Coronary angioplasty versus bypass grafting: Cost-Benefit considerations

CONTEMPORARY ISSUES IN CARDIOLOGY

0025-7125/95 $0.00 + .20

CORONARY ANGIOPLASTY VERSUS BYPASS GRAFTING Cost-Benefit Considerations Debra L. Sherman, MD, and Thomas J. Ryan, MD

Coronary heart disease strikes 1.5 million people per year and is the leading cause of death in the United States, killing 923,000 per year. Because of earlier detection and better treatment for ischemic heart disease, the death rate from myocardial infarction (MI) has decreased by 32.4% from 1981 to 1991. ' Despite the declining mortality rate, the number of revascularization procedures has increased, likely reflecting the increased survival rate after MI. More than half of all revascularization procedures performed in the world are done in the United States. In 1988, 381,500 percutaneous transluminal coronary angioplasties (PTCAs) and 345,000 coronary artery bypass grafts (CABGs) were done in the United States compared with 181,000 PTCAs and 255,000 CABGs in the remainder of the world. 23 The choice of procedures is usually left up to the discretion of the cardiologist and cardiothoracic surgeon and is based on the anatomy of the coronary obstructions, patient preferences, and various patient characteristics. As the developed world becomes increasingly aware that medical resources are finite, there is a universal need for physicians, even in the United States, to consider the cost versus the benefit for these procedures. This article focuses on the overall cost of revascularization in the United States today, the comparative costs of PTCA versus CABG, the benefits of PTCA versus CABG, and the implications for revascularization in the 1990s. MEDICAL ECONOMICS

Comparison of medical costs is the most accurate way to evaluate the resources expended for different procedures. Costs are divided, however, among

From the Evans Memorial Department of Clinical Research, Boston University Medical Center; and the Departments of Cardiology (DLS) and Medicine (TJR), Boston University School of Medicine, Boston, Massachusetts MEDICAL CLINICS OF NORTH AMERICA VOLUME 79· NUMBER 5· SEPTEMBER 1995

1085

1086

SHERMAN & RYAN

multiple resources, including hospital personnel, supplies, and medical facilities, making it difficult to calculate the actual cost for a specific procedure. Information regarding medical charges is more easily obtainable. Comparison of charges, however, is an inaccurate way of assessing the consumption of resources. Charges do not reflect true hospital costs because they not only include hospital costs, but also take into account free care for the uninsured, uncollected fees from third-party payers, purchase of new equipment, upkeep of existing equipment, and allocation for additional services. 19 Other important considerations when comparing medical costs include the particular time period studied as well as the geographic location. Costs from differing years cannot be compared directly owing to inflation. The hospital setting is also important because costs tend to be higher in academic centers than private community hospitals because house staff account for an additional expense. Costs also may vary depending on the region of the United States, secondary to differing costs for labor and supplies. 19 Cost-effectiveness and cost-benefit are methods of economic analysis. The more commonly used cost-effectiveness analysis is the ratio of costs to benefits or outcomes. Typically the change in life expectancy for different treatments is considered the benefit and is expressed as dollars per year of life saved. Costbenefit analysis is used less commonly and requires the conversion of benefits to a monetary value. The two methods for accomplishing this are the humancapital approach and the willingness-to-pay approach. With the human capital approach, life expectancy is converted into one's contribution to society over that period of time. It is usually assessed in the form of wages, which has the obvious disadvantage of underestimating the value of retired people, homemakers, and those who do not return to employment. With the willingness-to-pay approach, the amount an individual is willing to pay for a given benefit is assessed. The drawback of this method is that those who have more money are usually willing to pay more for a particular benefit. Although these methods provide a means of economic comparison between different forms of treatment, they all have serious medical and social shortcomings, making them difficult to incorporate into the practice of medicine. 19 COST OF REVASCULARIZATION Overall Cost of Revascularization in the United States

With the substantial increase in success rates of PTCA from 65% in the 1970s to 95% in the 1990s,2l the number of interventional procedures performed has dramatically risen from 10,000 in 1981 to almost 400,000 in 1992." In addition, the number of interventional cardiologists has increased from 2000 in 1985 to more than 8000 in 1992.31 In 1990, the United States spent $110 billion for cardiovascular diseases 6 and were projected to spend $128 billion in 1994.2 PTCA is responsible for approximately $8 billion per year, with restenosis costing approximately $2 billion per year. 32 Initial Cost of Revascularization

The initial cost of PTCA is less than CABG and has been verified in many studies. In one of the first comparisons, Black and coworkers3 reviewed costs for CABG and PTCA in 200 patients with multivessel coronary artery disease.

