Cardiovascular disease remains the commonest cause of death in New Zealand, being responsible for 11,293 (39%) of the 28,636 total deaths in 2004.1 Ischaemic heart disease was responsible for 6313 (22%) of these deaths. An acute coronary syndrome (ACS) is an unstable and potentially life-threatening presentation of ischaemic heart disease, and is a spectrum of clinical conditions: unstable angina pectoris (UAP), non-ST-segment elevation myocardial infarction (NSTEMI) and ST-segment elevation myocardial infarction (STEMI).Effective treatment strategies, summarised in international2-5 and local6,7 guidelines, exist and are able to significantly improve the morbidity and mortality of this condition.In May 2002, the Cardiac Society of New Zealand ACS Audit Group performed a comprehensive audit over a 14-day period which described patient numbers, presentation type and management during hospital admission. This audit demonstrated low levels of investigations, evidence-based treatments, and revascularisation.8 In addition, there was inequitable management as patients admitted to a hospital without cardiac interventional facilities received fewer investigations and less revascularisation than patients admitted to interventional centres.9Significant efforts were subsequently made by clinicians to improve both the medical and invasive management of patients with ACS. Local practice consensus guidelines were written,6,7 and efforts were made to try to improve access to treatments of proven benefit.10-15 Clinicians from interventional and non-interventional centres in New Zealand aimed to facilitate the transfer of appropriate patients from Non-Interventional to Interventional hospitals: the so-called hub and spoke approach to management.16In May 2007, the Cardiac Society of New Zealand ACS Audit Group performed a further 14-day assessment of ACS in New Zealand to record current management, to discover if significant changes had been made from 2002, and to identify areas where further improvements in service delivery may be indicated. Once again we chose to undertake the Audit during a 2-week autumn period, to be consistent with the 2002 Audit, and to minimise the known influence of seasonal change on the number of ACS patients.17Methods Data collectionThe established ACS Audit Group network from 2002 was used, consisting of one physician for every hospital in New Zealand that admitted ACS patients. Most centres also co-opted one or more research nurses or registrars to assist with data collection for the study. Since 2002, the Green Lane Hospital Cardiovascular Service in Auckland had moved to the site of the Auckland Public Hospital, and these two cardiac services had combined as the Auckland City Hospital service. In addition, Waitakere Hospital in Auckland had opened a coronary care unit, and Kaitaia, Dargaville and Rawene in Northland were now actively planning to admit patients with an ACS presentation. Other smaller hospitals in New Zealand were not actively trying to admit such patients. Therefore, in 2007 there were 39 hospitals admitting ACS patients, compared to 36 hospitals in 2002. The data collection form recorded patient demographics, initial and discharge diagnosis, medication use in hospital and at discharge, as well as investigations undertaken and invasive treatments received by patients. The dataset collected in 2007 was similar to 2002, with additional information obtained to help assess pre-hospital presentation and aspects of PCI practice. The inclusion criterion for the audit was a patient admitted overnight with a suspected or definite acute coronary syndrome. Following admission and investigations, a discharge diagnosis was subsequently determined by the local clinical team who confirmed the diagnosis of an ACS, as a STEMI, NSTEMI or UAP, or determined a non-ACS presentation resultant on investigations undertaken in hospital and the patients clinical course. A 2-week audit period was accepted as a compromise between the need to collect sufficient patient numbers to obtain an accurate representative cohort versus the ability of mainly unfunded clinicians and nurses to collect the consecutive patient data. We collected data