Acute coronary syndrome (ACS) is the major cause for hospital admission and mortality. The frequency, management and outcomes of acute coronary syndromes vary in different registries. The key factor predicting the outcome is age. The specific type of acute coronary syndrome (unstable angina vs. non-STEMI vs. STEMI) included in each given registry must always be kept in mind when analysing the outcome of any given registry. Some registries mix all MIs (STEMI and non-STEMI) or even report all three forms of ACS together; some registries enroll patients with suspected ACS, while other ones only enroll patients with confirmed ACS, etc.
Incidence of acute coronary syndromes.
Only a few registries report the true incidence of acute coronary syndromes (i.e., cover completely a defined geographic region with a known total population). A carefully done regional acute MI registry in Denmark (1) revealed 727 hospitalised patients with confirmed MI (mean age 73 years) during a two-year follow-up in a region with 139,000 inhabitants: 58% were non-ST-elevation MI, 37% ST-elevation MI, and 5% bundle branch block type MI’s. The calculated annual incidence of hospitalised MI cases is thus 2,612 per million.
A change in diagnostic criteria for acute MI increased the hospitalisation rate for acute MI in Denmark from 2,560 to 3,091/mil. pop. (2). Early mortality decreased from 26% in 1994 to 17.5% in 2002.
In the recently published CZECH registry (3), the annual incidence of hospitalised patients with confirmed MI was 1,960/mil. pop., i.e., substantially less than in Denmark. The CZECH registry was the first registry reporting the incidence of confirmed acute coronary syndromes (all three forms): 3,248 patients/mil. pop.
Treatment strategies for acute coronary syndromes.
The CZECH registry (3) confirmed that widespread use of the network system for primary PCI as the default reperfusion strategy dramatically increased the overall availability of reperfusion therapy: 87% STEMI patients enrolled in the CZECH registry were treated by reperfusion (primary PCI in 83%, CABG in 3% and thrombolysis in only 1% of cases).
Reperfusion therapy was used in 54% of ST elevation infarctions in Denmark (1), while 25% ST-elevation MI patients were admitted within 12 hours of symptom onset, but were not treated by reperfusion. The remaining 21% of ST-elevation MI patients arrived late (after 12 hours). An early invasive strategy (CAG) was performed in 48% of non-ST-elevation MI patients. Patients with ST-elevations receiving any reperfusion therapy died in 10.5%, while no reperfusion therapy was associated with 34% mortality. Patients with non-ST-elevation MI?s undergoing early CAG died in 10% whereas those who did not died in 49.5% of cases!
In the Euro Heart Survey (4), CAG was undertaken in 56% of STEMI and 52% of NSTE-ACS. PCI was performed in 72% of STEMI patients undergoing CAG, the figure being 49% in NSTE-ACS patients undergoing CAG.
In the GRACE registry (5, 6, 7), CAG was performed in 55% of STEMI (followed by PCI in 73%), 53% of NSTEMI (followed by PCI in 53%), and in 42% of UAP patients (followed by PCI in 43%). Reperfusion therapy was used in 62% of STEMI patients (with primary PCI in 15%, rescue PCI after failed thrombolysis in 3%, and thrombolysis in only 44%). The primary PCI-to-thrombolysis ratio was thus 28 : 72. In the Italian BLITZ survey (8), the ratio of primary PCI to thrombolysis was 23 : 77.
The Euro Heart Survey (9) showed that 4.5% of patients hospitalized with ACS in Europe underwent CABG during the same hospital stay. Their in-hospital mortality was 3.7%.
Reperfusion therapy is used in 60-70% of STEMI patients in Canada, with a vast majority of procedures being thrombolysis, and only few primary PCIs (10).
There were 17,926 high-risk NSTE ACS patients in the CRUSADE registry (11, 12) enrolled in 248 US hospitals with catheterization and revascularization facilities. Patients treated with early invasive management were more likely to be treated with medications and interventions recommended by the ACC/AHA guidelines and had a lower risk of in-hospital mortality after adjusting for differences in clinical characteristics and after comparing propensity-matched pairs (2.5% vs 3.7%, P<.001).
In-hospital CAG was performed in the UK registry (13) in only 10% of patients and only 29% of CAG patients underwent PCI.
In the Israeli registry (14), 69% of patients had CAG during their hospital stay and 58% of these underwent PCI. Discharge medication included aspirin in 89%, ticlopidine/clopidogrel in 48%, ACE inhibitors in 60%, beta-blockers in 75% and statins in 67% of patients.
In the Canadian registry (15), CAG was performed in 39.6% of patients and PCI in 41% of the CAG group.
The large US registry (16) analyzed 255,256 patients with acute myocardial infarction enrolled in the NRMI-2,3,4 in the US between 1994 and 2002: 20% had ST depression (mean age, 71 years) and 80% had ST elevation or bundle branch block (mean age, 66 years) on their admission ECG. Females made up 40% of the entire registry population, and individuals with 25%. CAG at any time during hospital stay was performed in 62% of STEMI patients. Primary PCI was performed in 22% of STEMI patients, and the ratio of primary PCI to thrombolysis was 37 : 63.
