Has Improvement in PTCA Intervention Affected Long-term Prognosis?
The NHLBI PTCA Registry Experience
Background The NHLBI Percutaneous Transluminal Coronary Angioplasty (PTCA) Registry followed 1345 consecutive patients with first PTCA between 1977 and 1981 (registry 1) and 2136 consecutive patients with PTCA between 1985 and 1986 (registry 2). Changes in patient selection and in immediate and 1-year outcome are presented. This report extends to 5 years the comparison of the effects of early and more recent management with PTCA.
Methods and Results Sixteen participating centers entered consecutive patients who had angioplasty for the first time between 1977 and 1981 and between 1985 and 1986. Patients with recent myocardial infarction (MI) were excluded. Vessel disease was defined according to the Coronary Artery Surgery Study. Successful dilatation required ≥20% reduction in luminal narrowing and <50% lumen diameter stenosis after intervention. Routine annual follow-up was conducted by telephone interview. The product-limit method was used to estimate freedom from untoward events, Cox regression analysis to model relative risk and adjusted relative risk of events between the two registries, and logistic regression when the exact time of outcome (such as recurrence of symptoms) was not known. Long-term event rates were computed by vessel disease for all patients and for the cohort of patients with initially successful PTCA. After adjustment for extent of disease, diabetes, prior bypass surgery (CABG), hypertension, age, and sex, the 5-year risk of death was similar in the two registry cohorts. However, rates of MI, CABG, and a combined outcome measure of death, MI, and/or CABG were significantly lower in the registry 2 cohort both for all patients and for patients who were initially treated successfully. Use of repeated PTCA was higher, and freedom from symptoms without adverse events was significantly better in the latter cohort.
Conclusions Compared with registry 1, the management of the registry 2 cohort resulted in lower 5-year morbid event rates and reduced CABG operations. Mortality rates remained similar. When symptomatic status was considered in combination with events, a significantly better outcome was seen overall and in the initially successful cohort. In registry 2, repeated PTCA was used with much greater frequency early after the initial procedure.
The multicenter NHLBI Percutaneous Transluminal Coronary Angioplasty (PTCA) Registry prospectively followed two large cohorts of consecutive patients, those undergoing PTCA from 1977 through 1981 and those undergoing PTCA from 1985 through 1986. Both registries involved the same experienced centers listed in the Appendix. Methods of patient enrollment and data collection were kept constant from one registry to the next. Earlier reports described the change in PTCA practice from the first to the second registry and its impact on immediate and 1-year outcomes.1 2 3 The current analysis extends the comparisons to 5-year outcomes.
Because several long-term follow-up reports are based on patients who were successfully treated with PTCA and because the success rate of PTCA changed so dramatically from the first registry to the second, the present report presents long-term results both for all patients and for patients with initially successful PTCA procedures.
Eligibility criteria for the registry and characteristics of the patient population were described previously.1 2 3 Briefly, consecutive patients of participating PTCA registry sites treated by PTCA for the first time outside the setting of a myocardial infarction (MI) were entered in the registry. All subjects gave informed consent for routine annual follow-up after the initial procedure and for reporting of intercurrent clinical events. The study protocol was approved by the Institutional Review Board for Biomedical Research at the University of Pittsburgh, the coordinating center for the two registries. Patient confidentiality was ensured by using alphanumeric codes to identify patient records; patient names were not included in the databases.
Since the previous publications,1 2 complete data for 334 consecutive patients from Emory University Hospital became available and were added to the registry 2 database so that the total number of registry 2 patients was 2136. Correspondingly, the first cohort was also reconstituted with 190 Emory patients who were not included in the previous reports that compared registries 1 and 2. Thus, the size of the complete registry 1 cohort is 1345.
Patients were classified as having single-, double-, or triple-vessel disease according to the definition of the Coronary Artery Surgery Study.4 Patients with 50% or greater stenosis of the left main coronary artery were classified as having double- or triple-vessel disease.
