European Heart Journal (2001) 22, 1183–1197doi:10.1053/euhj.2000.2497, available online at onRe...
1184                 C. V. Patil et al.Table 1                  Comparative trials of CABG and medical treatment in stable...
Review     1185Table 2    Comparative trials of PTCA and medical therapy                                                Un...
1186    C. V. Patil et al.                                                            Overall combined risk of deathOveral...
Review     1187compared to PTCA, as provided by thallium scintigra-          associated with both improved survival and ev...
1188    C. V. Patil et al.outcome may be obtained with functionally complete              approximately 1%, stents were us...
Review     1189                  100                  90                  80                  70   Survival (%)           ...
1190    C. V. Patil et al.haemodynamic, electrophysiological or renal effects[62].        failure, co-morbid conditions and...
Review    1191Table 4    Non-randomized studies of multivessel stenting                                         Mean      ...
1192    C. V. Patil et al.(n=605) and stenting (n=600) in stable and unstable              The glycoprotein (GP) IIb/IIIa ...
Review     1193sanguineous cardioplegia lead to improved survival in          of subacute thrombosis[120]. As aspirin itse...
1194    C. V. Patil et offers more complete revascularization with long                   [4] The VA Coronary Artery ...
Review     1195       Angioplasty Versus Coronary Artery Bypass Surgery in                   coronary artery disease (a re...
1196    C. V. Patil et al. [56] Stein B, Weintraub WS, Gebhart S et al. Short and long                  infarction (the UL...
Multivessel coronary artery disease: current ...
Upcoming SlideShare
Loading in...5

Multivessel coronary artery disease: current ...


Published on

  • Be the first to comment

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Multivessel coronary artery disease: current ...

  1. 1. European Heart Journal (2001) 22, 1183–1197doi:10.1053/euhj.2000.2497, available online at onReview Article Multivessel coronary artery disease: current revascularization strategies C. V. Patil, E. Nikolsky, M. Boulos, E. Grenadier and R. Beyar Rambam Medical Center and Rappaport Faculty of medicine-Technion, Haifa, Israel Introduction of revascularization in multivessel coronary artery disease.The clinical presentation, extent and severity of cor-onary artery disease, left ventricular function and associ-ated co-morbid conditions influence the choice of initialmanagement of a patient with coronary artery disease. Coronary artery bypass graft surgeryIn 1964, Garrett, Dennis, and DeBakey first used cor- compared with medical therapyonary artery bypass grafting (CABG) to treat coronaryartery disease[1]. Subsequently, in the 1970s, the indica- Shortly after the introduction of CABG, three large,tions for coronary revascularization, CABG vs medical multicentre, randomized trials began enrolling patientstherapy, were the major focus of investigation[2–6]. These to compare a strategy of initial CABG with a strategy ofdata enabled the development of comprehensive guide- initial medical therapy in stable angina pectoris. Theselines for indications of coronary revascularization[7]. are the largest randomized trials of revascularization vsPercutaneous transluminal coronary angioplasty medical therapy to date and they have maintained their(PTCA) was introduced by Dr Andreas Gruentzig in major clinical role in evidence-based decision making1977 as a technique for the treatment of proximal, regarding coronary revascularization. These trials arenon-calcified, concentric lesions involving a single cor- summarized in Table 1.onary artery[8]. With improvements in angioplasty The meta-analysis[9] of seven randomized trials,equipment and techniques, the use of PTCA was comparing CABG and medical therapy, contains theexpanded to more complex lesions and in patients with majority of patients from three large trials[2–4]. Thesemultivessel disease. In the mid-1980s and 1990s, the patients represent a small proportion of patientsfocus of investigation shifted toward the preferred screened from randomizing sites in different trials. Mostmethod of revascularization — CABG vs PTCA. In (91·1%) of these patients had either multivessel or leftrecent years, there have been major improvements in main coronary artery disease. They were predominantlyboth surgical and medical treatments. For instance middle-aged male patients, while a few were womenpercutaneous and surgical revascularization, the use of (3·2%), some were elderly (7·3% over 60 years of age)stents, arterial bypass conduits, minimally invasive and some had left ventricular dysfunction (7·2% withcoronary artery surgery have all progressed. Medical ejection fraction less than 0·40). Only 10% of thetreatment, too, such as adjunctive pharmacotherapy patients received internal mammary grafts and only 20%(glycoprotein IIb/IIIa inhibitors) and aggressive lipid were treated with antiplatelet agents. There was a sig-lowering have reached optimum levels. The purpose of nificant crossover in the medical treatment group duringthis review is to discuss the current role of various modes the long-term follow-up, with 41% of these patients undergoing CABG surgery within 10 years. The meta-analysis[9] supports the general concept thatKey Words: Multivessel coronary artery disease, percutaneous the greater the amount of myocardium jeopardized, duetransluminal coronary angioplasty, coronary artery bypass to more extensive or proximal coronary disease,surgery, revascularization, multivessel stent implantation. the greater the improvement in prognosis with surgery.Revision submitted 10 October 2000, and accepted 11 October The CABG group had significantly lower mortality than2000. the medical treatment group (Fig. 1) at 5 years (10·2 vsCorrespondence: Dr Chandrashekhar V. Patil, Department of 15·8%, P=0·0001) and at 10 years (26·4% vs 30·5%,Cardiology, Rambam Medical Center, Haifa, Israel POB 9602, P=0·03). The benefit of CABG surgery was greater in31096. patients with extensive coronary disease, such as left0195-668X/01/221183+15 $35.00/0 2001 The European Society of Cardiology
  2. 2. 1184 C. V. Patil et al.Table 1 Comparative trials of CABG and medical treatment in stable angina Left main Frequency % of Male Angina EjectionStudy Number Age disease of IMA screened (%) Class III/IV fraction (%) (%)VACS[4] 686 19 No limit 100 58 13·1 0·25 0ECSS[2] 767 na 65 100 42 1·8 0·50 0CASS[3] 780 5 65 90 0 14 0·35 16VACS=Veterans Administrations Cooperative Study; ECSS=European Coronary Surgery Study;CASS=Coronary Artery Surgery Study; IMA=internal mammary artery; na=not available. 0·50 All studies 0·40 Medical treatment CABGTotal mortality 0·30 N = 1325 N = 1324 0·20 0·10 0 2 4 6 8 10 12 Years from randomization (years)Figure 1 Comparative survival curves of patients undergoing initial CABG (– – –)or initial medical treatment (——) for stable angina from a meta-analysis of sevenrandomized trials, showing survival benefit diminishing after initial 10 years. (FromYusuf S, Zucker D, Peduzzi P, et al. Effects of coronary artery bypass graft surgeryon survival: overview of 10-year results from randomized trials by the CoronaryArtery Bypass Graft Surgery Trialists Collaboration. Lancet 1994; 344: 563–70.Reproduced with permission. Copyright by The Lancet Ltd. 1994).main coronary artery disease, three- and two-vessel There are four randomized trials comparing PTCAdisease with left anterior descending coronary artery and medical therapy in patients with stable angina[10–13].involvement. Factors such as left ventricular dysfunc- These trials have included predominantly single vesseltion, advanced age, abnormal exercise test, and previous disease patients. The second Veterans Affairs ACME[11]myocardial infarction, did not individually affect the trial of double vessel disease has included 101 patientsrelative benefit. Since the mortality rate was higher in randomized into two groups; however, the number ispatients with lower ejection fractions, those with abnor- too small to draw any conclusion. In the RITA-2[12]mal left ventricular function derived greater absolute trial, 40% of a total of 1018 patients had multivesselbenefit. disease; however, there is no separate subgroup analysis There is no evidence from randomized trials that of multivessel disease in this trial. So the data from trialsbypass surgery reduces the subsequent occurrence of Q comparing PTCA and medical therapy is limited andwave myocardial infarction compared with medical inconclusive for multivessel disease. These trials aretherapy. The Veterans Affairs (VA) investigators[4] summarized in Table 2.speculated that, the survival benefit of CABG was due to In the Asymptomatic Cardiac Ischemia Pilot (ACIP)improved survival in patients who suffered a myocardial study[14], 588 chronic stable angina patients with suitableinfarction during long-term follow-up in surgical coronary anatomy for revascularization were random-patients (possibly with patent grafts functioning as col- ized to three strategies: angina-guided drug therapy,laterals) as compared to medically treated patients. The angina plus angina plus ischaemia-guided drugbenefit of surgery diminished with time in all patient therapy and revascularization (angioplasty or by-subsets, possibly as a result of graft attrition, progres- pass surgery — 92 and 78 patients underwent PTCAsion of disease in unbypassed vessels and cross-overs and CABG, respectively). Two years after randomiz-from medical to surgical therapy[2–4]. ation, total mortality (P0·02), combined death andEur Heart J, Vol. 22, issue 14, July 2001