CORONARY ANGIOPLASTY VERSUS BYPASS GRAFTING

1087

More patients in the CABG group had three-vessel disease than in the PTCA group. The mean cost, which included hospital costs and physician fees, was 60% less in the PTCA group than in the CABG group. One major limitation of this study was that costs were calculated from hospital and physician charges, which might not accurately reflect true costs. Cohen and associates 4 compared costs and charges of PTCA, newer interventional techniques, and CABG. In contrast to the previous study, physician costs were not included in the analysis. Once again, more of the CABG patients had multivessel disease. The mean cost of PTCA was found to be 75% less than that of CABG. In the Argentine randomized trial of PTCA versus CABG in multivessel disease, Rodriguez and coworkers25 found that the cost for PTCA was 66% less for both complicated and uncomplicated procedures than that of CABG. In contrast to the previous two studies, there was no difference in extent of coronary disease between the two groups. The lower initial cost for PTCA appears to be due to lower procedural and physician costs; shorter hospital stay; fewer days in intensive care; and less use of ancillary services, including laboratory, blood bank, pharmacy, and radiology.3,4 Cost of Revascularization at One Year

Unfortunately, 1 to 6 months after PTCA, approximately one third of patients have restenosis of the dilated artery and often require repeat revascularization procedures for recurrent angina. Approximately 12% to 20% of vein grafts are occluded after 1 year,!! but virtually all internal mammary arteries are patent. Therefore, more patients undergoing PTCA require repeat revascularization procedures in the first year than patients who undergo CABG, resulting in additional costs for the PTCA patients. Despite the higher number of procedures in the PTCA patients, the total cost of PTCA is still less than CABG at 1 year. Black and coworkers 3 found that PTCA was 52% less costly than CABG at 1 year despite the need for repeat revascularization in 18% of the PTCA patients (CABG in 3 and PTCA in 15) compared with 1% of the CABG patients (PTCA in 1). In the Argentine trial of coronary angioplasty versus CABG (ERACI), Rodriguez and coworkers25 found at 1 year 20 of 63 patients in the PTCA group required repeat revascularization (PTCA in 9 and CABG in 11) compared with 2 of 64 patients in the CABG group (PTCA in both). The cost of PTCA for the first year including the initial procedure and complications, however, was nearly half that of CABG. Van den Brand and associates 33 came to similar conclusions after following 1937 patients who had either CABG or PTCA. Again, there were many more patients in the PTCA group who had single-vessel disease compared with the CABG group (91% versus 18%). After 1 year, the cost of PTCA was 36% less than CABG despite the need for repeat revascularization in 29% of PTCA patients compared with 2% of CABG patients. Lubitz and colleagues!8 examined the rate of rehospitalization in Medicare patients undergoing PTCA or CABG during 1986 and 1987. The I-year rehospitalization rate per 1000 was 629 in the CABG group and 863 in the PTCA group compared with 607 for the overall population. Although further hospitalization may be secondary to other concomitant illness in this elderly population, approximately 45% of the CABG patients and 66% of the PTCA patients required rehospitalization for reasons relating to the initial procedure. Nineteen percent of CABG patients and 34.6% of PTCA patients were rehospitalized within 1 year of the procedure for a cardiac-related event defined as a cardiac procedure

1088

SHERMAN & RYAN

(cardiac catheterization, PTCA, or CABG), ischemic heart disease, arrhythmia, congestive heart failure, infection related to the procedure, and vascular events. In the year following the procedure, Medicare payments for rehospitalization for cardiac-related events were $1207 per patient in the CABG group and $3508 per patient in the PTCA group. The payments for the initial procedure, however, were more than twice as much for CABG compared with PTCA ($19,776 versus $8827). Long-Term Cost of Revascularization