In the current report, successful dilatation was defined as a reduction of ≥20% in luminal diameter stenosis by visual assessment and a post-PTCA stenosis of <50% luminal diameter narrowing for the attempted lesions. This criterion differs from that of previous reports that required only >20% reduction in percent stenosis. While the definition for lesion success for this report differs from previous registry reports, it is the definition recommended by the Joint International Society and Federation/World Health Organization Task Force on Coronary Angioplasty.5 Angiographic success per patient was defined as successful dilatation of all attempted lesions; partial success was defined as successful dilatation of at least one but not all lesions attempted. Clinical success was defined as complete or partial angiographic success without in-hospital death, MI, or coronary artery bypass graft (CABG) surgery.
MIs were documented by at least two of the following: clinical symptoms, ECG evidence (Q-wave criteria of a definite MI according to the Minnesota Code6 ), and enzyme changes (more than double the upper normal limits of creatinine kinase and/or the presence of creatinine kinase–MB). Infarctions after hospitalization for initial PTCA were recorded whether occurring alone, during a repeated PTCA, or during subsequent CABG. Hospital records were examined to verify that ECG findings, enzyme test results, and clinical symptoms were consistent with the registry definition of MI. CABG surgery after angioplasty was also considered to be an untoward event. Repeated angioplasty was defined as a procedure performed after discharge from hospitalization for initial PTCA. Patients were considered to have angina at 5 years if they experienced angina within 30 days before the follow-up contact. In the few cases where 6- but not 5-year angina status was available in registry 1, that information was used for the analysis.
At each participating site, patients had annual telephone interviews conducted by the site coordinator. Information on hospitalizations for MI or repeated revascularization, symptoms, activity level, employment status, angina, cardiac catheterizations, and daily medications used at the time of contact was collected and reported on standard forms. Additional data were collected if a patient reported catheterization, a repeated PTCA, or CABG after the initial angioplasty.
In registry 2, there were 585 patients who underwent at least one repeated PTCA procedure within 5 years of initial angioplasty. Of these patients, 563 (96%) had complete angiographic information available. In registry 1, no angiographic data were collected for repeated angioplasty procedures after the first year of follow-up. Therefore, comparison of lesions at the time of repeated revascularization between registries 1 and 2 is limited to 1 year. Information collected at the time of the repeated PTCA was used to determine whether the repeated procedure was performed on the same arterial segments as in the initial PTCA.
To compare patients in registries 1 and 2, differences in means for continuous variables were assessed by t tests and differences in proportions of categorical variables by χ2 tests. Statistical significance was defined as a two-sided probability value less than .05. Life table analysis with the product-limit method7 was used to estimate rates of survival and freedom from morbid events. Univariate Cox8 regression analysis was used to estimate the crude relative risk of registry 2 versus registry 1 for each untoward event, with time to event as the outcome variable and registry cohort as the only independent variable. The adjusted relative risk of registry 2 versus registry 1 for each adverse event was estimated with multivariate Cox regression analysis. Baseline patient-specific factors including age, sex, history of MI, hypertension, diabetes, hypercholesterolemia, congestive heart failure, vessel disease, dominance, ejection fraction, prior bypass, angina status, length of chest pain, and high surgical risk were considered for adjustment. The indicator variable for registry cohort was forced to stay in the model. The Cox regression analysis was then performed, with backward elimination used to identify significant adjustment factors. Factors with significance level of 0.1 or less remained in the final model. The corresponding 95% CI of each estimated risk was also displayed. If the CI excludes unity, then the hypothesis of equal risk of the two registry cohorts is rejected.
Because of a lack of exact dates for the recurrence of angina, composite end points, including events and symptomatic status at the 5-year follow-up, were modeled by multivariate logistic regression with the odds ratio of registry 2 to registry 1 as measures of comparison. The strategy for variable selection in the adjusted logistic models was the same as for the Cox regression models.
Data on MI and repeated revascularization at 5 years were available for 96.5% of patients in registry 2 and 96% of patients in registry 1; vital status, however, was known for 98.2% and 96% of patients in registries 1 and 2, respectively. Of the patients in registry 2 without information on 5-year events, 4 were out of the country, 33 withdrew consent to be followed, and 45 were lost to follow-up. The corresponding figures in registry 1 were 13, 14, and 30. Data on angina status at 5 years were complete in 96.2% of registry 2 and 95% of registry 1 patients.
