Author + information
- Received December 31, 1998
- Revision received March 25, 1999
- Accepted May 10, 1999
- Published online September 1, 1999.
- Paul D. McGrath, MD, MSc, FACC∗,* (, )
- David J. Malenka, MD, FACC†,
- David E. Wennberg, MD, MPH∗,†,
- Samuel J. Shubrooks Jr., MD, FACC§,
- William A. Bradley, MD, FACC‡,
- John F. Robb, MD, FACC†,
- Mirle A. Kellett Jr., MD, FACC∗,
- Thomas J. Ryan Jr., MD, FACC∗,
- Michael J. Hearne, MD, FACC∥,
- Bruce Hettleman, MD, FACC†,
- John R. O’Meara, MD, FACC∗,
- Peter VerLee, MD, FACC¶,
- Matthew W. Watkins, MD, FACC#,
- Winthrop D. Piper, MSc†,
- Gerald T. O’Connor, PhD, DSc, FACC†,
- for the Northern New England Cardiovascular Disease Study Group∗∗
- ↵*Reprint requests and correspondence: Dr. Paul D. McGrath, Center for Outcomes Research and Evaluation, Maine Medical Center, 22 Bramhall Street, Portland, Maine 04102.
We sought to evaluate the changing outcomes of percutaneous coronary interventions (PCIs) in recent years.
The field of interventional cardiology has seen considerable growth in recent years, both in the number of patients undergoing procedures and in the development of new technology. In view of recent changes, we evaluated the experience of a large, regional registry of PCIs and outcomes over time.
Data were collected from 1990 to 1997 on 34,752 consecutive PCIs performed at all hospitals in Maine (two), New Hampshire (two) and Vermont (one) supporting these procedures, and one hospital in Massachusetts. Univariate and multivariate regression analyses were used to control for case mix. Clinical success was defined as at least one lesion dilated to <50% residual stenosis and no adverse outcomes. In-hospital adverse outcomes included coronary artery bypass graft surgery (CABG), myocardial infarction and mortality.
Over time, the population undergoing PCIs tended to be older with increasing comorbidity. After adjustment for case mix, clinical success continued to improve from a low of 88.2% in earlier years to a peak of 91.9% in recent years (p trend <0.001). The rate of emergency CABG after PCI fell in recent years from a peak of 2.3% to 1.3% (p trend <0.001). Mortality rates decreased slightly from 1.2% to 1.1% (p trend 0.007).
There has been a significant improvement in clinical outcomes for patients undergoing PCIs in northern New England, including a significant decline in the need for emergency CABG.
The practice of percutaneous coronary interventions (PCIs) continues to evolve. Since Andreas Gruentzig performed the first human coronary angioplasty in September 1977 (1), the field has rapidly expanded to an estimated 1 million PCIs performed worldwide each year (2). Over the past 20 years, the increasing experience of operators (3,4)coupled with the advent of newer technologies, including coronary stents (5)and a variety of adjuvant drug therapies (6–8), have permitted more successful procedures and decreased the morbidity associated with PCIs (9). In May of 1994, the Palmaz-Schatz stent was recommended for approval by the Food and Drug Administration advisory panel for elective use in selected patients to prevent restenosis after angioplasty (10). Later that year, results from the Benestent Study Group (11)and the Stent Restenosis Study (12)were published. The use of coronary stents in PCIs subsequently increased dramatically, and it is estimated that nearly 50% of all PCIs now involve stents (2).
Although many randomized trials have described improving outcomes for PCIs, there are few large observational trials examining the generalizability of these results (13,14). Several randomized studies have documented improvement with respect to abrupt closure and restenosis (11,12). A recent study comparing the National Heart, Lung, and Blood Institute (NHLBI) Registry (1985 to 1986) with the New Approaches to Coronary Intervention (NACI) Registry (1990 to 1994) suggests progress remains to be seen with respect to long-term survival after PCIs (15).
The Northern New England Cardiovascular Disease Study Group (NNECVDSG) is a voluntary research consortium composed of clinicians, research scientists and hospital administrators. The focus of the group is to foster continuous improvement in the quality of care of patients with cardiovascular disease in northern New England through the analysis of process and outcome data combined with the timely feedback of data to clinicians (16,17). We used our prospective, multicenter, clinical database to evaluate the overall improvement in outcomes for patients after PCIs in our region from 1990 to 1997.
