Canadian-American Differences in the Management of Acute Coronary Syndromes in the GUSTO IIb Trial
One-Year Follow-Up of Patients Without ST-Segment Elevation
Background—Little information exists concerning practice patterns between Canada and the United States in the management of myocardial infarction (MI) patients without ST-segment elevation and unstable angina.
Methods and Results—We examined the practice patterns and 1-year outcomes of 2250 US and 922 Canadian patients without ST-elevation acute coronary syndromes in the Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb trial. The US hospitals more commonly had on-site facilities for angiography and revascularization. These procedures were performed more often and sooner in the United States than Canada, whereas Canadian patients were more likely to undergo noninvasive stress testing. The length of initial hospital stay was 1 day longer for Canadian than US patients. Recurrent and refractory ischemia was more common in Canada. One-year mortality was comparable between the 2 countries. However, at 6 months, even after baseline differences were accounted for, the (re)MI rate was significantly higher in Canadian than US patients with unstable angina (8.8% versus 5.8%, P=0.039), as was the composite rate of death or (re)MI (13.1% versus 9.1%, P=0.016).
Conclusions—One-year mortality was comparable between Canada and the United States in both MI and unstable angina cohorts despite higher intervention rates in the United States. However, outcomes at 6 months among patients with unstable angina differed. Whereas more frequent coronary interventions were not associated with reduced recurrent MI or death among MI patients without ST elevation, they may favorably affect outcomes in patients with unstable angina.
Recent reports attest to major differences between Canada and the United States in the process of care for patients with acute ST-segment–elevation myocardial infarction (MI).1 2 3 4 5 6 Although revascularization has been associated with improved functional status in US patients,1 2 5 7 this strategy has little impact on mortality at 13 5 or 3 years.1 Interestingly, little parallel information exists in patients comprising the majority of coronary care admissions, ie, MI without ST elevation and unstable angina (UA).8 9 The Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb trial, which compared the efficacy of hirudin and heparin in acute coronary syndromes (ACS), afforded us an opportunity to examine this issue.
Our objective was to evaluate Canadian-American differences in the process of health care among patients with non–ST-elevation ACS and their impact on the principal study outcomes, ie, death and MI.
The GUSTO IIb trial has been previously described.10 Briefly, 12 142 patients within 12 hours of chest discomfort onset with an abnormal ECG were randomized to receive 72 hours of intravenous heparin or hirudin therapy. Patients were first stratified into those with ST elevation (n=4131) and those without ST elevation (n=8011); the latter group was then further subdivided into non–ST-elevation MI and UA, depending on whether their index event was an MI. Patients were followed up for 30 days, 6 months, and 1 year after randomization. We evaluated the non–ST-elevation cohort of 2250 US patients (982 with non–ST-elevation MI and 1268 with UA) and 922 Canadian patients (428 with non–ST-elevation MI and 494 with UA).
Baseline characteristics, treatment, complications, and clinical outcomes of the US and Canadian patients were compared. MI or reinfarction [(re)MI] has been defined previously.11 Recurrent ischemia was defined as symptoms with either (1) new ST-segment deviation or definite T-wave inversion in ≥2 contiguous leads or (2) new onset of pulmonary edema or development of cardiac murmur or hypotension thought by the physician to represent myocardial ischemia while the patient was hospitalized. Refractory ischemia consisted of symptoms of ischemia with ECG changes persisting for ≥10 minutes despite the use of medical therapy. Emergency angiography was defined as angiography urgently performed for clinical instability.
Both the hirudin and heparin treatment arms were combined, and US and Canadian patients were then compared. Descriptive statistics were summarized as medians with 25th and 75th percentiles for continuous variables, and the Mann-Whitney U test was used for comparisons between groups. For categorical variables, the data were summarized in percentages, and Fisher’s exact test or the χ2 test was used to assess group differences. Kaplan-Meier survival estimates and the Cox proportional-hazards regression model were used to compare time to the first occurrence of the end points. Cox regression with time-dependent covariates was used to assess the effects of revascularization on the end points. One-year mortality data for the US and Canadian cohorts were 96% and 98% complete, respectively.
Univariate analysis was first performed to identify potentially important baseline characteristics associated with the composite of death or (re)MI at 6 months and mortality at 1 year. Multivariate Cox and logistic regression models were developed through backward, stepwise variable selection procedures to assess the effect of baseline characteristics and geographical location on the composite of death or (re)MI at 6 months and on 1-year mortality. The models were assessed by use of the Hosmer-Lemeshow test and the C index (or the area under the receiver-operator characteristic curves).12 All tests were 2 sided, with a 5% level of significance. All analyses were performed with SPSS (version 9.0).