Differences in baseline characteristics and in PTCA strategy between the two registries1 changed little after the addition of patients from Emory University Hospital. Briefly, patients in the cohort from 1985 through 1986 were on average 5 years older (58 versus 53 years, P<.001) and were twice as likely to have double-vessel disease and three times as likely to have triple-vessel disease (31.2% versus 16.2% and 20.9% versus 6.9%, respectively) as the registry 1 cohort. They had significantly more cardiac risk factors, including abnormal left ventricular function, prior MI, prior CABG, congestive heart failure, diabetes, and hypertension. Patients in registry 2 also had a longer history of chest pain (28 versus 17 months, P<.001) and reported unstable angina more frequently. As Table 1⇓ shows, significantly more patients in registry 2 underwent multilesion or multivessel PTCA. The frequency of PTCA attempts of bypass grafts was similar.
With the new definition, lesion success was 82% in registry 2 compared with only 58% in registry 1. This large difference was due primarily to the high percentage (20%) of lesions that could not be passed with the early devices used in registry 1. As a result, patients in registry 1 were less likely to have partial and complete angiographic success and clinical success. The rates in registry 1 were 58%, 56%, and 55% compared with 87%, 75%, and 83%, respectively, in registry 2. In-hospital mortality (1% for both registries) and nonfatal MI (4.4% in registry 2 and 4.9% in registry 1) were similar between the two registries. As expected, a large difference was found between registries in the rate of CABG within the initial hospitalization period. Surgery during initial admission was performed in 25% of patients in registry 1 (6% underwent emergency and 19% underwent elective CABG after an unsuccessful PTCA) compared with 6% (4% emergency and 2% elective) in registry 2 (P<.001).
Discharge medications differed significantly between registry 2 and registry 1, especially use of aspirin (85% versus 34%), persantine (76% versus 33%), and calcium channel blockers (78% versus 11%). Use of β-blocker was 24% versus 14%, respectively.
The Figure⇓ presents the cumulative crude 5-year event rates by vessel disease for both registries for all patients and for the patients with initially successful PTCA. The crude 5-year cumulative mortality rate for patients with single-vessel disease was 6.8% in registry 2 and 4.4% in registry 1. For patients with double-vessel disease, the corresponding rates were 11.2% and 7.6%; for triple-vessel disease, the rates were 14% and 17.9%, respectively. Successful cohorts with single-vessel disease experienced 6.0% and 3.3% mortality; with double-vessel disease, 10.9% and 6.9%; and with triple-vessel disease, 9.7% and 7.8% in registries 2 and 1, respectively. After the first 6 months from the initial PTCA (7 to 60 months), mortality rates were similar in the patients with initially successful PTCA and in all patients for both registry cohorts, except for the small number of patients with triple-vessel disease in registry 1. In the patients with successful initial PTCA in the registry 2 cohort, the annual mortality rate was nearly constant over 5 years, amounting each year to about 1.2% for single-vessel disease and 2.0% for multivessel disease.
Crude 5-year MI rates were highest for patients with triple-vessel disease, 20.3% and 25.8%; the rates were 15.1% and 17.7% for double-vessel disease and 10.6% and 13.0% for single-vessel disease for registry 2 versus registry 1 patients. Thirty to forty percent of the MIs were reported periprocedurally during initial hospitalization. Again, although patients with multivessel disease continued to have nearly twice the MI rates as those with single-vessel disease, after 6 months from the first PTCA, initial success no longer affected MI rates.
The 5-year cumulative CABG rate was greatly influenced by whether patients were from registry 1 or 2 and by the extent of vessel disease. In registry 2, the CABG rate was 25.9% for triple-vessel disease, 23% for double-vessel disease, and 15% for single-vessel disease; approximately half of the surgeries occurred during initial hospitalization. Corresponding rates in registry 1 were 54.3%, 50.5%, and 35.7%, and roughly 70% to 85% of these occurred during initial hospitalization. Again, after the first 6 months, CABG rates did not depend on initial success, only on vessel disease. In the patients with successful initial PTCA in registry 2, about half of the subsequent CABG procedures were performed during the first year of follow-up. The annual rates thereafter were about 1%, 2%, and 3% for single-, double-, and triple-vessel disease, respectively. Repeated PTCA was more frequent in registry 2 than in registry 1. Respective 5-year cumulative repeated PTCA rates were 26.4% versus 17.3% for single-vessel disease, 28.7% versus 13.2% for double-vessel disease, and 34.8% versus 26.8% for triple-vessel disease. When only patients with successful PTCA were considered, the corresponding rates were all higher.
