https://orcid.org/0000-0001-7630-907X
Abstract
Pharmaco-invasive percutaneous coronary intervention (PI-PCI) is recommended for patients with ST-elevation myocardial infarction (STEMI)who are unable to undergo timely primary PCI (pPCI). The present study examined late outcomes after PI-PCI (successful reperfusion followed by scheduled PCI or failed reperfusion and rescue PCI)compared with timely and late pPCI (>120 min from first medical contact).
All patients with STEMI presenting within 12 h of symptom onset, who underwent PCI during their initial hospitalization at Liverpool Hospital (Sydney), from October 2003 to March 2014, were included. Amongst 2091 STEMI patients (80% male), 1077 (52%)underwent pPCI (68% timely, 32% late), and 1014 (48%)received PI-PCI (33% rescue, 67% scheduled). Mortality at 3 years was 11.1% after pPCI (6.7% timely, 20.2% late) and 6.2% after PI-PCI (9.4% rescue, 4.8% scheduled); P < 0.01. After propensity matching, the adjusted mortality hazard ratio (HR) for timely pPCI compared with scheduled PCI was 0.9 (95% CIs 0.4–2.0) and compared with rescue PCI was 0.5 (95% CIs 0.2–0.9). The adjusted mortality HR for late pPCI, compared with scheduled PCI was 2.2 (95% CIs 1.2–3.1)and compared with rescue PCI, it was 1.5 (95% CIs 0.7–2.0).
Patients who underwent late pPCI had higher mortality rates than those undergoing a pharmaco-invasive strategy. Despite rescue PCI being required in a third of patients, a pharmaco-invasive approach should be considered when delays to PCI are anticipated, as it achieves better outcomes than late pPCI.
A propensity-matched comparison of late outcomes amongst patients who underwent timely primary PCI, late primary PCI, or a pharmaco-invasive strategy; STEMI, ST-Elevation Myocardial Infaction; PCI, percutaneous coronary intervention; pPCI,primary percutaneous coronary intervention.
See the editorial comment for this article ‘Reperfusion in ST-elevation myocardial infarction: delays have dangerous ends’, by J. J. Coughlan and Borja Ibanez, https://doi.org/10.1093/eurheartj/ehac723.
Introduction
For patients with ST-elevation myocardial infarction (STEMI), ‘timely’ primary percutaneous coronary intervention (pPCI) is guideline-recommended to enhance myocardial salvage and improve survival rates.1–3 European guidelines recommend pPCI to occur within 120 min of first medical contact (FMC).1,3,4 Due to the location of some patients developing STEMI with respect to pPCI centres, where timely pPCI is not an option, fibrinolytic therapy is recommended.1,3,5,6 Such barriers to ‘timely’ pPCI commonly occur in Australia and North America.,1,4,6–8
The meta-analysis of Keeley et al.9 reported the superiority of pPCI compared with fibrinolytic therapy, when followed by variable rates of coronary angiography, though The Comparison of primary Angioplasty and Pre-hospital fibrinolysis In acute Myocardial Infarction (CAPTIM) study, being the first trial comparing pPCI to a pharmaco-invasive strategy,10 was not included. This strategy involved emergency angiography and rescue PCI for those who failed pharmacologic reperfusion11 and, for others, transfer to a PCI facility for early angiography, and PCI if indicated.1,3 The safety and efficacy of pharmaco-invasive PCI (PI-PCI) compared with pPCI have been supported by the Strategic Reperfusion Early After Myocardial Infarction Trial (STREAM-1) and registry data from Alberta (Canada) and France8,12,13 Indeed, recent registries from France and Norway reported better outcomes among patients who underwent PI-PCI compared with late pPCI.14,15
In routine practice, pPCI is often performed beyond recommended time targets, perhaps due to the high failure rates of reperfusion with fibrinolytic therapy, thus the need for rescue PCI in a third of patients16 with its perceived bleeding risks as well as unrealistic assessments of the probable times for performance of pPCI.14,17 Thus, we aimed to compare late mortality amongst unselected patients with STEMI undergoing PI-PCI to those undergoing pPCI, including at <90 min and <120 min from FMC and later. Rates of MI, stent thrombosis (ST), target vessel revascularization (TVR), stroke, and bleeding were also analysed.
Methods
Study population
Consecutive patients with STEMI who presented within 12 h of symptom onset and underwent PCI during their initial hospitalization at the Liverpool Hospital (South Western Sydney, Australia) between October 2003 and March 2014 were included (Figure 1). Patients presenting with STEMI in the South Western Sydney Local Health District (SWSLHD) prior to May 2006 (Period 1) received initial fibrinolytic therapy with weight adjusted tenecteplase (unless contraindicated) and rescue PCI was performed for failed pharmacological reperfusion, as previously described.18 During this period, only patients who presented to the Liverpool Hospital Emergency Department, during working hours underwent pPCI. Between June 2006 and May 2010 (Period 2), pPCI was offered to all patients presenting to Liverpool Hospital. From June 2010 (Period 3), patients who called an ambulance within ∼45 min drive-time and met STEMI criteria on their paramedic-obtained ECG were transported directly for pPCI.19 In Period 3, patients presenting to EDs of non-PCI centres or calling ambulances when situated >45 min drive away received fibrinolytic therapy and were transferred for early scheduled angiography and PCI, often called a ‘drip and ship’ strategy. Baseline clinical, angiographic, and procedural data were recorded prospectively in the cardiology department database at Liverpool Hospital as previously described.20 Patients with prior coronary artery bypass grafting and cardiogenic shock were included. Follow-up after PCI was approved as a quality assurance project by the SWSLHD human research ethics committee (QA2008/034).
Study population diagram.
Percutaneous coronary intervention procedures
Unless contraindicated, aspirin (300 mg) was given immediately at presentation or pre-PCI and continued indefinitely thereafter at 100 to 150 mg/day. P2Y12 inhibitors were administered as follows: a clopidogrel loading dose of 300 or 600 mg was given either ‘upstream’ or peri-procedurally and continued at 75 mg/day post-PCI for all PI-PCI and for pPCI. After prasugrel was introduced in 2009, this was recommended for pPCI, using a loading dose of 60 mg and 10 mg/day post-PCI. Ticagrelor (introduced in 2012) was given as a loading dose of 180 mg either upstream or at the time of PCI and continued at 90 mg twice daily post-PCI. All three P2Y12 inhibitors were recommended to be continued for at least 12 months post-PCI.1,3 A bolus of unfractionated heparin (60–100 U/kg) or a bolus and infusion of bivalirudin (recommended for 2–4 h) were given at the start of PCI procedures. Bare metal stents or drug-eluting stents (DES) were deployed according to previously reported criteria,6 with second-generation DES becoming available in October 2009. Over the study period, the usual management of patients with multi-vessel disease, in the absence of cardiogenic shock, was to treat the culprit lesion only at the time of the acute event, unless multi-vessel PCI was performed due to the ambiguity of the culprit lesion(s). Intravenous glycoprotein IIb/IIIa inhibitors, either tirofiban, abciximab, or eptifibatide, were administered at the operator’s discretion. Statins, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and β-blockers were recommended, unless contraindicated.
Definitions
Criteria for STEMI were chest pain of ≥20 min and ST-segment elevation ≥1 mm in two contiguous leads (or ≥2 mm in two contiguous leads V1-V3) or new left bundle-branch block, together with a rise and/or fall in levels of cardiac biomarkers {creatine kinase [>2× upper reference limit (URL)], creatine kinase-MB (> URL), or troponin T (> URL)}.21,22
Cardiogenic shock was defined as refractory hypotension (systolic blood pressure <90 mmHg lasting ≥1 h) and end-organ hypoperfusion, with or without mechanical support.23 Serum creatinine was collected from Liverpool Hospital patient databases and used to calculate the estimated glomerular filtration rate (eGFR) of patients with the Chronic Kidney Disease Epidemiology Collaboration equation.24,25 Significant angiographic coronary lesions were defined as ≥70% (≥50% for left main stenosis). Lesions were classified according to the American College of Cardiology/American Heart Association (ACC/AHA) criteria.26 Angiographically, successful PCI was defined as <20% stenosis post-stenting (or <50% post-balloon angioplasty) and Thrombolysis in Myocardial Infarction (TIMI) flow 3.27
Outcomes
Clinical follow-up was performed by research staff (nurses and/or physicians). Patients, their cardiologists, or general practitioners were contacted by email or telephone to ascertain recurrent cardiac symptoms requiring hospitalization, need for coronary re-vascularization, MI, or stroke. Mortality was obtained from medical records, physicians, and next of kin. ST and restenosis were angiographically verified.
