Medical Therapy vs. Revascularization: Results of Two Important Studies
The second randomized intervention treatment of Angina (RITA-2) trial sought to address whether early angioplasty or medical therapy was more beneficial in individuals with coronary artery disease (CAD), in whom clinical factors did not mandate a definite revascularization strategy. Patients with stable angina of no more than moderate intensity were randomized by their physician to receive conventional angioplasty or medical therapy. Results were analyzed on an intention-to-treat basis; at the time of this interim report, all patients had been followed at least six months, with a median of 2.7 years.
Approximately 1000 subjects were randomized to percutaneous transluminal coronary recanalization (PTCA) or medical therapy, culled from 2700 eligible patients seen in catheterization laboratories of 20 institutions in the United Kingdom and Ireland. Patients with severe angina, depressed left ventricular function, or other high-risk conditions were eligible by physician discretion. Protocol exclusions included prior revascularization, left main disease, or subjects in whom early revascularization was deemed necessary. The study had difficulty in recruitment and did not achieve target enrollment. The authors conclude that this was due to physicians refusing to randomize; also, approximately one-half of eligible patients refused to enroll. Those assigned to PTCA underwent conventional balloon dilatation within three months of randomization; multivessel angioplasty could be staged. Procedural success was achieved with a reduction of lesion severity of 20% and less than 50% residual stenosis with good distal flow in 93%.
Following PTCA, clinicians were encouraged to stop antianginal therapy, whereas in the medical therapy cohort, angina drugs were chosen by individual physicians and not by protocol; only a small number required triple therapy during the study. The combined primary endpoint was all-cause death and nonfatal myocardial infarction (MI). Subjects underwent a symptom limited stress test at three, six, 12, and 36 months after randomization. The baseline characteristics of the treatment groups were quite similar. Severe angina was present in only 20%, and another 20% reported no angina at the time of randomization. One out of six individuals had been admitted with unstable angina within three months; half had a prior MI.
Results: Ninety-two percent of randomized individuals received PTCA within five weeks. A large majority had single-vessel disease (335/471, 60%), and 33% and 7% had two- and three-vessel disease, respectively. Occluded vessels were also dilated, with a 70% success rate. Stents were used only in the late years of the trial and in a minority of procedures. Following randomization, 2.2% of the PTCA group and 1.4% of the medical group died (P = NS). However, there were twice as many non-fatal MIs in the PTCA group, partly explained by procedure related infarcts. The combined endpoint of death or MI occurred in 6.3% of the PTCA group and 3.3% of medical therapy subjects, an almost two-fold greater risk for PTCA (P = 0.02). During follow-up, 8% and 6% of the PTCA and medical subjects received a coronary artery bypass graft (CABG), respectively. Combined risk for death, MI, or CABG after two years was greater in the PTCA group (12.3%) than in medically treated patients (7.1%). Non-randomized PTCA occurred in 101 randomized to medicine. A minority of individuals also needed a CABG in both groups. The long-term risk of requiring subsequent revascularization (cross-over) in the medical therapy group was 15.4% vs. 14.9% in the initial PTCA cohort. Subsequent occurrence of unstable angina was comparable in both groups. However, the degree of angina severity and exercise duration was more favorable in the PTCA group, particularly early after randomization; over time there was attenuation in these treatment differences between PTCA and medical therapy. Many patients with worsening symptoms subsequently underwent revascularization. Angina was reduced in both groupsmore so in the PTCA cohort. As expected, medical treatment involved more antianginal drugs; most PTCA patients remained on at least one agent. Exercise time increased more in PTCA than in medical therapy; in six months, the PTCA group had greater improvement over baseline than medical therapy, although both cohorts improved. Subjects with moderately severe baseline angina and short exercise times improved considerably more with PTCA vs. medical therapy. Conversely, patients with no to mild angina at baseline demonstrated no significant difference between PTCA and medical therapy over the study period. Exercise duration at six months was also related to baseline treadmill performance and angina grade. The authors conclude that the relatively low-risk patients did well with either therapy, although the PTCA group had a greater risk of death or nonfatal MI by two to three years. Most of this difference was procedural. The increased risk of infarction in the PTCA group was somewhat surprising. Medically treated patients had very few infarcts. Thus, "RITA-2 provides no evidence to support the widely held belief that successful PTCA of severe coronary stenosis reduces the risk of myocardial infarction." Revascularization has the greatest benefit in individuals who are most limited at baseline; on the other hand, patients without severe angina benefited little from PTCA. The authors state that revascularization in such individuals "may reasonably be deferred." Finally, the investigators suggest that contemporary angioplasty with IIb/IIIa blockers, coronary stents, and widespread use of lipid lowering might further improve these results, particularly in the angioplasty group. (RITA-2 trial participants. Lancet 1997;350:461-468.)
