Transcatheter aortic valve implantation (TAVI) has become an option for patients with symptomatic severe aortic stenosis whose co-morbidities place them at high surgical risk. However, little is known regarding treatment allocation. From May 2008 to May 2011, all high-risk patients with symptomatic severe aortic stenosis referred to an experienced single-center TAVI clinic were reviewed. A total of 170 consecutive patients were evaluated. Of these, 58 (34%) were accepted for TAVI (mean age 81 ± 8 years). Thirty-three patients (19%) were accepted for conventional aortic valve replacement (AVR; mean age 83 ± 6 years). Sixty-two patients (37%) were treated conservatively (mean age 83 ± 6 years). Seventeen patients (10%) died awaiting complete assessment. At 30 days, all-cause mortality was 10% in the TAVI group, 3% in the conventional AVR group, and 32% in the conservatively treated group. Multivariate-adjustment identified the absence of chronic obstructive pulmonary disease (hazard ratio 0.30, 95% confidence interval 0.09 to 0.98, p <0.05) and the absence of frailty (hazard ratio 0.19, 95% confidence interval 0.07 to 0.55, p <0.01) as independent predictors of conventional AVR. In conclusion, of the high-risk patients with severe aortic stenosis referred for TAVI at a large single center, approximately 1/2 were accepted for intervention (conventional AVR or TAVI), and roughly 1/3 were treated conservatively.
Transcatheter aortic valve implantation (TAVI) has become a viable alternative for patients with symptomatic severe aortic stenosis (AS) considered to be high surgical risk. However, as this procedure becomes available, only some of those referred may be considered appropriate for TAVI. Accordingly, we sought to characterize the referral pattern and treatment allocation among patients referred for TAVI at a large tertiary cardiac care center.
Methods
From May 2008 and May 2011, all high-risk patients with symptomatic severe AS referred to the Hamilton Heath Sciences TAVI clinic (Hamilton, Ontario, Canada) were reviewed. The TAVI clinic accepts referrals from cardiologists, internists, or cardiac surgeons who consider patients to be at high risk on the basis of co-morbidities that would preclude conventional aortic valve replacement (AVR). Patients must be symptomatic, with documented New York Heart Association functional class II to IV symptoms, angina, or syncope, and must have echocardiographic documentation of severe AS. Hamilton Heath Sciences is the only cardiac surgical center for a large geographic area of >1.8 million residents of southern Ontario. Patients are reviewed by members of the TAVI heart team, comprising 2 interventional cardiologists and 2 cardiac surgeons. Patients must be turned down for conventional AVR by the 2 cardiac surgery physicians to be considered for TAVI. Consensus conferences are held regularly to ensure eligibility.
All patients undergo a complete history and physical examination along with electrocardiography and serologic laboratory investigations. Transthoracic echocardiography is routinely performed as part of the consultation. On the basis of the American College of Cardiology and American Heart Association guidelines, if there is an indication for AVR, patients are referred for coronary angiography, aortic root angiography. and angiography of the peripheral vasculature (iliofemoral) to assess access for TAVI. Transesophageal echocardiography is performed to assess valvular morphology and aortic annular size. Computed tomography of the aorta and peripheral vasculature are performed in selected patients. Upon completion of workup, the TAVI heart team (all 4 members) discusses each patient for appropriate management (AVR vs TAVI vs medical management) on the basis of a consensus decision. Surgical AVR is performed by a TAVI heart team cardiac surgeon. TAVI is performed by all 4 members of the TAVI heart team using either the transfemoral (preferred) or transapical approach. Edwards valves (Edwards Lifesciences, Inc., Irvine, California) are used in all patients.
Demographics, co-morbidities, echocardiographic measurements, and treatment decisions are routinely collected prospectively. New York Heart Association functional class IV is defined as symptoms with any physical activity or even at rest. Angiographic coronary artery disease is defined as any stenosis ≥50% in the coronary tree. Porcelain aorta is defined as extensive circumferential calcification of the ascending aorta seen on fluoroscopy and/or computed tomography. Frailty as a dichotomous variable is assessed on the basis of the overall impression of the TAVI heart team. Echocardiographic assessments collect mean aortic valve gradient and calculated aortic valve area. The left ventricular ejection fraction is calculated using the biplane method of disks. Pulmonary hypertension is defined as a pulmonary systolic pressure >60 mm Hg estimated by Doppler echocardiography. Severe mitral regurgitation is graded at the discretion of the echocardiographer.
Continuous variables are presented as mean ± SD. Categorical variables are presented as percentages. A logistic regression model was developed to determine demographics and clinical characteristics predictive of referral to conventional surgical AVR. The dependent variable was “received conventional AVR” (yes or no). Age (years), chronic obstructive pulmonary disease (yes or no), previous coronary artery bypass surgery (yes or no), porcelain aorta (yes or no), and frailty (yes or no) were included as independent variables in the regression model. Adjusted odds ratios for variables were determined from this model. A p value <0.05 was considered statistically significant. All analyses were performed using PASW version 20 (SPSS, Inc., Chicago, Illinois).