Selected baseline characteristics are given in Table 1⇓. Patients in the US cohort were more often hypertensive and more likely to have had prior angioplasty. Age and sex distributions between the 2 countries were similar within each category of ACS, as were hypercholesterolemia, Killip class >1, and time to treatment.
In Table 2⇓, medication use is summarized. The in-hospital use of ACE inhibitors, lipid-lowing agents, and aspirin was similar between the 2 countries. Intravenous nitroglycerin, intravenous β-blockers, calcium channel blockers, and digitalis were much more frequently administered to US patients, whereas Canadian patients more often received oral β-blockers. At discharge, US patients were prescribed digitalis more often and β-blockers less often than Canadians.
The characteristics of the participating institutions are depicted in Figure 1⇓, indicating that more US than Canadian hospitals were equipped with angiography, angioplasty, and bypass surgery facilities. Table 3⇓ shows the in-hospital use of noninvasive/invasive procedures and the elapsed time to the invasive procedures. Stress tests were more commonly performed in Canada than in the United States and usually involved stress ECG testing alone, whereas stress echocardiogram and perfusion imaging were performed less often in Canada. The frequency of positive stress tests was similar between the 2 countries in the MI cohort but higher for Canadian patients with UA. In contrast, US patients with MI more often had Swan-Ganz catheters, ventilators, and intra-aortic balloon pumping than their Canadian counterparts; among patients who did not undergo bypass surgery, Swan-Ganz catheters and intra-aortic balloon pumping were also undertaken more frequently in US patients with MI. However, such intercountry differences were not seen in the UA cohort. Angiography and revascularization were performed approximately twice (≥1.7-fold) as commonly in the United States as in Canada (all P≤0.003). Although these procedures were more frequently used among US patients with MI compared with UA (all P≤0.035), this difference was not evident in Canada. Canadian patients, on the other hand, more often had emergency angiography regardless of whether they had an MI or UA on admission. Time to angiography and revascularization was approximately one half as long in the United States as in Canada.
Among patients treated in hospitals with on-site revascularization facilities, the US patients had a higher rate of revascularization than their Canadian counterparts (53% in the United States versus 40% in Canada; see Figure 2⇓). This difference was even greater for patients treated in hospitals lacking on-site facilities: 41% in the United States versus 16% in Canada. Compared with patients who presented to hospitals with on-site facilities, US patients treated in hospitals without on-site facilities had approximately a one-fourth–less chance of undergoing revascularization; the comparable figure was more than one half for Canada. On-site facilities were also associated with shorter time to procedure in both countries.
Canadian patients stayed in hospital 1 day longer than their US counterparts and had longer stays in intensive or coronary care units but shorter stays in step-down units (Table 4⇓). The incidence of stroke, shock, and congestive heart failure was similar between the 2 countries. Moderate to severe bleeding occurred more often in the United States, whereas the rates of recurrent and refractory ischemia were higher in Canada (48% to 49% for recurrent ischemia; 8.7% to 11% for refractory ischemia) than in the United States (25% to 33% for recurrent ischemia; 2.4% to 5.1% for refractory ischemia). Although recurrent and refractory ischemia was substantially more common among US patients with MI than with UA, the rates of these ischemic events were similar in the Canadian cohort. There were no significant intercountry differences within the MI cohort in the rates of (re)MI up to 6 months and mortality through 1 year. Canadian patients with UA had higher rates of (re)MI (a 3% absolute or 52% relative increase in Canada) and of death or (re)MI (a 4.0% absolute or 44% relative increase in Canada) at 6 months compared with their US counterparts. These intercountry differences at 6 months persisted after adjustment for significant baseline covariates, including age, diabetes, previous angina, previous MI, previous bypass surgery, Killip class, and heart rate (Figure 3⇓). In addition, mortality at 30 days and 6 months for Canadian patients with UA was higher after adjustment for significant baseline characteristics; these mortality differences were attenuated by 1 year (Figure 3⇓).