Data were prospectively collected on 34,752 consecutive PCIs performed from January 1990 through December 1997 at all hospitals in Maine (2), New Hampshire (2)and Vermont (1)supporting these procedures, and one hospital in Massachusetts. Not all cardiologists contributed from the beginning of data collection, and some left the practice of invasive cardiology or the region before the end of the study. However, cardiologists were required to report on all cases performed at these institutions while participating in the study. Given the advent of stent technology during the study period, data were divided into early (1990 to 1993; 0% of cases involving stents), transitional (1994 to 1995; 3.9% of cases involving stents) and current years (1996 to 1997; 48.5% of cases involving stents). The transitional period is intended to mark the widespread introduction of stent technology to the field of PCIs.
Data for each PCI were recorded on separate forms at the time of the procedure. Demographic and comorbidity data included age, gender, previous PCIs, previous coronary artery bypass graft surgery (CABG), previous myocardial infarction (MI), presence of diabetes, history of peripheral vascular disease and history of chronic obstructive pulmonary disease. Previous MI was defined as a MI occurring at least three weeks before the intervention.
Clinical data included the indication for the procedure, the priority of the procedure (emergency, urgent and nonurgent), whether the patient was on intravenous (IV) nitroglycerin or IV heparin, whether the patient had an intraaortic balloon pump (IABP) in place before the intervention and ejection fraction (EF). Angiographic data included the presence of a >50% left main stenosis, whether the patient had >70% stenoses in the left anterior descending (LAD), circumflex, right coronary artery or in a bypass graft, dominance, the number of lesions attempted, interventional devices used and the location and complexity of the treated lesion(s). Lesion complexity was defined using the American College of Cardiology definitions for type A, B and C lesions (18), with B lesions broken down to B1 and B2 lesions starting in 1994 (19). Percent stenoses before and after inflations were recorded by the primary operator using visual estimates or calipers depending on the local standard.
Successful and adverse outcomes occurring during the PCI hospitalization were assessed. Clinical success was defined as at least one lesion dilated with <50% residual stenosis and no adverse outcomes (20,21). Adverse outcomes assessed during the hospitalization included any CABG (emergency or nonemergency CABG), new MI or death. Emergency CABG was defined as surgery performed to treat acute closure, unstable angina or congestive heart failure (CHF) requiring IV nitroglycerin or IABP or tamponade resulting from the intervention. Nonemergency CABG was defined as surgery performed after an unsuccessful PCI in a stable patient during the same hospitalization. A new MI was defined as a clinical event, electrocardiogram (ECG) changes and a creatine phosphokinase (CPK) rise to greater than or equal to twice normal levels with positive isoenzymes.
To ensure that data were collected for all eligible patients and that the outcomes of CABG and death were correctly assessed, the data collection was validated. Hospitals provided lists of discharge abstracts for patients and catheterization laboratory logs for comparison with the registry to identify patients for whom forms were missing. Any missing information was then obtained. Hospital discharge abstracts served as the “gold” standard for validating the outcomes of CABG and death. Myocardial infarctions were not separately validated.
All analyses were carried out using Statistical Analysis Software version 6.11 (Cary, North Carolina) or STATA Statistical Software, Release 5.0 (College Station, Texas). Treatment variables including preprocedure IV heparin, IV nitroglycerin, thrombolytics and IABP were coded as “not present” if the value was missing under the assumption that a recording would have been made had the therapy been present. Procedures with missing values for gender were excluded from the multivariate analysis. Ejection fraction was not reported in 41% of cases. For these patients, values for EF were imputed based on a prediction model generated from the remaining data set (Appendix A). Pearson chi-square tests were used to assess the univariate association between potential predictors and the outcome variables of clinical success, new MI, CABG and death (22). All variables demonstrating a univariate association with the event of interest at a p value < 0.10 (without adjustment for multiple comparisons) were considered potential independent variables for inclusion in the multivariate analyses. Multivariate analyses were conducted using logistic regression models (23). Age and gender were included in all models. Adjusted rates for successful and adverse outcomes were calculated using the beta estimates from the logistic regression model and direct standardization (24).
