There is currently no consensus on the duration of hospitalization required after transfemoral transcatheter aortic valve implantation (TAVI). We report the feasibility and safety of early discharge after TAVI with the Edwards SAPIEN-XT prosthesis. From 2009 to 2013, 337 patients underwent transfemoral TAVI with the Edwards SAPIEN-XT prosthesis using local anesthesia and were discharged home either early (≤3 days, Early Discharge group, n = 121) or after 3 days (Late Discharge group, n = 216). The primary end point of the study combined death and rehospitalization from discharge to 30-day follow-up. Patients in the Early Discharge group were less symptomatic (New York Heart Association class ≥III: 64.5% vs 75.5%, p = 0.01) and had less renal failure (creatinine: 102.1 ± 41.0 vs 113.3 ± 58.9 μmol/L, p = 0.04), atrial fibrillation (33.1% vs 46.3%, p = 0.02), and previous balloon aortic valvuloplasty (11.6% vs 23.1%, p = 0.01) and were more likely to have a pacemaker before TAVI (16.5% vs 8.3%, p = 0.02). Pre-existing pacemaker (p = 0.05) and the absence of acute kidney injury (p = 0.02) were independent predictors of an early discharge, whereas previous balloon aortic valvuloplasty (p = 0.03) and post-TAVI blood transfusions (p = 0.002) were independent predictors of late discharge. The primary end point occurred in 4 patients (3.3%) in the Early Discharge group and in 11 patients (5.1%) in the Late Discharge group (p = 0.58). In conclusion, the results of our study suggest that early discharge after transfemoral TAVI using the Edwards SAPIEN-XT prosthesis is feasible and safe in selected patients.
Since the First-In-Man transcatheter aortic valve implantation (TAVI) in 2002, our group has been proposing a progressive simplification of TAVI procedures while keeping patient safety as the main objective. We soon promoted a “minimalist approach” for all transfemoral cases using local anesthesia and fluoroscopic guidance for valve implantation without intraprocedural transesophageal echocardiography. Since the launch of the SAPIEN-XT device in October 2009 and the concomitant decrease in sheath size to 16 F/20 F, most operators have switched to a pure percutaneous approach. This further strengthened the use of the “minimalist approach” for us and many other operators and early discharge progressively emerged as a reasonable option in the absence of procedural complications. There is currently no consensus on the duration of monitoring required after TAVI. Prolonged hospitalization often appears unnecessary, whereas shorter length of stay with early return home may be beneficial, particularly in old and frail patients, and might also be cost effective. We, therefore, aimed to evaluate the feasibility and safety of early discharge after transfemoral TAVI.
Methods
We included all consecutive patients implanted with the Edwards SAPIEN-XT valve through a transfemoral approach (Edwards Lifesciences Inc., Irvine, California) from October 2009 to November 2013. The Heart Team established the indication of TAVI according to the current guidelines. All patients were included in national or international registries, after approval of the corresponding Ethics Committee, and provided signed informed consent.
The technique of transfemoral SAPIEN-XT valve implantation using local anesthesia has been previously described and reported by our group. After the procedure, all the patients were monitored in the intensive care unit (ICU) for at least 24 hours. Electrocardiography was recorded daily until discharge, and transthoracic echocardiography was performed before discharge. Patients were eligible for early discharge if the procedure was not performed in emergency and if there were no major complications during the procedure and the first 24 hours of monitoring in ICU (i.e., stroke, myocardial infarction, major vascular complication requiring transfusions, life-threatening bleeding, acute kidney injury stage 2 or 3). Careful electrocardiographic monitoring in ICU allowed for early conduction disturbances detection. A pacemaker was systematically implanted in patients with persistent complete atrioventricular block 24 hours after TAVI. Patients with persistent de novo complete left bundle branch block were also closely monitored and were not eligible for early discharge. At discharge, patients were advised to return to our department in case of worsening of dyspnea, syncope, or any abnormal symptom. Thirty-day outpatient visit and echocardiography were systematically performed in all patients at our institution.
Baseline, procedural, and in-hospital data were entered prospectively in the Rouen TAVI registry. Complications and procedural outcomes were defined according to the Valvular Academic Research Consortium-2 consensus statement. Clinical and transthoracic echocardiographic parameters were obtained at baseline, discharge, and 1 month.