  3. 3. Review 1185Table 2 Comparative trials of PTCA and medical therapy UnstableTrial Number Diseased vessels angina Comment (%)ACME[10] 212 SVD 0 PTCA more effective in angina relief and improving exercise time but with higher cost and complicationsACME-DVD[11] 101 DVD 0 Lack of superiority of PTCA as compared to medical therapyRITA-2[12] 1018 40% MVD 10 PTCA more effective in angina relief and improving exercise time, excess early hazard due to procedural complicationsMASS[13] 214 Prox LAD 0 Randomization to IMA bypass surgery, PTCA and medical therapy, no difference in primary end-point between PTCA and medical therapyACME=Angioplasty compared to Medicine; ACME-DVD=ACME trial in double vessel disease; RITA-2=Second RandomizedIntervention Treatment of Angina; MASS=Medicine, Angioplasty or Surgery Study; SVD=single vessel disease; DVD=double vesseldisease; MVD=multivessel disease; Prox LAD=proximal left anterior descending coronary artery; IMA=internal mammary artery.myocardial infarction (P0·04) and a composite of dominantly male patients and there were a significantdeath, myocardial infarction and recurrent cardiac number with unstable angina. The majority of thehospitalization (P0·001) were significantly lower in the patients in the surgical arm received internal mammaryrevascularization group than other two groups. The artery grafts. Most of these trials were conducted in thebenefit of revascularization was concentrated in those late 1980s and therefore stents were not used in thewith proximal left anterior descending coronary artery PTCA arm of these trials (summarized in Table 3). Instenosis (P=0·013 for difference between relative risks). these trials, only 5·2% of 91 730 patients screened wereThere was similar benefit in patients with three-vessel randomly assigned to a study group. Nearly two-thirdsdisease compared with those with one- or two-vessel of patients were excluded for angiographic reasons, suchdisease (P=0·015 for difference between relative risks). as left main coronary artery disease, chronic total occlu- sion, diffuse disease or inability to achieve ‘functional’ complete revascularization. In addition, most of the PTCA vs CABG in multivessel disease patients had well-preserved left ventricular systolic func- tion. As a result, patients who had previously beenThe trials comparing CABG with medical therapy[2–4] shown to benefit most from surgical revascularizationhave established a role for revascularization in the were excluded[28,29].treatment of multivessel coronary artery disease. But, Despite differences in study designs, inclusion andthe preferred method of revascularization remains a exclusion criteria, the results of these trials are remark-major issue of debate. There are several observational ably consistent. Two meta-analyses, including five of sixstudies[15–18], seven randomized trials[19–25], and two trials (excluding BARI) have been completed. One wasmeta-analyses[26,27] comparing PTCA and CABG in on multivessel disease only with a follow-up of 1 to 3multivessel disease. All the randomized trials included years[26] (Fig. 2); the other was on predominantly mul-patients with two- or three-vessel disease, with the tivessel disease, with a mean follow-up of 2·7 years[27].exception of the RITA trial which included 45% The overall risk of death was similar between the CABGsingle-vessel disease patients. These trials enrolled pre- and the PTCA patients (3·7% vs 3·9%, P=0·67). TheTable 3 Comparative trials of PTCA and CABG in multivessel disease Mean Unstable Ejection % of Male IMATrial Number age angina fraction screened (%) (%) (years) (%) (mean)TOULOUSE[19] 152 na na na na na naERACI[20] 127 17 58 85 83 61 naGABI[21] 358 4 59 81 20 56 37EAST[22] 392 8 62 73·5 65·9 61 90RITA[23] 1011 5 57 81 55 na 76CABRI[24] 1054 4·6 61 78 14·5 63 81BARI[25] 1829 7 62 73·5 64 57 82ERACI=Argentine trial of PTCA versus CABG; GABI=German Angioplasty Bypass SurgeryInvestigation; EAST=Emory Angioplasty versus Surgery Trial; RITA=Randomized InterventionTreatment of Angina Trial; CABRI=Coronary Angioplasty versus Bypass RevascularizationInvestigation; BARI=Bypass Angioplasty Revascularization Investigation; IMA=internalmammary artery graft; na=not available. Eur Heart J, Vol. 22, issue 14, July 2001
  4. 4. 1186 C. V. Patil et al. Overall combined risk of deathOverall risk of death or non-fatal MIStudy Year Total patients Study Year Total patients Odds ratio, Odds ratio, 95% confidence interval 95% confidence intervalERACI 1993 127 ERACI 1993 127RITA 1993 1011 RITA 1993 1011CABRI 1994 1054 CABRI 1994 1054GABI 1994 359 GABI 1994 359EAST 1994 392 EAST 1994 392Total 2943 Total 2943O.R. = 0·92, p = 0·67 0·1 1 10 O.R. = 1·03, p = 0·81 0·1 1 10 Lower in Lower in Lower in Lower in CABG PTCA CABG PTCAFigure 2 Overall risks of death and a composite of death and non-fatal myocardial infarction at 1 to 3years of follow-up from a meta-analysis of five randomized trials comparing CABG and PTCA formultivessel coronary artery disease. CABRI=Coronary Angioplasty versus Bypass RevascularizationInvestigation; EAST=Emory Angioplasty versus Surgery Trial; ERACI=Argentine trial of PTCA versusCABG; GABI=German Angioplasty Bypass Surgery Investigation; MI=myocardial infarction;OR=odds ratio; PTCA=percutaneous transluminal coronary angioplasty; RITA=Randomized Inter-vention Treatment of Angina Trial (Reprinted from Sim I, Gupta M, McDonald K, Bourassa M,Hlatky MA. A meta-analysis of randomized trials comparing coronary artery bypass grafting withpercutaneous transluminal coronary angioplasty in multivessel coronary artery disease. Am J Cardiol1995; 76: 1025–29. Copyright (1995) Reproduced with permission from Excerpta Medica, Inc.).combined risk of death and non-fatal myocardial infarc- available in 102 CABG patients and 117 PTCA patients.tion was also similar between CABG and PTCA groups The saphenous vein bypass grafts had an occlusion rate(10·1% vs 9·8%, P=0·81). The rate of additional inter- of 12·6% at 6 months, which is comparable to otherventions (PTCA and/or CABG) in the first year of studies[31,32]. Although the protocol excluded patientsfollow-up was 33·7% and 3·3% in patients randomized with total occlusion, 37% of the native arteries wereto PTCA and CABG, respectively[27]. The surgical occluded 6 months after bypass grafting compared topatients had a long initial hospitalization; the PTCA only 2·5% after PTCA. The rate of occluded nativepatients were discharged early and resumed work early, arteries was higher in the case of patent grafts than inbut were subsequently hospitalized more often. the case of occlusion of the corresponding bypass graft. The BARI trial[25] is the largest trial comparing PTCA This kind of progression of disease to total occlusion hasand CABG in multivessel coronary artery disease and generally occurred proximal to the site of graft insertion.the only trial empowered to detect differences in mor- Native arteries with high grade lesions (70%) were onlytality. The technically suitable patients with clinically at a slightly higher risk of progressing to total occlusionsevere angina or objective evidence of ischaemia were after bypass surgery than were native arteries with lessrandomly assigned to an initial treatment strategy of severe lesions. Thus, a bypass to an artery with minimalCABG (n=914) or PTCA (n=915) and were followed disease might be potentially harmful, as it accelerates thefor an average of 5·4 years. The 5-year survival rate was progression of native disease; in cases of graft attrition,89·3% for those assigned to CABG and 86·3% for those it will be undoubtedly harmful.assigned to PTCA (P=0·19). The respective 5-year sur- Thus combined evidence comparing PTCA withvival rates free from Q wave myocardial infarction were CABG shows no difference in prognosis (death and80·4% and 78·7% (P=0·84). By 5 years after study entry, myocardial infarction) between these two revascular-8% of the patients assigned to CABG had undergone ization strategies. However, the treatments differadditional revascularization procedures, as compared markedly in the subsequent requirement for additionalwith 54% of those assigned to PTCA (P0·001); 69% of revascularization procedures and in the relief of angina.those assigned to PTCA did not subsequently undergo CABG patients were significantly more likely to beCABG. angina-free as compared to PTCA (80·7% vs 73·1%; OR The follow-up angiographic substudy of GABI shows 1·57, 95% CI 1·32 to 1·87, P0·00001)[26]. The objectivesome interesting results[30]. Follow-up angiograms were evidence of less ischaemic burden in CABG patients asEur Heart J, Vol. 22, issue 14, July 2001