The primary outcome was all-cause mortality at 3 years. Secondary outcomes included (i) all-cause mortality, (ii) re-infarction defined by chest pain lasting ≥20 min and accompanied by new electrocardiographic changes (Q waves >0.04 s or ST-segment elevation >0.1 mV) and/or further biomarker rise [creatine kinase (>2× URL), creatine kinase-MB (> URL, or troponin T (> URL)],28 (iii) TVR defined as ischaemia-driven repeat revascularization of the infarct related artery requiring repeat PCI or coronary artery bypass graft surgery, and (iv) ST confirmed by angiography as defined by the Academic Research Consortium.29 Rates of stroke, cardiovascular (CV) death, and bleeding according to the Bleeding Academic Research Consortium (BARC) criteria were also reported.
The composite endpoint was defined as a combination of all-cause mortality, re-infarction, TVR, and ST.
Data analysis
Categorical variables are presented as numbers and percentages, while continuous variables are displayed as mean ± standard deviation (SD) or medians with 25th and 75th percentiles. The Pearson χ2 test and the Fisher exact test were used to compare unpaired categorical variables, while the Student’s t-test and Mann–Whitney U test were used for continuous variables. Kaplan–Meier analyses were used to compare mortality rates with significance between groups (pPCI vs. PI-PCI) assessed by the log rank (Mantel-Cox) test. Patients lost to follow-up were considered censored. Cox regression analyses were used to calculate hazard ratios (HRs) with 95% confidence intervals (CIs). Proportional hazard assumptions were tested based on Schoenfeld residuals and were not violated for any analyses. Variables with P < 0.1 in univariable analyses were included in a multivariable Cox regression analysis model to evaluate the influence of relevant variables on all-cause mortality. Variables included were: age ≥65 years, female gender, diabetes mellitus, hypertension, hyperlipidaemia, current smoking, family history of ischaemic heart disease, era of presentation, pre-PCI cardiogenic shock, anterior MI [defined as left main coronary artery or left anterior descending coronary artery (LAD) culprit lesions], ostial lesions, calcified lesions, ACC/AHA lesion classification type C, left ventricular ejection fraction , presence of post-procedural contrast-induced nephropathy, and eGFR <60 mL/min/1.73 m2.
To account for the bias inherent in the selection of either a pharmaco-invasive strategy or pPCI, propensity matching was undertaken using all baseline clinical characteristics including, infarct location, symptom onset to FMC time, and era of presentation, as variables. Matching was performed using 1:1 nearest neighbour matching (without replacement) within 0.1 of the standard deviation. Balance was assessed by calculating the standardized mean difference (SMD). A SMD of <0.1 was obtained for all covariates to ensure well-balanced cohorts. The four propensity-matched cohorts were formed: scheduled PCI and timely pPCI; scheduled PCI and late pPCI; rescue PCI and timely pPCI; and rescue PCI and late pPCI.
A sensitivity analysis was conducted, including patients who presented within 3 h of symptom onset to exclude the impact of late presenters (Appendix B). A propensity-matched analysis excluding patients who received late pPCI was also performed (Appendix C). To assess the possibility of competing risks, we estimated proportional sub-distribution hazard models using a Fine and Grey modified Cox proportional hazard regression (Appendix D).
All statistical analyses were performed with IBM SPSS version 27 (SPSS, Inc., Chicago, IL) and R version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria). All P-values used 2-tailed tests, and P-values of <0.05 were considered statistically significant.
Results
Clinical and procedural characteristics
From October 2003 to March 2014, 2091 patients with STEMI presenting within 12 h of symptom onset survived to undergo pPCI (n = 1077) or PI-PCI (n = 1014). Of those undergoing PI-PCI, 33% required rescue PCI, and 67% had early PCI after successful pharmacological reperfusion. Among those who underwent pPCI, 38% had a FMC to first device time (FDT)of <90 min, 30% had this time between 90–119 min (called ‘timely’ pPCI), and 32% had a FMC to FDT of ≥120 min (called ‘late’ pPCI). The median symptom onset to balloon time was 215 [154–324] min in patients treated by pPCI and 326 [252–483] min in those that required rescue PCI (P < 0.01). The median follow-up time was 42 [26–63] months, with 58 patients (2.8%) lost to follow-up within 6 months, and 5.4% (112 patients) lost to clinical follow-up at any time.
Patients undergoing PI-PCI compared with those undergoing pPCI were younger, had higher eGFR, less cardiogenic shock pre-PCI, less complex disease, less proximal LAD culprits, had shorter segments stented, and received less glycoprotein IIb/IIIa inhibitors (P < 0.01); other clinical, angiographic, and procedural characteristics are shown in Tables 1 and 2.
Baseline clinical characteristics
| All (n = 2091) | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Age (years) | 59 ± 12 | 61 ± 12 | 60 ± 12 | 61 ± 13 | 59 ± 12 | 59 ± 12 | 59 ± 12 | 0.013 |
| Age ≥ 75 years | 275 (13) | 95 (13) | 63 (18) | 158 (15) | 85 (12) | 32 (10) | 117 (12) | 0.023 |
| Male (sex) | 1664 (80) | 622 (85) | 229 (66) | 851 (79) | 535 (78) | 279 (84) | 814 (80) | 0.479 |
| Body mass index (kg/m2) | 27 [25–31] | 27 [24–32] | 28 [25–31] | 27 [25–31] | 27 [25–31] | 28 [25–30] | 27 [24–31] | 0.413 |
| Diabetes | 416 (20) | 142 (19) | 80 (23) | 222 (21) | 139 (20) | 55 (17) | 194 (19) | 0.385 |
| Current smoker | 811 (39) | 320 (44) | 105 (30) | 425 (39) | 260 (38) | 126 (38) | 386 (38) | 0.479 |
| Hypertension | 899 (43) | 312 (43) | 171 (50) | 483 (45) | 291 (43) | 125 (38) | 416 (41) | 0.065 |
| Family history of CHD | 402 (19) | 137 (19) | 69 (20) | 206 (19) | 136 (20) | 60 (18) | 196 (19) | 0.888 |
| Hyperlipidaemiab | 1096 (52) | 360 (49) | 185 (54) | 545 (51) | 397 r (58) | 154 (47) | 551 (54) | 0.087 |
| Cardiogenic shock pre-PCI | 113 (5) | 43 (6) | 35 (10) | 78 (7) | 20 (3) | 15 (5) | 35 (3) | <0.001 |
| Previous PCI | 70 (3) | 13 (2) | 35 (10) | 48 (4) | 15 (2) | 7 (2) | 22 (2) | <0.001 |
| Previous CABG | 29 (1) | 0 (0) | 13 (4) | 13 (1) | 13 (2) | 3 (1) | 16 (2) | 0.904 |
| Anaemiac | 154 (7) | 20 (3) | 34 (10) | 55 (5) | 73 (11) | 26 (8) | 99 (10) | 0.196 |
| Renal function on admission | ||||||||
| ȃeGFR ≥90 | 569 (27) | 215 (29) | 71 (21) | 286 (27) | 185 (27) | 98 (30) | 283 (28) | 0.593 |
| ȃeGFR 60–89 | 1057 (51) | 340 (46) | 141 (41) | 481 (45) | 399 (58) | 177 (53) | 576 (57) | 0.250 |
