Percutaneous balloon aortic valvuloplasty (PBAV) is a procedure used for palliation, bridging to surgery, and as an integral step in the procedure for percutaneous aortic valve replacement. Older patients with severe aortic stenosis are thought to have greater risk for adverse perioperative events than younger patients. The aim of this study was to evaluate the outcomes of patients aged >80 years and those aged ≤80 years who underwent PBAV to identify factors associated with adverse clinical outcomes. This was a retrospective study of 111 consecutive patients with severe symptomatic aortic stenosis who underwent retrograde PBAV at Massachusetts General Hospital from December 2004 to December 2008. Forty-nine patients (44%) were men, and the mean age for the whole group was 82 ± 8 years. Patients were divided into 2 age groups: those aged >80 years (n = 73) and those aged ≤80 years (n = 38). Procedural outcomes, complications, and in-hospital adverse events were compared. Multivariate logistic regression was used for the adjusted analysis. Nearly 90% of patients were in New York Heart Association class III or IV. Patients aged >80 years had lower baseline ejection fractions (43.5% vs 56.1%, p <0.01) and smaller aortic valve areas (0.59 vs 0.73 cm 2 , p <0.01). Although the 2 age groups had a similar percentage of aortic valve area increase (55.5% vs 45.2%, p = 0.28), those aged >80 years had smaller post-PBAV aortic valve areas (0.89 vs 1.02 cm 2 , p <0.05). Overall, in-hospital mortality was 8.1%, with no significant differences between the groups. Advanced age was not an independent predictor of in-hospital death, myocardial infarction, stroke, cardiac arrest, or tamponade; however, patients aged >80 years had a significantly higher incidence of intraprocedural emergent intubation and cardiopulmonary resuscitation compared to the younger group. New York Heart Association class was the only independent predictor of worse in-hospital outcomes. In conclusion, compared to younger patients, those aged >80 years had less favorable preprocedural characteristics for PBAV but similar overall in-hospital clinical outcomes. Patients aged >80 years had significantly higher incidence of emergent intubation and cardiopulmonary resuscitation during PBAV.
Percutaneous balloon aortic valvuloplasty (PBAV) can be performed with reasonable safety in patients in their 70s and 80s, and a few investigators have reported successful outcomes in small series of nonagenarian patients. Nevertheless, age has been shown to be a risk factor for worse short- and long-term outcomes in PBAV. Accordingly, physicians may have apprehension about performing PBAV in patients at advanced ages, because they are frequently frail, with multiple co-morbidities, poor functional status, and vascular access issues. In 1991, the Mansfield Scientific Aortic Valvuloplasty Registry reported adverse in-hospital outcomes of 492 patients aged <70, 70 to 80, and >80 years that ranged from 4.2% to 9.4% but were statistically similar. Although the improvements in PBAV equipment over the past 2 decades have been minor, there are techniques that are now used routinely that allow the procedure to be performed more safely. We present a systematic comparison of patients aged >80 years to those aged ≤80 years in the contemporary era, in which rapid ventricular pacing, percutaneous suture-mediated arterial closure, smaller French catheters, the avoidance of double balloons, and routine periprocedural echocardiography have been used to help improve the overall safety and technical aspects of PBAV.
Methods
This was a retrospective cohort study of patients with severe, symptomatic calcific aortic stenosis (aortic valve area [AVA] <1.0 cm 2 determined by echocardiography) who underwent nonemergent retrograde percutaneous PBAV at Massachusetts General Hospital from December 22, 2004, to December 15, 2008. Patients who were in cardiogenic shock and those requiring mechanical ventilatory support before the procedure were excluded. We also excluded patients with bicuspid aortic valves, previous PBAV, moderate or severe aortic regurgitation, and severe peripheral vascular disease that precluded retrograde PBAV.
Patients were categorized into those aged >80 and those aged ≤80 years and were compared by clinical, procedural, and hemodynamic characteristics and outcomes. Patient data were obtained through hospital records and the catheterization laboratory database. Patients were designated as having hypertension or dyslipidemia if they were given that diagnosis by their physicians or were taking blood pressure or lipid-lowering agents. Coronary artery disease was defined as having a documented diagnosis or history of coronary revascularization.