Restenosis occurs again in approximately 30% of patients who undergo repeat PTCA, usually within 6 months of the procedure, after which the restenosis rate is approximately 0.7% per year. 36 By comparison, 12% to 20% of saphenous vein grafts are obstructed by 1 year after CABG, and the patency declines by 0.2% per year thereafter. By 10 to 12 years post CABG, nearly 40% to 60% of saphenous vein grafts are occluded. 26 Weintraub and colleagues35 followed patients with symptomatic multivessel disease who were randomized to revascularization with CABG or PTCA. Baseline characteristics between the two groups were similar. After 3 years, there was no difference in mortality, myocardial infarction, or amount of ischemic myocardium. Only 13% of the CABG patients, however, required an additional revascularization procedure, compared with 55% of the PTCA patients. The initial hospital cost was 44% lower in the PTCA group, but after 3 years there was no significant difference in the total costs between the two groups owing to the need for repeat interventions in the PTCA patients for recurrent angina. The Bypass Angioplasty Revascularization Intervention Substudy of Economics and Quality of Life (BARI SEQOL) is a 5year study that analyzes the short-term and long-term costs of PTCA and CABG in patients with multivessel disease. Results will be available in 1996. BENEFIT OF PERCUTANEOUS TRANSLUMINAL CORONARY ANGIOPLASTY VERSUS CORONARY ARTERY BYPASS GRAFT

Until the introduction of PTCA in 1977, CABG was the only option available for revascularization in patients with symptomatic coronary artery disease (CAD). Both CABG and PTCA are effective for relieving angina; however, the choice of procedure varies in different patient groups depending on coronary anatomy, ejection fraction, other comorbid illnesses, and patient preferences. Understanding the advantages and disadvantages of each procedure is crucial when deciding on treatment strategies for different patients. Established Benefits of Percutaneous Transluminal Coronary Angioplasty

The main advantage of PTCA is that it is much less invasive than CABG. Revascularization is achieved percutaneously with catheters introduced through the femoral or brachial artery. There is no need for general anesthesia, endotracheal intubation, cardiopulmonary bypass, or sternotomy, resulting in shorter hospital stays (approximately 3 days for PTCA versus 9 days for CABG) and less time in intensive care (approximately 1 day for PTCA versus 3 days for

CORONARY ANGIOPLASTY VERSUS BYPASS GRAFTING

1089

CABe).4 The success rate for PTCA is high and averages approximately 89% for all lesions. In addition, the risks are low, with a mortality rate of less than 1% and an MI rate of approximately 3.7%, with an equal number requiring emergent CABe for abrupt vessel closure. 36 The major disadvantages of PTCA include the 30% incidence of restenosis, the higher rates of incomplete revascularization compared with CABe owing to difficulty in accessing certain lesions, low success rates for dilating chronically totally occluded vessels, and the possibility of worsening 50% to 70% stenoses because of future restenosis. 9 In one of the first trials to compare PTCA to medical therapy for singlevessel disease, Parisi and colleagues22 followed 212 patients randomized to either strategy. At 6 months, 64% of PTCA patients versus 46% of medically treated patients were free of angina (P
CABe is a far more invasive procedure than PTCA and entails physically bypassing obstructive coronary lesions using saphenous veins or internal mammary arteries as conduits. The heart is accessed through a midsternotomy incision, and cardiopulmonary bypass and cardioplegia are required. The shortterm risks are higher than that of PTCA with a mortality rate of 2.3% and a postoperative MI rate of 5.8%.36 In contrast to PTCA, the long-term benefits of CABe for prolonging life have been well established for certain groups of patients, which include those with left-main CAD and those with three-vessel CAD and a reduced ejection fraction.'4, 15, 34 The Veterans Administration Cooperative Study was one of the first to compare surgical with medical treatment of CAD in men with stable angina. They found a mortality benefit with surgery at 5 to 7 years in patients with leftmain CAD, impaired left ventricular function, or three-vessel CAD. In addition, several clinical parameters were associated with increased mortality in the group treated medically, including class III or IV angina, hypertension, previous MI, and ST-segment depression on electrocardiogram. 34 The Coronary Artery Surgery Study (CASS) compared surgical with medical treatment in patients with mild angina or a history of MI. A significant survival advantage with surgery was noted in patients with impaired ejection fraction and three-vessel CAD.14, 15 In addition, patients with severe angina had better relief of symptoms and improved exercise tolerance compared with the group treated medically.27 Comparison of Coronary Artery Bypass Graft and Percutaneous Transluminal Coronary Angioplasty for Multivessel Coronary Artery Disease