| ȃeGFR 30–59 | 403 (19) | 120 (16) | 92 (27) | 212 (20) | 123 (18) | 68 (21) | 191 (19) | 0.121 |
| ȃeGFR 15–29 | 43 (2) | 8 (1) | 15 (4) | 23 (2) | 9 (1) | 11 (3) | 20 (2) | 0.062 |
| ȃeGFR <15 | 11 (1) | 2 (<1) | 3 (1) | 5 (<1) | 5 (1) | 1 (<1) | 6 (1) | 0.276 |
| Aspirin prior to MI | 1224 (59) | 444 (61) | 203 (59) | 647 (60) | 383 (56) | 194 (59) | 577 (57) | 0.052 |
| Symptom onset to balloon time (min) | 230 [162–354] | 169 [134–254] | 263 [213–375] | 196 [154–324] | N/A | 326 [252–483] | 326 [252–483] | <0.001 |
| Symptom onset to first medical contact time (min) | 90 [51–184] | 91 [55–158] | 94 [52–215] | 92 [54–186] | 89 [50–149] | 87 [46–216] | 88 [49—182] | 0.552 |
| LVEF (%) | 51 ± 13 | 50 ± 11 | 48 ± 13 | 49 ± 12 | 54 ± 12 | 50 ± 12 | 53 ± 13 | <0.001 |
| Anterior infarction | 892 (43) | 304 (42) | 150 (43) | 454 (42) | 285 (42) | 153 (46) | 438 (43) | 0.603 |
| All (n = 2091) | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Age (years) | 59 ± 12 | 61 ± 12 | 60 ± 12 | 61 ± 13 | 59 ± 12 | 59 ± 12 | 59 ± 12 | 0.013 |
| Age ≥ 75 years | 275 (13) | 95 (13) | 63 (18) | 158 (15) | 85 (12) | 32 (10) | 117 (12) | 0.023 |
| Male (sex) | 1664 (80) | 622 (85) | 229 (66) | 851 (79) | 535 (78) | 279 (84) | 814 (80) | 0.479 |
| Body mass index (kg/m2) | 27 [25–31] | 27 [24–32] | 28 [25–31] | 27 [25–31] | 27 [25–31] | 28 [25–30] | 27 [24–31] | 0.413 |
| Diabetes | 416 (20) | 142 (19) | 80 (23) | 222 (21) | 139 (20) | 55 (17) | 194 (19) | 0.385 |
| Current smoker | 811 (39) | 320 (44) | 105 (30) | 425 (39) | 260 (38) | 126 (38) | 386 (38) | 0.479 |
| Hypertension | 899 (43) | 312 (43) | 171 (50) | 483 (45) | 291 (43) | 125 (38) | 416 (41) | 0.065 |
| Family history of CHD | 402 (19) | 137 (19) | 69 (20) | 206 (19) | 136 (20) | 60 (18) | 196 (19) | 0.888 |
| Hyperlipidaemiab | 1096 (52) | 360 (49) | 185 (54) | 545 (51) | 397 r (58) | 154 (47) | 551 (54) | 0.087 |
| Cardiogenic shock pre-PCI | 113 (5) | 43 (6) | 35 (10) | 78 (7) | 20 (3) | 15 (5) | 35 (3) | <0.001 |
| Previous PCI | 70 (3) | 13 (2) | 35 (10) | 48 (4) | 15 (2) | 7 (2) | 22 (2) | <0.001 |
| Previous CABG | 29 (1) | 0 (0) | 13 (4) | 13 (1) | 13 (2) | 3 (1) | 16 (2) | 0.904 |
| Anaemiac | 154 (7) | 20 (3) | 34 (10) | 55 (5) | 73 (11) | 26 (8) | 99 (10) | 0.196 |
| Renal function on admission | ||||||||
| ȃeGFR ≥90 | 569 (27) | 215 (29) | 71 (21) | 286 (27) | 185 (27) | 98 (30) | 283 (28) | 0.593 |
| ȃeGFR 60–89 | 1057 (51) | 340 (46) | 141 (41) | 481 (45) | 399 (58) | 177 (53) | 576 (57) | 0.250 |
| ȃeGFR 30–59 | 403 (19) | 120 (16) | 92 (27) | 212 (20) | 123 (18) | 68 (21) | 191 (19) | 0.121 |
| ȃeGFR 15–29 | 43 (2) | 8 (1) | 15 (4) | 23 (2) | 9 (1) | 11 (3) | 20 (2) | 0.062 |
| ȃeGFR <15 | 11 (1) | 2 (<1) | 3 (1) | 5 (<1) | 5 (1) | 1 (<1) | 6 (1) | 0.276 |
| Aspirin prior to MI | 1224 (59) | 444 (61) | 203 (59) | 647 (60) | 383 (56) | 194 (59) | 577 (57) | 0.052 |
| Symptom onset to balloon time (min) | 230 [162–354] | 169 [134–254] | 263 [213–375] | 196 [154–324] | N/A | 326 [252–483] | 326 [252–483] | <0.001 |
| Symptom onset to first medical contact time (min) | 90 [51–184] | 91 [55–158] | 94 [52–215] | 92 [54–186] | 89 [50–149] | 87 [46–216] | 88 [49—182] | 0.552 |
| LVEF (%) | 51 ± 13 | 50 ± 11 | 48 ± 13 | 49 ± 12 | 54 ± 12 | 50 ± 12 | 53 ± 13 | <0.001 |
| Anterior infarction | 892 (43) | 304 (42) | 150 (43) | 454 (42) | 285 (42) | 153 (46) | 438 (43) | 0.603 |
Values are expressed as mean ± SD, n (%), or median [IQR].
CABG, coronary artery bypass graft; CHD, coronary heart disease; eGFR, estimated glomerular filtration rate (ml/min/1.73 m2); PCI, Percutaneous coronary intervention; pPCI, primary PCI, PI-PCI, pharmaco-invasive PCI, MI, myocardial infarction, LVEF, left ventricular ejection fraction.
P-value comparing pPCI with PI-PCI.
Hyperlipidaemia defined as previous diagnosis of hypercholesterolaemia, including treatment with lipid-lowering agents or fasting low-density lipoprotein cholesterol of ≥130 mg/dl or total cholesterol of ≥200 mg/dl.
Anaemia is defined as a haemoglobin of <120 g/L in females, <130 g/L in males.
Baseline clinical characteristics
| All (n = 2091) | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Age (years) | 59 ± 12 | 61 ± 12 | 60 ± 12 | 61 ± 13 | 59 ± 12 | 59 ± 12 | 59 ± 12 | 0.013 |
| Age ≥ 75 years | 275 (13) | 95 (13) | 63 (18) | 158 (15) | 85 (12) | 32 (10) | 117 (12) | 0.023 |
| Male (sex) | 1664 (80) | 622 (85) | 229 (66) | 851 (79) | 535 (78) | 279 (84) | 814 (80) | 0.479 |
| Body mass index (kg/m2) | 27 [25–31] | 27 [24–32] | 28 [25–31] | 27 [25–31] | 27 [25–31] | 28 [25–30] | 27 [24–31] | 0.413 |
| Diabetes | 416 (20) | 142 (19) | 80 (23) | 222 (21) | 139 (20) | 55 (17) | 194 (19) | 0.385 |
| Current smoker | 811 (39) | 320 (44) | 105 (30) | 425 (39) | 260 (38) | 126 (38) | 386 (38) | 0.479 |
| Hypertension | 899 (43) | 312 (43) | 171 (50) | 483 (45) | 291 (43) | 125 (38) | 416 (41) | 0.065 |
| Family history of CHD | 402 (19) | 137 (19) | 69 (20) | 206 (19) | 136 (20) | 60 (18) | 196 (19) | 0.888 |
| Hyperlipidaemiab | 1096 (52) | 360 (49) | 185 (54) | 545 (51) | 397 r (58) | 154 (47) | 551 (54) | 0.087 |
| Cardiogenic shock pre-PCI | 113 (5) | 43 (6) | 35 (10) | 78 (7) | 20 (3) | 15 (5) | 35 (3) | <0.001 |
| Previous PCI | 70 (3) | 13 (2) | 35 (10) | 48 (4) | 15 (2) | 7 (2) | 22 (2) | <0.001 |
| Previous CABG | 29 (1) | 0 (0) | 13 (4) | 13 (1) | 13 (2) | 3 (1) | 16 (2) | 0.904 |
| Anaemiac | 154 (7) | 20 (3) | 34 (10) | 55 (5) | 73 (11) | 26 (8) | 99 (10) | 0.196 |
| Renal function on admission | ||||||||
| ȃeGFR ≥90 | 569 (27) | 215 (29) | 71 (21) | 286 (27) | 185 (27) | 98 (30) | 283 (28) | 0.593 |
| ȃeGFR 60–89 | 1057 (51) | 340 (46) | 141 (41) | 481 (45) | 399 (58) | 177 (53) | 576 (57) | 0.250 |
| ȃeGFR 30–59 | 403 (19) | 120 (16) | 92 (27) | 212 (20) | 123 (18) | 68 (21) | 191 (19) | 0.121 |
| ȃeGFR 15–29 | 43 (2) | 8 (1) | 15 (4) | 23 (2) | 9 (1) | 11 (3) | 20 (2) | 0.062 |
| ȃeGFR <15 | 11 (1) | 2 (<1) | 3 (1) | 5 (<1) | 5 (1) | 1 (<1) | 6 (1) | 0.276 |
| Aspirin prior to MI | 1224 (59) | 444 (61) | 203 (59) | 647 (60) | 383 (56) | 194 (59) | 577 (57) | 0.052 |
| Symptom onset to balloon time (min) | 230 [162–354] | 169 [134–254] | 263 [213–375] | 196 [154–324] | N/A | 326 [252–483] | 326 [252–483] | <0.001 |
| Symptom onset to first medical contact time (min) | 90 [51–184] | 91 [55–158] | 94 [52–215] | 92 [54–186] | 89 [50–149] | 87 [46–216] | 88 [49—182] | 0.552 |
| LVEF (%) | 51 ± 13 | 50 ± 11 | 48 ± 13 | 49 ± 12 | 54 ± 12 | 50 ± 12 | 53 ± 13 | <0.001 |
| Anterior infarction | 892 (43) | 304 (42) | 150 (43) | 454 (42) | 285 (42) | 153 (46) | 438 (43) | 0.603 |
| All (n = 2091) | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Age (years) | 59 ± 12 | 61 ± 12 | 60 ± 12 | 61 ± 13 | 59 ± 12 | 59 ± 12 | 59 ± 12 | 0.013 |
