Author + information
- Marco Roffi, MD*,
- Peter Wenaweser, MD†,
- Stephan Windecker, MD†,
- Haresh Mehta, MD†,
- Franz R. Eberli, MD*,
- Christian Seiler, MD, FACC†,
- Martin Fleisch, MD†,
- Ali Garachemani, MD†,
- Giovanni B. Pedrazzini, MD‡,
- Otto M. Hess, MD, FACC† and
- Bernhard Meier, MD, FACC†,* ()
- ↵*Reprint requests and correspondence:
Dr. Bernhard Meier, Swiss Cardiovascular Center Bern, University Hospital, 3010 Bern, Switzerland.
Objectives In this study, we sought to assess safety of symptom-limited exercise stress tests the day after coronary stenting.
Background Isolated cases of coronary stent thrombosis have been linked to early exercise stress testing, thereby questioning the safety of unrestricted physical activity after the coronary procedure.
Methods At a single center, 1,000 patients were randomized to a symptom-limited stress test the day after coronary stenting or no stress test. The antiplatelet regimen consisted of acetylsalicylic acid and postprocedural ticlopidine or clopidogrel. The primary end point of the study was the incidence of clinical stent thrombosis at 14 days. The secondary end point was the occurrence of access site complications.
Results Clinical stent thrombosis occurred in five patients (1%) undergoing stress test and in five patients (1%) randomized to no stress test (p = 1.0). Access site complications were detected in 4% and 5.2% of cases, respectively (p = 0.37).
Conclusions Symptom-limited exercise stress testing the day after coronary stenting does not increase the risk of clinical stent thrombosis or access site complications. Further investigations on safety of early vigorous exercise after coronary stenting in a non-supervised setting are warranted.
Isolated cases of acute ischemic events linked to exercise stress testing have been described after coronary stenting (1–4). An exercise-induced systemic prothrombotic state in conjunction with the intrinsic thrombogenicity of the stent have been postulated as trigger mechanisms. These reports have led to uncertainty among patients and physicians regarding resumption of normal physical activity after coronary stenting. With modern antiplatelet therapy, stent thrombosis is a rare event, albeit associated with high mortality (5). In order to address safety of unrestricted physical activity after coronary stenting, we randomized patients to stress test or no stress test the day after the coronary procedure.
At the Swiss Cardiovascular Center in Bern, exercise stress testing after percutaneous coronary revascularization is frequently performed. Between September 1997 and June 2001, a total of 1,000 patients (788 males, 212 females) were randomized to a symptom-limited exercise stress test the day after coronary stenting or no stress test. Clinical exclusion criteria were recent or ongoing myocardial infarction (MI) (i.e., within one week), symptomatic heart failure, severe valvular heart or pulmonary disease, or inability to exercise. Periprocedural exclusion criteria included symptomatic main vessel or side branch occlusions, access site-related complications, persistent chest pain or new ST-segment deviation, or final coronary flow less than TIMI (Thrombolysis In Myocardial Infarction) grade 3. Access site complications were defined as one of the following: pseudoaneurysm, arteriovenous fistula, deep vein thrombosis or pulmonary embolism, need for surgical repair of the access site, need for blood transfusion, or the presence of a hematoma requiring prolonged or repeat compressive measures to achieve hemostasis, prolonged hospital stay, or surgical revision. Coronary disease was considered significant if the stenoses were visually estimated at ≥50%. Suboptimal procedural results included residual stenosis >50%, non–flow-limiting dissections, and side-branch compromise in the absence of flow impairment.
All the patients had femoral access unless contraindicated. Percutaneous coronary intervention was performed using 6F guiding catheters and 5,000 to 10,000 U of unfractionated heparin, without monitoring of the periprocedural activated clotting time. Acetylsalicylic acid 500 mg intravenously was administered before the procedure in patients who were not undergoing long-term therapy. Oral ticlopidine (250 mg twice a day for two weeks) or clopidogrel (300 mg followed by 75 mg a day for two to four weeks) was prescribed after stenting. The use of adjunctive platelet glycoprotein (GP) IIb/IIIa receptor inhibitors was left to the discretion of the operator. Sheaths were pulled 4 h or more after the intervention without previous coagulation check, and manual compression was applied. Patients were usually ambulated the next day.