Many randomized trials comparing the efficacy of PTCA and CABe in patients with multivessel CAD have been reported, The Randomised Intervention Treatment of Angina (RITA) triaP4 followed 1011 patients with one-vessel, two-vessel, or three-vessel CAD who were revascularized with either PTCA or CABe. It is the largest of the randomized trials reported to date and the only one to include patients with single-vessel CAD. The CABe patients required

1090

SHERMAN & RYAN

longer initial hospitalization and recuperation. After 2.5 years of follow-up, however, there was no difference in death or nonfatal MI between the two groups. Within 2 years of randomization, approximately 38% of the PTCA patients required repeat revascularization, suffered a MI, or died compared with 11% of the CABG patients (P <0.001). In addition, 21.5% of the CABG patients versus 31.3% of the PTCA patients experienced angina (P <0.007) and required antianginal medications more often than the CABG patients. No difference in physical activity or employment status was found at 2 years. These patients will be followed for a total of 5 years. The Argentine Randomized Trial of Percutaneous Transluminal Coronary Angioplasty Versus Coronary Artery Bypass Surgery in Multivessel Disease (ERACI) triaPS reported on 127 patients with two-vessel or three-vessel CAD randomized to PTCA or CABG. After 1 year, there was no difference in death or nonfatal MI between the two groups. More patients in the PTCA group, however, suffered from angina and required repeat revascularization compared with the CABG group (32% versus 3.2%, P <0.001). Results from 3- and 5-year follow-up will be reported. The German Angioplasty Bypass Surgery Investigators (GABI)'2 issued a report comparing PTCA and CABG in 359 randomized patients with multivessel CAD followed for 1 year. In contrast to the previous studies, these investigators found a significantly higher risk of death and MI in the CABG group (13.6%) compared with the PTCA group (6%), possibly owing to the longer period of time between randomization and treatment in the surgical patients. In addition, there was no difference in freedom from angina or improvement in exercise capacity between the two groups. The major difference was the need for future revascularization. Forty-four percent of the PTCA patients required further interventions as opposed to 6% of the CABG patients, and patients undergoing PTCA took antianginal medications more often than patients undergoing CABG at both 6 and 12 months. Investigators from the Emory Angioplasty Versus Surgery Trial (EAST)'1i also reported their 3-year experience of PTCA versus CABG in 392 randomized patients with multivessel disease. They found no difference in primary end points, which included death, Q-wave MI, and ischemic defect on thallium scan. As with the other studies, the major difference between the two groups was the need for further revascularization procedures. Only 1% of the CABG patients required additional surgery compared with 22% of the PTCA patients. Those undergoing subsequent PTCA included 13% of CABG patients compared with 41 % of PTCA patients. In addition, more PTCA patients suffered from angina and required antianginal medications at 3 years compared with patients initially undergoing CABG. There was no difference in physical activity and employment after 3 years. Mark and coworkers20 published a prospective, nonrandomized study comparing medical treatment, PTCA, and CABG in 9263 patients using the Duke Cardiovascular Disease Databank. The purpose was to compare 5-year survival in patients with ischemic heart disease who underwent different treatment modalities. In patients with one-vessel disease, there was a slight trend toward improved survival with PTCA (95%) compared with CABG (93%) and medical therapy (94%). With two-vessel disease, there was no difference in survival between the PTCA (91%) and CABG (91%) groups. PTCA and CABG, however, offered a survival advantage compared with medical therapy (86%). For threevessel disease, patients who underwent CABG had improved survival (89%) compared with PTCA (81%) and medical therapy (72%). Because prognosis of CAD depends not only on the number of vessels with significant stenoses, but

