Five-Year Outcome in Patients With Isolated Proximal Left Anterior Descending Coronary Artery Stenosis Treated by Angioplasty or Left Internal Mammary Artery Grafting
A Prospective Trial
Background—Percutaneous transluminal coronary angioplasty (PTCA) and coronary artery bypass surgery (CABG) improve the clinical status of patients with isolated proximal left anterior descending coronary artery stenosis. At 2 years, only additional revascularization was more frequently required after PTCA.
Methods and Results—We monitored 134 patients randomized to PTCA (n=68) or CABG (n=66) for ≤5 years. End points were death, myocardial infarction, need for additional revascularization, clinical status, and medical treatment. At 5 years, 6 patients (9%) had died in the PTCA group versus 2 (3%) in the CABG group (P=0.12). One patient in each group died of a cardiac cause. Myocardial infarction was more frequent after PTCA (15% versus 4%; P=0.0001), but Q-wave infarction was not (6% in the PTCA group versus 3% in the CABG group; P=0.8). Additional revascularization was required in 38% of patients in the PTCA group versus 9% in the CABG group (P=0.0001). Functional status was comparable, with 6% of patients after PTCA and 3% after CABG in functional class III or IV. Finally, after PTCA or CABG, 62% and 91% of patients, respectively, were free of events (P=0.0001).
Conclusions—The 5-year prognosis of patients with isolated proximal left anterior descending coronary artery stenosis is good. Both PTCA and CABG improve clinical status, but revascularization was needed more frequently after PTCA. There is an excess incidence of non–Q-wave myocardial infarction in the PTCA group that does not affect the vital or symptomatic outcome.
Many trials comparing PTCA and CABG are still in progress.1 2 3 4 5 Early results have been reported in a meta-analysis,6 but long-term results are still awaited for the majority of these trials; the long-term outcome has been published only for the Randomised Intervention Treatment of Angina (RITA) trial.7 We conducted a prospective randomized trial in patients with isolated proximal left anterior descending coronary artery (LAD) stenosis. The intermediate results of our study have been published previously.8 At 2 years, comparison of the 2 strategies (PTCA versus CABG) did not show significant differences for the major end points, such as death and incidence of myocardial infarction. Significant and comparable symptom improvement was provided by both techniques. Only the need for additional revascularization was more frequent with PTCA. Because many of the complications and events after CABG appear only late after surgery, the long-term follow-up of this cohort of patients is more appropriate to elucidate the place of these 2 strategies in the management of patients with isolated stenosis of the proximal LAD. We report herein 5-year follow-up data.
We will briefly describe the inclusion criteria; the study protocol, randomization design, and percutaneous and surgical techniques have been described previously.8
Patients with isolated proximal LAD stenosis and an ejection fraction >0.50 were eligible for the trial provided they had documented clinical or silent ischemia; we excluded patients with refractory unstable angina or previous anterior Q-wave myocardial infarction. The study protocol was approved by the hospital ethics committee. A consensus on the feasibility of PTCA or CABG had to be obtained from 2 cardiologists and 2 cardiac surgeons before randomization.
Randomization was performed according to Zelen’s design.9 The standard treatment was CABG. Patients allocated this treatment were not informed of the PTCA option. Patients in the PTCA group were fully informed about the trial and were requested to provide informed consent. However, the patients in the CABG arm who requested PTCA because they were aware of this technique underwent PTCA after having been fully informed of the trial. CABG had to be performed within 30 days after randomization and PTCA within 10 days.
PTCA was performed by the right or left femoral route with an 8F guiding catheter. Adjunct medication consisted of 500 mg of aspirin and 15 000 U of heparin. Standard angioplasty techniques with over-the-wire or monorail balloons were used. Inflation was performed as recommended by the manufacturer. PTCA was performed until an acceptable result (<50% residual stenosis) was obtained. The management of threatened or acute closure was left to the discretion of the operator, with the knowledge that stents were available (Wallstent; Schneider). After the procedure, patients were placed in an intensive care unit. Creatine kinase (CK) activity and ECG monitoring were performed after the procedure, at 6 hours, and the following day.
Surgery was performed by median sternotomy. Cardiopulmonary bypass was established by means of a single canula. St Thomas II cardioplegia and cooled Ringer’s solution were used to induce cardiac arrest. The left internal mammary artery (LIMA) pedicle was transferred into the surgical field through a lateral incision in the pericardium. The LAD was incised for 0.5 to 1.0 cm. Vessel anastomosis was assured by means of 2 double-armed 7.0 polypropylene wires sutured in continuous stitches. The pedicle was tacked to the heart surface with 2 interrupted 6.0 polypropylene sutures to prevent twisting of the pedicle and tension of the suture line. Visual assessment was assured by the use of 4 to 5× magnifying glasses.
Data were collected by direct contact with the patient and his or her home physician. Repeat cardiac investigations and angiography were not mandatory at 6 months but were highly encouraged in case of recurrence of typical or atypical symptoms. Repeat PTCA was recommended for patients in the PTCA group who had restenosis, but CABG was presented as a possible alternative. For patients allocated CABG who needed additional revascularization, both approaches were presented, but the final choice was left to the patient and his or her physician.