  5. 5. Review 1187compared to PTCA, as provided by thallium scintigra- associated with both improved survival and event-freephy in the Emory Angioplasty vs Surgery trial supports survival (P=0·04). Moreover, the mortality rate amongthe subjective reports of angina relief[22]. The late patients experiencing a myocardial infarction duringfollow-up of these trials is of considerable importance follow-up was significantly higher for patients with lessbecause restenosis after PTCA is a relatively early complete revascularization. The beneficial effects in aphenomenon, whereas graft attrition, especially with subset with severe angina and a low ejection fractionvein grafts after CABG produces its major effects several were attributed to the improvement in myocardial per-years after the initial operation[33]. fusion and function from complete revascularization in hibernating myocardium. One important item to note from this study was that the survival curves diverge after around 2–3 years of follow-up. In another surgical Important issues in multivessel PTCA study[37], it was concluded that postoperative residual lesions in the left anterior descending or left circumflex Completeness of revascularization and coronary artery were the most important predictors of culprit lesion dilatation approach mortality. In the same study, it was concluded that postoperative survival correlated better with the extentFunctionally complete revascularization is defined as the of residual disease than with the pre-operative extent ofsuccessful dilatation of all significant stenoses in vessels coronary disease. The 5-year survival rates of all well-that are of sufficient size (1·50mm in diameter) and revascularized patients with good left ventricular func-supplying viable myocardium. The definition of signifi- tion were similar, about 90% regardless of the extent ofcant stenosis varies from 50% to 70% in surgical and pre-operative disease.PTCA literature, respectively. Vessels serving either In the case of PTCA, revascularization is less com-infarcted territory, as demonstrated by akinesis on the plete as compared to a surgical series. In the NHLBIleft ventricular angiogram, or that were too small were PTCA registry[38], complete revascularization wasnot considered functionally important. Anatomically achieved in 23% of patients with double-vessel diseasecomplete revascularization is defined as no residual but in only 9% of those with triple-vessel disease. Thesignificant stenosis in any major epicardial vessels[34]. most common causes of incomplete revascularizationThus, functionally complete revascularization can be were culprit lesion dilatation strategy and chronic totalanatomically complete or incomplete. The impact of occlusion.completeness or revascularization in multivessel disease In the observational series of PTCA[34,38–41], the com-is more obvious from surgical series. pleteness of revascularization and long-term outcome The coronary artery surgery study (CASS) registry[35] varied depending on the improvement in angina andis the largest observational study comparing 4-year reduced the need for repeat revascularization, with com-cumulative survival patterns in multivessel disease in- plete revascularization in some studies[39,41]. But therevolving 20 088 medically treated patients. Survival var- was no survival advantage with complete revasculariz-ied according to the severity of coronary artery disease, ation, as in surgical series. The reason maybe the shorterfrom 92% in the single-vessel disease to 60% in three- length of follow-up and the comparatively low-risk ofvessel disease, with significant obstruction of the left the PTCA patients as compared to the surgical series. Inmain coronary artery. Survival also varied according to the randomized trials of PTCA and CABG[19,25], theleft ventricular function: with survival of 92% with an revascularization strategies also varied. In some trialsejection fraction greater than 50%, and 58% with an the objective was complete anatomical (GABI), com-ejection fraction of less than 35%. In a further analysis plete functional (ERACI), or equivalent (RITA) revas-of the impact of the extent of coronary artery disease cularization, but in EAST and CABRI an initial strategyand left ventricular function on survival, it was con- of partial revascularization was permitted. The rates ofcluded that left ventricular function is a more important complete revascularization were significantly higher inpredictor of survival than the number of diseased CABG patients than PTCA patients in all trials. Forvessels. example, in the BARI trial[25], complete revasculariz- In the retrospective analysis of 3372 non-randomized ation was achieved in 57% and 91% of patients in thesurgical patients from the CASS surgical registry[36] who PTCA and CABG arms, respectively. However, in nonehad three-vessel disease, adjusted event-free survival was of the trials was there any difference in the hard end-not influenced by completeness of revascularization in a points, such as death or myocardial infarction. One ofgroup with Canadian class I or II angina. In patients the explanations given by EAST investigators was thatwith class III and IV angina, those with more complete even though in the EAST trial, successful index segmentrevascularization were more likely to be asymptomatic revascularization (50% residual stenosis) was moreor free of severe angina, and a trend towards improve- common at 3 years with CABG as compared to PTCAment in event free survival (P=0·074) was observed. The (89% vs 76%; P0·01), the index segments free of severeeffect of complete revascularization was particularly disease with physiological priority (70%) did not differevident in patients with severe angina and an ejection between two groups (95% vs 93%, P=ns).fraction 35%. In these patients, complete revasculariz- In spite of these conflicting data, complete revascular-ation, defined as grafts to three or more vessels, was ization remains a desirable goal, and a satisfactory Eur Heart J, Vol. 22, issue 14, July 2001
  6. 6. 1188 C. V. Patil et al.outcome may be obtained with functionally complete approximately 1%, stents were used in around 65% ofrevascularization. This may not apply, however, to patients undergoing percutaneous interventions, the leftpatients with limited reserve, such as left ventricular internal mammary artery was used as a conduit indysfunction and severe ischaemia. Similarly, significant 95% of bypass surgery and multivessel disease patientsresidual disease in the left anterior descending coronary constituted approximately 53% of the total trialartery, as part of an incomplete revascularization strat- population.egy, is not an acceptable approach[42]. The use of stents The culprit lesion can be identified by using thein multivessel coronary artery disease may not cause a combined evidence of transient ECG changes duringsubstantial increase in the rate of complete revascular- ischaemia and angiographic findings (complex ulceratedization[94], reflecting the continued practice of function- plaque with thrombus and severity of the lesion).ally complete revascularization and the fact that the Although these patients are incompletely revascularized,acute procedural success rate of chronic total occlusion many will continue to be asymptomatic. Some patientsis not affected by stents. will require subsequent interventions in the form of Even though dilatation of the culprit lesion is part of PTCA or CABG, but they will be at less risk on a morean incomplete revascularization strategy, which may be elective basis[53]. Similarly, in spite of differing coronaryless than ideal, this approach, may be of value in some anatomy, the infarct-related artery is the only target incases. Among patients with stable angina pectoris and primary PTCA, except in patients whose haemodynamicmultivessel disease, only in 50% of patients can a culprit status remains poor despite restoration of the patency oflesion be identified[43]. The long-term effect of this that artery[54].approach in these patients is unknown[44,45]. However, inorder to achieve symptomatic relief, in chronic stableangina, this strategy may be valuable in some selected Diabetic patientspatients. The natural history of unstable angina varies with the The risk of death from cardiovascular causes is threerisk stratification based on the severity of angina and the times higher for diabetic than non-diabetic men, evenunderlying clinical circumstances. The risk of death and after adjustment for age and other cardiovascular risknon-fatal myocardial infarction at 3 months follow-up factors[55]. The long-term event rates after PTCAranges from 5% in the lowest risk group to 25% to 35% and CABG are also higher than in non-diabeticin the highest risk group[46,47]. Similarly major com- patients[56–58]. The medically treated diabetic patientsplications following percutaneous interventions vary comprised 19·5% of the total BARI trial population[25].according to the baseline clinical severity of the Five-year survival (Fig. 3) was 65·5% among patientspatients[48,49]. In one study[48], the composite risk of with treated diabetes who were assigned to PTCA, asdeath and myocardial infarction after PTCA was 2·2%, compared with 80·6% among diabetics assigned to3·5% and 5·1% in unstable angina Braunwald class I, II CABG, (P=0·003). The 2-year results for 122 diabeticand III, respectively. Among all patients with unstable patients in the Coronary Angioplasty vs Bypassangina, multivessel disease is found in approximately Revascularization Investigation (CABRI)[24] also sug-60% of patients[50]. In the VANQWISH (Veterans gest that CABG improved survival to the same extent asAdministration Non-Q Wave Infarction Strategies In in BARI trial. The more extensive disease among dia-Hospital) trial comparing ‘invasive’ and ‘conservative’ betics and their greater tendency to have a restenosisstrategies, the bypass surgery mortality of 15% was after angioplasty were the mechanisms proposed by theunacceptably high[51]. So in patients with acute coronary BARI investigators. Late coronary occlusion causingsyndrome, dilatation of only the ischaemia-causing fatal myocardial infarction is also one of the mechan-vessel (the ‘culprit’ lesion) has been recommended as an isms proposed to explain the worse long-term survival ininitial approach to stabilize the patient’s condition[44,45]. the PTCA group[59]. In another study comparing an early invasive with an In a retrospective observational study at Emoryearly non-invasive strategy in the unstable coronary- University hospitals[60], involving diabetic patients withartery disease FRISC II trial[52], 2457 patients were multi-vessel disease (PTCA, n=1057 and CABG,randomly assigned to invasive or non-invasive treatment n=2088), in both groups, survival was approximatelyand 3 months of dalteparin or placebo. The composite 50% by 10 years of follow-up. In the insulin-requiringof death or myocardial infarction occurred in 10·4% vs subgroup, in contrast to the total group, therapy with14·1% of patients (risk ratio=0·74; 95% CI=0·60–0·92; PTCA was a risk factor for long-term mortality with aP=0·005) at 1 year in the invasive and non-invasive hazard ratio of 1·35 (95% CI=1·01 to 1·79, P=0·045)strategies, respectively. In 100 patients with moderate to relative to CABG. In the patients treated with oralhigh risk unstable coronary-artery disease, at 1 year hypoglycaemic agents, the choice of revascularizationfollow-up, an invasive strategy saves 1·7 lives, prevents was not a correlate of survival. The higher long-termtwo non-fatal myocardial infarctions and 20 readmis- mortality rate for insulin-requiring diabetic patientssions, and provides earlier and better symptom relief at treated with angioplasty was explained by two possi-the cost of 15 more CABG and 21 more PTCA pro- bilities in this study: (1) diabetes itself has patho-cedures. These results are particularly important physiological effects that increase the mortality rate afterconsidering that the in-hospital surgical mortality was PTCA as compared to CABG, or (2) insulin-requiringEur Heart J, Vol. 22, issue 14, July 2001
