The Safety of Treadmill Exercise Stress Testing in Patients with Abdominal Aortic Aneurysms

Methods

Patients

By using diagnoses at autopsy and well-established diagnostic, medical, and procedural indexing systems of all hospital and outpatient care, we obtained a list of all patients seen at the Mayo Clinic in Rochester, Minnesota, between 1982 and 1995 who had abdominal aortic aneurysms and underwent treadmill exercise stress testing by exercise electrocardiography, exercise myocardial perfusion imaging (thallium or sestamibi), or exercise echocardiography [13]. With this method, we identified 556 patients who had had 922 treadmill exercise tests. Inclusion criteria were 1) an abdominal aortic aneurysm at least 4.0 cm in diameter and 2) a first exercise stress test done before aneurysmectomy but after a diagnosis of abdominal aortic aneurysm or an exercise stress test done before but within 3 months of a new diagnosis of abdominal aortic aneurysm.

Of the 922 stress tests identified, 18.1% were excluded because the aneurysm was less than 4 cm in diameter, 8.6% were excluded because the exercise test was done more than 3 months before diagnosis of the aneurysm, 41.0% were excluded because the test was done after aneurysmectomy, and 3.9% were excluded because it was not the first treadmill exercise test done after diagnosis of the aneurysm. Initial exercise stress tests of 262 patients were included in the study.

Measurements

Medical records were reviewed for comorbid conditions, medications, aneurysm size, and stress test results. For standardization of exercise protocols, exercise duration was converted to metabolic equivalents (METs) [14]. Adverse events that could have been related to the exercise stress test were defined as symptoms that suggested increasing aneurysm size, such as abdominal or back pain, rupture or leaking of the aneurysm, myocardial infarction, or death within 72 hours after stress testing. All patients had follow-up evaluation within 72 hours after stress testing. All testing and monitoring were done by previously described standard protocols [15, 16]. Stress test reports were reviewed for other complications, including hypotension and arrhythmia.

Statistical Analysis

Data are presented as the mean ±SD and as percentages. Exact binomial CIs were calculated. Aneurysms more than 6 cm in diameter and aneurysms less than 6 cm in diameter were evaluated as a priori subgroups because of increased risk for rupture of larger aneurysms and because of the single case report [12].

Results

The mean age of the 262 patients was 69 ± 7 years. Before 55% of the exercise tests were done, presence of coronary disease was documented by a previous abnormal exercise test result, previous coronary angiography that showed stenosis greater than 50% in at least one vessel, previous percutaneous transluminal coronary angioplasty, previous myocardial infarction, or previous coronary artery bypass graft surgery. In another 22 patients (8%), coronary artery disease was clinically suspected without documentation. Baseline patient characteristics are presented in Table 1.

The mean aneurysm diameter was 5.5 ± 1.1 cm, and the median diameter was 5.3 cm (Figure 1). Aneurysms were evaluated by ultrasonography in 57% of patients, computed tomography in 37%, angiography in 5%, and magnetic resonance imaging in 1%. In addition, 17 patients (6%) had a thoracic aortic aneurysm (6.0 ± 1.9 cm), and 1 patient had a 3-cm aneurysm in the hypogastric artery.

View larger version:

• In this window
• In a new window

Figure 1. Numbers above the bars are the number of patients in each group. Distribution of the cross-sectional diameter of abdominal aortic aneurysms at the time of exercise stress testing.

Patients underwent symptom-limited treadmill exercise stress tests; 14% had exercise electrocardiography, 65% had thallium exercise myocardial perfusion imaging, 9% had sestamibi exercise myocardial perfusion imaging, and 12% had exercise echocardiography. The mean time from the imaging study to the exercise test was 19 ± 136 days (median, 3 days). With exercise, the patients achieved a double product of 22 509 ± 5973 and 7.4 ± 2.1 METs; 90% of patients achieved more than 5 METs. In 125 tests (48%), a heart rate greater than 85% of the age-predicted maximum was achieved. Forty percent of the patients who did not exceed the 85% age-predicted maximum heart rate were receiving β-blocker therapy.