The primary end points were cardiac-related death, myocardial infarction, and the need for additional revascularization. Standard clinical, biochemical, and ECG criteria were used to define myocardial infarction. A CK rise at twice the normal value without new Q wave on the ECG was considered a non–Q-wave infarction. Secondary end points were angina functional status and antianginal drug regimen. An interim analysis was done at 2 years and the final analysis at 5 years after inclusion in the trial. Clinical status was assessed by means of the Canadian Cardiovascular Society classification, and the need for antianginal treatment was documented. Reinterventions were classified as (1) target-vessel revascularization (LAD) or (2) revascularization in a vessel other than the LAD.
The sample size to achieve 95% statistical power (α=0.01, β=0.05) was calculated based on previous reports of the effects of PTCA and CABG on patients with single-artery coronary disease. We assumed, based on the composite primary end point, that 35% of patients treated by PTCA would reach an end point compared with only 5% for the CABG group; we calculated that a total of 132 patients were required for the study.
All analyses were made according to the intention-to-treat principle. All tests of significance were 2-tailed (Fisher’s and χ2 test for discrete variables and Mann-Whitney, Friedman, and Kruskal-Wallis tests for continuous variables). The proportions of patients without any primary end points or repeat angiography were estimated by the Kaplan-Meier method and compared by means of a log-rank test. These primary end points were also compared by means of odds ratios and a simultaneous Fisher’s test. A statistical probability of <0.05 was considered to indicate significance.
Among 5119 patients investigated for ischemic heart disease at our institution between 1989 and 1993, 142 met the inclusion criteria, and 134 were ultimately randomized. Of these, 68 were assigned to PTCA and 66 to CABG. Baseline characteristics were not different between the 2 groups and are shown in Table 1⇓. In the PTCA group, all patients underwent angioplasty, and 64 (94%) had an uneventful in-hospital course. Fifty-nine patients in the CABG group underwent surgery, and the operation was successful in 58 (98%). At 2 years, 8 patients (12%) had suffered a cardiac death or myocardial infarction in the PTCA group versus 3 (4%) in the surgical group (P=0.09). Twenty-six patients (38%) in the PTCA group had an additional revascularization versus 6 (9%) after CABG (P<0.01).
The incidence of adverse events at 5 years is shown in Table 2⇓. Six patients (9%) died in the PTCA group versus 2 in the CABG group (2%). However, cardiac death was equally frequent, with only 1 death in each group (1%). The cause of noncardiac death was cancer in 4 patients (3 patients in the PTCA group and 1 in the CABG group), pulmonary embolism (1 patient in the PTCA group), and mesenteric infarction (1 patient in the PTCA group). Myocardial infarction was more frequent after PTCA and occurred in 10 patients (15%) compared with 3 (4%) in the CABG group (P=0.0001). However, the incidence of Q-wave infarction was not significantly different: 6% for the PTCA group versus 3% for the CABG group (P=0.8). Six (9%) of the PTCA patients suffered a non–Q-wave infarction due to an acute closure during the procedure or unstable angina with CK rise related to restenosis; these infarctions were documented by CK rise only. Finally, taken together, cardiac death and myocardial infarction were more frequent after PTCA (P=0.0004), with a relative risk of 2.6 (95% CI, 1.1 to 5.4) (Figure 1⇓ and Table 2⇓).
There were significantly more repeat interventions (Table 2⇑) in the PTCA group, with an incidence of 36% versus 9% in the CABG group (P=0.002; relative risk, 4.2; 95% CI, 2.8 to 5.6). In the PTCA group, 26% of the patients had a repeat revascularization on the LAD and 12% on another vessel compared with 4.5% and 4.5%, respectively, in the CABG group (P=0.002) (Table 3⇓). Finally, the LAD did not require revascularization during follow-up in 74% of the patients in the PTCA group compared with 95% in the CABG group (P=0.04) (Table 3⇓). The proportion of patients free of events was 62% in the PTCA group (P=0.00001) (Figure 2⇓) and 91% in the CABG group.
Clinical status revealed no difference between the 2 groups. Seventy-four percent of patients in the PTCA group and 71% in the CABG group were in functional class I. Only 6% in the PTCA group and 3% in the CABG group were in class III or IV (Table 4⇓).
At 5 years, most of the patients were taking either no drug, aspirin alone, or aspirin plus an antianginal drug (Table 5⇓). There were no differences in drug use between the 2 groups.
Control of risk factors was encouraged, but only 22% of the patients in both groups were taking lipid-lowering drugs, and mean cholesterol blood level was similar at 5.9 mmol/L (95% CI, 5.6 to 6.2 mmol/L). In the PTCA group, 89% of the smokers had stopped smoking compared with 71% in the CABG group. Only 12% in the PTCA group and 6% in the CABG group were smoking after 5 years of follow-up.