  7. 7. Review 1189 100 90 80 70 Survival (%) 60 50 40 30 20 Patients with treated diabetes, P = 0·003 10 All other patients, P = 0·73 0 1 2 3 4 5 6 Years after randomizationPatients with treated diabetesCABG 180 161 93PTCA 173 139 69All other patientsCABG 734 696 449PTCA 742 701 468Figure 3 Survival among patients in the BARI trial who were beingtreated for diabetes at baseline ( ) and all other patients (——).Patients assigned to CABG are indicated by solid lines, and thoseassigned to PTCA by broken lines. The numbers of patients at risk areshown below the graph at baseline, and at 3 and 5 years. (From TheBypass Angioplasty Revascularization Investigation (BARI) Investi-gators: Comparison of coronary bypass surgery with angioplasty inpatients with multivessel disease. N Engl J Med 1996; 335: 217–25.Reproduced with permission. Copyright 1996 Massachusetts MedicalSociety. All rights reserved).diabetes may be a marker for more severe disease, in Stagingwhich surgery is a better option than angioplasty. How-ever, in another large non-randomized Duke registry[61] Multivessel coronary interventions not infrequentlyno advantage was observed with CABG as compared involve complex anatomy, such as long chronic totalwith PTCA among patients with diabetes. In the EAST occlusions, severely calcified lesions requiring adjunctivetrial there was no difference in 5 year survival between rotational atherectomy or difficult bifurcation lesions59 diabetic patients[22]. requiring complex angioplasty techniques. The patient’s So the survival advantage of CABG in diabetic baseline characteristics may sometimes be unfavourable;patients with multivessel revascularization varied in dif- they may be old and frail with poor left ventricularferent studies. There was a survival advantage for both function, or renal impairment. In these situations it isinsulin and oral hypoglycaemic treated patients in the better to follow the ‘dentist approach’ to the coronaryBARI and CABRI trials. However, in the observational revascularization by staging the procedure. It can be aEmory registry only insulin-requiring patients had an true elective prospective staging or unintentional stagingadvantage, and in the Duke registry there was no due to premature termination of the procedure due tobenefit. Diabetic patients constitute a separate subset of unexpected problems.patients and the issue of multivessel revascularization The foremost reason for staging is patient safety. Ifneeds to be addressed in prospective randomized trials in the result of the first lesion in suboptimal, it is generallythese patients. Until that time, it is advisable to maintain safer to tackle remaining lesions in another setting. Inaggressive risk factor modification, tight glycaemic con- spite of the availability of better contrast media, such astrol, intensive non-invasive screening for asymptomatic on-ionic monomeric or dimeric and ionic low osmolarrestenosis and a lower threshold for CABG in these contrast agents, the amount of the contrast agent shouldpatients who require coronary revascularization[59]. not exceed 5 ml . kg 1 due to possible undesirable Eur Heart J, Vol. 22, issue 14, July 2001
  8. 8. 1190 C. V. Patil et al.haemodynamic, electrophysiological or renal effects[62]. failure, co-morbid conditions and diabetes identifyPatient compliance and operator fatigue should also be patients likely to have higher costs after eitherconsidered especially in restless, uncomfortable patients angioplasty or bypass surgery[68,69].and if the procedure is prolonged. If PTCA is a directcontinuation of the diagnostic angiography, then stagingis an attractive option. Due to differing reimbursing Unconventional indicationspolicies, and cost increment with staging, the decision tostage is also influenced by economic considerations. Patients with left main coronary artery (LMCA) disease When the staging strategy is chosen, the culprit lesion have a poor prognosis[35] with medical treatment andis obviously the first to undergo PTCA. The time they derive maximum benefit from coronary arteryinterval for planning the second procedure varies from surgery[3,4]. A report of the American College ofdays to a few months, according to operator discretion, Cardiology/American Heart Association task forceas there are no available data to support one particular regards unprotected left main stenosis as an absolutetiming. One approach is staging in the same hospital- contraindication for PTCA[70]. The proximity of thisization that can be either after 24 h, leaving the sheath in vessel to the aorta, its large calibre and lack of tortuosityplace, or towards the end of the hospital stay. This makes it an inviting target for PTCA[71]. In Gruentzig’speriod is still within the time frame of subacute stent original report of the initial five patients undergoingthrombosis. The second approach is to time the second PTCA, three underwent dilatation of the left mainprocedure at around 4 to 8 weeks, well after the first coronary artery[8]. However, early reports of PTCA oflesion is stabilized with complete endothelization. Yet the left main demonstrated a high procedural complica-another approach is to stage the patient at 3–6 months, tion rate and short-term mortality. The initial NHLBIif the patient is reasonably asymptomatic after the first Registry[72] reported 19 patients with PTCA to LMCAprocedure, with the proposed goal of detecting and with a success rate of only 68%. In the largest report[73]treating the restenosis in the same setting. However, in of PTCA to LMCA in 127 patients, the proceduralfollowing these approaches, caution should be exercised mortality was 9·1% and 3 year survival was 36% in thein terms of the doubling of the reinterventions due to an unprotected left main (not protected by previous CABG‘oculostenotic reflex’ in the angiographic follow-up arm, to one or more branches or from extensive right-to-leftas compared to the clinical follow-up arm in the collaterals). The need for subsequent revascularizationBENESTENT II trial[63,64]. was 42%. In the BARI trial, 17% of patients underwent staged One hundred and seven patients from 25 centres werePTCA for the initial PTCA. In the preliminary report reported in the ULTIMA registry[74]. They were treatedof the Arterial Revascularization Therapies Study either electively (n=91) or for acute myocardial infarc-(ARTS)[65,66], 10% of the first 482 patients were staged. tion (n=16) for unprotected left main (ULMT) coronaryThe reasons given for staging were, patient fatigue in stenoses. Of patients treated electively, 25% were consid-60%, amount of contrast used in 30%, operator fatigue ered inoperable, and 27% were considered at high riskin 19%, long fluroscopic time in 19% and ‘other’ in for bypass surgery. Primary treatment included stents24·5% (in some patients more than one cause). (50%), directional atherectomy (24%), and balloon In a recent comparison between staged and non- angioplasty (20%). Results varied considerably, depend-staged procedures, it was shown that staging at an ing on presentation and treatment. For patients withinterval of 4–8 weeks is safe and results in more com- acute myocardial infarction, technical success wasplete revascularization vs a non-stop approach[65]. achieved in 75%, and for elective patients, technical success was achieved in 98·9%. Longer-term event (death, infarction, or bypass surgery)-free survival was Cost correlated with ejection fraction (P0·001) and was inversely related to presentation with progressive or restEven though the clinical efficacy of any treatment angina (P0·001). Surgical candidates with ejectionmodality is the most important factor in choosing fractions d40% had an in-hospital survival of 98% andamong the multiple options, cost-effectiveness remains a 9-month event-free survival of 86 5%, whereasthe important factor. The RITA trial indicates that patients with ejection fractions 40% had 67% andalthough PTCA is initially less expensive, the cost of the 22 12% in-hospital and 9-month event-free survivals,PTCA strategy rises over 2 years to around 80% of the respectively. Nine hospital survivors (10·6%) experi-cost of a strategy of CABG[28]. Similar findings were enced cardiac death within 6 months of hospital dis-reported for GABI[29]. In the BARI trial, the cost of the charge; they were primarily patients with unstableinitial procedure was 35% lower for angioplasty than angina and the authors suggested follow-up angiogra-surgery. However, the long-term cost at the 3 year phy 6 to 8 weeks after PTCA to prevent early post-follow-up, was only 4% lower than bypass surgery, hospital discharge cardiac death. Forty out of a patientpredominantly due to additional revascularization pro- total of 277, who underwent an emergency percutaneouscedures required among patients with three-vessel dis- coronary intervention of unprotected left main stenosisease[67]. Two-vessel disease identifies patients likely to for acute myocardial infarction were discussed in aincur lower costs after angioplasty, whereas age, heart further report of this registry[75]. Most of these patientsEur Heart J, Vol. 22, issue 14, July 2001