from 0000 hours on Monday 14 May to 2400 hours on Sunday 27 May 2007 (13-26 May in 2002). Following input from all 39 centres, ethical approval was obtained from the Multi-region Ethics Committee. As an audit of current practice, individual patient consent was not required. The Ethics Committee permitted the collection of patient names and National Health Index (NHI) numbers to assist with accurate data collection. Data (including revascularisation procedures) from patients subsequently transferred to another institution are attributed to their original admitting hospital. Patients readmitted within the 2 weeks have all admissions included in the data; they only represented a small percentage of the overall patient number. Ethnicity was self-reported at hospital admission. Troponin was measured at all 39 hospitals. In May 2007 there were 11 different analysers across New Zealand provided by 5 major companies: Roche (5), Abbott (3), Bayer (1), Dade Behring (1) and Beckman Coulter (1). In order to divide NSTEMI and UAP by means of a positive troponin18 we defined normal or abnormal troponin levels using the cut-off for positive troponins as troponin T [Roche]: Modular E170, Elecsys 1010, Elecsys 2010, COBAS 601, Cardiac Reader, 22650.03ug/L, troponin I [Abbott]: Axysm 22650.04ug/l, troponin I [Abbott]: Architect 22650.03ug/l, troponin I [Abbott]: i-stat 22650.08ug/l, troponin I [Bayer]: Advia centaur 22650.04ug/l, troponin I [Dade Behring]: Dimension 22650.1ug/l. Hypertension and dyslipidaemia were defined as patients on treatment, or with a previous clinical diagnosis. Patients with diabetes mellitus were those on diet control, oral hypoglycaemic, or insulin treatment. Cardiogenic shock was defined as: a systolic blood pressure of <90 mmHg for at least 30 minutes, or the need for supportive measures to maintain a systolic blood pressure of 226590 mmHg with end organ hypoperfusion.19 Sustained ventricular tachycardia was defined as >30 seconds of ventricular tachycardia, or requiring electrical cardioversion. StatisticsContinuous data are summarised as median and interquartile range. Differences in frequencies were tested using chi-squared procedures or Fishers exact test as appropriate. SAS (SAS Institute Inc, v9.1) was used to perform the analyses. All tests were two-tailed and a 5% significance level was used. Results 1003 patients with a suspected or definite ACS were admitted to 39 New Zealand hospitals and enrolled in the ACS audit over the 14-day period (Figure 1). Eight patients (7 once and 1 twice) were readmitted within the 2 weeks, all to the same hospital. 134 patients were transferred from their admitting hospital to another institution for further management (128 [96%] to an intervention centre). Over the 2 weeks, 3 hospitals had no ACS admissions, 8 hospitals admitted 40 or more patients of which 2 hospitals (Auckland City and Christchurch) admitted more than 120 patients, from their own catchment area. Patient demographics: The median age was 67 (IQR 56-78) years. Forty-two percent of patients were female, 77% Caucasian, 9.2% Maori, 3.3% Pacific Islander, 2.7% Indian, 2.0% Asian, 0.9% from another ethnic group and in 4.8% the ethnicity was unspecified. Patient demographics were slightly changed from those in 2002 except that there were fewer Caucasian patients (77% vs 83%, p<0.05), more Maori (9.2% vs 6.7%, p<0.05) and more Pacific Island ethnicity patients (3.3% vs 1.5%, p<0.05) in 2007. In addition, more patients were now identified as being dyslipidaemic (50% vs 35%, P<0.05) and more had previously undergone CABG surgery (9.9% vs 4.6%, p<0.05) (Table 1). Figure 1. New Zealand ACS hospitals and patient numbers: 1003 patients admitted to 39 hospitals Table 1. Baseline patient demographic data for 2007 (n=1003) and comparison with the 2002 Audit baseline demographic data (n=930) Variables 2002 2007 (n=930) (n=1003) Median age [years] (IQR) 70 (58-78) 67 (56-78) Sex (male) 535 (58%) 580 (58%) Ethnicity Caucasian M 0101ori Pacifica Indian Asian Not reported Others 768 (83%) 62 (6.7%) 14 (1.5%) 17 (1.8%) 8 (0.9%) 56 (6.0%) 5 (0.5%) 775 (77%)* 92 (9.2%)* 33 (3.3%)* 27 (2.7%) 20 (2.0%) 47 (4.8%) 9 (0.9%) Smoking Current Previous