CORONARY ANGIOPLASTY VERSUS BYPASS GRAFTING

1091

also on their anatomic locations, they calculated mortality ratios based on different levels of CAD severity. For patients with one-vessel disease (not involving left anterior descending [LAD] artery) and less severe two-vessel disease (no LAD involvement), PTCA and medical therapy appeared to be superior to CABe, with PTCA showing a slight benefit over medical therapy. With more severe one-vessel disease and two-vessel CAD (both involving LAD lesions), CABe and PTCA were equivalent in terms of survival at 5 years. Both CABe and PTCA were slightly better than medical therapy. With severe two-vessel disease (involving >95% proximal LAD) and three-vessel disease, 5-year survival in patients undergoing CA Be was far superior to PTCA and medical therapy. Mortality with PTCA and medical therapy was similar in this group. This study supports previous studies, in which patients with severe two-vessel disease and three-vessel disease have improved survival with CABe compared with medical therapy. With less severe CAD, there was a small, statistically insignificant benefit in terms of survival with PTCA. Therefore, the treatment decision in these patients should take into account symptoms, quality of life, and patient preferences. Complete Compared with Incomplete Revascularization

Surgical studiesl3, 28 have shown that patients who were completely revascularized had lower mortality rates, less angina, and better functional status 5 to 10 years after surgery compared with those who were incompletely revascularized. With PTCA, there is the higher incidence of incomplete revascularization. The reasons for incomplete revascularization, however, may be different for the two procedures. With PTCA, certain lesions are purposely not dilated because they are inaccessible, totally occluded, only mildly stenotic «60% obstructed), or supplying a nonviable segment of myocardium. This strategy is undertaken because complications, such as abrupt closure and vessel dissection, increase as more vessels are dilated. In addition, insignificant lesions may become significant if restenosis occurs. With CABe, vessels may be left ungrafted if there is diffuse coronary disease with small distal vessels or if the vessel supplies a segment of nonviable myocardium. Faxon and associates 7 studied immediate and I-year outcome of patients undergoing PTCA for multivessel CAD who were incompletely revascularized. A patient was considered adequately revascularized if all lesions to viable myocardium were dilated. Inadequate revascularization resulted when arteries with significant stenoses serving viable myocardium were not dilated. Incomplete adequate revascularization was compared with incomplete inadequate revascularization. After 1 year, there were more cardiac events and coronary bypass operations in those who had incomplete inadequate revascularization compared with those who had incomplete adequate revascularization. In the EAST trial,16 99.1% of coronary stenoses per patient in the CABe group were revascularized at initial presentation versus 75.1% in the PTCA group. At 1 and 3 years, the percentage of revascularized segments per patient dropped to 88.1% and 86.7% in the CABe group and 58.8% and 69.9% in the PTCA group. These data indicate that at all times during this study CABe patients received more complete revascularization than the PTCA patients. There was no difference in death and MI between the two groups. The investigators, however, did not compare adequate versus inadequate revascularization. In the ERACI study,>5 complete revascularization was accomplished more often in the CABe group (88%) than the PTCA group (51%). Complete adequate

1092

SHERMAN & RYAN

revascularization was achieved in 89.2% of the PTCA patients. Vessels not dilated were those with chronic total occlusions or those supplying infarcted myocardium. They found no difference in mortality and MI after 1 year between the two groups. In addition, there was no difference in the frequency of angina at 6 months between PTCA patients who underwent complete anatomic revascularization versus those who had incomplete adequate revascularization. IMPLICATIONS FOR TREATMENT

Both PTCA and CABG are effective treatments for relieving angina, but each has different short-term and long-term benefits and risks, which have already been discussed in detail. There are also considerable cost differences between the two procedures, with PTCA costing less for the initial procedure and at 1 year but not significantly different from CABG at 3 years owing to the need for repeat revascularization in the PTCA group. Although PTCA initially costs less than CABG, angioplasty is more expensive than medical treatment. Increasingly, patients are treated with PTCA who in the past would have been treated medically.2o Which treatment is most cost-effective depends on many variables. Single-Vessel Coronary Artery Disease

