In patients with aortic stenosis who cannot have surgery, transcatheter aortic valve replacement using the Edwards SAPIEN valve has been shown to improve survival rate and is approved for commercial use in the United States. This study aims to assess the clinical profile, procedural characteristics, and in-hospital complications in patients treated with a commercial SAPIEN valve outside the clinical trial context. We retrospectively analyzed 69 consecutive patients who underwent transcatheter aortic valve replacement with a commercial SAPIEN valve compared with 55 Placement of AoRTic traNscathetER valves (PARTNER) trial patients from cohort B enrolled in the same institution by the same Heart Team. Compared with the commercial group, patients in the PARTNER cohort B had higher mean Society of Thoracic Surgeons score (10 ± 5 vs 9 ± 4, p = 0.04) and a lower rate of peripheral arterial disease (19% vs 44%, p = 0.004). Most patients in the commercial group had the procedure under conscious sedation (83% vs 66%, p = 0.03). Planned surgical cut down for vascular access was rare in the commercial group (1.4% vs 46%, p <0.001). The overall rates of major vascular complications, life-threatening or major bleeding, and blood transfusions were lower in commercial group (7.2% vs 27%, p = 0.003; 2.9% vs 16%, p = 0.01; and 28% vs 60%, p <0.001, respectively). In-hospital all-cause mortality (5.8% vs 9.1%, p = 0.51) and stroke rates (7.2% vs 14.5%, p = 0.19) were not statistically different between groups. The median length of hospitalization (p <0.001) and postprocedural length of stay (p = 0.01) was shorter in the commercial group. In conclusion, transfemoral commercial use of the Edwards SAPIEN valve for inoperable patients shows similar in-hospital mortality and stroke rates compared with PARTNER cohort B. The refinements in the procedure such as more conscious sedation, experience of the operators, and careful vascular planning in the commercial group led to lesser vascular and bleeding complications and shorter length of stay.
Transcatheter aortic valve replacement (TAVR) has become an effective treatment for severe aortic stenosis in patients deemed at prohibitive or excessive surgical risk. In the randomized, controlled Placement of AoRTic traNscathetER valves (PARTNER) cohort B study, TAVR with a balloon-expandable Edwards SAPIEN valve (Edwards Lifesciences Corporation, Irvine, California) was superior to standard therapy in inoperable patients with severe aortic stenosis who were unsuitable for surgical aortic valve replacement based on anatomic factors or surgical risk. Based on the results of this trial (November 2, 2011), the Food and Drug Administration approved the Edwards SAPIEN transcatheter heart valve (sizes 23 and 26 mm) for commercial use through a transfemoral delivery system in patients with severe, symptomatic, native aortic valve stenosis who have been determined by a cardiac surgeon to be inoperable for open surgical aortic valve replacement and in whom existing co-morbidities would not preclude the expected benefit from correction of the aortic stenosis. Although randomized trials are best at evaluating the treatment efficacy, they are limited by rigid inclusion and exclusion criteria. This leads to the exclusion of important patient subgroups and limits the generalizability of the results to an all-comer population. In contrast, previous postapproval studies from Europe have indicated that patients referred to commercial TAVR are at less risk than those included in the clinical trials. The goal of the present study was to characterize the clinical profile, procedural characteristics, and in-hospital outcomes of patients who underwent TAVR with a commercial SAPIEN valve outside the clinical trial context.
Methods
The study was approved by the institutional review board of the MedStar Health Research Institute (Washington, DC). From April 2012 to February 2013, 69 consecutive inoperable patients with symptomatic severe aortic stenosis who underwent TAVR with a commercial SAPIEN valve through the transfemoral route were included in the commercial group. The comparative group consisted of 55 inoperable patients from PARTNER cohort B who were treated with a SAPIEN valve at our institution from January 2008 to August 2011.
All potential candidates for TAVR were evaluated by a multidisciplinary team composed of an interventional cardiologist, clinical cardiologists, and cardiac surgeons (Heart Team) who determined the patient’s eligibility for TAVR. All patients had angiographic (aortogram, coronary, and aorto-iliac evaluation), echocardiographic, and computed tomography (chest without contrast and abdominal with contrast) assessments before the procedure. The clinical, electrocardiographic, echocardiographic, angiographic, and procedural details and clinical outcomes were prospectively recorded and entered into a registry by independent physicians from the Cardiovascular Research Institute data center at MedStar Washington Hospital Center (Washington, DC).