Table 2⇓ summarizes the crude 5-year cumulative event rates overall and the unadjusted and adjusted relative risks of events (registry 2 relative to registry 1). Adjustments were made for the number of diseased vessels, sex, age, ejection fraction, history of congestive heart failure, hypertension, diabetes, prior CABG, and unstable angina. For all patients undergoing PTCA, after adjustment for the excess baseline risk in registry 2, the risk of 5-year mortality was not significantly different from that in registry 1 (ie, the CI includes unity). However, rates of MI, of CABG, of the combined events of death-MI and death-MI-CABG, and of any repeated revascularization were all significantly lower in registry 2, while the rate of subsequent PTCA was significantly higher.
When only cohorts with initial success were compared, the reductions of the 5-year untoward event rates in registry 2 relative to registry 1 adjusted to inequalities were found to be more modest. While the lower MI rate, decreased use of subsequent CABG, and decreased combined events of death-MI-CABG in registry 2 remained statistically significant, the decreased need for any repeated revascularization seen overall was no longer evident. Of the two modes of revascularization, repeated PTCA was more frequently used (P=.14).
Table 3⇓ compares 5-year event- and symptom-free status in registries 1 and 2 for all patients and for patients with initial PTCA success. Events are presented with and without adjustment for inequalities of risk at baseline. Five-year survival without MI, CABG, or angina was 54% in registry 2 and 41% in registry 1, a significant difference even without adjustment for the excess risk in registry 2. Other outcome measures, such as survival without angina or survival without MI or angina, were also significantly better in registry 2, but only after adjustment for the excess risk. Similarly, for patients with initially successful angioplasty, superior results in registry 2 were seen for all end points after adjustments.
At 5 years, 73.1% of registry 2 patients reported using aspirin, 50.6% used calcium antagonist, 24.0% used β-blockers, and 14.0% used nitroglycerin. In registry 1, use of aspirin and calcium antagonists was much lower.
Investigation of the repeated PTCAs during the 5-year follow-up revealed clear trends in registry 2 (Table 4⇓). In the first 6 months, when more than half of all repeated PTCAs occurred, more than two thirds of the procedures were performed on target lesions, 21% on new lesions, and 10% on both new and target lesions. Between 6 months and 1 year, 56% of repeated procedures were done on target lesions, 27% on new lesions, and 19% on both types of lesions. After 1 year, the trend was completely reversed: 24% of the procedures were done for the target lesion only, 65% for new lesions only, and 11% for both. In registry 1, with only 1-year detailed angiographic information available, the first 6-month trend was similar to that of registry 2, but between 6 months and 1 year, very few new lesions (14%) were attempted with repeated PTCA.
This report presents 5-year outcome in the earlier (1977 through 1981) and later (1985 through 1986) cohorts of PTCA Registry patients from the same 16 participating centers. At the time of analysis, both cohorts completed at least a 5-year follow-up. In registry 1, the crude 5-year mortality rate was 6.4%, the MI rate was 15%, the CABG rate was 39%, and the rate of repeated PTCA was 17%. In registry 2, the crude 5-year rates were 9.8% for mortality, 14% for MI, 19% for CABG, and 28% for repeated PTCA. All untoward events except repeated PTCA were considerably higher for patients with double- and triple-vessel disease. In registry 1, these excesses amounted to twofold to fourfold more death, 40% to 100% more MI, and 40% to 50% more CABG. Similarly, the excesses in registry 2 were 60% to 100%, 40% to 80%, and 50% to 70% for corresponding events. The 5-year repeated PTCA rates in registry 1 were 17%, 13%, and 27% for single-, double-, and triple-vessel disease; the corresponding rates in registry 2 were 26%, 29%, and 35%. These percentages would have been higher if the registries defined additional trips to the catheterization laboratory during initial hospitalization as repeated PTCA.
