEF (%)	53.7 ± 14.0	51.4 ± 14.7	0.99 [0.98 – 0.99]	<0.001
Pulmonary hypertension (mmHg)	44.6 ± 13.9	47.8 ± 14.4	1.01 [1.01 – 1.02]	<0.001
Surgical approach	761 (30.5%)	167 (28.8%)	0.93 [0.77 – 1.12]	0.44
Left approach routes	860 (34.6%)	206 (35.5%)	1.07 [0.89 – 1.27]	0.49
Cross-over	19 (0.6%)	9 (1.1%)	1.61 [0.76 – 3.38]	0.21
Transapical approach	460 (16.5%)	181 (25.0%)	1.47 [1.23 –1.76]	<0.001
General anesthesia	2111 (68.1%)	586 (72.9%)	1.16 [0.98 – 1.37]	0.09
2 VALVES	46 (1.5%)	26 (3.2%)	2.23 [1.47 – 3.38]	<0.001
TEE	1888 (60.9%)	509 (63.3%)	1.01 [0.87 – 1.18]	0.88

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Comparison Among Patients ≥75 Years Having Percutaneous Coronary Angioplasty Using Drug-Eluting Stents Versus Bare Metal Stents
2. Usefulness of Regional Distribution of Coronary Artery Calcium to Improve the Prediction of All-Cause Mortality
3. Relation of Aortic Valve Calcium to Chronic Kidney Disease (from the Chronic Renal Insufficiency Cohort Study)
4. Erratum for Nazare et al. “Usefulness of Measuring Both Body Mass Index and Waist Circumference for the Estimation of Visceral Adiposity and Related Cardiometabolic Risk Profile (from the INSPIRE ME IAA Study).” Am J Cardiol 2015;115:307–315

Stay updated, free articles. Join our Telegram channel

Join