| Age ≥ 75 years | 275 (13) | 95 (13) | 63 (18) | 158 (15) | 85 (12) | 32 (10) | 117 (12) | 0.023 |
| Male (sex) | 1664 (80) | 622 (85) | 229 (66) | 851 (79) | 535 (78) | 279 (84) | 814 (80) | 0.479 |
| Body mass index (kg/m2) | 27 [25–31] | 27 [24–32] | 28 [25–31] | 27 [25–31] | 27 [25–31] | 28 [25–30] | 27 [24–31] | 0.413 |
| Diabetes | 416 (20) | 142 (19) | 80 (23) | 222 (21) | 139 (20) | 55 (17) | 194 (19) | 0.385 |
| Current smoker | 811 (39) | 320 (44) | 105 (30) | 425 (39) | 260 (38) | 126 (38) | 386 (38) | 0.479 |
| Hypertension | 899 (43) | 312 (43) | 171 (50) | 483 (45) | 291 (43) | 125 (38) | 416 (41) | 0.065 |
| Family history of CHD | 402 (19) | 137 (19) | 69 (20) | 206 (19) | 136 (20) | 60 (18) | 196 (19) | 0.888 |
| Hyperlipidaemiab | 1096 (52) | 360 (49) | 185 (54) | 545 (51) | 397 r (58) | 154 (47) | 551 (54) | 0.087 |
| Cardiogenic shock pre-PCI | 113 (5) | 43 (6) | 35 (10) | 78 (7) | 20 (3) | 15 (5) | 35 (3) | <0.001 |
| Previous PCI | 70 (3) | 13 (2) | 35 (10) | 48 (4) | 15 (2) | 7 (2) | 22 (2) | <0.001 |
| Previous CABG | 29 (1) | 0 (0) | 13 (4) | 13 (1) | 13 (2) | 3 (1) | 16 (2) | 0.904 |
| Anaemiac | 154 (7) | 20 (3) | 34 (10) | 55 (5) | 73 (11) | 26 (8) | 99 (10) | 0.196 |
| Renal function on admission | ||||||||
| ȃeGFR ≥90 | 569 (27) | 215 (29) | 71 (21) | 286 (27) | 185 (27) | 98 (30) | 283 (28) | 0.593 |
| ȃeGFR 60–89 | 1057 (51) | 340 (46) | 141 (41) | 481 (45) | 399 (58) | 177 (53) | 576 (57) | 0.250 |
| ȃeGFR 30–59 | 403 (19) | 120 (16) | 92 (27) | 212 (20) | 123 (18) | 68 (21) | 191 (19) | 0.121 |
| ȃeGFR 15–29 | 43 (2) | 8 (1) | 15 (4) | 23 (2) | 9 (1) | 11 (3) | 20 (2) | 0.062 |
| ȃeGFR <15 | 11 (1) | 2 (<1) | 3 (1) | 5 (<1) | 5 (1) | 1 (<1) | 6 (1) | 0.276 |
| Aspirin prior to MI | 1224 (59) | 444 (61) | 203 (59) | 647 (60) | 383 (56) | 194 (59) | 577 (57) | 0.052 |
| Symptom onset to balloon time (min) | 230 [162–354] | 169 [134–254] | 263 [213–375] | 196 [154–324] | N/A | 326 [252–483] | 326 [252–483] | <0.001 |
| Symptom onset to first medical contact time (min) | 90 [51–184] | 91 [55–158] | 94 [52–215] | 92 [54–186] | 89 [50–149] | 87 [46–216] | 88 [49—182] | 0.552 |
| LVEF (%) | 51 ± 13 | 50 ± 11 | 48 ± 13 | 49 ± 12 | 54 ± 12 | 50 ± 12 | 53 ± 13 | <0.001 |
| Anterior infarction | 892 (43) | 304 (42) | 150 (43) | 454 (42) | 285 (42) | 153 (46) | 438 (43) | 0.603 |
Values are expressed as mean ± SD, n (%), or median [IQR].
CABG, coronary artery bypass graft; CHD, coronary heart disease; eGFR, estimated glomerular filtration rate (ml/min/1.73 m2); PCI, Percutaneous coronary intervention; pPCI, primary PCI, PI-PCI, pharmaco-invasive PCI, MI, myocardial infarction, LVEF, left ventricular ejection fraction.
P-value comparing pPCI with PI-PCI.
Hyperlipidaemia defined as previous diagnosis of hypercholesterolaemia, including treatment with lipid-lowering agents or fasting low-density lipoprotein cholesterol of ≥130 mg/dl or total cholesterol of ≥200 mg/dl.
Anaemia is defined as a haemoglobin of <120 g/L in females, <130 g/L in males.
Angiographic and Procedural Characteristics
| All n = 2091 | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-value a | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Culprit artery | ||||||||
| ȃLeft main | 17 (1) | 6 (1) | 6 (2) | 12 (1) | 2 (<1) | 3 (1) | 5 (<1) | 0.117 |
| ȃLAD | 946 (45) | 325 (44) | 163 (47) | 488 (45) | 295 (43) | 163 (49) | 458 (45) | 0.938 |
| ȃLCx | 269 (13) | 94 (13) | 51 (15) | 145 (13) | 100 (15) | 24 (7) | 124 (12) | 0.390 |
| ȃRCA | 827 (40) | 292 (40) | 124 (36) | 416 (39) | 278 (41) | 133 (40) | 411 (41) | 0.338 |
| ȃBypass graft | 22 (1) | 7 (1) | 6 (2) | 13 (1) | 5 (1) | 4 (1) | 9 (1) | 0.483 |
| B2/C Lesion (ACC/AHA) | 1479 (71) | 548 (75) | 269 (78) | 817 (76) | 436 (64) | 227 (69) | 662 (65) | <0.001 |
| Proximal LAD | 655 (31) | 262 (36) | 99 (29) | 361 (34) | 186 (27) | 108 (33) | 294 (29) | 0.034 |
| Lesions at bifurcation | 331 (16) | 115 (16) | 44 (13) | 159 (15) | 119 (17) | 53 (16) | 172 (17) | 0.174 |
| Lesion calcifications | 176 (8) | 46 (6) | 34 (10) | 80 (8) | 68 (10) | 28 (8) | 96 (9) | 0.085 |
| Ostial lesion | 117 (6) | 36 (5) | 28 (8) | 64 (6) | 30 (4) | 23 (7) | 53 (5) | 0.498 |
| Pre-PCI coronary artery stenosis | 95 [90–100] | 100 [95–100] | 100 [95–100] | 100 [95–100] | 90 [80–99] | 95 [90–100] | 90 [88–99] | <0.001 |
| Culprit lesion length (mm) | 18 [15–28] | 20 [15–26] | 20 [15–30] | 20 [15–28] | 18 [13–25] | 18 [14–25] | 18 [13–26] | 0.005 |
| Lesion length ≥20mm | 1004 (48) | 361 (49) | 179 (52) | 540 (50) | 310 (45) | 154 (47) | 464 (46) | 0.039 |
| >1 stent, n (%) | 586 (28) | 212 (29) | 130 (38) | 342 (32) | 170 (25) | 74 (22) | 244 (24) | 0.001 |
| Stent diameter | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 0.102 |
| Glycoprotein IIb/IIIa Inhibitor | 836 (40) | 277 (38) | 170 (49) | 447 (42) | 232 (34) | 157 (47) | 389 (38) | <0.001 |
| TIMI flow pre-PCI | ||||||||
| ȃ0 | 918 (44) | 459 (63) | 221 (64) | 680 (63) | 73 (11) | 165 (50) | 238 (23) | <0.001 |
| ȃ1 | 215 (10) | 116 (16) | 39 (11) | 155 (14) | 34 (5) | 26 (8) | 60 (6) | <0.001 |
| ȃ2 | 354 (17) | 84 (11) | 45 (13) | 129 (12) | 168 (25) | 57 (17) | 225 (22) | 0.001 |
| ȃ3 | 604 (29) | 73 (10) | 40 (12) | 113 (11) | 408 (60) | 83 (25) | 491 (48) | <0.001 |
| TIMI grade 3 flow after PCI | 2011 (96) | 705 (96) | 331 (96) | 1036 (96) | 661 (97) | 315 (95) | 975 (96) | 0.490 |
| Angiographic success | 1980 (95) | 696 (95) | 324 (94) | 1020 (95) | 654 (96) | 306 (92) | 960 (95) | 0.757 |
| All n = 2091 | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-value a | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Culprit artery | ||||||||
| ȃLeft main | 17 (1) | 6 (1) | 6 (2) | 12 (1) | 2 (<1) | 3 (1) | 5 (<1) | 0.117 |
| ȃLAD | 946 (45) | 325 (44) | 163 (47) | 488 (45) | 295 (43) | 163 (49) | 458 (45) | 0.938 |
| ȃLCx | 269 (13) | 94 (13) | 51 (15) | 145 (13) | 100 (15) | 24 (7) | 124 (12) | 0.390 |
| ȃRCA | 827 (40) | 292 (40) | 124 (36) | 416 (39) | 278 (41) | 133 (40) | 411 (41) | 0.338 |
| ȃBypass graft | 22 (1) | 7 (1) | 6 (2) | 13 (1) | 5 (1) | 4 (1) | 9 (1) | 0.483 |
| B2/C Lesion (ACC/AHA) | 1479 (71) | 548 (75) | 269 (78) | 817 (76) | 436 (64) | 227 (69) | 662 (65) | <0.001 |
| Proximal LAD | 655 (31) | 262 (36) | 99 (29) | 361 (34) | 186 (27) | 108 (33) | 294 (29) | 0.034 |
| Lesions at bifurcation | 331 (16) | 115 (16) | 44 (13) | 159 (15) | 119 (17) | 53 (16) | 172 (17) | 0.174 |
| Lesion calcifications | 176 (8) | 46 (6) | 34 (10) | 80 (8) | 68 (10) | 28 (8) | 96 (9) | 0.085 |
| Ostial lesion | 117 (6) | 36 (5) | 28 (8) | 64 (6) | 30 (4) | 23 (7) | 53 (5) | 0.498 |
| Pre-PCI coronary artery stenosis | 95 [90–100] | 100 [95–100] | 100 [95–100] | 100 [95–100] | 90 [80–99] | 95 [90–100] | 90 [88–99] | <0.001 |