Registry 2 patients were 5 years older; 19% had triple-vessel and 31% had double-vessel disease compared with 7% and 14% in registry 1. Significantly more registry 2 patients reported prior bypass, prior MI, poor left ventricular function, and more complex coronary lesions. Thus, comparison of crude event rates between the two registries is biased. To achieve a fair comparison, statistical adjustment for the differing patient characteristics was necessary. After adjustment, mortality rates of the two cohorts were similar, whereas untoward event rates, including MI, bypass surgery, death-MI-CABG, and need for any repeated revascularization, were significantly lower in registry 2. When performed in registry 2, repeated revascularizations tended to be repeated PTCA, which was significantly higher than in registry 1, particularly during the first year after the initial PTCA. While more than 80% of the procedures repeated in the first year involved the target lesion, this rate was reduced to 30% after 1 year.
Compared with event rates for all patients, event rates in the successfully treated cohorts were much lower in both registries and were similar to event rates reported by others.9 10 11 12 13 The better 5-year prognosis in the successfully treated cohorts in both registries was due almost entirely to the fact that patients with unsuccessful PTCA (by definition) incurred adverse periprocedural events. Event rates 7 to 60 months after the initial PTCA were nearly identical in the entire cohorts and in the groups who had successful angioplasty. Indeed, multivessel disease patients who had initially unsuccessful PTCA did not experience substantially higher late mortality than patients who had an initially successful PTCA. Among successfully treated patients, the more recent cohort fared significantly better in terms of 5-year MI, CABG, and death-MI-CABG. Reductions in the events occurred at the price of greater use of repeated PTCA.
The adjusted odds of achieving both event- and angina-free status were significantly higher in registry 2, regardless of initial success. This improvement is likely to be related to the more liberal use of repeat PTCAs and medications, to more extensive initial revascularization, and possibly to lifestyle changes. Unmeasured patient, lesion, or procedural characteristics that could not be controlled in the analysis, as well as temporal changes in intensity of ascertainment, also could have contributed to the improved results. It is interesting to note that whatever the reasons, they influenced long-term morbidity and angina but not mortality.
By its concept and nature, a therapeutic registry has both strengths and limitations. According to specifications in the manual of operations, the PTCA Registry has relied since its inception on the voluntary efforts of individual investigators to accumulate clinical and angiographic data. Because of funding constraints, there was no central laboratory to review all angiograms from the registries. Clinical events, particularly MIs, were not adjudicated by an independent review committee. Thus, it is possible that the intensity of ascertainment of MIs has increased with time. This would have resulted, however, in an increased rather than a decreased diagnosis of MI, thereby lending strength to the finding of the reduction of such events in the second registry compared with the first. Because routine, annual follow-up contact was made by telephone, risk factor measurements, such as blood lipids and blood pressure, were not performed as part of the registries. It is also unfortunate that information on attempts at cardiac rehabilitation and risk factor intervention, the importance of which was not well recognized in the late 1970s, was not collected.
Finally, applicability of the long-term PTCA Registry results to guide treatment selection for symptomatic coronary artery disease is limited by the lack of control groups, comparable patients who were managed by medical treatment or by CABG.14 Comparison with CABG for selected patients with multivessel disease is forthcoming from randomized clinical trials. Although the Angioplasty Compared to Medicine (ACME) trial15 compared PTCA with medical therapy for single-vessel disease, to the best of our knowledge, no randomized study is currently under way to compare long-term clinical course after intervention with PTCA to noninvasive management for double- or triple-vessel coronary artery disease.
The strength of the PTCA Registry mechanism is the continuity of the participating centers, the commitment of the investigators and clinic coordinators, and the prospective nature of data collected with a uniform protocol. Monthly submission of logs of all consecutive first PTCA procedures per center in registry 2 allowed verification that all patients were entered in the registry during recruitment. For this reason, even the PTCA arms of the more contemporary randomized clinical trials of PTCA versus CABG16 17 18 19 20 21 22 23 24 25 or PTCA versus other devices26 27 28 will not be a replacement for the long-term PTCA Registry experience. The registry includes all patients; the clinical trials exclude certain patients. For example, except for RITA,19 20 single-vessel disease and prior procedures are excluded from the CABG studies, and multivessel and graft disease is excluded from the device studies.