COMMENT BY JONATHAN ABRAMS, MD
RITA-2 subjects represent an important group of individuals commonly seen by practitioners, i.e., significant stenoses of one or two coronary vessels with good left ventricular function and mild-to-moderate angina. There are no clear cut survival advantages in the medical therapy vs. CABG trials in such individuals, and therefore, it is reasonable to ask whether revascularization with angioplasty would be preferable to medical therapy. The results are concordant with the VA ACME trial (single vessel only) that indicated no survival advantage of PTCA over medical therapy, more procedural complications and repeat PTCA in the angioplasty cohort, and a greater improvement in exercise time and less anginal symptoms in the angioplasty group, but with progressive attenuation in these differences noted as early as six months. Ultimate need for bypass surgery in ACME was comparable in both groups. The RITA-2 trial demonstrated somewhat similar results, with a small but definite increase in major complications (death or MI) in the angioplasty patients but a trade-off of improved angina suppression and greater exercise duration.
As with most major trials in angina comparing two strategies, clear cut differences between any two approaches diminish over the several years following randomization. RITA-2 provides fuel for those who believe that good clinical judgment rather than a routine aggressive approach with revascularization is appropriate to guide clinical decision making in patients with angina and documented CAD. RITA-2 patients had to have at least one lesion amenable to balloon angioplasty; procedural results appear to be comparable to those obtained in the literature. Moderate anginal symptoms with diminished baseline exercise time favors PTCA, whereas few or mild symptoms, even in the presence of a positive exercise test, do not clearly benefit from PTCA, which may result in procedural complications and MI and does not make individuals feel better. Of significance, the vulnerable plaque hypothesis is supported by this data, in that there were very few MIs in the patients with stable angina pectoris.
Part II
In the early years following widespread thrombolytic therapy for acute myocardial infarction (AMI), it was thought that early revascularization of the infarct artery would decrease morbidity and mortality. However, large, randomized, clinical trials (e.g., SWIFT, TIMI-IIb) demonstrated conclusively that such individuals could be carefully observed, with angiography and subsequent revascularization reserved for those individuals with recurrent ischemia. In such trials, however, a substantial portion of such individuals did not undergo revascularization, and there was considerable crossover to angiography and revascularization. The Danish Trial in Acute Myocardial Infarction (DANAMI) hypothesized that following thrombolytic therapy, a strategy of routine revascularization in subjects with clinical or inducible post-MI ischemia, would improve survival as well as clinical status. One thousand patients were randomized to two groups, revascularization with PTCA or CABG vs. standard medical therapy, as directed by the clinician. Those enrolled were relatively young (< 69 years of age), and could not have had prior revascularization or MI. Virtually all received streptokinase within 12 hours of symptoms, and almost all had a symptom limited bicycle exercise test before discharge. Patients with spontaneous ischemia (rest angina) more than 36 hours after admission or those with a positive exercise test (with or without chest pain) were randomized to PTCA or medication. The primary endpoint was a combination of death, reinfarction, or admission for unstable angina. Secondary endpoints related to the presence and severity of angina pectoris. Patients were followed from one to four years (median 2.4). In the PTCA group, angiography was performed within two weeks of exercise testing; a positive angiogram was manifest by more than 50% narrowing in a coronary artery. PTCA or CABG was selected on the basis of the severity of coronary disease or number of stenoses by protocol. In the conservative arm, medical treatment consisted of antianginal drugs "prescribed according to local practice." Individuals who developed severe angina pectoris were subsequently referred for arteriography and possible revascularization. Patients were enrolled from 43 Danish hospitals, with angiography being performed at five selected institutions. The primary hypothesis was that revascularization would result in a 33% reduction in the primary endpoint.
The results were impressive, with a robust decrease in the combined primary endpoint in the revascularization cohort but no reduction in mortality. Thus, recurrent AMI and unstable angina were markedly reduced during follow-up in the invasive treatment group, with a relative risk reduction of 47% for reinfarction, 39% for unstable angina, and 33% for the combined endpoint (P = 0 < 0.004-0.00001). Mortality was low at median follow-up of 2.4 years equalling approximately 4%, indicating the relatively low-risk nature of this population in spite of the presence of ischemia.