Results
From May 2008 to May 2011, 170 patients with high-risk symptomatic severe AS were referred to the TAVI clinic for assessment ( Figure 1 ). Fifty-eight patients (34%) were accepted for TAVI (mean age 81 ± 8 years). Thirty-three patients (19%) were accepted for conventional AVR (mean age 83 ± 6 years). Sixty-two patients (37%) were treated conservatively (mean age 83 ± 6 years). Seventeen patients (10.0%) died while awaiting assessment for TAVI. The time period for these patients awaiting complete assessment (including investigations) was on average 4 to 6 weeks. During this time, mortality was due mainly to sudden cardiac death (no deaths were procedurally related). Descriptive characteristics for each group are listed in Table 1 . The mean Society of Thoracic Surgeons surgical risk score for TAVI patients was 12.8 ± 8.0 compared with 6.5 ± 4.4 for conventional AVR patients, placing these patients at high surgical risk (predicted risk for operative mortality ≥15% for a Society of Thoracic Surgeons score ≥10). The calculated aortic valve area and mean aortic valve gradients are listed in Table 2 . Treatment assignment of patients who survived until complete TAVI assessment is shown in Figure 2 . The reasons for conservative treatment are outlined in Figure 3 . Patients’ declining the assigned valve procedures despite indications for valve treatment as due mainly to concerns of excessive risk. Patient ineligibility for valve replacement because of life expectancy <1 year was attributable to noncardiac co-morbidities. Of patients who underwent TAVI, 71% (n = 41) underwent the transfemoral approach, whereas 29% (n = 17) underwent the transapical approach. Thirty-day all-cause mortality was 10% (n = 6) in TAVI patients and 3% (n = 1) in the conventional AVR patients. Not unexpectedly, the conservatively treated patients had the highest rate of death (32% [n = 20]). Given the low 30-day all-cause mortality of conventional AVR patients, multivariate adjustment was performed to determine predictors for conventional AVR. After adjustment for age, chronic obstructive pulmonary disease, previous coronary artery bypass grafting, porcelain aorta, frailty, and pulmonary hypertension, only 2 variables were predictive of who underwent conventional AVR in this group: chronic obstructive pulmonary disease (adjusted odds ratio 0.30, 95% confidence interval 0.09 to 0.98, p <0.05) and frailty (odds ratio 0.19, 95% CI 0.07 to 0.55, p <0.01) ( Table 3 ). A classification table showed that the model correctly predicted the dependent variable 80% of the time. A Hosmer-Lemeshow test was nonsignificant (p = 0.24), indicating a good fit of the data. On the basis of Nagelkerke’s R 2 value, approximately 25% of the variance was explained by the included factors.
| Variable | Conservative (n = 62) | AVR (n = 33) | TAVI (n = 58) |
|---|---|---|---|
| Age (yrs) | 82.5 ± 5.5 | 82.7 ± 6.4 | 81.3 ± 7.6 |
| Men | 53% | 61% | 47% |
| Body mass index (kg/m 2 ) | 25.8 ± 5.4 | 26.5 ± 4.8 | 25.8 ± 5.6 |
| Diabetes mellitus | 26% | 36% | 31% |
| Hypercholesterolemia | 58% | 67% | 81% |
| Hypertension | 79% | 67% | 88% |
| Current smokers | 8% | 6% | 0% |
| New York Heart Association class IV | 18% | 6% | 24% |
| Atrial fibrillation | 36% | 46% | 36% |
| Angiographic coronary disease | — | 64% | 67% |
| Previous myocardial infarction | 18% | 24% | 33% |
| Previous percutaneous coronary intervention | 8% | 6% | 22% |
| Previous coronary artery bypass grafting | 24% | 21% | 36% |
| Chronic obstructive pulmonary disease | 29% | 9% | 36% |
| Cerebrovascular disease | 16% | 27% | 24% |
| Peripheral artery disease | 13% | 6% | 14% |
| Creatinine (μmol/L) | 130 ± 106 | 102 ± 53.5 | 115 ± 95.3 |
| Dialysis | 0% | 0% | 3% |
| Porcelain aorta | 10% | 0% | 14% |
| Frail | 48% | 18% | 55% |
| Variable | Conservative (n = 62) | AVR (n = 33) | TAVI (n = 58) |
|---|---|---|---|
| Aortic valve area (cm 2 ) | 0.76 ± 0.22 | 0.66 ± 0.19 | 0.62 ± 0.14 |
| Mean gradient (mm Hg) | 38.4 ± 16.5 | 47.1 ± 14.8 | 48.7 ± 12.2 |
| Left ventricular ejection fraction | 47% | 52% | 50% |
| Left ventricular ejection fraction <40% | 31% | 15% | 19% |
| Pulmonary hypertension | 37% | 24% | 36% |
| Severe mitral regurgitation | 13% | 0% | 16% |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