Randomization occurred at the end of the coronary procedure, in the absence of exclusion criteria. Symptom-limited treadmill stress testing was performed the following day, before discharge, using a modified Bruce protocol. A physician obtained the follow-up by phone interview at 14 days. The primary end point of the study was clinical stent thrombosis at 14 days, defined by one of the following: angiographically documented stent thrombosis, sudden death, any MI, or urgent target vessel revascularization. The secondary end points consisted of access site complications, as previously described.
Assuming a stent thrombosis rate at 14 days of 1% in the control group (5), and a clinical margin of inferiority of an additional 1.65%, a sample size of 450 patients per arm was required to demonstrate non-inferiority of the stress test arm, with a power of 80% and a 0.05 level of significance. The total sample size of 500 patients per arm accounts for patients missing the stress test. Continuous variables are expressed as mean ± standard deviation, and differences between the two groups of patients were tested with the Student ttest. Categorical data are presented as percentages, and comparisons among the groups were made with the chi-squared test. A value of p < 0.05 was considered statistically significant. Results were analyzed on an intention-to-treat basis.
A total of 498 patients were randomized to stress test group, and 502 patients to a no stress test group. The two groups were well matched for baseline and procedural characteristics (Table 1). A greater proportion of patients with unstable angina (UA) in the non–stress-test group was the only imbalance detected. Among patients randomized to stress testing, 50 patients (10%) did not undergo the test but were followed up as part of the stress-test group. The only coronary event detected among them was an acute stent thrombosis that occurred 2 h after the intervention, requiring emergent percutaneous intervention. Other reasons not to undergo a stress test were related (18 cases) or unrelated (16 cases) to the coronary procedure. In an additional 16 cases, no reason was given for skipping the test (Fig. 1). Among patients undergoing a stress test, the mean attained working capacity was 9.6 ± 2.5 MET (metabolic equivalent; corresponding to a respiratory oxygen uptake of 3.5 ml/min/kg of body weight), the heart rate 130 ± 21 beats/min, and the rate-pressure product (heart rate × systolic blood pressure) 23,500 ± 6,000. Adverse events during stress test included a syncope at maximal workload in the absence of rhythm disturbances, and a transient ST-segment elevation in a patient with known coronary spasm. No repeat angiograms were performed in these two patients, and they remained free of cardiac events during follow-up.
Follow-up was completed in 99.5% of cases. The primary end point, clinical stent thrombosis within 14 days, occurred in five cases each of patients undergoing stress test (1%) or randomized to no stress test (1%) (p = 1) (Table 2). Events occurred on day 0 (2 h after intervention), day 1, day 3, day 6, and day 9 in the stress-test group; and on day 0 (1 and 12 h after intervention), day 4, and day 6 in the no-stress-test group. In one patient randomized to the no stress test group, stent thrombosis could not be timed. It was documented incidentally during an elective angiogram performed on day 9 for a staged revascularization. One patient had anterior ST-segment elevation and left anterior descending coronary artery stent thrombosis 30 min after a negative stress test. The indication for stenting had been UA. He was treated with acetylsalicylic acid and ticlopidine. He had not received GP IIb/IIIa receptor inhibitors during the procedure. Despite immediate successful recanalization, the patient developed anterior MI. Vascular access site complications were comparable in both groups (4.0% with and 5.8% without stress test, p = 0.37) (Table 2).
No difference was observed in the clinical stent occlusion rate in patients receiving ticlopidine (1.5%) or clopidogrel (0.8%) (p = 0.34). Patients who received platelet GP IIb/IIIa receptor inhibitors during the coronary procedure had a stent thrombosis rate of 1.4%. The occlusion rate of those who did not was 0.9% (p = 0.60). Among 254 patients undergoing coronary stenting for UA, eight patients (3.1%) had stent thrombosis, compared with two patients (0.3%) among 741 stable patients (p < 0.01). A total of 89 patients had suboptimal periprocedural result and an incidence of stent thrombosis of 3.4%, compared with 0.8% among 906 patients with good angiographic results (p = 0.02). When compared with patients who did not have an event, those with stent thrombosis presented more frequently with UA, received more stents, had a significantly greater total stent length, and more frequently had a suboptimal angiographic result after the procedure (Table 3).