For symptomatic single-vessel CAD, few would argue that PTCA or medical therapy is the treatment of choice, given the good long-term prognosis of these patients and effectiveness of the treatment strategies. Although more expensive than medical therapy, PTCA has been shown to be more effective for relieving angina and increasing exercise capacity in patients with single-vessel disease. 22 The cost-effectiveness ratio (incremental cost per additional life-year saved) has been calculated at $75,000 when compared with medical treatment for singlevessel LAD disease, mild angina, and normal left ventricular function. 19 In addition, patients with single-vessel disease not involving the proximal LAD have a lower mortality when treated medically or with PTCA compared with CABG.20 The choice of PTCA or medical treatment for patients with singlevessel CAD should depend on relief of symptoms, patient preference, and quality-of-life issues. Multivessel Coronary Artery Disease

The treatment for multivessel CAD is more complicated than single-vessel CAD. Clearly there are patients who have a survival advantage with surgery, such as those with three-vessel CAD and diminished ejection fraction and those with left-main stenosis. Many patients with multivessel CAD, however, can be treated effectively with either CABG or PTCA with no difference in death and MI rate after 1 to 3 years of follow-up.12, 16, 24, 25 The incremental cost per additional life-year saved for CABG versus medical therapy suggests that surgery is more cost-effective in patients with left-main and three-vessel disease than in patients with two-vessel disease. The cost-effectiveness ratios for multivessel CAD are as follows: severe angina and left-main disease, $7000; severe angina and three-vessel disease, $13,000; and mild angina and two-

CORONARY ANGIOPLASTY VERSUS BYPASS GRAFTING

1093

vessel disease, $55,000. 10 Thus, as coronary disease becomes more severe, CABe becomes more cost-effective. Because approximately 50% of saphenous vein grafts are occluded 10 years after CABe, many surgical patients eventually require revascularization. 26 Consequently, PTCA is sometimes used in an attempt to delay surgery. French and associates 'O found that this approach may be justified. They studied the longterm efficacy of CABe in patients less than 40 years of age and found that the median time to recurrent angina or MI was 6 years, revascularization or cardiac transplant was 9.6 years, and late death was 14 years. The investigators concluded that the long-term outcome in young patients undergoing CABe was poor. The use of internal mammary arteries as conduits, aspirin, and lipidlowering therapy improved long-term mortality. PTCA may also be justified in certain high-risk groups, such as the elderly or patients with prior bypass surgery for relief of angina in those who have failed medical therapy. Strain and coworkers29 compared CABe with PTCA in patients over the age of 80 with three-vessel disease. Patients in the PTCA group had less complete revascularization with 1.35 ± 0.5 dilated compared with 3.5 ± 0.8 bypassed vessels in the surgical group (P = 0.01). Short-term outcomes were similar in each group with a significantly shorter length of stay in the PTCA group. Dahiya and coworkers5 examined the issue of revascularization in patients with prior CABe. They compared PTCA with reoperation in patients matched for age, ejection fraction, and extent of CAD. Patients treated with PTCA had improved survival at 1 year (94% PTCA versus 81% CABe, P = 0.001) and at 6 years (74% PTCA versus 68% CABe, P = 0.001). The higher mortality in the CABe group may have been due to a higher in-hospital mortality rate after repeat CABe (11.1%) compared with PTCA (0.6%) (P <0.001). There was no difference in angina in the two groups; however, the PTCA patients required more repeat revascularization procedures at 5 years (69% PTCA versus 10% CABe, P <0.001). One way to improve bypass graft patency rates long-term is to use either one or both internal mammary arteries. These vessels appear to be resistant to the intimal hyperplasia and atherosclerosis seen in vein grafts. At 10 to 12 years postsurgery, patency rates have been found to exceed 90%. In addition, patients who have an internal mammary artery grafted to their LAD have a 35% lower mortality than patients who receive only vein grafts. 17 The main disadvantage of using internal mammary arteries for grafting is the longer operating time required for mobilizing the vessel. As a result, it is not usually used for emergent surgery. SUMMARY

Both PTCA and CABe are effective strategies for coronary revascularization. The initial cost of PTCA is 60% to 75% less than that of CABe. PTCA patients, however, often require repeat procedures secondary to restenosis and incomplete revascularization. Despite this, the cost of PTCA is still approximately half that of CABe at 1 year but approaches that of CABe at 3 years. The BARI SEQOL trial will be available in 1996 and will analyze cost differences as well as quality of life for PTCA versus CABe up to 5 years after revascularization. Patients with single-vessel disease can be treated effectively with PTCA or medications. Although PTCA is more expensive, patients have less angina and better exercise tolerance. Many patients with single-vessel disease are now treated with PTCA, who in the past would have been treated medically. Un-