Admittedly, the reported PTCA Registry results do not reflect state-of-the-art intervention, for it is impossible to have both long-term follow-up and the latest in procedural outcome. Yet the 76% clinical success rate (by quantitative angiographic estimation) obtained by PTCA in the 1991-1992 Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT)26 is consistent with the 85% success rate (by caliper visual estimation) obtained in a comparable group of registry 2 patients. Possible changes in lesion morphology targeted for more current intervention are known to affect initial results29 but not long-term outcome.30 We speculate, therefore, that regardless of the improvements in PTCA technology since 1985, it is unlikely that they had a greater impact on outcome than the improvements observed from registry 1 to registry 2. Provided that patient and lesion characteristics are similar today, we could expect a higher degree of initial success and a similar or somewhat improved long-term clinical outcome in successfully treated patients compared with that seen in the successful cohort of registry 2. Perhaps if current revascularization practice placed more emphasis on simultaneous risk factor interventions, this could have a beneficial impact on long-term prognosis.
Main Investigators and Institutions
Boston University Medical Center/University Hospital, Boston, Mass: Alice K. Jacobs, MD, and David P. Faxon, MD (former) (principal investigators), and Caroline Melidossian (coordinator); Emory University Hospital, Atlanta, Ga: Spencer B. King III, MD (principal investigator), Neal Scott, MD, and John Douglas, MD (associate investigators), and Dee Anderson (coordinator); Georgetown University Hospital, Washington, DC: Kenneth M. Kent, MD (former) (principal investigator), and Katie Kehoe (coordinator); Massachusetts General Hospital, Boston, Mass: Peter C. Block, MD (former) (principal investigator), and Elizabeth Block (coordinator); the Mayo Clinic, Rochester, Minn: David R. Holmes, Jr, MD (principal investigator), Guy S. Reeder, MD, and J.F. Bresnahan, MD (associate investigators), and LaVon Hammes and Susie Brevig (coordinators); Medical Center Hospital, Houston, Tex: Mahdi Al-Bassam, MD (principal investigator), and Debbie Lance (coordinator); Medical College of Pennsylvania, Philadelphia, Pa: Lamberto G. Bentivoglio, MS, MD (principal investigator), and Eileen Shappell (coordinator); Medical College of Virginia, Richmond, Va: Michael J. Cowley, MD (principal investigator), George W. Vetrovec, MD (associate investigator), and Kim Kelly (coordinator); Miami Heart Institute, Miami, Fla: James Margolis, MD (principal investigator), Arthur J. Gosselin, MD (former), and Hazel Yon (coordinator); Montreal Heart Institute, Montreal, Quebec, Canada: Martial G. Bourassa, MD (principal investigator), and Micheline Labbe and Claudette Faille (coordinators); NHLBI, Bethesda, Md: Richard O. Cannon, MD (principal investigator), and Rita Mincemoyer and Anette Stein (coordinators); Seton Medical Center, Daly City, Calif: Richard K. Myler, MD (principal investigator), and Mary Murphy (coordinator); Rhode Island Hospital/Brown University, Providence, RI: David O. Williams, MD (principal investigator), and Shirley Emin (coordinator); St. Francis Regional Medical Center, Wichita, Kan: Joseph P. Galichia, MD (principal investigator), and Pat Patterson and Jo Robinson (coordinators): St. Luke’s Episcopal Hospital, Houston, Tex: Terry Ferguson, MD (principal investigator), Louis L. Leatherman, MD (former), Arthur J. Springer, MD, and Samantha Beba (associate investigators), and Brenda Lambert (coordinator); and St. Luke’s Hospital, Milwaukee, Wis: Gerald Dorros, MD (principal investigator), and Joann Brandt (coordinator).
Data Coordinating Center
University of Pittsburgh, Pittsburgh, Pa: Katherine M. Detre, MD, DrPH, and Sheryl F. Kelsey, PhD (principal investigators), Wanlin Yeh, MS (statistician), Verna Niedermeyer (data coordinator), Polly Swanson and John Winegarten (data managers), Kevin Kip (graduate student researcher), Donna Gibbons (administrative secretary), and Rita Wolk (secretary).