Pre- and postprocedural echocardiography was routinely performed in all study patients. Standard right- and left-sided cardiac catheterizations were performed. Cardiac output was measured using the thermodilation technique. In the presence of left-to-right shunting or significant tricuspid regurgitation, cardiac output was measured according to the Fick method using assumed oxygen consumption. Simultaneous transaortic valve gradients were measured routinely in all patients using a 6Fr double-lumen pigtail catheter. AVA was calculated according to the Gorlin formula. Coronary angiography was performed before PBAV. When indicated, percutaneous coronary intervention was performed before PBAV.
PBAV was performed retrograde from the common femoral artery (12Fr) using a standard technique. The balloon catheter (Z-Med, NuMed, Inc., Nicholville, New York; or Maxi LD, Cordis Corporation, Bridgewater, New Jersey) diameter was selected for a diameter less than the aortic annular diameter determined by echocardiography. All patients received intravenous heparin to achieve an activated clotting time ≥250 seconds before PBAV.
The balloon was manually inflated for 5 to 10 seconds. Rapid ventricular pacing was not routinely used at our center and was used only at the discretion of the operator. For patients receiving rapid pacing, a rate of 180 to 200 beats/min was used produce a decrease in systolic blood pressure of ≤50 mm Hg during balloon inflation. Right- and left-sided cardiac catheterization was performed at the end of the procedure in every patient. Post-PBAV AVA, aortic valve gradient, and cardiac output were compared with those obtained at baseline.
The primary end point of the study was the composite of in-hospital major adverse cardiovascular events (MACEs), composed of hospital death, myocardial infarction, stroke, hemorrhagic tamponade, and cardiac arrest requiring cardiopulmonary resuscitation. Secondary end points included the individual components of the primary end point in addition to intraprocedural hypotension requiring intravenous vasopressors, intraprocedural endotracheal intubation, and the composite of all intraprocedural adverse events. Myocardial infarction was defined as an increase of creatinine kinase ≥3 times the upper limit of normal measured <24 hours after the procedure or any new pathologic Q wave on electrocardiography. Acute kidney injury was defined as an increase in creatinine ≥0.5 mg/dl over baseline at 48 hours.
All patients were routinely assessed for vascular complications, including retroperitoneal bleeding, pseudoaneurysm, arterial-to-venous fistula, arterial dissection, and access-site major bleeding (hemoglobin decrease ≥5 g/dl or any access bleeding requiring transfusion).
Categorical variables were compared using chi-square analysis or Fisher’s exact test for nonparametric data. Continuous variables were compared using Student’s t test. Confounding and effect modification were evaluated using the Mantel-Haenszel method. Multivariate logistic regression was performed for the composite of in-hospital and intraprocedural complications. The clinical and hemodynamic factors were evaluated for their relative risk for association with the composite outcome. Factors that were associated with the composite outcome in the unadjusted comparisons to a significance level of p <0.10 were included in the adjusted model. Age >80 years was forced back into the model regardless of significance. Additionally, variables that differed between older patients and younger patients were tested to see if they significantly affected the relative risk for age in the model. The odds ratios derived from the logistic regression are reported as relative risks given the small number of events. Model fit was evaluated using likelihood ratio testing. Analyses were performed using Intercooled Stata version 9.2 (StataCorp LP, College Station, Texas).
Results
From December 22, 2004, to December 15, 2008, 127 patients underwent retrograde PBAV at Massachusetts General Hospital. Of these, 81 patients were aged >80 years at the time of the procedure, and 46 were aged ≤80 years. The oldest patient was 98 years old, and the youngest was 52 years old. There were 17 patients aged 90 to 98 years. Sixteen patients presenting with cardiogenic shock, requiring intra-aortic balloon pump support, intravenous vasopressors, or mechanical ventilatory support were excluded. In total, 111 patients were studied, 73 of whom were aged >80 years.
Baseline demographic and clinical characteristics are listed in Table 1 . Only 25 patients underwent PBAV for critical aortic stenosis as a bridge to surgery, while 86 patients had significant symptoms due to aortic stenosis and were not considered surgical candidates. The mean age for the study group was 82 ± 8.1 years. Ninety percent of patients presented for the procedure with New York Heart Association (NYHA) class III or IV symptoms, with no difference between the 2 groups. Patients aged >80 years had slightly higher body mass indexes, a lower rate of diabetes, and lower average renal glomerular filtration rates than those aged ≤80 years. The additive and logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE) values were higher in patients aged >80 years, given their older ages and higher creatinine levels ( Table 1 ).