  9. 9. Review 1191Table 4 Non-randomized studies of multivessel stenting Mean ACC/AHA Emergency Procedural Repeat Event freeStudy Number follow-up type B2/C CABG success revasc survival months % % % % %Moussa et al.[92] 100 21 60 2 97 30 66Laham et al.[93] 103 13 ns 0 97 17 80Mathew et al.[94] 175 12 74·5 0 98·3 18·3 79·8de Servi et al.[95] 377 12 60 1·3 91·7 10·8 72Kornowsky et al.[96] 398 12 45 0·5 96 21 78ACC/AHA=American College of Cardiology/American Heart Association; revasc=revascularization; event free survival=free of death,myocardial infarction and revascularization; na=not available.were in cardiogenic shock, requiring aggressive mechan- and C lesions. As the studies in small vessels[84,85],ical support. The in-hospital death rate was 55% and the saphenous vein grafts[86], restenotic lesions[87], chronic12-month survival rate was 42%. However, for a similar total occlusions[88], and acute myocardial infarc-clinical presentation of acute myocardial infarction, left tions[89,90] are showing the superiority of stenting overmain stenosis and shock (left main shock syndrome) the balloon angioplasty, the stent indications are growing,in-hospital mortality of medical therapy and CABG is and currently upto 80% of interventions in many centresvery high and comparable, respectively[76–78]. Hence the are accomplished by stent placement[91].authors concluded that percutaneous revascularization There are several recent non-randomized studies[92–96]can be a preferred strategy in this setting. of multivessel stenting. They are summarized in Table 4. Recently there have been several reports of left The multivessel coronary stenting in these studies wasmain stenting with 100% procedural success and no associated with a high angiographic success rate, a lowin-hospital mortality[71,79,80]. In these studies, the clinical periprocedural complication rate, and a high event-freerestenosis rate was 20%[71] and 17%[79] and event-free survival rate. The rate of emergency CABG was lower insurvival 80%[71]. However, considering the small number these trials, (ranging from 0% to 2%) as compared toof patients in these studies, and their non-randomized 6·3% in the BARI trial[25] and 9·6% in EAST trial[22].nature, left main interventions can be advocated only to Severe dissection is the most common cause of abruptprotect the left main coronary artery and in patients vessel closure, and stents are the most effective bailoutwith unacceptable surgical risks or severe co-morbid device, significantly reducing the need for emergencyconditions. There is concern about late sudden death CABG. These series included a relatively high pro-after unprotected left main stenting[74,81]. portion of complex lesions, and ACC/AHA type B2 and C lesions constituted 45% to 74·5% in different studies. In one study[91], 53% of patients had undergone saphe- Current trends nous vein graft stenting. However, in these studies there was no increase in completeness of revascularization. In Percutaneous revascularization one study[94], complete revascularization was achieved in only 57·1%; the reasons given were chronic total occlu-Stents sion and a policy of functionally adequate revasculariz-Coronary stent implantation has become the major ation. In another study[95], out of 377 multivessel diseasemode of myocardial revascularization after the STRESS patients, only one vessel was treated in 321 patients, dueand BENESTENT landmark trials[82,83], which proved to a significant number with unstable angina (33%) of athe efficacy of Palmaz–Schatz stents in the prevention of culprit lesion and chronic total occlusions (40·3%).restenosis. The target lesions in these trials were discrete However, repeat revascularization during follow-up was(15 mm), large vessels (3 mm) with certain anatomi- very low (10·8%).cal exclusions, such as multiple lesions, diffuse disease, There was a reduced need for repeat revascularizationthrombus, ostial location and tortuous vessels. There procedures with multivessel stenting (ranging fromwas a higher clinical success, reduced restenosis (around 10·8% to 30%) as compared to 37% to 55% in theone third) and a reduction in the need of target vessel historical PTCA arm of PTCA vs CABG trials. Therevascularization in the stent patients as compared to study of Moussa et al.[92] which had a maximum repeatthe angioplasty patients. While these trials were critically revascularization rate of 30%, was an angiographicimportant in shaping our current revascularization strat- follow-up study with follow-up angiography in 89% ofegies, they included highly selected lesions, constituting patients, which might have precipitated an ‘oculosten-less than 20% of lesions in current clinical practice. In otic reflex’. However, in spite of these encouraging data,the BENESTENT II trial[64–65], with the heparin-coated patient selection and the non-randomized nature of thePalmaz–Schatz stent, the restenosis rate was reduced to study prevents drawing any firm conclusions.16% as compared to 30% in the balloon group, even Recently, the Arterial Revascularization Therapythough this trial included 56% of ACC/AHA type B2 Study (ARTS), a randomized trial comparing CABG Eur Heart J, Vol. 22, issue 14, July 2001
  10. 10. 1192 C. V. Patil et al.(n=605) and stenting (n=600) in stable and unstable The glycoprotein (GP) IIb/IIIa inhibitors haveangina with at least two de novo lesions amenable to demonstrated a reduction in death, myocardial infarc-either CABG or stenting (employing the Cordis Crown tion, and the need for repeat revascularization inand Crossflex stent) with agreement between the surgeon percutaneous coronary revascularization as a ‘classand cardiologist was published[97]. The primary end- effect’ (trials with all of these agents show benefit):point of the study was major adverse cardiac and however, heterogeneity in the treatment effect is presentcerebrovascular event (MACCE)-free survival at 1 year. with the strongest evidence for the use of abciximab.The main exclusion criteria were overt congestive heart Kong et al.[105] reported a meta-analysis of 16 random-failure and intention to treat 1 totally occluded major ized controlled clinical trials of GP IIb/IIIa inhibitors toepicardial vessel. Two-thirds of patients in both groups assess the effect on death, myocardial infarction andhad two-vessel coronary artery disease and a third had revascularization in 32 135 patients. For the compositethree-vessel disease; left ventricular function was pre- end-point, there was highly significant benefit (P0·001)served (mean LVEF 60·5%). One third of patients in favouring GP IIb/IIIa inhibitors. At 6 months thereboth groups had unstable angina. There was a statisti- were 28 fewer events per 1000 patients undergoingcally significant reduction in MACCE in the CABG interventions. In a pooled data of 1462 diabetic patientsgroup at 1 year (12·2% vs 26·3%). This was primarily from EPIC, EPILOG and EPISTENT trials, mortalitydue to a marked difference in the repeat revasculariz- in diabetics who underwent multivessel intervention wasation (3·5% in the CABG group as opposed to 16·9% in reduced from 7·7% to 0·9% (P=0·018) with use ofthe stent group). The hard events such as death (2·8% vs abciximab[106]. The primary benefit of IIb/IIIa inhibitors2·5%), myocardial infarction (4·0% vs 5·3%), and strokes is in reducing the coronary events associated with the(2·0% vs 1·7%) was not significantly different between procedure, peri-procedural reduction of myocardialthe CABG and stent groups, respectively. The investi- necrosis and abrupt closure. No additional benefit isgators performed a cost analysis of the two strategies. apparent in the subsequent period from 2 days afterInitial costs were lower in the stent group, with the intervention until 30 days[107]. Thus glycoprotein IIb/average cost in the surgical group amounting IIIa inhibitors are a major development in adjunctiveto 10 742 Euros (1 Euro=approximately 1 US$) as pharmacotherapy enhancing the safety of the procedure.opposed to 6484 Euros to the stented group. Follow-up With the introduction of sensor-tipped angioplastycosts were higher in the stent group (4216 vs 2903 guide wires, Doppler or pressure based information,Euros). However, the overall cost was lower in the stent such as coronary flow reserve and myocardial fractionalgroup by approximately 3000 Euros over the 1-year flow reserve, can be derived regarding the significance ofperiod. Thus in this trial, there was no significant stenosis, facilitating physiologically based decisionsdifference in ‘hard events’ (death, myocardial infarction, regarding the need for intervention for an individualstroke) with either revascularization technique at 1-year lesion during a multivessel-multilesion procedure. As thefollow-up. Previous trials of CABG vs percutaneous number of lesions treated increases in multivessel dis-intervention have shown, on average, a 30%–40% differ- ease, the chances of restenosis increases[92]. So, tacklingence in MACCE rates in favour of the surgical arms. only physiologically significant lesions should be anThe ARTS trial[97] shows that with multivessel stenting, optimum policy. Recent studies suggest that intra-this difference can be narrowed to only 14%. The SOS coronary physiological information complements infor-(Stent OR Surgery) trial is another ongoing study com- mation derived from coronary angiography in bothparing the two treatment modalities in multivessel decision making and to optimize the result of thedisease. coronary intervention[108], which should be more appli- cable to multivessel disease. Similarly