The Edwards SAPIEN heart valve system consists of a trileaflet bovine pericardial valve and a balloon-expandable stainless steel support frame. All TAVR procedures were performed in a hybrid operating room. Conscious sedation was provided by a cardiac anesthesiologist who was present and ready with airway equipment and induction agents for rapid intubation if necessary. The decision to provide general anesthesia with intubation or conscious sedation was made on a case-by-case basis and left to anesthesiologist’s discretion. The femoral artery was accessed percutaneously or by surgical cut down with exposure of the common femoral artery. In cases using the complete percutaneous approach, after wire introduction, the arteriotomy access site was prepared with either a single Prostar XL device (Abbott Vascular Devices, Redwood City, California) or 2 Perclose ProGlide 6F suture mediated closure devices (Abbott Vascular Devices, Redwood City, California). A 22Fr or 24Fr delivery sheath was used depending on the selected size of the valve (23 or 26 mm). The final placement and function of the aortic valve prosthesis was assessed by angiography and intraoperative transesophageal echocardiography (TEE). Adjunctive pharmacologic therapy included heparin during the procedure and dual antiplatelet therapy (aspirin and clopidogrel) for 3 to 6 months after the procedure.
For all clinical end points, the standard definitions for transcatheter aortic valve implantation in the consensus report from the Valve Academic Research Consortium were used. The clinical in-hospital end points examined included total mortality, any stroke, major vascular, bleeding complications, and the hospital length of stay.
Continuous variables are summarized as mean ± SD or median and interquartile range, as appropriate and were compared using Student t test or Wilcoxon rank sum test, depending on variable distribution. Categorical variables were compared using chi-square or Fisher’s exact test. A 2-sided alpha level of 0.05 was used for all superiority testing. All statistical analyses were performed using SAS software, version 9.3 (SAS Institute, Cary, North Carolina).
Results
All 124 study patients (mean age, 83 ± 8 years) had severe, symptomatic, native valve aortic stenosis with a mean gradient of 48 ± 14 mm Hg, maximal velocity of 4.3 ± 0.6 m/s, mean aortic valve area of 0.66 ± 0.11 cm 2 , mean Society of Thoracic Surgeons (STS) score of 9.5 ± 5, and mean ejection fraction of 53 ± 13%. Patients’ baseline clinical characteristics and the preprocedural echocardiographic findings are listed in Table 1 . Compared with patients in the commercial group, PARTNER cohort B had a higher rate of Caucasians (86% vs 69%, p = 0.03) and higher mean STS score (10 ± 5 vs 9 ± 4, p = 0.04). In contrast, patients in the commercial group had a higher rate of peripheral arterial disease (44% vs 19%, p = 0.004) and also included 6% of dialysis-dependent patients who were excluded from the PARTNER trial.
| Variable | Commercial Valve (n = 69) | PARTNER Cohort B (n = 55) | p |
|---|---|---|---|
| Age (yrs) | 83 ± 8 | 83 ± 7 | 0.87 |
| Men | 61 (42/69) | 46 (25/55) | 0.09 |
| European American | 69 (40/58) | 86 (44/51) | 0.03 |
| STS score | 9 ± 4 | 10 ± 5 | 0.04 |
| STS score >8 | 52 (32/62) | 66 (35/53) | 0.12 |
| STS score >15 | 10 (6/62) | 15 (8/53) | 0.38 |
| Porcelain aorta | 10 (7/69) | 7 (4/55) | 0.75 |
| Systemic hypertension ∗ | 93 (64/69) | 93 (51/55) | 1.0 |
| Diabetes mellitus | 33 (23/69) | 31 (17/55) | 0.77 |
| Insulin-treated diabetes mellitus | 17 (12/69) | 7 (4/55) | 0.09 |
| Chronic obstructive pulmonary disease | 28 (19/68) | 33 (18/54) | 0.52 |
| Coronary artery disease | 72 (43/60) | 87 (28/32) | 0.09 |
| Dyslipidemia | 85 (57/67) | 76 (42/55) | 0.22 |
| Peripheral arterial disease | 44 (28/63) | 19 (10/52) | 0.004 |
| Previous cerebrovascular accident or transient ischemic attack | 15 (10/67) | 24 (13/55) | 0.22 |
| Arrhythmia | 40 (27/68) | 51 (28/55) | 0.21 |
| Atrial fibrillation and/or flutter | 33 (23/69) | 51 (28/55) | 0.05 |
| Pacemaker or implanted cardiac defibrillator | 13 (8/63) | 10 (3/31) | 1.0 |
| Body mass index (kg/m 2 ) | 28 ± 6 | 29 ± 8 | 0.60 |
| Baseline creatinine (mg/dl) | 1.34 ± 0.93 | 1.34 ± 0.55 | 0.99 |
| Dialysis | 6 (4/67) | 0 | 0.13 |
| Previous intervention | |||
| Previous angioplasty | 30 (20/66) | 26 (14/54) | 0.60 |
| Previous coronary artery bypass grafting | 29 (20/66) | 24 (13/54) | 0.54 |
| Preprocedural echocardiography | |||
| Left ventricular ejection fraction (%) | 53 ± 13 | 53 ± 14 | 0.95 |
| Left ventricular ejection fraction <40% | 22 (15/67) | 24 (13/53) | 0.78 |
| Mean aortic valve gradient, mm Hg | 47 ± 14 | 49 ± 15 | 0.68 |
| Max aortic valve gradient, mm Hg | 71 ± 14 | 67 ± 23 | 0.64 |
| Aortic valve area, cm 2 | 0.69 ± 0.1 | 0.61 ± 0.1 | 0.002 |
| Mean pulmonary artery systolic pressure, mm Hg | 56 ± 20 | 53 ± 21 | 0.61 |
∗ History of systemic hypertension diagnosed and/or treated with medication or currently being treated with diet and/or medication by a physician.