We have shown in this comparative study of PTCA and CABG for proximal isolated LAD stenosis that significant clinical improvement can be achieved with both treatments at a mean follow-up of 5 years. The risk of cardiac death is similar in the 2 groups and is very low (1%) compared with what has been reported previously in similar cohorts of patients.11 12 13
However, the overall incidence of myocardial infarction was higher after PTCA. This is mainly due to the higher number of non–Q-wave infarctions, most of which were due to periprocedural complications, although their incidence is comparable to those reported previously.14 A majority of the infarction cases that we observed only resulted in a CK rise, with no impact on left ventricular function; they are reported as myocardial infarctions, as defined in our initial protocol. Our data differ from those of the RITA I trial, in which no excess of myocardial infarction was noted in the angioplasty group.7 However, a more detailed analysis of the results of the present study shows that for patients with single-vessel disease, there was a clear trend toward more myocardial infarction in the angioplasty patients. The difference was not significant but probably would have been significant with the occurrence of just 1 more event. This makes our results and those of the RITA I trial very concordant. However, if we exclude patients with periprocedural infarction in our trial, there are no differences between the 2 strategies during follow-up with regard to myocardial infarction.
The need for additional revascularization was significantly higher in the PTCA group, which was mainly related to the occurrence of restenosis in 32% of patients. More than 75% of these procedures were performed on the LAD. In the CABG group, subsequent revascularization was rarely required (9%), and in half of the cases, it was related to progression of the disease in other vessels. Surprisingly, the incidence of non-LAD revascularization was significantly higher in the PTCA group (12% versus 4.5%). Patients in the PTCA group more often underwent control angiography and were therefore more prone to have revascularization because of a detection bias. Indeed, the progression of the disease cannot explain this difference; there is no obvious reason to have more rapid progression in the PTCA group. In particular, incidence and control of risk factors for ischemic heart disease were similar in both groups. Lipid-level control was insufficient compared with current standards but was comparable in both groups. Finally, only 4.5% of the initial cohort randomized to CABG underwent additional revascularization of the LAD. This confirms the high value of arterial conduits to revascularize the LAD. Our results are very similar to those reported recently by Hennessy et al.14 Very similar conclusions also arise from the Medicine, Angioplasty or Surgery Study (MASS),15 with a higher incidence of adverse events in the angioplasty and medical group owing to more additional revascularizations. The authors did not observe an increased incidence of myocardial infarction after angioplasty, but the true incidence of this event was not reported. They mention an 8% incidence of occluded LAD at 2 years, but whether these occlusions were silent or not is not indicated. In the present study, the proportion of asymptomatic patients was high, with no differences between treatment groups. This clinical improvement may also be due in part to the medical treatment, but again, this was similar in both groups, as was the quality of risk factor control.
Limitation of the Present Trial
One limitation of this study is its relatively low power to demonstrate that a difference between the 2 groups for an event is actually “statistically significant.”
On the basis of the above-mentioned data, we suggest that patients with isolated proximal LAD stenosis who need to be revascularized should be fully informed about the various approaches. The final choice should integrate the clinical status, morphology of the stenosis, and expressed preference of a fully informed patient. Indeed, there is an opportunity for a shared decision-making process to assist physician and patients in making the decision.
- Received December 7, 1998.
- Revision received April 14, 1999.
- Accepted April 14, 1999.
- Copyright © 1999 by American Heart Association
Hamm C, Reimers J, Ischinger T, Rupprecht H, Berger J, Bleifeld W. A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease: German Angioplasty Bypass Surgery Investigation. N Engl J Med. 1994;331:1037–1043.
Henderson R, Pocock S, Sharp J, Nanchahal K, Sculpher M, Buxton M, Hampton H, for the Randomised Intervention Treatment of Angina (RITA I) trial participants. Long-term results of RITA-1 trial: clinical and cost comparisons of coronary angioplasty and coronary-artery bypass grafting. Lancet. 1998;352:1419–1425.
Mark D. Assessment of prognosis in patients with coronary artery disease. In: Roubin G, Califf R, O’Neil W. Interventional Cardiovascular Medicine: Principles and Practice. New York, NY: Churchill Livingstone; 1994:163–166.
Detre K, Holmes D, Holubkov R, Cowley M, Bourassa M, Faxon D, Dorros G, Bentivoglio L, Kent K, Myler RK. Incidence of consequences of periprocedural occlusion: the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angiography registry. Circulation. 1990;82:739–750.
Hennessy T, Codd M, Donnelly S, Hartigan C, McCann H, McCarthy C, Neligan M, Wood A, Luke D, McGovern E, Aherne T, Sugrue D. Long term clinical outcome following coronary artery bypass grafting for isolated stenosis of the left anterior descending coronary artery. Eur Heart J. 1998;19:447–457.
Hueb W, Bellotti G, Almeida De Olivera S, Shiguemituzo A, de Albuquerque C, Jatene A, Pileggi F. The Medicine, Angioplasty or Surgery Study (MASS): a prospective, randomized trial of medical therapy, balloon angioplasty or bypass surgery for single proximal left anterior descending artery stenoses. J Am Coll Cardiol. 1995;26:1600–1605.