Comparisons of mortality according to revascularization status revealed that patients without versus those with revascularization had substantially higher 1-year mortality in both countries, especially in the United States. The higher mortality in patients without revascularization accounted for >70% of the total 1-year mortality in both countries. Moreover, US MI patients without revascularization had 1.5-fold–higher mortality than their Canadian counterparts. Comparisons of baseline characteristics of patients without versus with revascularization showed that patients without revascularization in both countries were older (median age: 73 versus 63 years in the United States, P<0.001; 67 versus 62 years in Canada, P<0.001), more often had diabetes (29% versus 20% in the United States, P=0.02; 22% versus 15% in Canada, P=0.18), and had more prior MI (35% versus 25% in the United States, P=0.008; 39% versus 24% in Canada, P=0.006). A time-dependant Cox regression model in which revascularization was treated as a time-dependent covariate also confirmed that revascularization was associated with a significantly lower 1-year mortality (hazard ratio, 0.64; 95% CI, 0.47 to 0.87) after adjustment for significant risk factors such as age, diabetes, previous MI, previous bypass surgery, Killip class, heart rate, MI on admission, and in-hospital recurrent ischemia.
In Figure 4⇓, cumulative event-free survival curves at 6 months (left) and 1-year survival (right) for patients with MI and UA are presented. There was no intercountry difference in the composite rate of death or (re)MI in the MI cohort at 6 months, but US patients with UA had significantly higher event-free survival than their Canadian counterparts (P=0.012); this difference remained significant after adjustment for baseline characteristics with Cox regression (P=0.029). One-year survival was comparable between the United States and Canada in both MI and UA cohorts. Intracountry comparisons showed that US patients with UA versus MI had significantly better survival at both 6 months and 1 year (P<0.001), but this pattern was not evident in Canada.
The results of this study provide novel insights into the process of medical care between Canada and the United States as it relates to the contemporary management of patients with non–ST-elevation MI and UA. In agreement with the findings of the Organization to Assess Strategies for Ischemic Syndromes (OASIS) Registry study,9 angiography, angioplasty, and bypass surgery in our study were more common in the United States. Although these differences were related in part to the availability of on-site facilities at participating institutions,13 14 15 16 similar intercountry differences in the rates of revascularization procedures were evident in patients initially admitted to hospitals without such facilities. It is important to note that the transfer rate for hospitals without revascularization facilities was nearly 3-fold higher in the United States than in Canada (54% versus 19%). The differences in the practice environment and physicians’ attitudes between the United States and Canada,6 coupled with nonclinical influences, likely led to a lower transfer threshold in the United States. Thus, it follows that the lack of on-site facilities was less inhibitory on angiography and revascularization use in the United States compared with Canada. Not surprisingly, the presence of on-site facilities resulted in shorter waiting times for angiography and revascularization. Consistently longer waiting times existed before such procedures in Canada compared with the United States, and this, in conjunction with the higher rates of recurrent and refractory ischemia, likely accounted for a higher incidence of emergency angiography and the 1-day-longer hospital stay in Canada.
Although there was no statistically significant difference in 1-year mortality in the UA cohort between the 2 countries, our study reveals, for the first time, significant intercountry differences in the rate of (re)MI (a 3% absolute or 52% relative increase in Canada) and in the composite rates of death and (re)MI (a 4.0% absolute or 44% relative increase in Canada) at 6 months. These results are in agreement with the recently published Fragmin and Fast Revascularization During Instability in Coronary Artery Disease (FRISC II) study, which showed that the composite end point of death and (re)MI at 6 months was 9.4% in the early invasive group compared with 12.1% in the noninvasive group.17 Importantly, unlike the Veterans Affairs Non–Q-Wave Infarction Strategies in Hospital (VANQWISH) Study,18 the cause of mortality in MI patients in our study was not primarily mediated by perioperative morbidity and mortality because the adverse outcome were higher in patients who did not have revascularization. Also, in agreement with our findings in the MI cohort, the TIMI IIIB study demonstrated that there were no significant differences in the rates of death or (re)MI through 1 year among patients with non–Q-wave MI and UA who were randomly assigned to an invasive versus conservative strategy.19 20 The revascularization rates in the TIMI IIIB (60% in invasive versus 40% in conservative strategy) were higher than our rates (49% in the United States versus 25% in Canada). One-year mortality in TIMI IIIB was approximately one-half that found in our study, presumably related to less stringent ECG criteria, younger age, and more exacting exclusion criteria.