intravascularAdjunctive therapies and techniques ultrasound assists in optimizing acute gain, symmetry,Even through stents reduce the relative rate of restenosis and apposition of intracoronary stents even when post-in comparison with balloon angioplasty, the occurrence stent high-pressure balloon inflation achieves an optimalof stent restenosis is substantial due to frequent use of angiographic result and may be associated with lowstents and its use in complex and long lesion subsets. rates for target vessel revascularization and majorWhen in-stent restenosis occurs, conventional treat- adverse cardiac events at follow-up[109].ments are of limited value and except for focal stentrestenosis, repeat in-stent restenosis occurs in 54% to66% of patients after treatment[98–100]. As the principal CABGcause of in-stent restenosis is neointimal tissue prolifer-ation in ultrasound studies[101], the speculation was that Recently there has been significant improvement inlocal radiation might be useful in this situation. The myocardial protection techniques, less-invasive opera-studies using 192Ir as a source of gamma radiation[98–100] tive techniques and arterial conduits of CABG. Thefor in-stent restenosis and studies using beta radiation to CABG is associated with myocardial stunning which isboth de novo and in-sent restenosis[102–104] demonstrate associated with morbidity and mortality, especially withimpressive benefit in the reduction of repeat restenosis in acutely and chronically depressed myocardial function.the range of 50% to 60%. Thus intracoronary radiation It is demonstrated that some techniques, such as ventedis a step ahead in the percutaneous interventions. cardiopulmonary bypass and substrate enhancedEur Heart J, Vol. 22, issue 14, July 2001
  11. 11. Review 1193sanguineous cardioplegia lead to improved survival in of subacute thrombosis[120]. As aspirin itself is a weakhigh-risk cases[110]. The choice of graft is very important antiplatelet agent, the other potent antiplateletin CABG because long-term graft patency is closely drugs, GP IIb/IIIa inhibitors, are major advances inrelated to cardiac mortality. Barner et al.[111] reported a pharmacotherapy.superior graft patency of the internal mammary artery The feasibility of the use of recombinant genesof 83% over saphenous vein graft patency of 41% at 10 or growth factors to enhance myocardial collateralyears in 1000 patients. There is also a reduction in blood vessels has been proved in animals. It will beperi-operative mortality with internal mammary artery interesting to see the results of the large prospectiveapplication and its use is recommended as myocardial trials of these factors in humans of both direct intramyo-protection in high-risk patients[112]. The employment of cardial injection or catheter based transendocardialbilateral internal mammary artery grafts is associated injection[121].with less recurrent angina, fewer reoperations and atrend towards better survival[113,114]. Use of free rightinternal mammary artery has also been shown to haveexcellent short and long-term results[115]. Acar et al.[116] Newer modalitiesrecently reported 84%, 5-year patency in the radial Both surgical (transmyocardial revascularization) andartery as a conduit. For other arterial grafts, such as catheter based myocardial laser revascularization (per-gastro-epiploic and inferior epigastric arteries, long-term cutaneous myocardial revascularization) are potentialresults are not available. alternatives that are currently being evaluated in Less invasive surgery has been used recently to reduce patients in whom conventional methods fail[122]. Boththe morbidity of conventional CABG. It is of three types result in a reduction in anginal symptoms intypes[110]. (1) Off-bypass surgery is performed with patients with intractable medically refractory anginamedian sternotomy, reducing cardiac motion by mech- pectoris not amenable to CABG and PTCA, mostanical and pharmacological means. (2) Minimally likely related to myocardial inflammation, secondaryinvasive direct coronary artery bypass (MIDCAB) is stimulation of growth factors and denervation of theperformed through a small left anterior thoracotomy myocardium. There are three randomized publishedwithout cardiopulmonary bypass. Usually only left trials[123–125] prospectively comparing transmyocardialanterior descending coronary artery and diagonal revascularization and medical treatment in 655 patients.branches, but occasionally two vessels, are grafted by There was significant improvement of subjective end-this approach. It is associated with rapid recovery and points such as angina class and quality of life withless morbidity, but long-term patency remains to be transmyocardial revascularization in all trials. There wasdetermined. (3) Closed chest, port access, video-assisted no significant difference in survival between medicallyCABG can be performed with femoro-femoral cardio- treated and transmyocardial revascularization patients.pulmonary bypass and cardioplegic arrest with limited There were differing results of objective end-points inincision. All these innovations may improve short-term different trials. There was improvement of exercisemorbidity, but the long-term results, in terms of graft tolerance in one trial[125], whereas treadmill exercise timepatency need to be carefully assessed in the future. and 12 min walk distance were not different in another trial[123]. There was improvement of perfusion defects in one trial[124], but no improvement in another trial[125]. In Medical treatment general, the improvement in angina appears to be out of proportion to any demonstrable improvement inThere have been significant advances in lipid lowering myocardial perfusion[126].therapy, antiplatelet therapy, beta-blocker therapy and The percutaneous myocardial revascularization ap-gene therapy. Clinical trials in patients with coronary proach applies laser energy to create channels from theheart disease and with and without high cholesterol have endocardial side of the left ventricular wall. The earlydemonstrated consistently that statins reduce coronary results of percutaneous myocardial revasculariz-events by 30% and also reduce mortality significantly. ation indicate that the reduction in anginal symptomsThe angiographic follow-up trials of lipid lowering show are similar to that with transmyocardial revasculariz-the slowing of progression, induction of regression, ation and improvement in exercise duration[127–130].reduction of new lesion formation and most importantly However, recently a study with Direct Myocardial‘plaque stabilization’;[117]. In one trial of 341 asympto- Revascularization (DMR), randomizing laser to placebomatic or mild to moderately symptomatic chronic stable treatment could not detect a beneficial effect for DMR inangina patients randomized to high dose atorvastatin comparison to the placebo group.(80 mg) and PTCA, the atorvastatin group had a 36%lower event rate than PTCA patients (P=0·048) at 18months follow-up[118]. Aspirin is currently the standardsecondary prevention therapy in all subsets of patients Conclusionwith established heart disease[119]. The addition of ticlo-pidine and clopidogrel to aspirin along with optimum The gold standard for multivessel revascularizationstent deployment had significantly reduced the incidence remains conventional coronary artery bypass surgery, as Eur Heart J, Vol. 22, issue 14, July 2001
  12. 12. 1194 C. V. Patil et offers more complete revascularization with long [4] The VA Coronary Artery Bypass Surgery Cooperative Studylasting symptom relief and with fewer reinterventions Group. Eighteen-year follow-up in the Veterans Affairs Cooperative Study of Coronary Artery Bypass Surgery forthan after PTCA. The surgical approach has a wide stable angina. Circulation 1992; 86: 121–30.applicability, such as left main disease, chronic total [5] Mock MB, Fisher LD, Holmes Dr Jr., et al. Comparison ofocclusions, diffuse disease and anatomically complex effects of medical and surgical therapy on survival in severedisease that would have been problematic and techni- angina pectoris and two-vessel coronary artery disease withcally demanding with PTCA. However, surgery has its and without left ventricular dysfunction: A Coronary Artery Surgery Study Registry study. Am J Cardiol 1988; 61:own limitations, such as the use of general anaesthesia, 1198–1203.extracorporal circulation, ventilation and associated [6] Kaiser GC, Davis KB, Fisher LD et al. Survival followingmorbidity of major surgery. It is also less easily imple- coronary artery bypass grafting in patients with severemented in acute coronary syndromes, less easily repeat- angina pectoris (CASS): An observational study. J Thorac Cardiovasc Surg 1985; 89: 513– and associated with longer hospital stay and [7] Subcommittee on Coronary Artery Bypass Surgery. ACC/difficulties in work resumption[65]. AHA guidelines and indications for coronary artery bypass In contrast, PTCA has many advantages. It is less graft surgery. Circulation 1991; 83: 1125–73.invasive, and more readily applicable to acute coronary [8] Gruentzig AR, Senning A, Siegenthaler WE. Nonoperativesyndromes and sick patients. In addition, it is more dilatation of coronary artery stenosis: Percutaneous translu- minal coronary angioplasty. N Engl J Med 1979; 301: 61–8.easily repeatable (dentist approach) with shorter [9] Yusuf S, Zucker D, Peduzzi P et al. Effects of coronaryhospitalization. However, it is associated with less symp- artery bypass graft surgery on survival: overview of 10-yeartomatic relief, more repeat procedures and possibly less results from randomized trials by the Coronary Arteryfavourable outcomes in diabetic patients. Recent devel- Bypass Graft Surgery Trialists Collaboration. Lancet 1994;opments in stents have introduced this device as an 344: 563–70. [10] Parisi