| Culprit lesion length (mm) | 18 [15–28] | 20 [15–26] | 20 [15–30] | 20 [15–28] | 18 [13–25] | 18 [14–25] | 18 [13–26] | 0.005 |
| Lesion length ≥20mm | 1004 (48) | 361 (49) | 179 (52) | 540 (50) | 310 (45) | 154 (47) | 464 (46) | 0.039 |
| >1 stent, n (%) | 586 (28) | 212 (29) | 130 (38) | 342 (32) | 170 (25) | 74 (22) | 244 (24) | 0.001 |
| Stent diameter | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 0.102 |
| Glycoprotein IIb/IIIa Inhibitor | 836 (40) | 277 (38) | 170 (49) | 447 (42) | 232 (34) | 157 (47) | 389 (38) | <0.001 |
| TIMI flow pre-PCI | ||||||||
| ȃ0 | 918 (44) | 459 (63) | 221 (64) | 680 (63) | 73 (11) | 165 (50) | 238 (23) | <0.001 |
| ȃ1 | 215 (10) | 116 (16) | 39 (11) | 155 (14) | 34 (5) | 26 (8) | 60 (6) | <0.001 |
| ȃ2 | 354 (17) | 84 (11) | 45 (13) | 129 (12) | 168 (25) | 57 (17) | 225 (22) | 0.001 |
| ȃ3 | 604 (29) | 73 (10) | 40 (12) | 113 (11) | 408 (60) | 83 (25) | 491 (48) | <0.001 |
| TIMI grade 3 flow after PCI | 2011 (96) | 705 (96) | 331 (96) | 1036 (96) | 661 (97) | 315 (95) | 975 (96) | 0.490 |
| Angiographic success | 1980 (95) | 696 (95) | 324 (94) | 1020 (95) | 654 (96) | 306 (92) | 960 (95) | 0.757 |
Values are expressed as n (%), or median [IQR].
ACC/AHA, American College of Cardiology/American Heart Association; LAD, Left Anterior descending coronary artery; LCx, Left circumflex artery; RCA, Right coronary artery; MI, myocardial infarction; PCI, Percutaneous coronary intervention; TIMI, Thrombolysis In Myocardial Infarction.
p-value comparing pPCI with PI-PCI.
Angiographic and Procedural Characteristics
| All n = 2091 | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-value a | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Culprit artery | ||||||||
| ȃLeft main | 17 (1) | 6 (1) | 6 (2) | 12 (1) | 2 (<1) | 3 (1) | 5 (<1) | 0.117 |
| ȃLAD | 946 (45) | 325 (44) | 163 (47) | 488 (45) | 295 (43) | 163 (49) | 458 (45) | 0.938 |
| ȃLCx | 269 (13) | 94 (13) | 51 (15) | 145 (13) | 100 (15) | 24 (7) | 124 (12) | 0.390 |
| ȃRCA | 827 (40) | 292 (40) | 124 (36) | 416 (39) | 278 (41) | 133 (40) | 411 (41) | 0.338 |
| ȃBypass graft | 22 (1) | 7 (1) | 6 (2) | 13 (1) | 5 (1) | 4 (1) | 9 (1) | 0.483 |
| B2/C Lesion (ACC/AHA) | 1479 (71) | 548 (75) | 269 (78) | 817 (76) | 436 (64) | 227 (69) | 662 (65) | <0.001 |
| Proximal LAD | 655 (31) | 262 (36) | 99 (29) | 361 (34) | 186 (27) | 108 (33) | 294 (29) | 0.034 |
| Lesions at bifurcation | 331 (16) | 115 (16) | 44 (13) | 159 (15) | 119 (17) | 53 (16) | 172 (17) | 0.174 |
| Lesion calcifications | 176 (8) | 46 (6) | 34 (10) | 80 (8) | 68 (10) | 28 (8) | 96 (9) | 0.085 |
| Ostial lesion | 117 (6) | 36 (5) | 28 (8) | 64 (6) | 30 (4) | 23 (7) | 53 (5) | 0.498 |
| Pre-PCI coronary artery stenosis | 95 [90–100] | 100 [95–100] | 100 [95–100] | 100 [95–100] | 90 [80–99] | 95 [90–100] | 90 [88–99] | <0.001 |
| Culprit lesion length (mm) | 18 [15–28] | 20 [15–26] | 20 [15–30] | 20 [15–28] | 18 [13–25] | 18 [14–25] | 18 [13–26] | 0.005 |
| Lesion length ≥20mm | 1004 (48) | 361 (49) | 179 (52) | 540 (50) | 310 (45) | 154 (47) | 464 (46) | 0.039 |
| >1 stent, n (%) | 586 (28) | 212 (29) | 130 (38) | 342 (32) | 170 (25) | 74 (22) | 244 (24) | 0.001 |
| Stent diameter | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 0.102 |
| Glycoprotein IIb/IIIa Inhibitor | 836 (40) | 277 (38) | 170 (49) | 447 (42) | 232 (34) | 157 (47) | 389 (38) | <0.001 |
| TIMI flow pre-PCI | ||||||||
| ȃ0 | 918 (44) | 459 (63) | 221 (64) | 680 (63) | 73 (11) | 165 (50) | 238 (23) | <0.001 |
| ȃ1 | 215 (10) | 116 (16) | 39 (11) | 155 (14) | 34 (5) | 26 (8) | 60 (6) | <0.001 |
| ȃ2 | 354 (17) | 84 (11) | 45 (13) | 129 (12) | 168 (25) | 57 (17) | 225 (22) | 0.001 |
| ȃ3 | 604 (29) | 73 (10) | 40 (12) | 113 (11) | 408 (60) | 83 (25) | 491 (48) | <0.001 |
| TIMI grade 3 flow after PCI | 2011 (96) | 705 (96) | 331 (96) | 1036 (96) | 661 (97) | 315 (95) | 975 (96) | 0.490 |
| Angiographic success | 1980 (95) | 696 (95) | 324 (94) | 1020 (95) | 654 (96) | 306 (92) | 960 (95) | 0.757 |
| All n = 2091 | pPCI (n = 1077, 52%) | PI-PCI (n = 1014, 48%) | P-value a | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732, 68%) | Late pPCI (n = 345, 32%) | All pPCI (n = 1077) | Scheduled PCI (n = 683, 67%) | Rescue PCI (n = 331, (33%) | All PI-PCI (n = 1014) | |||
| Culprit artery | ||||||||
| ȃLeft main | 17 (1) | 6 (1) | 6 (2) | 12 (1) | 2 (<1) | 3 (1) | 5 (<1) | 0.117 |
| ȃLAD | 946 (45) | 325 (44) | 163 (47) | 488 (45) | 295 (43) | 163 (49) | 458 (45) | 0.938 |
| ȃLCx | 269 (13) | 94 (13) | 51 (15) | 145 (13) | 100 (15) | 24 (7) | 124 (12) | 0.390 |
| ȃRCA | 827 (40) | 292 (40) | 124 (36) | 416 (39) | 278 (41) | 133 (40) | 411 (41) | 0.338 |
| ȃBypass graft | 22 (1) | 7 (1) | 6 (2) | 13 (1) | 5 (1) | 4 (1) | 9 (1) | 0.483 |
| B2/C Lesion (ACC/AHA) | 1479 (71) | 548 (75) | 269 (78) | 817 (76) | 436 (64) | 227 (69) | 662 (65) | <0.001 |
| Proximal LAD | 655 (31) | 262 (36) | 99 (29) | 361 (34) | 186 (27) | 108 (33) | 294 (29) | 0.034 |
| Lesions at bifurcation | 331 (16) | 115 (16) | 44 (13) | 159 (15) | 119 (17) | 53 (16) | 172 (17) | 0.174 |
| Lesion calcifications | 176 (8) | 46 (6) | 34 (10) | 80 (8) | 68 (10) | 28 (8) | 96 (9) | 0.085 |
| Ostial lesion | 117 (6) | 36 (5) | 28 (8) | 64 (6) | 30 (4) | 23 (7) | 53 (5) | 0.498 |
| Pre-PCI coronary artery stenosis | 95 [90–100] | 100 [95–100] | 100 [95–100] | 100 [95–100] | 90 [80–99] | 95 [90–100] | 90 [88–99] | <0.001 |
| Culprit lesion length (mm) | 18 [15–28] | 20 [15–26] | 20 [15–30] | 20 [15–28] | 18 [13–25] | 18 [14–25] | 18 [13–26] | 0.005 |
| Lesion length ≥20mm | 1004 (48) | 361 (49) | 179 (52) | 540 (50) | 310 (45) | 154 (47) | 464 (46) | 0.039 |
| >1 stent, n (%) | 586 (28) | 212 (29) | 130 (38) | 342 (32) | 170 (25) | 74 (22) | 244 (24) | 0.001 |
| Stent diameter | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 3.0 [2.75–3.5] | 0.102 |
| Glycoprotein IIb/IIIa Inhibitor | 836 (40) | 277 (38) | 170 (49) | 447 (42) | 232 (34) | 157 (47) | 389 (38) | <0.001 |
| TIMI flow pre-PCI | ||||||||
| ȃ0 | 918 (44) | 459 (63) | 221 (64) | 680 (63) | 73 (11) | 165 (50) | 238 (23) | <0.001 |
| ȃ1 | 215 (10) | 116 (16) | 39 (11) | 155 (14) | 34 (5) | 26 (8) | 60 (6) | <0.001 |
| ȃ2 | 354 (17) | 84 (11) | 45 (13) | 129 (12) | 168 (25) | 57 (17) | 225 (22) | 0.001 |
| ȃ3 | 604 (29) | 73 (10) | 40 (12) | 113 (11) | 408 (60) | 83 (25) | 491 (48) | <0.001 |
| TIMI grade 3 flow after PCI | 2011 (96) | 705 (96) | 331 (96) | 1036 (96) | 661 (97) | 315 (95) | 975 (96) | 0.490 |
| Angiographic success | 1980 (95) | 696 (95) | 324 (94) | 1020 (95) | 654 (96) | 306 (92) | 960 (95) | 0.757 |
Values are expressed as n (%), or median [IQR].