1094

SHERMAN & RYAN

doubtedly, this change has added to the increasing cost of health care. Although certain patient groups, such as those with three-vessel disease and low ejection fraction and left-main disease, have a significant mortality advantage when revascularized surgically, many patients with symptomatic two-vessel and threevessel disease can be treated either with CABG or PTCA with no difference in mortality and MI. To reach this equivalent outcome, however, PTCA patients require more interventional procedures. As a result, at 3 years, there is no cost savings with PTCA. Physicians in the United States have been able to choose the mode of revascularization for patients based on clinical judgment and preference, which has been financed by third-party payers. Given the escalating costs of health care in a country with limited resources, physiCians failing to consider costs and benefits may find their choices limited secondary to lack of funding and restrictive policies. Future treatment of CAD will most likely be influenced by aggressive lipid-lowering therapy to prevent secondary cardiac events and possibly by gene therapy to prevent restenosis. References 1. American Heart Association: Heart attack-still America's no. 1 killer, stroke-still America's no. 3 killer: More must be done in funding federal research. Monograph, Dallas, American Heart Association, 1994 2. American Heart Association: Heart and Stroke Facts: 1994 Statistical Supplement. Dallas, American Heart Association, 1994 3. Black AJR, Roubin CS, Sutor C, et al: Comparative costs of percutaneous transluminal coronary angioplasty and coronary artery bypass grafting in multivessel coronary artery disease. Am J Cardiol 62:809-811, 1988 4. Cohen DJ, Breall JA, Ho KKL, et al: Economics of elective coronary revascularization: Comparison of costs and charges for conventional angioplasty, directional atherectomy, stenting and bypass surgery. J Am Coil Cardiol 22:1052-1059, 1993 5. Dahiya RS, O'Keefe JH, Ligon R, et al: PTCA vs re-operation for patients with prior bypass surgery [abstr]. Circulation 90:1-334, 1994 6. Dustan HP, Caplan LR, Curry CL, et al: Report of the task force on the availability of cardiovascular drugs to the medically indigent. Circulation 85:849-860, 1992 7. Faxon DP, Chalilli K, Jacobs AK, et al: The degree of revascularization and outcome after multivessel coronary angioplasty. Am Heart J 123:854-859, 1992 8. Feinleib M, Havlik RJ, Cillum RP, et al: Coronary heart disease and related procedures: National hospital discharge survey data. Circulation 78:13-18, 1989 9. Filart RA, Ryan TJ: Comparison of coronary angioplasty with bypass surgery for multivessel disease. Coronary Artery Disease 4:1039-1047, 1993 10. French JK, Scott DS, Smith WM, et al: Coronary bypass grafting in patients aged <40 years has a poor long-term outcome [abstr]. Circulation 90:1-252, 1994 11. Crondin CM, Campeau L, Thornton JC, et al: Coronary artery bypass grafting with saphenous vein. Circulation 79(suppl 1):24, 1989 12. Hamm CW, Reimers J, Ischinger T, CABI Investigators: A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease (CABI). N Engl J Med 331:1037-1043,1994 13. Jones EL, Craver JM, Cuyton RA, et al: Importance of complete revascularization in performance of the coronary bypass operation. Am J Cardiol 51:7-12, 1983 14. Kaiser Cc, Davis KB, Fisher LD, et al: Survival following coronary artery bypass grafting in patients with severe angina pectoris (CASS). J Thorac Cardiovasc Surg 89:513-524, 1985 15. Killip T, Passamani E, Davis K, and the CASS Principal Investigators and Their Associates: Coronary Artery Surgery Study (CASS): A randomized trial of coronary bypass surgery: Eight years follow-up and survival in patients with reduced ejection fraction. Circulation 72(suppl V):V102-109, 1985 16. King SB, Lembo NJ, Hall EC, East Investigators: A randomized trial comparing

CORONARY ANGIOPLASTY VERSUS BYPASS GRAFTING

17. 18. 19. 20. 21. 22. 23. 24. 25.