AF, Folland ED, Hartigan P, for the Veterans Affairsimportant tool for revascularization provided the vessel ACME investigators. A comparison of angioplasty withis larger than 2·75 mm. Longer follow-ups with the medical therapy in the treatment of single-vessel coronarycurrent randomized trials of CABG, comparing stents in artery disease. N Engl J Med 1992; 326: 10–16.multivessel disease, will define in detail how the differ- [11] Folland ED, Hartigan PM, Parisi AF, for the Veteransence between surgical and percutaneous techniques are Affairs ACME Investigators. Percutaneous transluminal cor- onary angioplasty versus medical therapy for stable anginamodified by the introduction of the stent. pectoris: Outcomes for patients with double-vessel versus In recent years, neither mode of revascularization has single-vessel coronary artery disease in a Veterans Affairsbeen utilized in a mutually exclusive way. Often, they cooperative randomized trial. J Am Coll Cardiol 1997; 29:can be complementary. Now the question is changing 1505–11.from, which mode of treatment is better to which [12] RITA-2 Trial Participants. Coronary angioplasty versus medical therapy for angina: The Second Randomized Inter-sequence and what combination. The ‘hybrid’ pro- vention Treatment of Angina (RITA-2) trial. Lancet 1997;cedure, such as MIDCAB to left anterior descending 350: 461–8.coronary artery and PTCA to other vessels in patients [13] Hueb WA, Bellotti G, de Oliveira SA et al. The Medicine,with multivessel disease may be safer and more effective Angioplasty, or Surgery Study (MASS): A prospective, ran-than either alone[131]. domized trial of medical therapy, balloon angioplasty, or bypass surgery for single proximal left anterior descending Finally the importance of risk factor reduction and artery stenoses. J Am Coll Cardiol 1995; 26: 1600–5.medical treatment, including aggressive lipid lowering, [14] Davies RF, Goldberg D, Forman S et al. for the ACIPcannot be over-emphasized. Long-term follow-up pro- Investigators. Asymptomatic Cardiac Ischemia Pilot (ACIP)vides scarce evidence[122] for a reduction in future acute Study two year follow up: Outcomes of patients randomized to initial strategies of medical therapy versus revasculariz-coronary events with revascularization. However, there ation. Circulation 1997; 95: 2037– ample evidence in the literature for a reduction with [15] O’Keefe JH Jr., Allan JJ, McCallister BD et al. Angioplastymedical treatment. Comprehensive medical therapy vs versus bypass surgery for multivessel coronary artery diseaserevascularization is a re-challenge to our current prac- with left ventricular ejection fraction less than or equal totice. Current practice may need reevaluation due to the forty percent. Am J Cardiol 1993; 71: 897–901.recent important developments in both medical therapy [16] Hochberg MS, Gielchinsky I, Parsonnet V et al. Coronary angioplasty versus coronary bypass: Three-year follow-up ofand revascularization procedures. a matched series of 250 patients. J Thorac Cardiovasc Surg 1989; 97: 496–503. [17] Vacek JL, Rosamond TL, Stites HW et al. Comparison of References percutaneous transluminal coronary angioplasty versus cor- onary artery bypass grafting for multivessel coronary artery [1] Garrett HE, Dennis EW, DeBakey ME. Aortocoronary disease. Am J Cardiol 1992; 69: 592–7. bypass with saphenous vein graft Seven-year follow up. [18] Jones RH, Kesler K, Phillips HR et al. Long-term survival JAMA 1973; 223: 792–94. benefits of coronary artery bypass grafting and percutaneous [2] European Coronary Surgery Study Group. Long-term re- transluminal angioplasty in patients with coronary artery sults of prospective randomized study of coronary artery disease. J Thorac Cardiovasc Surg 1996; 111: 1013–25. bypass surgery in stable angina pectoris. Lancet 1982; 2: [19] Puel J, Karouny E,. Marco F et al. Angioplasty versus 1173–80. surgery in multivessel disease: immediate results and in- [3] Alderman EL, Bourassa MG, Cohen LS et al. Ten-year hospital outcome in a randomized prospective study. Circu- follow-up of survival and myocardial infarction in the ran- lation 1992; 86 (Suppl I): 372. domized Coronary Artery Surgery Study. Circulation 1990; [20] Rodriguez A, Boullon F, Perez-Balino N et al. Argentine 82: 1629–46. Randomized Trial of Percutaneous Transluminal CoronaryEur Heart J, Vol. 22, issue 14, July 2001
  13. 13. Review 1195 Angioplasty Versus Coronary Artery Bypass Surgery in coronary artery disease (a report from the 1985–1986 Multivessel Disease (ERACI): In-hospital results and 1-year NHLBI PTCA Registry). Am J Cardiol 1992; 70: 174–8. follow-up. J Am Coll Cardiol 1993; 22: 1060–7. [39] Vandormael MG, Chaitman BR, Ischinger T et al. Immedi-[21] Hamm CW, Reimers J, Ischinger T et al. for the German ate and short-term benefit of multilesion coronary angi- Angioplasty Bypass Surgery Investigation. A randomized oplasty: influence of degree of revascularization. J Am Coll study of coronary angioplasty compared with bypass surgery Cardiol 1985; 6: 983–91. in patients with symptomatic multivessel coronary disease. N [40] Reeder GS, Holmes DR Jr, Detre K et al. Degree of Engl J Med 1994; 331: 10377–43. revascularization in patients with multivessel coronary dis-[22] King SB, Lembo NJ, Weintraub WS et al. A randomized ease: A report from the National Heart, Lung, and Blood trial comparing coronary angioplasty with coronary bypass Institute Percutaneous transluminal Coronary Angioplasty surgery. N Engl J Med 1994; 331: 1044–50. Registry. Circulation 1988; 77: 638–44.[23] RITA Trial Participants. Coronary angioplasty vs coronary [41] Mabin TA, Holmes DR Jr, Smith SC et al. Follow up clinical artery bypass surgery: The Randomized Intervention Treat- results in patients undergoing percutaneous transluminal ment of Angina (RITA) Trial. Lancet 1993; 341: 573–80. coronary angioplasty. Circulation 1985; 71: 754–60.[24] CABRI Trial Participants. First-year results of CABRI [42] Diver DJ, Gersh BJ. Efficacy of percutaneous transluminal (Coronary Angioplasty versus Bypass Revascularization coronary angioplasty: randomized trials of myocardial revas- Investigation). Lancet 1995; 346: 1179–84. cularization. In Topol EJ, ed. Textbook of interventional[25] The Bypass Angioplasty Revascularization Investigation cardiology. Philadelphia: W. B. Saunders Co. 1999: 209–37. (BARI) Investigators. Comparison of coronary bypass sur- [43] Ambrose JA, Winters SL, Stern A et al. Angiographic gery with angioplasty in patients with multivessel disease. N morphology and the pathogenesis of unstable angina Engl J Med 1996; 335: 217–25. pectoris. J Am Coll Cardiol 1985; 5: 609–16.[26] Sim I, Gupta M, McDonald K, Bourassa M, Hlatky MA. A [44] Wohlgelernter D, Cleman M, Highman HA et al. Percu- meta-analysis of randomized trials comparing coronary taneous transluminal coronary angioplasty of the ‘culprit artery bypass grafting with percutaneous transluminal lesion’ for management of unstable angina pectoris in coronary angioplasty in multivessel coronary artery disease. patients with multivessel coronary artery disease. Am J Am J Cardiol 1995; 76: 1025–9. Cardiol 1986; 58: 460–4.[27] Pocock SJ, Henderson RA, Rickards AF et al. Meta-analysis [45] de Feyter PJ, Serruys PW, Arnold A et al. Coronary of randomized trials comparing coronary angioplasty with angioplasty of the unstable angina related vessel in patients bypass surgery. Lancet 1995; 346: 1184–9. with multivessel disease. Eur Heart J 1986; 7: 460–7. [46] Miltenburg AJM, Simoons ML, Veerhoek RJ, Bossuyt[28] Sculpher MJ, Seed P, Henderson RA et al. Health service costs of coronary angioplasty and coronary artery bypass PMM. Incidence and follow up of Braunwald subgroups in surgery: the Randomized Intervention Treatment of Angina unstable angina pectoris. J Am Coll Cardiol 1995; 25: (RITA) trial. Lancet 1994; 344: 927–30. 1286–92. [47] Calvin JE, Klein LW, VandenBerg BJ et al. Risk strati-[29] Hamm CW, Berger J, Kalmar P. Coronary angioplasty fication in unstable angina: Prospective validation of the compared with bypass grafting. N Engl J Med 1995; 332: Braunwald classification. JAMA 1995; 273: 136–41. 889. [48] Serruys PW, Herrman JPR, Simon R et al. on behalf of[30] Rupprecht HJ, Hamm C, Ischinger T, Dietz U, Reimers J, the HELVETICA study group. A comparison of hirudin Meyer J, on behalf of the GABI study group. Angiographic with heparin in the prevention of restenosis after coronary follow-up results of a randomized study on angioplasty angioplasty. Helvetica Investigators. N Engl J Med 1995; versus bypass surgery (GABI trial). Eur Heart J 1996; 17: 333: 757–63. 1192–8. [49] Ruygrok PN, de Jaegere PPT, Verploegh J et al. Immediate[31] Campeau L, Enjalbert M, Lesperance J et al. Atherosclerosis outcome following coronary angioplasty. A contemporary and late closure of aortocoronary saphenous vein grafts: single center audit. Eur Heart J 1995; 16 (Suppl L): 24–9. sequential angiographic studies at 2 weeks, 1 year, 5 to 7 [50] Gersh BJ, Braunwald E, Rutherford JD. Chronic coronary years, and 10 to 12 years after surgery. Circulation 1983; 68 heart disease. In: Braunwald E, ed. Heart disease: A textbook (Suppl II): 1–7. of cardiovascular medicine. Philadelphia: W.B. Saunders Co.