ACC/AHA, American College of Cardiology/American Heart Association; LAD, Left Anterior descending coronary artery; LCx, Left circumflex artery; RCA, Right coronary artery; MI, myocardial infarction; PCI, Percutaneous coronary intervention; TIMI, Thrombolysis In Myocardial Infarction.
p-value comparing pPCI with PI-PCI.
Clinical outcomes
Overall mortality rates at 3 years were 6.2% among patients treated by PI-PCI and 11.1% after pPCI (P < 0.01). Late mortality rates among those patients who underwent late and timely pPCI were 20.2% and 6.7%, respectively (P < 0.01). Mortality after rescue PCI was 9.4%, compared with 4.8% in those having PCI after successful pharmacological reperfusion (P < 0.01). For FMC to device times of <90, 90–119, and ≥120 min, mortality rates of pPCI were 6.4%, 7.4%, and 20.2%, respectively. Unadjusted Kaplan–Meier survival rates are depicted in Figure 2, and time-specific mortality rates are shown in Table 3.
Unadjusted survival in patients with ST-elevation myocardial infarction who underwent timely primary percutaneous coronary intervention, late primary percutaneous coronary intervention or a pharmaco-invasive strategy. Unadjusted mortality at 1, 3 and 5 years is reported after primary percutaneous intervention or pharmaco-invasive PCI (A) and scheduled PCI, rescue PCI, timely primary PCI or late primary PCI(B).
Mortality rates based on time period and reperfusion strategy
| pPCI | PI-PCI | Total | P-value | ||||
|---|---|---|---|---|---|---|---|
| n (%) | Mortality, n (%) | n (%) | Mortality, n (%) | n | Mortality, n (%)b | ||
| Period 1 (October 2003 to May 2006) | 121 (24) | 25 (20.7) | 393 (77) | 24 (6.1) | 514 | 49 (9.5) | P = 0.001 |
| Period 2 (June 2006 toMay 2010) | 415 (52) | 50 (12.0) | 386 (48) | 27 (7.0) | 801 | 77 (9.6) | P = 0.020 |
| Period 3 (June 2010 to March 2014) | 541 (70) a | 44 (8.1) | 235 (30) | 11 (4.7) | 776 | 55 (7.1) | P = 0.101 |
| pPCI | PI-PCI | Total | P-value | ||||
|---|---|---|---|---|---|---|---|
| n (%) | Mortality, n (%) | n (%) | Mortality, n (%) | n | Mortality, n (%)b | ||
| Period 1 (October 2003 to May 2006) | 121 (24) | 25 (20.7) | 393 (77) | 24 (6.1) | 514 | 49 (9.5) | P = 0.001 |
| Period 2 (June 2006 toMay 2010) | 415 (52) | 50 (12.0) | 386 (48) | 27 (7.0) | 801 | 77 (9.6) | P = 0.020 |
| Period 3 (June 2010 to March 2014) | 541 (70) a | 44 (8.1) | 235 (30) | 11 (4.7) | 776 | 55 (7.1) | P = 0.101 |
pPCI primary percutaneous coronary intervention; PI-PCI, pharmaco-invasive percutaneous coronary intervention.
197 had pre-hospital cardiac catheterization laboratory activation and 344 pPCI after emergency department presentation.
Compared with Periods 1 and 2 overall mortality post-PCI for ST-elevation myocardial infarction was lower in Period 3 (9.6% and 7.1% respectively, P = 0.024).
Mortality rates based on time period and reperfusion strategy
| pPCI | PI-PCI | Total | P-value | ||||
|---|---|---|---|---|---|---|---|
| n (%) | Mortality, n (%) | n (%) | Mortality, n (%) | n | Mortality, n (%)b | ||
| Period 1 (October 2003 to May 2006) | 121 (24) | 25 (20.7) | 393 (77) | 24 (6.1) | 514 | 49 (9.5) | P = 0.001 |
| Period 2 (June 2006 toMay 2010) | 415 (52) | 50 (12.0) | 386 (48) | 27 (7.0) | 801 | 77 (9.6) | P = 0.020 |
| Period 3 (June 2010 to March 2014) | 541 (70) a | 44 (8.1) | 235 (30) | 11 (4.7) | 776 | 55 (7.1) | P = 0.101 |
| pPCI | PI-PCI | Total | P-value | ||||
|---|---|---|---|---|---|---|---|
| n (%) | Mortality, n (%) | n (%) | Mortality, n (%) | n | Mortality, n (%)b | ||
| Period 1 (October 2003 to May 2006) | 121 (24) | 25 (20.7) | 393 (77) | 24 (6.1) | 514 | 49 (9.5) | P = 0.001 |
| Period 2 (June 2006 toMay 2010) | 415 (52) | 50 (12.0) | 386 (48) | 27 (7.0) | 801 | 77 (9.6) | P = 0.020 |
| Period 3 (June 2010 to March 2014) | 541 (70) a | 44 (8.1) | 235 (30) | 11 (4.7) | 776 | 55 (7.1) | P = 0.101 |
pPCI primary percutaneous coronary intervention; PI-PCI, pharmaco-invasive percutaneous coronary intervention.
197 had pre-hospital cardiac catheterization laboratory activation and 344 pPCI after emergency department presentation.
Compared with Periods 1 and 2 overall mortality post-PCI for ST-elevation myocardial infarction was lower in Period 3 (9.6% and 7.1% respectively, P = 0.024).
The rates of MI, late TVR, and ST were similar in each PCI strategy; the proportion of late deaths being CV (50%) was similar in each group (Table 4). Rates of stroke after PI-PCI and pPCI were 0.8% (0.3% after rescue PCI) and 1.1%, respectively (P = 0.10). Rates of BARC 3b-5 bleeding after PI-PCI and pPCI were 2.2% and 2.3%, respectively (P = 0.10); after rescue PCI, scheduled PCI, and ‘late’ pPCI, bleeding occurred in 3.9%, 1.3%, and 5.2% of patients, respectively (Table 4).