The recently published Euro Heart Survey on Acute Coronary Syndromes II (17) described the characteristics, treatment, and outcome of 6385 patients diagnosed with ACS in 190 medical centres in 32 countries. ACS with ST-elevation was the initial diagnosis in 47% of patients and non-ST-elevation in 48%. This survey showed greater use of recommended medications and coronary interventions. Among patients with ST-elevation, the use of primary reperfusion was 64% (primary PCI : thrombolysis ratio 59:41).
Outcomes of acute coronary syndromes.
The in-hospital mortality of myocardial infarctions in the Danish registry (1) was 14%, with one-year mortality of 28%. The strength of this Danish registry is independent and objective assessment of one-year mortality obtained from the “Civil Registration System“ in Denmark. The one-year mortality is thus higher than in other registries or trials: being 20.5% for ST-elevations (mean age was 69 years), 30.5% for non-ST-elevations (mean age 75 years) and as high as 54.8% for bundle branch block infarctions (mean age 80 years).
In a Swiss registry (18), in-hospital mortality varied between 2.4% and 11.8% (depending on the subgroup considered). Killip III and IV classes on admission were associated with mortality increased by factors 3.6 and 6, respectively. The Euro Heart Survey (4) enrolled 10,484 patients in 25 countries based on the discharge diagnosis of confirmed ACS. The admission diagnoses were STEMI in 42% and NSTE-ACS in 58%. The discharge diagnoses were Q-MI in 33%, non-Q MI in 25%, and UAP in 42%. In-hospital mortality was 7% for STEMI and 2.8% for NSTE-ACS. In the GRACE registry (5, 6, 7), the mean age of patients with STEMI was 64 years, with NSTEMI 68 years, and with UAP 66 years. In-hospital mortality was 4.5% (all ACS patients) and 7% for STEMI, 5% for NSTEMI, and 3% for UAP. Death rates between discharge and 6 months were 4.8% (STEMI), 6.2% (NSTEMI), and 3.6% (UAP).B
The BLITZ survey, performed in 2001 (8), enrolled 1,959 patients (mean age, 67 years) with acute MI admitted to 296 Italian coronary care units. Two thirds of cases were STEMI and one third NSTEMI. In-hospital mortality was 7.4% and 30-day mortality 9.4% for the entire cohort and 7.5% / 9.5% for STEMI, and 5.2% / 7.1% for NSTEMI. The second BLITZ survey (19) enrolled 1,888 NSTE-ACS patients (half had NSTEMI and half had UAP with negative myocardial necrosis markers) in 275 Italian hospitals over a 3-week period. The 30-day mortality was 2.4%. There was no significant difference between invasive (2.0% mortality) and non-invasive (2.9% mortality) hospitals. The only
independent predictors of outcome were age and Killip class.
The large CRUSADE registry (11, 12) enrolled 56,352 patients with NSTE-ACS in 310 US hospitals, with in-hospital mortality being 4.2%. The Euro Heart Survey (20) expressed mortality risk per age group as compared to patients below 55 years: the odds ratios of in-hospital mortality due to any type of ACS were 1.9 (age 55-64), 3.7 (age 65-74), 6.2 (age 75-84) and 14.5 (age 85 or higher). The UK registry (13) enrolled 1,046 NSTE-ACS patients (mean age, 66 years) in 56 UK hospitals. The overall 6-month mortality was 7.3%. In an Israeli CCU registry (14), 970 NSTE-ACS patients (mean age, 65 years) were included with 30-day mortality of 2.9%. A Canadian registry enrolled 5,312 patients (mean age, 66 years) with suspected ACS in 51 hospitals (15). The diagnosis of ACS was confirmed in 4,627 patients 87%). The overall in-hospital mortality was 2.4% for the entire cohort and 4.9% (Q-MI), 2.5% (NQMI), 0.4% (UAP) for each specific type of ACS. One-year mortality of hospital survivors was 6.5% (Q-MI), 10% (NQMI), and 5.4% (UAP). The US in-hospital mortality (16) was the same for ST-elevations (15.5%) as for ST-depressions (15.8%).
Practical implications.
Conclusions
The data in registries of acute coronary syndromes largely vary due to many factors: regional differences, type of patients enrolled, type of hospitals participating, treatment strategies used, etc. To provide reliable and comparable results, a registry should fulfill the following criteria:
1. To cover all hospitals (invasive and non-invasive) in a given region (to enroll 100% patients in that region during a given period). Unfortunately, most registries are based on voluntary participation of only selected (and usually better !) hospitals and thus cannot provide the true picture.
2. To enroll the most difficult patients: elderly (>80 years), Killip III-IV class or resuscitated (i.e. patients, who in many hospitals are admitted to resuscitation units and not to cardiology units).
3. To match the mortality data with an independent national database of death certificates.
Registries, fulfilling these criteria, usually provide significantly higher mortality data – but they represent the real life situation.
The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.