[32] Kroncke GM, Kosolcharoen P, Clayman JA et al. Five-year 1997: 1289–1365. changes in coronary arteries of medical and surgical patients [51] Boden WE, O’Rourke RA, Crawford MH et al. for the of the Veterans Administration randomized study of bypass Veteran Affairs Non-Q Wave Infarction Strategies in Hospi- surgery. Circulation 1988; 78 (Suppl I): 144–50. tal (VANQUISH) Trial Investigators. Outcome in patients[33] Solomon AJ, Gersh BJ. Management of chronic stable with acute non Q wave myocardial infarction randomly angina: medical therapy, percutaneous transluminal cor- assigned to an invasive as compared with a conservative onary angioplasty, and coronary artery bypass graft surgery: management strategy. N Engl J Med 1998; 338: 1785–92. Lessons from the randomized trials. Ann Intern Med 1998; [52] Wallentin L, Lagerqvist B, Husted S,. Kontny F, Stahle E, 128: 216–23. Swahn E. Outcome at 1 year after an invasive compared with[34] Faxon DP, Ghailli K, Jacobs AK et al. The degree of a non-invasive strategy in unstable coronary-artery disease: revascularization and outcome after multivessel coronary the FRISC II invasive randomized trial. FRISC II Investi- angioplasty. Am Heart J 1992; 123: 854–9. gators. Fast Revascularisation during Instability in Coronary[35] Mock MB, Ringqvist I, Fisher LD et al. Survival of medi- artery disease. Lancet 2000; 356: 9–16. cally treated patients in the coronary artery surgery study [53] de Feyter PJ. Percutaneous coronary intervention for un- (CASS) registry. Circulation 1982; 66: 562–8. stable angina. In: Topol EJ, ed. Textbook of interventional[36] Bell MR, Gersh BJ, Schaff HV et al. Effect of completeness cardiology. Philadelphia: W.B. Saunders Co., 1999: 238–55. of revascularization on long-term outcome of patients with [54] The global use of strategies to open occluded coronary three-vessel disease undergoing coronary artery bypass sur- arteries in acute coronary syndromes (GUSTO IIb) angi- gery. A report from the Coronary Artery Surgery Study oplasty substudy investigators. A clinical trial comparing (CASS) Registry. Circulation 1992; 86: 446–57. primary coronary angioplasty with tissue plasminogen acti-[37] Lawrie GM, Morris GC, Silvers A et al. The influence of vator for acute myocardial infarction. N Engl J Med 1997; residual disease after coronary bypass on the 5-year survival 336: 1621–28. rate of 1,274 men with coronary artery disease. Circulation [55] Stamler J, Vaccaro O, Neaton JD, Wentworth D. Diabetes, 1982; 66: 717–23. other risk factors and 12-year cardiovascular mortality for[38] Bourassa MG, Holubkov R, Yeh W et al. Strategy of men screened in the multiple risk factor intervention trial. complete revascularization in patients with multivessel Diabetes Care 1993; 16: 434–44. Eur Heart J, Vol. 22, issue 14, July 2001
  14. 14. 1196 C. V. Patil et al. [56] Stein B, Weintraub WS, Gebhart S et al. Short and long infarction (the ULTIMA experience). Am J Cardiol 1999; 83: term outcome of diabetic patients undergoing coronary 1513–17. angioplasty. Circulation 1995; 91: 979–89. [76] Chauhan A, Zubaid M, Ricci DR et al. Left main interven- [57] Kip KE, Faxon DP, Detre KM et al. for the Investigators of tion revisited: early and late outcome of PTCA and stenting. the NHLBI PTCA Registry. Coronary angioplasty in dia- Cathet Cardiovasc Diagn 1997; 41: 21–9. betic patients: the National Heart, Lung, and Blood Institute [77] Nakanishi K, Oba O, Schichijo T, Nakai M, Sudo T, Kimura Percutaneous Transluminal Coronary Angioplasty Registry. K. Study on risk factors and late results of coronary artery Circulation 1996; 94: 1818–25. bypass grafting for acute myocardial infarction. Nippon [58] Weintraub WS, Stein B, Gebhart SP, Craver JM, Jones EL, Kyobu Geka Gakkai Zasshi-J Jpn Assoc Thorac Surg 1997; Guyton RA. The impact of diabetes on the initial and long 45: 950–7. term outcomes of coronary artery bypass surgery (Abstr). [78] Quigley RL, Milano CAy, Smith LR, White WD, Rankin JS, 1995 92 Suppl I: I-643. Glower DD. Prognosis and management of anterolateral [59] O’Neill WW. Multivessel balloon angioplasty should be myocardial infarction in patients with severe left main dis- abandoned in diabetic patients! (Editorial comment). J Am ease and cardiogenic shock. The left main shock syndrome. Coll Cardiol 1998; 31: 20–2. Circulation 1993; 88: II 65–70. [60] Weintraub WS, Koisinski A, Culler S. Comparison of out- [79] Park SJ, Park SW, Hong MK et al. Stenting of unprotected come after coronary angioplasty and coronary surgery for left main coronary artery stenoses: immediate and late out- multivessel coronary artery disease in persons with diabetes. comes. J Am Coll Cardiol 1998; 31: 37–42. Am Heart J 1999; 138: S394–S399. [80] Fajadet J, Brunel P, Jordan C, Cassagneau B, Marco J. [61] Barsness GW, Peterson ED, Ohman EM et al. Relationship Stenting of unprotected left main coronary artery stenosis between diabetes mellitus and long-term survival after cor- without coumadin. J Am Coll Cardiol 1966; 27: 277A. onary bypass and angioplasty. Circulation 1997; 96: 2551–6. [81] Silvestri M, Barragan P, Simeoni JP, Roquebert PO, Bayet [62] Brinker JA. ‘Limitless’ contrast for coronary angiography? G, Sainsous J. Unprotected left main stenting: preliminary (Editorial comment). Cathet Cardiovasc Diagno 1997; 40: results and follow-up with first 41 patients. J Am Coll 247–8. Cardiol 1997; 29: 15A. [63] Serruys PW, Van Hout B, Bonnier H et al. for the Benestent [82] Fishman DL, Leon MB, Baim DS et al. A randomized study group. Randomized comparison of implantation of comparison of coronary-stent placement and balloon angi- heparin-coated stents with balloon angioplasty in selected oplasty in the treatment of coronary artery disease. N Engl J patients with coronary artery disease (Benestent II). Lancet Med 1994; 331: 496–501. 1998; 352: 673–81. [83] Serruys PW, de Jaegere P, Kiemeneij F et al. for the [64] Serruys PW, Unger F,. van Hout BA et al. on behalf of the BENESTENT study group. A comparison of balloon ARTS study group. The ARTS (Arterial Revascularization expandable-stent implantation with balloon angioplasty in Therapies Study): Backgrounds, goals and methods. Int J patients with coronary artery disease. N Engl J Med 1994; Cardiovasc Intervent 1999; 2: 41–50. 331: 489–95. [65] Nikolsky E, Halabi M, Roguin A et al. Staged versus [84] Savage MP, Fishman DL, Rake L et al. Efficacy of coronary one-step approach for multivessel percutaneous coronary stenting versus balloon angioplasty in small coronary interventions (Abstr). Am J Cardiol 2000; 86: 125. arteries. J Am Coll Cardiol 1998; 31: 307–11. [66] Foley DP. Multivessel stenting: staged vs non-staged [85] Akiyama T, Moussa I, Reimers B et al. Angiographic and approach: Perspective: Thoraxcentre, Rotterdam, The clinical outcome following coronary stenting of small vessels: Netherlands. Int J Cardiovasc Interventions 1999; 2: 71–8. A comparison with coronary stenting of large vessels. J Am [67] Hlatky MA, Rogers WJ, Johnstone I et al. for the BARI Coll Cardiol 1998; 32: 1610–18. Investigators. Medical care costs and quality of life after [86] Savage MP, Douglas JS, Fishman DL et al. A randomized randomization to coronary angioplasty or coronary bypass trial of coronary stenting and balloon angioplasty in the surgery. N Engl J Med 1997; 336: 92–9. treatment of aortocoronary saphenous vein bypass graft [68] Hlatky MA, Boothroyd DB, Brooks MM et al. Clinical disease. N Engl J Med 1997; 337: 740–7. correlates of the initial and long-term cost of coronary [87] Erbel R, Haude M, Hopp HW et al. Coronary-artery stent- bypass surgery and coronary angioplasty. Am Heart J 1999; ing compared with balloon angioplasty for restenosis after 138: 376–83. initial balloon angioplasty. Restenosis Stent Study Group. [69] Cohen DJ, Krumholz HM, Sukin CA et al. In hospital and N Engl J Med 1998; 339: 1672–8. one-year economic outcomes after coronary stenting or [88] Sirnes PA, Golf S, Myreng Y et al. Stenting in Chronic balloon angioplasty. Circulation 1995; 92: 2480–7. Coronary Occlusion (SICCO): A randomized, controlled [70] Ryan TJ, Faxon DP, Gunnar RM et al. Guidelines for trial of adding stent implantation after successful angi- percutaneous transluminal coronary angioplasty: a report of oplasty. J Am Coll Cardiol 1996; 28: 1444–51. the American College of Cardiology/American Heart Associ- [89] Stone GW, Brodie BR, Griffin JJ et al. Prospective, multi- ation task force on the assessment of diagnostic and center study of the safety and feasibility of primary stenting therapeutic cardiovascular procedures. Circulation 1988; 78: in acute myocardial infarction: in-hospital and 30-day results 486–502. of the PAMI stent pilot trial. Primary Angioplasty in Myo- [71] Wong P, Wong V, Tse KK et al. A prospective study of cardial Infarction Stent Pilot Trial Investigators. J Am Coll elective stenting in unprotected left main coronary disease. Cardiol 1998; 31: 23–30. Cathet Cardiovasc Intervent 1999; 46: 153–9. [90] Grines CL, Cox DA, Stone GW et al. Coronary angioplasty [72] Kent KM, Bentivoglio LG, Block PC et al. Percutaneous with or without stent implantation for acute myocardial transluminal coronary angioplasty: Report from the Registry infarction. Stent Primary Angioplasty in Myocardial Infor- of the National Heart, Lung, and Blood Institute. Am J mation Study Group. N Engl J Med 1999; 341: 1949–56. Cardiol 1982; 49: 2011–20. [91] Jang IK, Goldberg S. Stents: Indications and limitations. [73] O’Keefe JH Jr, Hartzler GO, Rutherford BD et al. Left main In: Topol EJ, ed. Textbook of interventional cardiology. coronary angioplasty: Early and late results of 127 acute and Philadelphia: W. B. Saunders Co., 1999: 602–13. elective procedures. Am J Cardiol 1989; 64: 144–7. [92] Moussa I, Reimers B, Moses J et al. Long-term angiographic [74] Ellis SG, Tamai H, Nobuyoshi M et al. Contemporary and clinical outcome of patients undergoing multivessel percutaneous treatment of unprotected left main coronary coronary stenting. Circulation 1997; 96: 3873–9. stenoses: initial results from a multicenter registry analysis [93] Laham RJ, Ho KKL, Baim DS, Kuntz RE, Cohen DJ, 1994–1996. Circulation 1997; 96: 3867–72. Carrozza JP Jr. Multivessel Palmaz-Schatz stenting: Early [75] Marso SP, Steg G, Plokker et al. Catheter-based reperfusion results and one-year outcome. J Am Coll Cardiol 1997; 30: of unprotected left main stenosis during an acute myocardial 180–5.Eur Heart J, Vol. 22, issue 14, July 2001