NHLBI, Bethesda, Md: Patrice Desvigne-Nickens, MD.
This study is supported by NIH grant HL-33292. We are grateful to Kevin Kip, graduate student researcher, for his valuable contribution to the analyses. The Appendix contains a list of all contributors.
- Received October 10, 1994.
- Revision received December 14, 1994.
- Accepted December 27, 1994.
- Copyright © 1995 by American Heart Association
Detre K, Holubkov R, Kelsey S, Cowley M, Kent K, Williams D, Myler R, Faxon D, Holmes D Jr, Bourassa M, Block P, Gosselin A, Bentivoglio L, Leatherman L, Dorros G, King S III, Galichia J, Al-Bassam M, Leon M, Robertson T, Passamani E, for the Coinvestigators of the NHLBI PTCA Registry. Percutaneous transluminal coronary angioplasty in 1985-1986 and 1977-1981. N Engl J Med. 1988;318:265-270.
Detre K, Holubkov R, Kelsey S, Bourassa M, Williams D, Holmes D Jr, Dorros G, Faxon D, Myler R, Kent K, Cowley M, Cannon R, Robertson T, for the Coinvestigators of the NHLBI PTCA Registry. One-year follow-up results of the 1985-1986 National Heart, Lung, and Blood Institute’s Percutaneous Transluminal Coronary Angioplasty Registry. Circulation. 1989;80:421-428.
Holmes DR Jr, Holubkov R, Vlietstra RE, Kelsey SF, Reeder GS, Dorros G, Williams DO, Cowley MJ, Faxon DP, Kent KM, Bentivoglio LG, Detre K, for the Coinvestigators of the NHLBI PTCA Registry. Comparison of complications during percutaneous transluminal coronary angioplasty from 1977 to 1981 and from 1985-1986: the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. J Am Coll Cardiol. 1988;12:1149-1155.
CASS Principal Investigators and their associates. Coronary Artery Surgery Study (CASS): a randomized trial of coronary artery bypass surgery: quality of life in patients randomly assigned to treatment groups. Circulation. 1983;68:951-960.
Bourassa MG, Alderman EL, Bertrand M, de la Fuente L, Gratsiansi A, Kaltenback M, King SB, Nobuyoshi M, Romaniuk P, Ryan TJ, Serruys PW, Smith HC, Sousa JE, Bothig S, Rapaport E. Report of the Joint ISFC/WHO Task Force on Coronary Angioplasty. Circulation. 1988;78:780-789.
Cox DR. Regression models and life tables (with discussion). J R Statist Soc B. 1972;74:187-220.
O’Keefe JH, Rutherford BD, McConahay DR, Johnson WL Jr, Giorgi LV, Ligon RW, Shimshak TM, Hartzler GO. Multivessel coronary angioplasty from 1980 to 1989: procedural results and long-term outcome. J Am Coll Cardiol. 1990;15:1097-1102.
Weintraub WS, Ghazzal ZMB, Cohen CL, Douglas JS Jr, Liberman H, Morris DC, King SB III. Clinical implications of late proven patency after successful coronary angioplasty. Circulation. 1991; 84:572-582.
Talley JD, Hurst JW, King SB III, Douglas JS Jr, Roubin GS, Gruentzig AR, Anderson HV, Weintraub WS. Clinical outcome 5 years after attempted percutaneous transluminal coronary angioplasty in 427 patients. Circulation. 1988;77:820-829.
Ellis SG, Cowley MJ, DiSciascio G, Deligonul U, Topol EF, Bulle TM, Vandormael MG, for the Multivessel Angioplasty Prognosis Study Group. Determinants of 2-year outcome after coronary angioplasty in patients with multivessel disease on the basis of comprehensive preprocedural evaluation: implications for patient selection. Circulation. 1991;83:1905-1914.
Henderson RA, Raskino C, Karani S, Sowton E. Comparative long-term results of coronary angioplasty in single and multivessel disease. Eur Heart J. 1992;13:781-786.
Mock MB, Smith HC, Mullany CJ. The ‘second generation’ NHLBI Percutaneous Transluminal Coronary Angioplasty Registry: have we established the role for PTCA in treating coronary artery disease? Circulation. 1989;80:700-702. Editorial.