The study population was divided into 2 groups based on tertile length of stay values. Length of stay was calculated from TAVI procedure (day 0) to discharge. Patients in the lowest tertile (length of stay ≤3 days, n = 121) constituted the Early Discharge group, and patients with a length of stay >3 days (n = 216) constituted the Late Discharge group. Qualitative variables were expressed as percentage, and quantitative variables as mean ± SD or median (25th to 75th interquartile range). Comparison of numerical variables was performed with the Student t test or Wilcoxon rank-sum test, depending on variable distribution. The chi-square test or Fisher’s exact test was used to compare qualitative variables. A logistic regression multivariable analysis was used to assess independent correlates of early discharge. The model was built on the basis of the univariate association between the variable and early discharge with a p = 0.05 and an elimination p = 0.10. All statistical tests were 2 sided. Differences were considered statistically significant at a p value ≤0.05. All data were analyzed using SPSS software (version 17.0; IBM, Armonk, New York).
Results
A total of 424 consecutive patients underwent TAVI using the SAPIEN-XT prosthesis at our center in the specified period. We excluded 65 patients implanted through transapical or transaortic routes, 14 transfemoral patients who died before 72 hours after implantation, and 8 patients who did not return home directly but were transferred to a rehabilitation center or another hospital. The final study population comprised 337 patients who returned home after discharge. The distribution of length of stay after TAVI is shown in Figure 1 . One hundred twenty-one (36%) patients were discharged within 3 days (Early Discharge group), whereas 216 (64%) were discharged >3 days after TAVI (Late Discharge group). The proportion of patients with a length of stay ≤3 days over the years is shown in Figure 2 . Interestingly, the incidence of early discharge gradually increased from 0% in 2009 to 53.2% in 2013. In parallel, the mean length of stay gradually decreased from 2009 to 2013 ( Figure 3 ).
Baseline characteristics of the studied population are listed in Tables 1 and 2 . Patients in the Early Discharge group had less renal failure and were more likely to have a history of pacemaker. Moreover, they had less history of atrial fibrillation and balloon aortic valvuloplasty and were less symptomatic. Other variables were comparable between the 2 groups.
| Variables | Overall population (n=337) | Early discharge (n=121) | Late discharge (n=216) | P |
|---|---|---|---|---|
| Age (years) | 84.0 ± 6.8 | 83.7 ± 6.9 | 84.2 ± 6.2 | 0.46 |
| Men | 145 (43.0%) | 58 (47.9%) | 87 (40.3%) | 0.17 |
| Body mass index (kg/m 2 ) | 26.6 ± 5.1 | 26.8 ± 5.1 | 26.5 ± 5.1 | 0.61 |
| Hypertension | 240 (71.2%) | 89 (73.6%) | 151 (69.9%) | 0.56 |
| Diabetes mellitus | 101 (30.0%) | 32 (26.4%) | 69 (31.9%) | 0.32 |
| Previous myocardial infarction | 34 (10.1%) | 12 (9.9%) | 22 (10.2%) | 0.94 |
| Previous percutaneous coronary intervention | 75 (22.2%) | 25 (20.7%) | 50 (23.1%) | 0.60 |
| Previous coronary bypass | 31 (9.2%) | 11 (9.1%) | 20 (9.3%) | 0.96 |
| Sinus rhythm | 238 (70.6%) | 94 (77.6%) | 144 (66.6%) | 0.04 |
| History of atrial fibrillation | 140 (41.5%) | 40 (33.1%) | 100 (46.3%) | 0.02 |