Three periods of time corresponding to the increased use of coronary stents were utilized to describe our patient population (early: January 1990 to December 1993; transitional: January 1994 to June 1995; current: July 1995 to December 1997). Table 1describes the clinical characteristics of the Northern New England patient population over the entire study period. More recent years have seen PCIs involving an older patient population overall. In addition to more patients with renal failure undergoing PCIs, similar proportions of patients suffered from concomitant diabetes mellitus, peripheral vascular disease and chronic obstructive pulmonary disease. A smaller proportion of patients in recent years had a history of previous MI or PCI, and a larger proportion had undergone a previous CABG or had a history of congestive heart failure.
The procedural characteristics and indications changed over time (Table 2). A larger proportion of cases were performed in an emergency setting, partially accounted for by the increased use of PCI as the primary therapy for MI. Cases tended to involve more complex lesions over time; there were more type B2 and type C lesions and more lesions in bypass grafts. An impressive change in the use of new devices was seen with no stents used in the early 1990s to nearly 50% of cases involving stents at the end of the study period.
Data regarding outcomes after PCIs are presented in Table 3. With respect to unadjusted rates, clinical success increased over the study period to 91.5% in recent years. Rates of MI and both emergency and nonemergency CABG declined over time. A slight increase in the rate of in-hospital mortality was seen over time.
Adjusted rates for these same in-hospital outcomes are also presented in Table 3. After adjusting for case mix, clinical success and rates of adverse outcomes continued to improve over the study period. In particular, rates of CABG occurring during the same hospitalization dropped from 3.1% to 1.8%, a 40% decrease in this particular adverse outcome. Adjusted emergency CABG rates declined from earlier rates of 2.1% to 2.3% to a low of 1.3% recently. A slight decrease in mortality was noted after controlling for case mix with a rate of 1.1% in more recent years.
The field of interventional cardiology has grown at a remarkable pace since its inception over 20 years ago. Accompanying this growth are improvements in patient care and outcomes. Improvements in technology have occurred both in devices and in adjunctive pharmacologic therapy; coronary stents (10–12)and glycoprotein IIb/IIIa inhibitors (7,8)are two prominent examples that have contributed to better patient outcomes. Northern New England has seen a dramatic decline in the rates of CABG during the same hospitalization after PCIs and an associated increase in the clinical success of these procedures in recent years. This has occurred in the setting of an aging patient population with increasing comorbidities.
We attribute the decline in the need for emergency CABG in our study to the availability of coronary stents. Review of the data presented in Table 3reveals that the rate of this outcome did not begin to decline until 1995, when stents were introduced, and did not dramatically fall until 1996 and 1997, when stents were involved in 45.2% and 62.4% of cases, respectively. Other authors have also described that coronary stent use reduces the incidence of emergency CABG associated with PCIs (25,26). In 1993, Lincoff et al. (25)described how patients treated by intracoronary stenting for abrupt vessel closure saw a reduction in the need for emergency bypass surgery when compared with matched controls (4.9% vs. 18%, p = 0.02).
However, other factors may be contributing to the improving outcomes, such as increasing operator volume and hospital experience, improvements in equipment and newer pharmacologic agents. Although use of glycoprotein IIb/IIIa inhibitors in our patients had only risen to 12.7% of cases during 1997, the Evaluation of 7E3 for the Prevention of Ischemic Complications (EPIC) Investigators noted a 35% reduction in the rate of the adverse clinical events after high-risk PCIs in patients receiving these agents (7). The use of the antiplatelet agent ticlopidine has also allowed for decreased complications associated with coronary stents (6). At the same time, increasing experience of operators and hospitals performing these procedures has been shown to contribute to these improved outcomes (3,4,27). Within our own registry, most operators performing PCIs in the early 1990s continued to perform PCIs through the end of the study period, increasing their overall experience over time.
Other investigators have provided information with respect to PCI outcomes in the setting of large, clinical registries from earlier years. Hannan et al. studied 62,670 PCIs in New York State during 1991 to 1994. They described a risk-adjusted in-hospital mortality rate of 0.9% (27)and a CABG rate of 3.43%. Similar event rates were described by the New Approaches to Coronary Intervention (NACI) Registry during 1990 to 1994 (15).