Our findings are at variance with the observational registry data derived from 14 US and 4 Canadian tertiary care centers.8 Slightly more frequent use of angioplasty in Canada but similar rates of angiography and bypass surgery and similar outcomes of death and MI at both 6-week and 1-year follow-up were observed. The basis for the differences between these data and others, including our own, likely relates to sampling differences, including a preponderance of UA (80%) compared with non–ST-elevation MI (20%) in this TIMI Registry study.
Our study suggests that revascularization was avoided in patients with a higher-risk profile in both countries, particularly in the US MI patients, and highlights the substantially worse outcome for MI compared with UA patients in the United States. Interestingly, US patients with UA had both lower rates of angiography and revascularization than MI patients and a lower incidence of death, MI, and recurrent ischemia. In contrast, the overall rates of angiography and revascularization were lower in Canada than in the United States and occurred with similar frequency in UA and MI patients. Moreover, the frequency of death and MI was similar between these 2 diagnostic cohorts in Canada and most closely aligned to the US MI patients’ outcomes. It is interesting to speculate as to whether the UA outcomes in Canada could have been improved by more frequent use of coronary interventions. Regardless of which country patients presented, the strikingly greater increase in mortality between 30 days and 1 year in the UA versus MI cohort (P<0.001) is noteworthy. New strategies, such as low-molecular-weight heparin, glycoprotein IIb/IIIa inhibitors, and appropriate use of improved revascularization procedures, may offer further benefit to such individuals.
Some caveats relating to our study are appropriate to acknowledge. Although a substantial number of institutions participated from both countries in this study, they are not necessarily representative of all US and Canadian hospitals. However, there was good balance between tertiary and secondary care institutions; hence, they most likely reflect contemporary practice patterns in both health care environments. The (re)infarction rates beyond 6 months were unavailable, and the revascularization procedures were nonrandomized.
In conclusion, our study has extended previous observations concerning striking intercountry differences in the approach to medical care of patients with acute ST-elevation MI to the larger cohort of patients with ACS, ie, with non–ST-elevation MI and UA, and offered insights into their implications. Whereas more frequent angiography and intervention do not appear to be associated with reductions in re-MI or death among MI patients without ST elevation, we provide new evidence that a more aggressive approach may improve clinical outcomes of patients with UA.
The GUSTO-IIb trial was funded by grants from Boehringer-Mannheim, Mannheim, Germany; Ciba-Geigy Corporation, Summit, NJ; and Advanced Cardiovascular Systems, Mountain View, Calif.
- Received February 4, 2000.
- Revision received April 17, 2000.
- Accepted April 19, 2000.
- Copyright © 2000 by American Heart Association
Yusuf S, Flather M, Pogue J, et al, for the OASIS (Organization to Assess Strategies for Ischemic Syndromes) Registry Investigators. Variations between countries in invasive cardiac procedures and outcomes in patients with suspected unstable angina or myocardial infarction without initial ST elevation. Lancet. 1998;352:507–514.
Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIa Investigators. Randomized trial of intravenous heparin versus recombinant hirudin for acute coronary syndromes. Circulation. 1994;90:1631–1637.
Armstrong PW, Fu Y, Chang WC, et al, for the GUSTO II Investigators. Acute coronary syndromes in the GUSTO IIB trial: Prognostic insights and impact of recurrent ischemia. Circulation. 1998;98:1860–1868.
Every NR, Larson EB, Litwin PE, et al, for the Myocardial Infarction Triage and Intervention Project Investigators. The association between on-site cardiac catheterization facilities and the use of coronary angiography after myocardial infarction. N Engl J Med. 1993;329:546–551.
Boden WE, O’Rourke RA, Crawford MH, et al, for the Veterans Affairs Non-Q-Wave Infarction Strategies in Hospital (VANQWISH) Trial Investigators. Outcomes in patients with acute non-Q-wave myocardial infarction randomly assigned to an invasive as compared with a conservative management strategy. N Engl J Med. 1998;338:1785–1792.
TIMI IIIB Investigators. Effects of tissue plasminogen activator and a comparison of early invasive and conservative strategies in unstable angina and non–Q-wave myocardial infarction: results of the TIMI IIIB trial. Circulation. 1994;89:1545–1556.
Anderson HV, Cannon CP, Stone PH, et al, for the TIMI IIIB investigators. One-year results of the thrombolysis in myocardial infarction (TIMI) IIIB clinical trial: a randomized comparison of tissue-type plasminogen activator versus placebo and early invasive versus early conservative strategies in unstable angina and non–Q-wave myocardial infarction. J Am Coll Cardiol. 1995;26:1643–1650.