| Characteristic | Age >80 Years (n = 73) | Age ≤80 Years (n = 38) | p Value |
|---|---|---|---|
| Baseline clinical characteristics | |||
| Men | 32 (43.8%) | 17 (44.7%) | 0.93 |
| Age (years) | 86.6 ± 4.4 | 74.1 ± 6.9 | <0.01 |
| Range | 81–98 | ||
| Body mass index (kg/m 2 ) | 26.7 ± 7.3 | 24.2 ± 5.6 | 0.03 |
| Race | 0.61 | ||
| White | 64 (94.1%) | 32 (91.4%) | |
| Black | 1 (1.5%) | 0 (0.0%) | |
| Asian | 1 (1.5%) | 1 (2.9%) | |
| Hispanic | 0 (0.0%) | 1 (2.9%) | |
| Other | 2 (2.9%) | 1 (2.9%) | |
| Diabetes mellitus | 19 (26.0%) | 17 (44.7%) | 0.05 |
| Diabetes therapy | 0.30 | ||
| Oral medications | 4 (22.2%) | 6 (5.9%) | |
| Insulin | 7 (38.9%) | 4 (35.3%) | |
| Diet | 7 (38.9%) | 3 (58.8%) | |
| Hypertension | 57 (78.1%) | 30 (78.9%) | 0.92 |
| Dyslipidemia | 58 (80.6%) | 32 (84.2%) | 0.64 |
| Tobacco use | 0.26 | ||
| Any previous | 32 (44.4%) | 23 (60.5%) | |
| Current | 4 (5.6%) | 1 (2.6%) | |
| Coronary artery disease | 40 (54.8%) | 23 (60.5%) | 0.56 |
| 3-vessel coronary artery disease | 11 (15.1%) | 5 (13.6%) | 0.79 |
| Previous myocardial infarction | 19 (26.0%) | 11 (28.9%) | 0.46 |
| Previous percutaneous coronary intervention | 13 (17.8%) | 6 (15.8%) | 0.79 |
| Previous coronary artery bypass grafting | 15 (20.6%) | 9 (23.7%) | 0.70 |
| Previous stroke | 32 (43.8%) | 13 (34.2%) | 0.33 |
| Family history of coronary disease | 2 (2.7%) | 3 (7.9%) | 0.21 |
| Previous heart failure | 55 (75.3%) | 26 (68.4%) | 0.44 |
| Chronic obstructive pulmonary disease | 25 (34.3%) | 13 (34.2%) | 0.99 |
| Peripheral vascular disease | 21 (28.8%) | 8 (21.1%) | 0.38 |
| Glomerular filtration rate <60 ml/min/1.73 m 2 (Modification of Diet in Renal Disease equation) | 52 (72.2%) | 20 (52.6%) | 0.04 |
| Previous renal failure | 6 (8.2%) | 5 (13.2%) | 0.41 |
| Currently on dialysis | 1 (20.0%) | 3 (60.0%) | 0.19 |
| Atrial fibrillation | 26 (35.6%) | 11 (28.9%) | 0.48 |
| Clinical presentation | |||
| Heart failure | 55 (75.3%) | 30 (78.9%) | 0.67 |
| NYHA class | 3.2 ± 0.7 | 3.4 ± 0.83 | 0.25 |
| I | 3 (2.7%) | 2 (5.3%) | 0.49 |
| II | 5 (6.9%) | 2 (5.3%) | 0.74 |
| III | 39 (53.4%) | 12 (31.6%) | 0.03 |
| IV | 27 (36.9%) | 22 (57.9%) | 0.04 |
| Any angina pectoris at presentation | 10 (13.7%) | 7 (18.4%) | 0.51 |
| Acute coronary syndrome | 3 (4.7%) | 7 (14.9%) | 0.06 |
| Scheduling type | 0.49 | ||
| Elective | 18 (24.7%) | 6 (15.8%) | |
| Urgent | 52 (71.2%) | 31 (81.6%) | |
| Emergent | 3 (4.1%) | 1 (2.6%) | |
| EuroSCORE (additive) | 12.5 ± 2.7 | 10.7 ± 3.6 | <0.01 |
| EuroSCORE (logistic) | 36.6 ± 0.2 | 27.9 ± 0.2 | 0.03 |
Patients aged >80 years had lower ejection fractions than the younger patients, but mean wall thickness and mean left ventricular cavity were similar. Cardiac output and pre-PBAV AVA were significantly lower in patients aged >80 years old ( Table 2 ).