Never Not reported 171 (18%) 379 (41%) 347 (37%) 33 (4%) 179 (17%) 389 (39%) 387 (39%) 48 (5%) Hypertension 442 (48%) 522 (55%) Diabetes mellitus 161(17%) 179 (19%) Dyslipidaemia 326 (35%) 474 (50%)* Prior MI 325 (35%) 318 (33%) Prior angiogram 325 (31%) 297 (31%) Prior PCI 105 (11%) 150 (16%) Prior CABG 27 (4.6%) 96 (9.9%)* Prior PVD 93 (10%) 76 (7.7%) Prior TIA/Stroke 119 (11%) 114 (12%) Prior AF 126 (13%) 138 (14%) *P<0.05 IQR: Interquartile range; MI: Myocardial infarction; PCI: Percutaneous coronary intervention; CABG: Coronary artery bypass grafting; PVD: Peripheral vascular disease; TIA: Transient ischaemic attack; AF: Atrial fibrillation. Patient diagnosis: Using both the admission clinical diagnosis and the measurement of a positive troponin level, we found that 86 (9%) patients presented with a STEMI, 413 (41%) with a NSTEMI, 329 (33%) with UAP, and 175 (17%) with another cardiac or medical diagnosis (Tables 2A, 2B, 2C, 2D). In 2007, compared with 2002, there was a higher percentage of NSTEMI patients (413 (41%) vs 287 (31%), P=0.0025)compared to UAP patients (Table 2B). Patient management: 69% of STEMI patients received reperfusion therapy. The majority received fibrinolytic therapy with only 13 (15%) of STEMI patients treated with primary PCI. This rate had, however, increased from 2002 (13 (15%) vs 3 (3%), p=0.0046) (Table 2A). Table 2A. Treatments and investigations of STEMI patients: 2002 and 2007 Variables STEMI P: 02 vs 07 2002 2007 N (%) 101 (11%) 86 (8.6%) 0.18 Treatments in hospital Fibrinolytic therapy Primary PCI Enoxaparin Daltaparin UF heparin No heparin*** Tirofiban Eptifibatide Abciximab Aspirin Clopidogrel 56 (55%) 3 (3.0%) 33 (34%) 6 (5.9%) 28 (28%) 40 (40%) 5 (5.0%) 2 (2.0%) 1 (1%) 87 (88%) 14 (14%) 47(55%) 13 (15%) 42 (49%) 0 28 (33%) 18 (21%) 2 (2.3%) 1 (1.2%) 2 (2.3%) 81 (94%) 63 (73%) 0.87 0.0046 0.014 0.024 0.72 <0.0001 0.37 0.68 0.45 0.082 <0.0001 Investigations in hospital Chest X-ray Echocardiogram Exercise test Angiogram No ETT/Angio No Echo/Angio PCI CABG surgery In-hospital deaths 89 (89%) 35 (35%) 18 (18%) 31 (31%) 57 (57%) 50 (50%)

Cardiovascular disease remains the commonest cause of death in New Zealand, being responsible for 11,293 (39%) of the 28,636 total deaths in 2004.1 Ischaemic heart disease was responsible for 6313 (22%) of these deaths. An acute coronary syndrome (ACS) is an unstable and potentially life-threatening presentation of ischaemic heart disease, and is a spectrum of clinical conditions: unstable angina pectoris (UAP), non-ST-segment elevation myocardial infarction (NSTEMI) and ST-segment elevation myocardial infarction (STEMI).Effective treatment strategies, summarised in international2-5 and local6,7 guidelines, exist and are able to significantly improve the morbidity and mortality of this condition.In May 2002, the Cardiac Society of New Zealand ACS Audit Group performed a comprehensive audit over a 14-day period which described patient numbers, presentation type and management during hospital admission. This audit demonstrated low levels of investigations, evidence-based treatments, and revascularisation.8 In addition, there was inequitable management as patients admitted to a hospital without cardiac interventional facilities received fewer investigations and less revascularisation than patients admitted to interventional centres.9Significant efforts were subsequently made by clinicians to improve both the medical and invasive management of patients with ACS. Local practice consensus guidelines were written,6,7 and efforts were made to try to improve access to treatments of proven benefit.10-15 Clinicians from interventional and non-interventional centres in New Zealand aimed to facilitate the transfer of appropriate patients from Non-Interventional to Interventional hospitals: the so-called hub and spoke approach to management.16In May 2007, the Cardiac Society of New Zealand ACS Audit Group performed a further 14-day assessment of ACS in New Zealand to record current management, to discover if significant changes had been made from 2002, and to identify areas where further improvements in service delivery may be indicated. Once again we chose to undertake the Audit during