The use of planned surgical cut down for vascular access (1.4% vs 46%, p <0.001) was lesser in the commercial group ( Figure 1 ). The percutaneous route was used for vascular access in all but 1 patient in the commercial group. Conscious sedation with local anesthesia (83% vs 66%, p = 0.03) was used more often in the commercial group. The crossover rate from conscious sedation to general anesthesia (16% vs 25%, p = 0.32) was similar in both groups.
Procedural characteristics are listed in Table 2 . Most patients from both groups underwent the TAVR procedure guided by intraprocedural TEE guidance (∼97%). TEE was well tolerated even with conscious sedation, showing a very low complication rate (1.9%). Technical success with device delivery (∼97%) and valve position (∼96%) was achieved in most patients from both groups. One patient in PARTNER cohort B had valve migration, which necessitated a second valve that was implanted without any untoward events. The median time of procedure (time of vascular access to guidewire removal) was similar between groups (80 minutes [63, 129] vs 94 minutes [61, 120], p = 0.82) as was median fluoroscopic time (19 minutes [16, 25] vs 19 minutes [14, 23], p = 0.51). Contrast volume use (110 ml [87, 140] vs 72 ml [54, 120], p <0.001) was greater in the commercial group compared with the PARTNER cohort B group.
| Variable | Commercial Valve (n = 69) | PARTNER Cohort B (n = 55) | p |
|---|---|---|---|
| Intraprocedural transesophageal echocardiogram | 97 (66/68) | 96 (53/55) | 1.0 |
| Valve size | |||
| 23 mm (22Fr introducer) | 57 (39/69) | 57 (32/56) | 1.0 |
| 26 mm (24Fr introducer) | 43 (30/69) | 43 (24/56) | 1.0 |
| Sedation during procedure | |||
| Conscious sedation | 83 (57/69) | 66 (36/55) | 0.03 |
| Converted to general anesthesia | 16 (9/57) | 25 (9/36) | 0.32 |
| Vascular access | |||
| Planned surgical cut down for access | 1 (1/69) | 46 (25/55) | <0.001 |
| Unplanned surgical closure | 3 (2/69) | 11 (6/55) | 0.14 |
| Closure device | 99 (68/69) | 54 (30/55) | <0.001 |
| Perclose closure device | 84 (58/69) | 45 (25/55) | <0.001 |
| Procedure length, minutes | 80 (63, 129) | 94 (61, 120) | 0.82 |
| Fluoroscopic time, minutes | 19 (16–25) | 19 (14–23) | 0.51 |
| Contrast volume, ml | 110 (87–140) | 72 (54–120) | <0.001 |
| Delivery success | 97 (63/65) | 98 (53/54) | 1.0 |
| Position success | 96 (63/66) | 96 (52/54) | 1.0 |
| Successful valve deployment | 97 (65/67) | 100 (56/56) | 1.0 |
In-hospital all-cause mortality (5.8% vs 9.1%, p = 0.51) and stroke rates (7.2% vs 14.5%, p = 0.19) were statistically not different between groups ( Table 3 ). No intraprocedural deaths were observed in this cohort. The rates of major vascular complications (7.2% vs 27%, p = 0.003) and life-threatening or major bleeding complications (2.9% vs 16%, p = 0.01) were lower in the commercial group. Similarly, the rate of blood transfusion (28% vs 60%, p <0.001) and the median number of units of blood transfused (1 [0, 2] vs 2 [2, 4], p = 0.003) was lower in the commercial group. The median length of hospitalization (6 days [4, 10] vs 10 days [6, 16], p <0.001), postprocedural length of stay (5 days [4, 6] vs 7 days [4, 13], p = 0.01), and the length of intensive care unit stay (1 days [1, 2] vs 2 days [1, 4], p = 0.001) were lower in the commercial group ( Figure 2 ).