Of the 503 patients randomized to the invasive strategy, 96% underwent arteriography, and 82% received a revascularization (53% PTCA, 29% CABG). Crossover to revascularization in the conservative cohort was not common, reaching 17% by study completion. Medical therapy consisted of beta blockers in 40-45%; calcium antagonists in 40%; and long acting nitrates in 25-30%. It is unclear how many patients were on double or triple therapy. Reinfarction rate at a median follow-up of 2.4 years was 10.5% in the conservative arm vs. 5.6% in the invasive treatment group; unstable angina was reduced from 29% in the conservative cohort to 18% in the invasive treatments. The authors conclude, "DANAMI is the first study to show significant benefit of revascularization in selected patients after AMI." In their discussion, they stress the relatively low mortality in the overall group, and they postulate that the results "probably apply to all AMI patients with ischemia," including those older than 70 years and those with prior revascularization or AMI. (Madsen JK, et al. Circulation 1997;96:748-755.)
COMMENT BY JONATHAN ABRAMS, MD
This report, previously presented at a national cardiology meeting, provides a useful road map for physicians taking care of patients with MI who receive thrombolytic therapy. It seems unlikely that the dominant use of t-PA in the United States would change the results of the Danish cohort, virtually all of whom were given streptokinase. The authors suggestion that higher risk individuals, such as older patients or those with a previous heart attack or history of revascularization, would also benefit seems reasonable, as such individuals are likely to have more extensive coronary artery disease and therefore, receive a larger benefit from revascularization. Of interest, these patients represent only an estimated 8% of all patients admitted for AMI in the 43 Danish institutions during this study, although 41% of those eligible were enrolled in DANAMI.
Thus, this is a relatively low-risk population, in spite of the fact that all had post MI angina or a positive stress test. Only one in six patients had angina without ECG changes, but almost 60% had asymptomatic exercise ischemia that qualified them for study entry. It may be argued that medical therapy was not optimal in the conservative group. No protocol was provided for drug administration, and less than 50% received a beta blocker. It is unclear what proportion of individuals in the conservative group were on two or more drugs. Nevertheless, hard endpoints of admission for unstable angina and recurrent MI would unlikely be substantially reduced by more widespread use of antianginal agents, such as beta blockers. These antianginal drug usage rates are not dissimilar from the United States.
This trial, in addition to the recent results of the VANQWISH study, analyzed in the context of the older thrombolytic trials, which randomized patients to early vs. late or routine vs. conservative angiography, provide a rational approach to the post-lytic MI patient. Clearly, high-risk individuals, manifest by early recurrent chest pain and/or ischemia, significant left ventricular dysfunction, extensive infarction, particularly anterior, shock or other important endpoints, should (and in the United States, do) proceed to angiography with revascularization when appropriate. However, in a substantial number of individuals who receive thrombolytic therapy, recurrent chest pain is not common, even on stress testing. DANAMI indicates that bicycle stress testing appears to be reasonably reliable to identify higher risk individuals who would benefit from revascularization as opposed to medical therapy. Other stress test modalities may or may not be better.
The mechanisms of decreased recurrent infarction in DANAMI are puzzling and are not identified in this trial nor speculated upon by the authors. It may be that increasing blood flow to the infarct artery, other vessels, or both improves endothelial function, stabilizes atherosclerotic plaque, and decreases the likelihood of recurrent infarction. This hypothesis needs to be tested. The current emphasis on plaque rupture arising from less than obstructive lesions is not concordant with the robust reduction in subsequent acute ischemic events seen in those individuals in DANAMI who underwent revascularization with either PTCA or CABG. An accompanying editorial (Guetta V, Topol EJ. Circulation 1997;96:713-715) emphasizes that practice in America is very different from Denmark, with most U.S. patients here undergoing routine angiography after an infarct.
In spite of the results of DANAMI, the editorialists conclude that angiography should be carried out for all patients with "moderate or large infarcts" but not necessarily in those with a "small uncomplicated infarction." They also point out that the increasing use of stenting as well as the application of IIb-IIIa platelet agents may provide an even greater improvement in morbidity and mortality in post-MI patients given thrombolytic therapy who subsequently undergo revascularization. Whether the more invasive American approach or the Danish strategy is better remains to be seen. It is useful to keep in mind the recent data comparing Canadian and United States practice in AMI patients, which indicates that in spite of substantially greater use of angiography and revascularization in the United States, long-term post-MI survival in both countries appear to be very similar. DANAMI and VANQWISH provide new information for an evidence-based practiceangiography and probable revascularization for all individuals with post-MI ischemia, as evidenced by formal stress testing or recurrent angina. For those subjects who have an unperturbed MI, without evidence of major LV dysfunction or recurrent ischemia, a policy of watchful waiting and appropriate medical therapy remains a sound and rational approach.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.