a 2-week autumn period, to be consistent with the 2002 Audit, and to minimise the known influence of seasonal change on the number of ACS patients.17Methods Data collectionThe established ACS Audit Group network from 2002 was used, consisting of one physician for every hospital in New Zealand that admitted ACS patients. Most centres also co-opted one or more research nurses or registrars to assist with data collection for the study. Since 2002, the Green Lane Hospital Cardiovascular Service in Auckland had moved to the site of the Auckland Public Hospital, and these two cardiac services had combined as the Auckland City Hospital service. In addition, Waitakere Hospital in Auckland had opened a coronary care unit, and Kaitaia, Dargaville and Rawene in Northland were now actively planning to admit patients with an ACS presentation. Other smaller hospitals in New Zealand were not actively trying to admit such patients. Therefore, in 2007 there were 39 hospitals admitting ACS patients, compared to 36 hospitals in 2002. The data collection form recorded patient demographics, initial and discharge diagnosis, medication use in hospital and at discharge, as well as investigations undertaken and invasive treatments received by patients. The dataset collected in 2007 was similar to 2002, with additional information obtained to help assess pre-hospital presentation and aspects of PCI practice. The inclusion criterion for the audit was a patient admitted overnight with a suspected or definite acute coronary syndrome. Following admission and investigations, a discharge diagnosis was subsequently determined by the local clinical team who confirmed the diagnosis of an ACS, as a STEMI, NSTEMI or UAP, or determined a non-ACS presentation resultant on investigations undertaken in hospital and the patients clinical course. A 2-week audit period was accepted as a compromise between the need to collect sufficient patient numbers to obtain an accurate representative cohort versus the ability of mainly unfunded clinicians and nurses to collect the consecutive patient data. We collected data from 0000 hours on Monday 14 May to 2400 hours on Sunday 27 May 2007 (13-26 May in 2002). Following input from all 39 centres, ethical approval was obtained from the Multi-region Ethics Committee. As an audit of current practice, individual patient consent was not required. The Ethics Committee permitted the collection of patient names and National Health Index (NHI) numbers to assist with accurate data collection. Data (including revascularisation procedures) from patients subsequently transferred to another institution are attributed to their original admitting hospital. Patients readmitted within the 2 weeks have all admissions included in the data; they only represented a small percentage of the overall patient number. Ethnicity was self-reported at hospital admission. Troponin was measured at all 39 hospitals. In May 2007 there were 11 different analysers across New Zealand provided by 5 major companies: Roche (5), Abbott (3), Bayer (1), Dade Behring (1) and Beckman Coulter (1). In order to divide NSTEMI and UAP by means of a positive troponin18 we defined normal or abnormal troponin levels using the cut-off for positive troponins as troponin T [Roche]: Modular E170, Elecsys 1010, Elecsys 2010, COBAS 601, Cardiac Reader, 22650.03ug/L, troponin I [Abbott]: Axysm 22650.04ug/l, troponin I [Abbott]: Architect 22650.03ug/l, troponin I [Abbott]: i-stat 22650.08ug/l, troponin I [Bayer]: Advia centaur 22650.04ug/l, troponin I [Dade Behring]: Dimension 22650.1ug/l. Hypertension and dyslipidaemia were defined as patients on treatment, or with a previous clinical diagnosis. Patients with diabetes mellitus were those on diet control, oral hypoglycaemic, or insulin treatment. Cardiogenic shock was defined as: a systolic blood pressure of <90 mmHg for at least 30 minutes, or the need for supportive measures to maintain a systolic blood pressure of 226590 mmHg with end organ hypoperfusion.19 Sustained ventricular tachycardia was defined as >30 seconds of ventricular tachycardia, or requiring electrical cardioversion. StatisticsContinuous data are summarised as median and interquartile range. Differences in frequencies were tested using chi-squared procedures or Fishers exact test as appropriate. SAS (SAS Institute Inc, v9.1) was used to perform the analyses. All tests were two-tailed and a 5% significance level was used. Results 1003 patients with a suspected or definite ACS were admitted to 39 New Zealand hospitals and enrolled in the ACS audit over the 14-day period (Figure 1). Eight patients (7 once and 1 twice) were readmitted within the 2 weeks, all to the same hospital. 