| Right bundle branch block | 43 (12.8%) | 19 (15.7%) | 24 (11.1%) | 0.22 |
| Pacemaker | 38 (11.3%) | 20 (16.5%) | 18 (8.3%) | 0.02 |
| Previous balloon aortic valvuloplasty | 64 (18.9%) | 14 (11.6%) | 50 (23.1%) | 0.01 |
| Peripheral artery disease | 35 (10.4%) | 15 (6.8%) | 20 (9.3%) | 0.37 |
| Porcelain aorta | 11 (3.3%) | 5 (4.1%) | 6 (2.8%) | 0.50 |
| Previous stroke | 19 (5.6%) | 7 (5.8%) | 12 (5.6%) | 0.93 |
| Creatinine (μmol/l) | 109.4 ± 53.5 | 102.1 ± 41.0 | 113.3 ± 58.9 | 0.04 |
| Creatinine clearance (ml/min) | 47.0 ± 22.4 | 50.9 ± 24.7 | 44.9 ± 20.8 | 0.02 |
| Chronic obstructive pulmonary disease | 59 (17.5%) | 20 (16.5%) | 39 (18.1%) | 0.72 |
| Chest irradiation | 18 (5.3%) | 8 (6.6%) | 10 (4.6%) | 0.44 |
| Neoplasia | 55 (16.3%) | 19 (15.7%) | 36 (16.7%) | 0.80 |
| NYHA III or more | 241 (72.6%) | 78 (64.5%) | 163 (75.5%) | 0.01 |
| Logistic EuroSCORE (%) | 16.9 ± 9.6 | 15.6 ± 9.6 | 17.6 ± 9.5 | 0.07 |
| Variables | Overall population (n=337) | Early discharge (n=121) | Late discharge (n=216) | P |
|---|---|---|---|---|
| Annulus diameter (mm) | 21.5 ± 2.3 | 21.6 ± 2.0 | 21.5 ± 2.5 | 0.93 |
| Mean aortic gradient (mm Hg) | 47.0 ± 17.9 | 47.0 ± 17.1 | 47.1 ± 18.8 | 0.96 |
| Aortic valve area (cm 2 ) | 0.64 ± 0.24 | 0.61 ± 0.25 | 0.65 ± 0.24 | 0.12 |
| Pulmonary artery systolic pressure (mm Hg) | 44.8 ± 14.9 | 44.4 ± 12.9 | 45.0 ± 15.8 | 0.77 |
| Pulmonary artery systolic pressure ≥ 60 mm Hg | 37 (11.0%) | 8 (6.6%) | 29 (13.4%) | 0.19 |
| Left ventricular ejection fraction (%) | 59.0 ± 16.3 | 59.7 ± 16.3 | 58.7 ± 16.3 | 0.61 |
| Left ventricular ejection fraction < 30% | 14 (4.2%) | 2 (1.6%) | 12 (5.6%) | 0.10 |
Procedural and in-hospital outcomes are listed in Tables 3 to 5 . Valve size and proportion of valve-in-valve procedures were similar in the 2 groups. Procedural duration and contrast volume were significantly lower in the Early Discharge group, whereas x-ray time was not significantly different. None of the patients in the Early Discharge group had stroke, life-threatening bleeding, acute myocardial infarction, or acute kidney injury stages 2 or 3 before discharge. Furthermore, patients in the Early Discharge group had less major bleeding/vascular complications and transfusions. Other variables were not significantly different between the 2 groups. ICU stay was significantly shorter in the Early Discharge group. Postprocedural echocardiographic characteristics are listed in Table 4 . Mean aortic gradient and aortic valve area were similar, whereas the proportion of patients with an aortic regurgitation ≥grade 2 was significantly lower in the Early Discharge group.
| Variables | Overall population (n=337) | Early discharge (n=121) | Late discharge (n=216) | P |
|---|---|---|---|---|
| Valve size (mm) | 0.23 | |||
| 23 | 154 (45.7%) | 52 (43.0%) | 102 (47.2%) | |
| 26 | 153 (45.4%) | 54 (44.6%) | 99 (45.8%) | |
| 29 | 30 (8.9%) | 15 (12.4%) | 15 (6.9%) | |
| Valve-in-valve | 5 (1.5%) | 1 (0.8%) | 4 (1.8%) | 0.66 |
| Procedural duration (minutes) | 102.4 ± 46.4 | 83.4 ± 43.8 | 110.6 ± 45.5 | 0.02 |
| X-ray time (minutes) | 19.6 ± 9.0 | 19.0 ± 11.7 | 20.0 ± 6.9 | 0.38 |
| Contrast volume (ml) | 194.4 ± 74.5 | 183.7 ± 63.7 | 200.6 ± 79.5 | 0.05 |
| Variables | Overall population (n=337) | Early discharge (n=121) | late discharge (n=216) | P |
|---|---|---|---|---|
| Mean aortic gradient (mm Hg) | 10.3 ± 4.5 | 10.3 ± 4.4 | 10.2 ± 4.5 | 0.78 |