Hamm CW, Ischinger T, Reimers J, Dietz U, Rupprecht H-J, for the GABI Study Group. Angioplasty vs bypass-surgery in patients with multivessel disease: in-hospital outcome in the GABI trial. Circulation. 1992;86(suppl I):I-373. Abstract.
Rupprecht HJ, Hamm CW, Ischinger T, Reimers J, Trautmann S, Meyer J, for the GABI Study Group. Angiographic follow-up of the German Angioplasty vs Bypass-Surgery Investigation (GABI Trial). Circulation. 1993;88(suppl I):I-506. Abstract.
Serruys PW, Breeman A. Coronary angioplasty: long-term follow-up results and detection of restenosis: guidelines for aviation cardiology, a European view. Eur Heart J. 1992;13(suppl H):76-88.
The RITA Trial. Lancet. 1993;341:1020. Letter.
King SB III, Lambo NJ, Hall EC, and the EAST Investigators. The Emory Angioplasty vs Surgery Trial (EAST): analysis of baseline patient characteristics. Circulation. 1990;82(suppl III):III-508. Abstract.
King SB, Lambo NJ, Weintraub WS, for the EAST Investigators. Results from the Emory Angioplasty vs Surgery Trial (EAST) compared to the eligible, non-randomized registry. J Am Coll Cardiol. 1994;24:469A. Abstract.
Rodriquez A, Boullon F, Perez-Balinoo N, Paviotti C, Liprandi MS, Palacios IF. Argentine Randomized Trial of Percutaneous Transluminal Coronary Angioplasty Versus Coronary Artery Bypass Surgery in Multivessel Disease (ERACI): in-hospital results and 1-year follow-up. J Am Coll Cardiol. 1993;22:1060-1067.
Rodriquez A, Ahualli P, Perez-Balino N, Mele E, Paviotti C, Liprandi MS, Palacios I. Argentine Randomized Trial of Percutaneous Transluminal Coronary Angioplasty Versus Coronary Artery Bypass Surgery in Multivessel Disease (ERACI): late cost and three-year follow-up results. J Am Coll Cardiol. 1994;24:469A. Abstract.
Protocol for the Bypass Angioplasty Revascularization Investigation. Circulation. 1991;84(suppl V):V-1-V-27.
Topol EJ, Leya F, Pinkerton CA, Whitlow PL, Hofling B, Simonton CH, Masden RR, Serruys PW, Leon MB, Williams DO, King SB III, Mark DB, Isner JM, Holmes DR, Ellis SG, Lee KL, Keeler GP, Berdan LG, Hinohara T, Califf RM, for the CAVEAT Study Group. A comparison of directional atherectomy with coronary angioplasty in patients with coronary artery disease. N Engl J Med. 1993;329:221-227.
Serruys PW, de Jaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, Emanuelsson H, Marco J, Legrand V, Materne P, Belardi J, Sigwart U, Colombo A, Goy JJ, van den Heuvel P, Delcan J, Morel M-A, for the BENESTENT Study Group. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med. 1994;331:489-495.
Fischman DL, Leon BM, Baim D, Schatz RA, Savage MP, Penn I, Detre K, Veltri L, Ricci D, Nobuyoshi M, Cleman M, Heuser R, Almond D, Teirstein P, Fish D, Colombo A, Brinker J, Moses J, Shaknovich A, Hirshfeld J, Bailey S, Ellis S, Rake R, Goldberg S, for the STRESS Trial Investigators. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med. 1994;331:496-501.
Ellis SG, Vandormael MG, Cowley MJ, DiSciascio G, Deligonul U, Topol EJ, Bulle TM. Multivessel Angioplasty Prognosis Study Group. Coronary morphologic and clinical determinants of procedural outcome with angioplasty of multivessel coronary disease: implications for patient selection. Circulation. 1990;82:1193-1202.
Fortin DF, Nelson CL, Gardner LH, Harrell F Jr, Frid DJ, Tcheng JE, Phillips HR III, Califf R, Mark DB, Stack RS. Pre-angioplasty lesion morphology: a case of diminishing effect on survival? J Am Coll Cardiol. 1993;21:35A.