134 patients were transferred from their admitting hospital to another institution for further management (128 [96%] to an intervention centre). Over the 2 weeks, 3 hospitals had no ACS admissions, 8 hospitals admitted 40 or more patients of which 2 hospitals (Auckland City and Christchurch) admitted more than 120 patients, from their own catchment area. Patient demographics: The median age was 67 (IQR 56-78) years. Forty-two percent of patients were female, 77% Caucasian, 9.2% Maori, 3.3% Pacific Islander, 2.7% Indian, 2.0% Asian, 0.9% from another ethnic group and in 4.8% the ethnicity was unspecified. Patient demographics were slightly changed from those in 2002 except that there were fewer Caucasian patients (77% vs 83%, p<0.05), more Maori (9.2% vs 6.7%, p<0.05) and more Pacific Island ethnicity patients (3.3% vs 1.5%, p<0.05) in 2007. In addition, more patients were now identified as being dyslipidaemic (50% vs 35%, P<0.05) and more had previously undergone CABG surgery (9.9% vs 4.6%, p<0.05) (Table 1). Figure 1. New Zealand ACS hospitals and patient numbers: 1003 patients admitted to 39 hospitals Table 1. Baseline patient demographic data for 2007 (n=1003) and comparison with the 2002 Audit baseline demographic data (n=930) Variables 2002 2007 (n=930) (n=1003) Median age [years] (IQR) 70 (58-78) 67 (56-78) Sex (male) 535 (58%) 580 (58%) Ethnicity Caucasian M 0101ori Pacifica Indian Asian Not reported Others 768 (83%) 62 (6.7%) 14 (1.5%) 17 (1.8%) 8 (0.9%) 56 (6.0%) 5 (0.5%) 775 (77%)* 92 (9.2%)* 33 (3.3%)* 27 (2.7%) 20 (2.0%) 47 (4.8%) 9 (0.9%) Smoking Current Previous Never Not reported 171 (18%) 379 (41%) 347 (37%) 33 (4%) 179 (17%) 389 (39%) 387 (39%) 48 (5%) Hypertension 442 (48%) 522 (55%) Diabetes mellitus 161(17%) 179 (19%) Dyslipidaemia 326 (35%) 474 (50%)* Prior MI 325 (35%) 318 (33%) Prior angiogram 325 (31%) 297 (31%) Prior PCI 105 (11%) 150 (16%) Prior CABG 27 (4.6%) 96 (9.9%)* Prior PVD 93 (10%) 76 (7.7%) Prior TIA/Stroke 119 (11%) 114 (12%) Prior AF 126 (13%) 138 (14%) *P<0.05 IQR: Interquartile range; MI: Myocardial infarction; PCI: Percutaneous coronary intervention; CABG: Coronary artery bypass grafting; PVD: Peripheral vascular disease; TIA: Transient ischaemic attack; AF: Atrial fibrillation. Patient diagnosis: Using both the admission clinical diagnosis and the measurement of a positive troponin level, we found that 86 (9%) patients presented with a STEMI, 413 (41%) with a NSTEMI, 329 (33%) with UAP, and 175 (17%) with another cardiac or medical diagnosis (Tables 2A, 2B, 2C, 2D). In 2007, compared with 2002, there was a higher percentage of NSTEMI patients (413 (41%) vs 287 (31%), P=0.0025)compared to UAP patients (Table 2B). Patient management: 69% of STEMI patients received reperfusion therapy. The majority received fibrinolytic therapy with only 13 (15%) of STEMI patients treated with primary PCI. This rate had, however, increased from 2002 (13 (15%) vs 3 (3%), p=0.0046) (Table 2A). Table 2A. Treatments and investigations of STEMI patients: 2002 and 2007 Variables STEMI P: 02 vs 07 2002 2007 N (%) 101 (11%) 86 (8.6%) 0.18 Treatments in hospital Fibrinolytic therapy Primary PCI Enoxaparin Daltaparin UF heparin No heparin*** Tirofiban Eptifibatide Abciximab Aspirin Clopidogrel 56 (55%) 3 (3.0%) 33 (34%) 6 (5.9%) 28 (28%) 40 (40%) 5 (5.0%) 2 (2.0%) 1 (1%) 87 (88%) 14 (14%) 47(55%) 13 (15%) 42 (49%) 0 28 (33%) 18 (21%) 2 (2.3%) 1 (1.2%) 2 (2.3%) 81 (94%) 63 (73%) 0.87 0.0046 0.014 0.024 0.72 <0.0001 0.37 0.68 0.45 0.082 <0.0001 Investigations in hospital Chest X-ray Echocardiogram Exercise test Angiogram No ETT/Angio No Echo/Angio PCI CABG surgery In-hospital deaths 89 (89%) 35 (35%) 18 (18%) 31 (31%) 57 (57%) 50 (50%)