Late Clinical Outcomes
| All (n = 2091) | pPCI (n = 1077) | PI –PCI (n = 1014) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732) | Late pPCI (n = 345) | All pPCI (n = 1077) | Scheduled PCI (n = 683) | Rescue PCI (n = 331) | All PI-PCI (n = 1014) | |||
| MI, n (%) | 135 (6.5) | 46 (6.3) | 24 (7.0) | 70 (6.5) | 41 (6.0) | 24 (7.3) | 65 (6.4) | 0.934 |
| Death, n (%) | 183 (8.8) | 49 (6.7) | 70 (20.2) | 119 (11.0) | 33 (4.8) | 31 (9.4) | 64 (6.2) | <0.001 |
| CV Death, n (%) | 92 (4.4) | 30 (4.1) | 33 (9.6) | 63(5.8) | 15 (2.2) | 14 (4.2) | 29 (2.9) | 0.001 |
| TVR, n (%) | 146 (7.0) | 44 (6.0) | 24 (7.0) | 68 (6.3) | 45 (6.6) | 33 (10.0) | 78 (7.7) | 0.751 |
| ST, n (%) | 102 (4.9) | 34 (4.6) | 12 (3.5) | 46 (4.2) | 33 (4.8) | 23 (6.9) | 56(5.5) | 0.985 |
| Stroke, n (%) | 20 (1.0) | 4 (0.5) | 8 (2.3) | 12 (1.1) | 7 (1.0) | 1 (0.3) | 8 (0.8) | 0.096 |
| Intracranial haemorrhage, n (%) | 3 (0.1) | 2 (0.3) | 0 (0) | 2 (0.2) | 0 (0) | 1 (0.3) | 1 (0.1) | 0.523 |
| BARC 3b-5 bleeding, n (%) | 52 (2.5) | 12 (1.6) | 18 (5.2) | 30 (2.3) | 9 (1.3) | 13 (3.9) | 22 (2.2) | 0.103 |
| All (n = 2091) | pPCI (n = 1077) | PI –PCI (n = 1014) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732) | Late pPCI (n = 345) | All pPCI (n = 1077) | Scheduled PCI (n = 683) | Rescue PCI (n = 331) | All PI-PCI (n = 1014) | |||
| MI, n (%) | 135 (6.5) | 46 (6.3) | 24 (7.0) | 70 (6.5) | 41 (6.0) | 24 (7.3) | 65 (6.4) | 0.934 |
| Death, n (%) | 183 (8.8) | 49 (6.7) | 70 (20.2) | 119 (11.0) | 33 (4.8) | 31 (9.4) | 64 (6.2) | <0.001 |
| CV Death, n (%) | 92 (4.4) | 30 (4.1) | 33 (9.6) | 63(5.8) | 15 (2.2) | 14 (4.2) | 29 (2.9) | 0.001 |
| TVR, n (%) | 146 (7.0) | 44 (6.0) | 24 (7.0) | 68 (6.3) | 45 (6.6) | 33 (10.0) | 78 (7.7) | 0.751 |
| ST, n (%) | 102 (4.9) | 34 (4.6) | 12 (3.5) | 46 (4.2) | 33 (4.8) | 23 (6.9) | 56(5.5) | 0.985 |
| Stroke, n (%) | 20 (1.0) | 4 (0.5) | 8 (2.3) | 12 (1.1) | 7 (1.0) | 1 (0.3) | 8 (0.8) | 0.096 |
| Intracranial haemorrhage, n (%) | 3 (0.1) | 2 (0.3) | 0 (0) | 2 (0.2) | 0 (0) | 1 (0.3) | 1 (0.1) | 0.523 |
| BARC 3b-5 bleeding, n (%) | 52 (2.5) | 12 (1.6) | 18 (5.2) | 30 (2.3) | 9 (1.3) | 13 (3.9) | 22 (2.2) | 0.103 |
MI, myocardial infarction; CV, cardiovascular; TVR, target vessel revascularization; ST, stent thrombosis; pPCI, primary percutaneous coronary intervention; PI-PCI, pharmaco-invasive percutaneous coronary intervention.
P-value for comparison between PI-PCI and PPCI.
Late Clinical Outcomes
| All (n = 2091) | pPCI (n = 1077) | PI –PCI (n = 1014) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732) | Late pPCI (n = 345) | All pPCI (n = 1077) | Scheduled PCI (n = 683) | Rescue PCI (n = 331) | All PI-PCI (n = 1014) | |||
| MI, n (%) | 135 (6.5) | 46 (6.3) | 24 (7.0) | 70 (6.5) | 41 (6.0) | 24 (7.3) | 65 (6.4) | 0.934 |
| Death, n (%) | 183 (8.8) | 49 (6.7) | 70 (20.2) | 119 (11.0) | 33 (4.8) | 31 (9.4) | 64 (6.2) | <0.001 |
| CV Death, n (%) | 92 (4.4) | 30 (4.1) | 33 (9.6) | 63(5.8) | 15 (2.2) | 14 (4.2) | 29 (2.9) | 0.001 |
| TVR, n (%) | 146 (7.0) | 44 (6.0) | 24 (7.0) | 68 (6.3) | 45 (6.6) | 33 (10.0) | 78 (7.7) | 0.751 |
| ST, n (%) | 102 (4.9) | 34 (4.6) | 12 (3.5) | 46 (4.2) | 33 (4.8) | 23 (6.9) | 56(5.5) | 0.985 |
| Stroke, n (%) | 20 (1.0) | 4 (0.5) | 8 (2.3) | 12 (1.1) | 7 (1.0) | 1 (0.3) | 8 (0.8) | 0.096 |
| Intracranial haemorrhage, n (%) | 3 (0.1) | 2 (0.3) | 0 (0) | 2 (0.2) | 0 (0) | 1 (0.3) | 1 (0.1) | 0.523 |
| BARC 3b-5 bleeding, n (%) | 52 (2.5) | 12 (1.6) | 18 (5.2) | 30 (2.3) | 9 (1.3) | 13 (3.9) | 22 (2.2) | 0.103 |
| All (n = 2091) | pPCI (n = 1077) | PI –PCI (n = 1014) | P-valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Timely pPCI (n = 732) | Late pPCI (n = 345) | All pPCI (n = 1077) | Scheduled PCI (n = 683) | Rescue PCI (n = 331) | All PI-PCI (n = 1014) | |||
| MI, n (%) | 135 (6.5) | 46 (6.3) | 24 (7.0) | 70 (6.5) | 41 (6.0) | 24 (7.3) | 65 (6.4) | 0.934 |
| Death, n (%) | 183 (8.8) | 49 (6.7) | 70 (20.2) | 119 (11.0) | 33 (4.8) | 31 (9.4) | 64 (6.2) | <0.001 |
| CV Death, n (%) | 92 (4.4) | 30 (4.1) | 33 (9.6) | 63(5.8) | 15 (2.2) | 14 (4.2) | 29 (2.9) | 0.001 |
| TVR, n (%) | 146 (7.0) | 44 (6.0) | 24 (7.0) | 68 (6.3) | 45 (6.6) | 33 (10.0) | 78 (7.7) | 0.751 |
| ST, n (%) | 102 (4.9) | 34 (4.6) | 12 (3.5) | 46 (4.2) | 33 (4.8) | 23 (6.9) | 56(5.5) | 0.985 |
| Stroke, n (%) | 20 (1.0) | 4 (0.5) | 8 (2.3) | 12 (1.1) | 7 (1.0) | 1 (0.3) | 8 (0.8) | 0.096 |
| Intracranial haemorrhage, n (%) | 3 (0.1) | 2 (0.3) | 0 (0) | 2 (0.2) | 0 (0) | 1 (0.3) | 1 (0.1) | 0.523 |
| BARC 3b-5 bleeding, n (%) | 52 (2.5) | 12 (1.6) | 18 (5.2) | 30 (2.3) | 9 (1.3) | 13 (3.9) | 22 (2.2) | 0.103 |
MI, myocardial infarction; CV, cardiovascular; TVR, target vessel revascularization; ST, stent thrombosis; pPCI, primary percutaneous coronary intervention; PI-PCI, pharmaco-invasive percutaneous coronary intervention.
P-value for comparison between PI-PCI and PPCI.
With respect to particular patient subgroups, the late mortality rates of patients with anterior STEMI and non-anterior STEMI after PI-PCI were 8.7% and 4.5% (P = 0.01), and after pPCI were 15.0% and 8.2% (P < 0.01), respectively. Among patients aged <60 and ≥60 years, late mortality rates after PI-PCI were 3.2% and 10.2% (P < 0.01), and after pPCI were 6.4% and 15.9% (P < 0.01), respectively. Among the 272 patients aged ≥75 years, mortality was 31.5% for those treated by PI-PCI and 25.4% for those treated with pPCI. Rates of stroke, intracranial haemorrhage, and major bleeding among this age group were similar with both PI-PCI and pPCI [0.9% vs. 1.9% (P = 0.09), 0% vs. 0.6% (P = 0.39) and 5.3% vs. 3.8% (P = 0.06), respectively]. Among the 15% of patients presenting 6–12 h from symptom onset, 40% underwent PI-PCI and 60% underwent pPCI; of the latter, 54% received timely pPCI.