Systolic blood pressure was 137 ± 21 mm Hg at rest and 174 ± 28 mm Hg with exercise. Diastolic blood pressure was 82 ± 11 mm Hg at rest and 84 ± 12 mm Hg with exercise. Heart rate increased from 75 ± 14 beats/min at baseline to 128 ± 22 beats/min with exercise.

The test was limited by symptoms that included fatigue, dyspnea, or chest pain in 229 patients (87%); claudication without other symptoms in 8 patients (3%); and claudication with other symptoms in 14 patients (5%). Six patients (2%) developed decreased systolic blood pressure ( −20± 15 mm Hg), 5 patients developed systolic hypertension (blood pressure at rest of 148 ± 18 mm Hg increased to 242 ± 19 mm Hg with exercise), 6 patients (2%) had orthopedic limitations, 5 patients (2%) had the test terminated by their physicians, and 5 patients (2%) had electrocardiographic changes that contributed to the decision to terminate the test.

No deaths or myocardial infarctions occurred during the 72 hours after stress testing. One patient with a 6.1-cm abdominal aortic aneurysm developed abdominal pain 12 hours after stress testing. Repeated computed tomography showed a contained rupture. No other patients reported abdominal or back pain after exercise stress testing. During the exercise stress test, 3 patients developed nonsustained ventricular tachycardia and 1 patient developed supraventricular tachycardia. No other negative outcomes were documented. Thus, for the overall population, the risk for rupture was 1 in 262 (0.4% [CI, 0.0% to 2.1%]), and no aneurysms ruptured in the 165 patients with aneurysms that were less than 6 cm in diameter (CI, 0.0% to 2.2%). For patients with aneurysms more than 6 cm in diameter, the risk for rupture was 1 in 97 (1% [CI, 0.0% to 5.6%]).

Discussion

Our study shows that in patients with abdominal aortic aneurysms, the incidence of acute adverse events caused by exercise stress testing is low. No significant complications, including death and aneurysm rupture, occurred among 165 patients with aneurysms less than 6 cm in diameter (most of our study sample). Thus, in patients with aneurysms less than 6 cm who are able to exercise, treadmill exercise stress testing seems safe and can be used if clinically indicated. In patients with aneurysms larger than 6 cm, the event rate was 1% (CI, 0.0% to 5.6%).

Cardiac stress testing is often recommended for preoperative evaluation of patients with abdominal aortic aneurysms. Both the high prevalence of coronary artery disease and the high risk for perioperative cardiac complications with this type of surgery warrant this recommendation [1, 2, 5, 7]. As expected, our patients with abdominal aortic aneurysms had a high incidence (55%) of coronary artery disease even before exercise stress testing.

Treadmill exercise stress testing is a time-honored method of cardiac evaluation and is the preferred method of cardiac stress testing. This test is readily available, efficient, and less expensive than other forms of cardiac testing. In addition, the patient's exercise capacity has important prognostic implications. In patients undergoing vascular surgery, the ability to achieve 85% of the predicted maximum heart rate is a significant prognostic factor for reduced postoperative cardiac complications even if the electrocardiographic stress test result is positive [17]. In addition, the American College of Cardiology and American Heart Association Task Force Report indicates that assessment of a patient's exercise capacity in METs significantly predicts subsequent cardiac events [7]. In patients who are able to exercise, exercise stress testing has an accuracy for detecting coronary artery disease similar to that of pharmacologic stress testing [18-20]. Therefore, for routine clinical practice, exercise stress testing is the first diagnostic approach and pharmacologic stress testing is reserved for patients who are unable to exercise adequately.

Our study suggests that exercise stress testing in patients with abdominal aortic aneurysms is associated with a low incidence (0.4%) of acute adverse events. However, this retrospective study is limited by selection bias. Many patients who have aneurysms-especially large aneurysms-were probably not referred for exercise stress testing. Despite this limitation, these data suggest that, especially for persons with aneurysms less than 6 cm in diameter, exercise stress testing is infrequently associated with acute complications. The safety of exercise stress testing in patients with large aneurysms (≥ 6 cm) is less certain because the event rate was 1% in this subgroup. After pharmacologic or exercise stress testing, patients who develop symptoms that could be related to aneurysm rupture should be promptly evaluated, although the risk for such events is low.