Addressing the safety of early physical exertion after coronary stenting is crucial for medical and legal reasons (e.g., counseling of patients with physically challenging activities). A total of four cases of stent occlusion linked to exercise testing have been reported in the literature (1–4). Among them, one patient was receiving coumadin (2); one was receiving acetylsalicylic acid, dipyridamole, and heparin while oral anticoagulation with acenocoumarol was started (1); one was receiving acetylsalicylic acid and ticlopidine (4); and in one case the antithrombotic regimen was not described (3). The latency period between stenting and the exercise stress test varied between two days and three months, and the occurrence of stent occlusion ranged from 15 min to 2 h after the stress test. These observations may lead to questions about the safety of heavy physical activity for an extended period of time after coronary stenting.
An exercise-induced systemic prothrombotic state in conjunction with the intrinsic thrombogenicity of the stent have been postulated as trigger mechanisms of stent thrombosis. Despite the paucity of evidence that exercise may be related to stent thrombosis, the belief that early stress tests after coronary stenting may be hazardous is not uncommon among physicians. According to the American College of Cardiology/American Heart Association guidelines, the safety of exercise stress testing after discharge in patients who have undergone percutaneous coronary revascularization has not been established. Notwithstanding, the procedure is considered a class IIa indication (i.e., weight of evidence/opinion is in favor of usefulness/efficacy) for activity counseling or exercise training as part of cardiac rehabilitation (6).
This randomized trial shows that an early symptom-limited exercise stress test does not increase the incidence of clinical stent thrombosis or access site complications. The clinical stent thrombosis rate observed was comparable with previous reports (5,7,8). As in previous observations, patients with UA and those with a suboptimal procedural result had an increased incidence of stent occlusion. In a pooled analysis of stent trials including more than 6,000 patients, patients with stent thrombosis had presented more frequently with UA (5). With respect to suboptimal procedural results, a recent retrospective analysis demonstrated that the vast majority of patients with stent thrombosis had at least one abnormal feature detected by intravascular ultrasound at the time of stent deployment (9). When compared with patients who did not have an event, those who had clinical stent thrombosis in our study received more stents per patient and had significantly longer stented coronary segments. This is in accordance with the existing literature showing that total stent length is an independent predictor of stent thrombosis (5).
Our data expand the findings of a nonrandomized series reporting no complication among 261 patients undergoing exercise stress test within 60 days of stent deployment (10). The one-stent thrombosis after the stress test observed in our study could arguably have been avoided by not performing the test. However, it is also possible that it would have occurred spontaneously or at the first strenuous physical activity. Outside the hospital setting, the medical management of this complication would have been more difficult.
Limitations of the study include the fact that roughly half of the 2,086 patients undergoing coronary stent implantation at our institution during the trial period were not enrolled, mainly because only selected physicians participated in the trial. In addition, events were recorded only up to 14 days. Yet, it is known that the vast majority of stent thromboses occur within this period of time. Moreover, patients or referring physicians may have under-reported access site-related complications occurring after hospital discharge. Finally, the study protocol did not mandate tracking of femoral access closure devices, because at the beginning of the trial they were not used at our institution. According to the practice during the time period of the study, all procedures were performed using 6F guiding catheters, and only a small minority of patients may have received closure devices. Therefore, our findings on access site complications may not apply to practices based on larger sheath use or extensive closure device utilization.
In our randomized trial enrolling 1,000 patients, symptom-limited exercise stress testing the first day after coronary stenting did not increase the risk for stent thrombosis or access site complications. Further investigations on safety of early vigorous exercise after coronary stenting in a non-supervised setting are warranted.
- myocardial infarction
- unstable angina
- American College of Cardiology Foundation
- Cutlip D.E.,
- Baim D.S.,
- Ho K.K.,
- et al.
- Gibbons R.J.,
- Balady G.J.,
- Beasley J.W.,
- et al.
- Albiero R.,
- Hall P.,
- Itoh A.,
- et al.
- Moussa I.,
- Oetgen M.,
- Roubin G.,
- et al.
- Uren N.G.,
- Schwarzacher S.P.,
- Metz J.A.,
- et al.