| Aortic valve area (cm 2 ) | 1.72 ± 0.64 | 1.73 ± 0.64 | 1.71 ± 0.65 | 0.69 |
| Aortic regurgitation ≥ Grade 2 | 78 (23.1%) | 20 (16.5%) | 58 (26.8%) | 0.03 |
| Pulmonary artery systolic pressure (mm Hg) | 29.6 ± 21.2 | 27.0 ± 19.7 | 31.0 ± 21.9 | 0.10 |
| Left ventricular ejection fraction (%) | 59.0 ± 16.3 | 60.7 ± 15.2 | 62.0 ± 14.5 | 0.47 |
| Variables | Overall population (n=337) | Early discharge (n=121) | Late discharge (n=216) | P |
|---|---|---|---|---|
| Stroke | ||||
| Major stroke | 4 (1.2%) | 0 | 4 (1.9%) | 0.14 |
| Minor stroke | 2 (0.5%) | 0 | 2 (0.9%) | 0.32 |
| Transient ischemic attack | 2 (0.5%) | 0 | 2 (0.9%) | 0.32 |
| Bleeding | ||||
| Life-threatening | 22 (6.5%) | 0 | 22 (10.2%) | <0.0001 |
| Major | 36 (10.7%) | 6 (4.9%) | 30 (13.8%) | 0.01 |
| Minor | 25 (7.4%) | 7 (5.8%) | 18 (8.3%) | 0.39 |
| Transfusions | 61 (18.1%) | 3 (2.4%) | 58 (26.8%) | <0.0001 |
| Acute kidney injury | 0.01 | |||
| Stage 3 | 2 (0.5%) | 0 | 2 (0.9%) | |
| Stage 2 | 1 (0.3%) | 0 | 1 (0.4%) | |
| Stage 1 | 50 (14.8%) | 8 (6.6%) | 42 (19.4%) | |
| Vascular complications | ||||
| Major | 52 (15.4%) | 7 (5.8%) | 45 (20.8%) | <0.0001 |
| Minor | 25 (7.4%) | 8 (6.6%) | 17 (7.8%) | 0.70 |
| Urgent vascular surgery | 4 (1.2%) | 0 | 4 (1.9%) | 0.30 |
| Failure of closure device | 16 (4.7%) | 1 (0.8%) | 15 (6.9%) | 0.002 |
| Covered stent | 28 (8.3%) | 4 (3.3%) | 24 (11.1%) | 0.01 |
| Peri-procedural myocardial infarction | 2 (0.6%) | 0 | 2 (0.9%) | 1 |
| Pacemaker | 20 (5.9%) | 4 (3.3%) | 15 (6.9%) | 0.13 |
| Days from procedure to Intensive Care Unit discharge | ||||
| Median (interquartile range) | 1 (1) | 1 (1) | 2 (2) | <0.0001 |
| Days from procedure to discharge | ||||
| Median (interquartile range) | 5 (4) | 3 (1) | 6 (4) | <0.0001 |
Univariate and multivariable analyses were performed to identify predictive factors of early discharge and are listed in Tables 6 and 7 . Pre-existing pacemaker before TAVI and absence of acute kidney injury were independent predictors of an early discharge, whereas previous balloon aortic valvuloplasty and post-TAVI blood transfusions were independent predictors of a late discharge.
| Variables | Univariate analysis | ||
|---|---|---|---|
| Odd ratio | 95% Confidence Interval | P value | |
| Transfusions | 0.07 | 0.02-0.23 | <0.0001 |
| Major vascular complication | 0.24 | 0.10-0.54 | 0.001 |
| Covered stent | 0.28 | 0.09-0.83 | 0.02 |
| Left ventricular ejection fraction ≤ 30% | 0.30 | 0.06-1.35 | 0.19 |
| Life-threatening or major bleeding | 0.32 | 0.13-0.80 | 0.01 |
| Previous balloon aortic valvuloplasty | 0.43 | 0.23-0.81 | 0.01 |
| Atrial fibrillation | 0.53 | 0.35-0.98 | 0.04 |
| NYHA ≥ 3 | 0.53 | 0.33-0.88 | 0.01 |
| Aortic regurgitation ≥ grade 2 post TAVI | 0.54 | 0.31-0.95 | 0.03 |
| Delta haemoglobin | 0.58 | 0.47-1.72 | 0.11 |
| Haemoglobin before TAVI | 0.90 | 0.77-1.04 | 0.15 |
| Creatinine clearance before TAVI | 0.99 | 0.98-1.00 | 0.02 |
| Delta creatinine | 0.99 | 0.98-1.00 | <0.0001 |
| NT-pro brain natriuretic peptide | 1.00 | 1.00-1.00 | 0.86 |
| Age | 1.01 | 0.98-1.04 | 0.46 |
| Logistic EuroSCORE | 1.02 | 1.00-1.05 | 0.07 |
| Diabetes mellitus | 1.32 | 0.81-2.17 | 0.27 |
| Man | 1.37 | 0.87-2.13 | 0.17 |
| Sinus rhythm | 1.72 | 1.03-2.86 | 0.04 |
| Previous pacemaker | 2.17 | 1.10-4.35 | 0.02 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