| Cardiac Parameter | Age >80 Years (n = 73) | Age ≤80 Years (n = 38) | p Value |
|---|---|---|---|
| Echocardiographic | |||
| Left ventricular hypertrophy | 42 (84.0%) | 34 (80.9%) | 0.70 |
| Septal thickness (mm) | 12.9 ± 2.6 | 12.9 ± 2.2 | 0.88 |
| Left ventricular end-diastolic diameter (mm) | 45.2 ± 8.5 | 46.2 ± 8.9 | 0.52 |
| Ejection fraction (%) | 56.1 ± 19.1 | 43.5 ± 18.6 | <0.01 |
| Pre-PBAV pressures (mm Hg) | |||
| Aortic systolic | 123.7 ± 25.6 | 122.9 ± 23.8 | 0.87 |
| Aortic mean | 80.9 ± 14.8 | 81.6 ± 15.5 | 0.82 |
| Left ventricular systolic | 174.6 ± 28.8 | 174.6 ± 29.3 | 0.99 |
| Left ventricular end-diastolic | 19.4 ± 8.0 | 21.7 ± 10.6 | 0.19 |
| Pulmonary artery mean | 30.1 ± 10.94 | 33.1 ± 12.9 | 0.19 |
| Pulmonary capillary wedge | 18.3 ± 8.0 | 20.2 ± 9.9 | 0.28 |
| Pre- and post-PBAV assessment | |||
| Cardiac output (thermodilution, pre) (L/min) | 3.8 ± 1.2 | 4.7 ± 1.4 | <0.01 |
| Cardiac output (thermodilution, post) (L/min) | 4.5 ± 1.3 | 3.8 ± 1.2 | <0.01 |
| Mean gradient (pre) (mm Hg) | 47.2 ± 15.2 | 46.5 ± 17.3 | 0.83 |
| Mean gradient (post) (mm Hg) | 28.5 ± 11.3 | 28.9 ± 12.4 | 0.88 |
| Δ Mean gradient (mm Hg) | −18.8 ± 10.8 | −17.7 ± 14.1 | 0.63 |
| Δ Mean gradient (% change) | −39.1% ± 17.3 | −34.8% ± 22.9 | 0.27 |
| AVA (pre) (cm 2 ) | 0.59 ± 0.18 | 0.73 ± 0.27 | <0.01 |
| AVA (post) (cm 2 ) | 0.89 ± 0.31 | 1.02 ± 0.37 | 0.05 |
| Δ AVA (cm 2 ) | 0.29 ± 0.25 | 0.30 ± 0.25 | 0.98 |
| Δ AVA (% change) | 55.5% ± 50.0 | 45.2% ± 38.8 | 0.28 |
| Increase in AVA by 25% | 56 (76.7%) | 26 (68.4%) | 0.35 |
The procedural characteristics were the same between the 2 groups ( Table 3 ). AVAs increased and mean gradients decreased significantly after PBAV in the 2 groups (pre- vs post-PBAV for each group, p <0.01). In light of the lower baseline AVA, the average post-PBAV AVA was lower in patients aged >80 years, but the absolute and percentage reductions in mean gradient and the absolute and percentage increases in AVA were not statistically different ( Table 3 ).
| Variable | Age >80 Years (n = 73) | Age ≤80 Years (n = 38) | p Value |
|---|---|---|---|
| Combined coronary and valvular procedure | 7 (9.7%) | 1 (2.6%) | 0.17 |
| Number of inflations | 2.4 ± 0.7 | 2.6 ± 1.2 | 0.18 |
| >2 inflations | 27 (38.0%) | 16 (44.4%) | 0.52 |
| Maximum balloon size | 21.6 ± 1.7 | 22.0 ± 1.6 | 0.23 |
| Suture-mediated arteriotomy closure | 26 (35.6%) | 12 (31.6%) | 0.67 |
| Rapid ventricular pacing | 43 (58.9%) | 21 (55.3%) | 0.71 |
Among patients aged >80 years, 10 required cardiopulmonary resuscitation, 7 needed intubation, and 1 died during the procedure. None of the patients aged ≤80 years of age experienced any of these in-hospital adverse events. This difference was statistically significant. All the patients requiring cardiopulmonary resuscitation did so because of pulseless electrical activity or asystole occurring after PBAV. One 83-year-old patient and 1 65-year-old patient developed bleeding into the pericardial space requiring pericardiocentesis for tamponade. The need for vasopressors was similar ( Table 4 ).