Although CABG rates have declined and clinical success has improved, the rate of overall in-hospital mortality has seen little change with the evolution of PCIs. King et al. compared registry data from 1985 to 1986 with data from 1990 to 1994 and noted no difference in one-year survival after PCIs, after adjustment for case mix (15). Other studies have reported in-hospital mortality rates of 1% to 2% (15,27)after PCIs, which is reflected by our own experience.
There are several limitations to our study. For the purposes of our study, we did not collect information with respect to the indication for the use of a coronary stent, that is, stent use as the primary intended device versus treatment for abrupt closure. This limits our ability to comment on a cause-and-effect relationship regarding stent use and a reduction in the rate of emergency CABG. In addition, although the adverse outcomes including CABG and in-hospital mortality were validated as part of this study, MIs after PCI were not validated. Only a slight decrease in adjusted mortality was seen over time. The statistical significance of this slight decrease may be a reflection of the large number of patients studied. Alternatively, our models may not completely account for the impact of the more critically ill patient population seen in recent years, including a higher proportion of patients with acute MI or cardiogenic shock. Our experience describes the outcomes within one region of the U.S. Previous studies have shown that New England practitioners are relatively conservative with the use of cardiac catheterization and PCIs when compared with other regions of the country (28). Whether our results may be extrapolated to the remainder of the country remains to be seen.
Over time, a significant reduction in major adverse clinical events has been seen in northern New England after PCIs. This has occurred in the setting of an aging patient population with greater preprocedure comorbidity. These improvements are likely related to the increasing experience of the operators, the advent of new technology and possibly, the sharing of outcomes data within our region to help guide decision making for our patient population. Our large regional database allows our physicians to better assess the outcomes of their patients using analyses of current data. Although more work must be done with respect to improving in-hospital mortality as well as determining how our patients fare after discharge, the practice of interventional cardiology has seen significant improvements in in-hospital outcomes over time.
Predicted ejection fraction = 71.98794 + [(−0.09095 × age) + (1.69334 × gender) − (0.19939 × left ventricular end diastolic pressure) − (4.20875 × history of previous MI) − (3.77906 × history of previous CABG) − 3.6]
↵∗∗ A list of members appears in .
☆ This study was supported in part by a grant from the Agency for Health Care Policy and Research (HS06813) and
was presented in part at the 71st Annual Scientific Session of the American Heart Association, Dallas, Texas, November 9–11, 1998.
- coronary artery bypass graft surgery
- creatine phosphokinase
- directional coronary atherectomy
- ejection fraction
- intraaortic balloon pump
- left anterior descending coronary artery
- myocardial infarction
- Northern New England Cardiovascular Disease Study Group
- percutaneous coronary intervention
- Received December 31, 1998.
- Revision received March 25, 1999.
- Accepted May 10, 1999.
- American College of Cardiology
- Gruentzig A.R.,
- Turina M.I.,
- Schneider J.A.
- Topol E.J.,
- Serruys P.W.
- Jollis J.G.,
- Peterson E.D.,
- Nelson C.L.,
- et al.
- McGrath P.D.,
- Wennberg D.E.,
- Malenka D.J.,
- et al.
- Hearn J.A.,
- King S.B.,
- Douglas-JS J.,
- et al.
- Bredlau C.E.,
- Roubin G.S.,
- Leimgruber P.P.,
- et al.
- ↵Summary of Minutes of the Meeting of the Circulatory System Devices Panel, Food and Drug Administration, Bethesda, MD, May 2–3, 1994. Bethesda, MD: Food and Drug Administration, 1994.
- King S.B.,
- Yeh W.,
- Holubkov R.,
- et al.
- Ryan T.J.,
- Faxon D.P.,
- Gunnar R.M.,
- et al.
- Ryan T.J.,
- Bauman W.B.,
- Kennedy J.W.,
- et al.
- Ellis S.G.,
- Omoigui N.,
- Bittl J.A.,
- et al.
- Fleiss J.
- Kleinbaum D.G.,
- Kupper L.L.,
- Morgenstern H.
- Kahn H.A.,
- Sempas C.T.
- Lincoff A.M.,
- Topol E.J.,
- Chapekis A.T.,
- et al.
- Roubin G.S.,
- Cannon A.D.,
- Agrawal S.K.,
- et al.
- Pilote L.,
- Califf R.M.,
- Sapp S.,
- et al.