26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

1095

coronary angioplasty with coronary bypass surgery (EAST). N Engl J Med 331:10441050,1994 Loop FD, Lytle BW, Cosgrove DM: New arteries for old. Circulation 79(suppl I):40, 1989 Lubitz JD, Gornick ME, Mentnech RM, et al: Rehospitalization after coronary revascularization among Medicare beneficiaries. Am J Cardiol 72:26-30, 1993 Mark DB: Medical economics and health policy issues for interventional cardiology. In Topol EJ (ed): Textbook of Interventional Cardiology, ed 2. Philadelphia, WB Saunders Company, 1994 Mark DB, Nelson CL, Califf RM, et al: Continuing evolution of therapy for coronary artery disease: Initial results from the era of coronary angioplasty. Circulation 89:20152025, 1994 Myler RK: Coronary and peripheral angioplasty: Historical perspective. In Topol EJ (ed): Textbook of Interventional Cardiology, ed 2. Philadelphia, WB Saunders Company, 1994 Parisi AF, Folland ED, Hartigan P, Veterans Affairs ACME Investigators: A comparison of angioplasty with medical therapy in the treatment of single-vessel coronary artery disease. N Engl J Med 326:10-16, 1992 Report of a Working Party of the British Cardiac Society: Coronary angioplasty in the United Kingdom. Br Heart J 66:325-331, 1991 RlT A Trial Participants: Coronary angioplasty versus coronary artery bypass surgery: The randomized intervention treatment of angina triaL Lancet 341:573-580, 1993 Rodriguez A, Boullon F, Perez-Balino N, ERACI Investigators: Argentine randomized trial of percutaneous transluminal coronary angioplasty versus coronary artery bypass surgery in multivessel disease (ERACI): In-hospital results and I-year follow-up. J Am Coli CardioI22:1060-1067, 1993 Rutherford JD, Braunwald E: Chronic ischemic heart disease. In Braunwald E (ed): Heart Disease, ed 4. Philadelphia, WB Saunders, 1992 Ryan Tl, Weiner DA, McCabe CH, et al: Exercise testing in the Coronary Artery Surgery Study randomized population. Circulation 72(suppl V):V31-38, 1985 Schaff HV, Bernard GJ, Pluth JR, et al: Survival and functional status after coronary artery bypass grafting: Results 10-12 years after surgery in 500 patients. Circulation 68(suppl II):II200-204, 1983 Strain JE, Stelzer PE, Subramanian VA, et al: CABG vs PTCA for octogenarians with three vessel disease [abstr]. Circulation 90:I-333, 1994 Topol EJ: Quality of care in interventional cardiology. In Topol EJ (ed): Textbook of Interventional Cardiology, ed 2. Philadelphia, WB Saunders, 1994 Topol EJ, Califf RM: Scorecard cardiovascular medicine: Its impact and future directions. Ann Intern Med 120:65-70, 1994 Topol El, Ellis SG, Cosgrove DM, et al: Analysis of coronary angioplasty practice in the United States with an insurance-claims data base. Circulation 87:1489-1497,1993 van den Brand M, van Halem C, van den Brink F, et al: Comparison of costs of percutaneous transluminal coronary angioplasty and coronary bypass surgery for patients with angina pectoris. Eur Heart J 11:765-771, 1990 Veterans Administration Coronary Artery Bypass Surgery Cooperative Study Group: Eleven-year survival in Veterans Administration randomized trial of coronary bypass surgery for stable angina. N Engl J Med 311:1333-1339,1984 Weintraub WS, Mauldin PD, Becker E, et al: The impact of additional procedures on the cost at three years of coronary angioplasty and coronary surgery in the EAST trial [abstr]. Circulation 90:I-91, 1994 Wong JB, Sonnenberg FA, Salem DN, et al: Myocardial revascularization for chronic stable angina: Analysis of the role of percutaneous transluminal coronary angioplasty based on data available in 1989. Am Coli Phys 113:852-871, 1990

Address reprint requests to Thomas J. Ryan, MD Section of Cardiology Boston University Medical Center 88 E. Newton Street Boston, MA 02118