Multivariable analyses
On multivariable analysis, the statistically significant predictors of 3-year mortality were pre-PCI cardiogenic shock, increasing age, TIMI flow < 3 post-PCI, eGFR <60 mL/min/1.73 m2 on admission, post-PCI bleeding (BARC 3b-5), anterior infarction, female sex, and pPCI (all P < 0.05) (Table 5).
Multivariate analysis of late outcomes
| Factor | Hazard ratio | 95% CI | P-value |
|---|---|---|---|
| Cardiogenic shock pre-PCI | 4.13 | 2.75–6.18 | <0.001 |
| eGFR <60 | 3.02 | 2.16–4.22 | <0.001 |
| Increasing age | 1.03 | 1.01–1.07 | <0.001 |
| Female sex | 1.44 | 1.04–2.00 | 0.029 |
| Anterior MIa | 1.53 | 1.15–2.05 | 0.004 |
| Hypercholesterolaemia | 0.73 | 0.52–1.09 | 0.070 |
| LVEF | 0.97 | 0.93–1.01 | 0.183 |
| Family history of IHD | 0.73 | 0.45–1.20 | 0.213 |
| TIMI flow pre-PCI | 0.91 | 0.79–1.06 | 0.241 |
| Diabetes | 1.22 | 0.84–1.78 | 0.289 |
| Ostial lesions | 1.30 | 0.77–2.20 | 0.323 |
| Era of presentationb | 0.92 | 0.76–1.10 | 0.422 |
| Lesion at bifurcation | 0.83 | 0.54–1.29 | 0.424 |
| Calcified lesions | 1.14 | 0.73–1.80 | 0.564 |
| ACC/AHA lesion classification C | 1.06 | 0.77–1.45 | 0.728 |
| Hypertension | 0.96 | 0.71–1.29 | 0.781 |
| Current smoker | 0.96 | 0.68–1.36 | 0.830 |
| Factor | Hazard ratio | 95% CI | P-value |
|---|---|---|---|
| Cardiogenic shock pre-PCI | 4.13 | 2.75–6.18 | <0.001 |
| eGFR <60 | 3.02 | 2.16–4.22 | <0.001 |
| Increasing age | 1.03 | 1.01–1.07 | <0.001 |
| Female sex | 1.44 | 1.04–2.00 | 0.029 |
| Anterior MIa | 1.53 | 1.15–2.05 | 0.004 |
| Hypercholesterolaemia | 0.73 | 0.52–1.09 | 0.070 |
| LVEF | 0.97 | 0.93–1.01 | 0.183 |
| Family history of IHD | 0.73 | 0.45–1.20 | 0.213 |
| TIMI flow pre-PCI | 0.91 | 0.79–1.06 | 0.241 |
| Diabetes | 1.22 | 0.84–1.78 | 0.289 |
| Ostial lesions | 1.30 | 0.77–2.20 | 0.323 |
| Era of presentationb | 0.92 | 0.76–1.10 | 0.422 |
| Lesion at bifurcation | 0.83 | 0.54–1.29 | 0.424 |
| Calcified lesions | 1.14 | 0.73–1.80 | 0.564 |
| ACC/AHA lesion classification C | 1.06 | 0.77–1.45 | 0.728 |
| Hypertension | 0.96 | 0.71–1.29 | 0.781 |
| Current smoker | 0.96 | 0.68–1.36 | 0.830 |
eGFR, estimated glomerular filtration rate (ml/min/1.73 m2); PCI, Percutaneous coronary intervention; MI, Myocardial infarction; TIMI, Thrombolysis In Myocardial Infarction; LVEF, left ventricular ejection fraction; CI, confidence interval; IHD, ischaemic heart disease; ACC/AHA, American college of cardiology/American heart association.
Defined as left anterior descending or left main culprit artery causing ST-segment elevation myocardial infarction.
Year of presentation: HR: 0.98 (0.93–1.03), P = 0.326.
Multivariate analysis of late outcomes
| Factor | Hazard ratio | 95% CI | P-value |
|---|---|---|---|
| Cardiogenic shock pre-PCI | 4.13 | 2.75–6.18 | <0.001 |
| eGFR <60 | 3.02 | 2.16–4.22 | <0.001 |
| Increasing age | 1.03 | 1.01–1.07 | <0.001 |
| Female sex | 1.44 | 1.04–2.00 | 0.029 |
| Anterior MIa | 1.53 | 1.15–2.05 | 0.004 |
| Hypercholesterolaemia | 0.73 | 0.52–1.09 | 0.070 |
| LVEF | 0.97 | 0.93–1.01 | 0.183 |
| Family history of IHD | 0.73 | 0.45–1.20 | 0.213 |
| TIMI flow pre-PCI | 0.91 | 0.79–1.06 | 0.241 |
| Diabetes | 1.22 | 0.84–1.78 | 0.289 |
| Ostial lesions | 1.30 | 0.77–2.20 | 0.323 |
| Era of presentationb | 0.92 | 0.76–1.10 | 0.422 |
| Lesion at bifurcation | 0.83 | 0.54–1.29 | 0.424 |
| Calcified lesions | 1.14 | 0.73–1.80 | 0.564 |
| ACC/AHA lesion classification C | 1.06 | 0.77–1.45 | 0.728 |
| Hypertension | 0.96 | 0.71–1.29 | 0.781 |
| Current smoker | 0.96 | 0.68–1.36 | 0.830 |
| Factor | Hazard ratio | 95% CI | P-value |
|---|---|---|---|
| Cardiogenic shock pre-PCI | 4.13 | 2.75–6.18 | <0.001 |
| eGFR <60 | 3.02 | 2.16–4.22 | <0.001 |
| Increasing age | 1.03 | 1.01–1.07 | <0.001 |
| Female sex | 1.44 | 1.04–2.00 | 0.029 |
| Anterior MIa | 1.53 | 1.15–2.05 | 0.004 |
| Hypercholesterolaemia | 0.73 | 0.52–1.09 | 0.070 |
| LVEF | 0.97 | 0.93–1.01 | 0.183 |
| Family history of IHD | 0.73 | 0.45–1.20 | 0.213 |
| TIMI flow pre-PCI | 0.91 | 0.79–1.06 | 0.241 |
| Diabetes | 1.22 | 0.84–1.78 | 0.289 |
| Ostial lesions | 1.30 | 0.77–2.20 | 0.323 |
| Era of presentationb | 0.92 | 0.76–1.10 | 0.422 |
| Lesion at bifurcation | 0.83 | 0.54–1.29 | 0.424 |
| Calcified lesions | 1.14 | 0.73–1.80 | 0.564 |
| ACC/AHA lesion classification C | 1.06 | 0.77–1.45 | 0.728 |
| Hypertension | 0.96 | 0.71–1.29 | 0.781 |
| Current smoker | 0.96 | 0.68–1.36 | 0.830 |
eGFR, estimated glomerular filtration rate (ml/min/1.73 m2); PCI, Percutaneous coronary intervention; MI, Myocardial infarction; TIMI, Thrombolysis In Myocardial Infarction; LVEF, left ventricular ejection fraction; CI, confidence interval; IHD, ischaemic heart disease; ACC/AHA, American college of cardiology/American heart association.
Defined as left anterior descending or left main culprit artery causing ST-segment elevation myocardial infarction.
Year of presentation: HR: 0.98 (0.93–1.03), P = 0.326.
Propensity-matched mortality adjusted HR were: 0.9 (95% CI 0.4–2.0; P = 0.71) for timely pPCI compared with scheduled PCI; 0.5 (95% CI 0.2–0.9; P = 0.03) for timely pPCI compared with rescue PCI; 2.2 (95% CI 1.2–3.1; P = 0.01) for late pPCI compared with scheduled PCI; and 1.5 (95% CI 0.7–2.0; P = 0.40) for late pPCI compared with rescue PCI.
For the late composite outcome of all-cause mortality, re-infarction, TVR, and ST, the propensity score adjusted HRs were 1.1 (95% CI 0.7–2.0; P = 0.65) and 0.7 (95% CI 0.4–1.2; P = 0.194), when timely pPCI was compared with scheduled PCI and rescue PCI, respectively. The composite endpoint occurred more often in patients with late pPCI compared with those undergoing scheduled PCI (adjusted HR 2.1, 95% CI 1.2–3.0; P = 0.01). Rates of the composite endpoint were comparable in the late pPCI and rescue PCI groups (adjusted HR 1.2, 95% CI 0.6–1.6; P = 0.57). Propensity-matched Kaplan–Meier curves for survival and event-free survival are depicted in Figure 3. The hazard ratios for survival and event-free survival are seen in Figure 4. Event-free survival is defined as freedom from the late composite outcome of all-cause mortality, re-infarction, TVR, and ST.
Kaplan–Meier curves for survival and event-free survival among patients who underwent timely primary percutaneous coronary intervention, late primary percutaneous coronary intervention, or a pharmaco-invasive strategy. (A) Two way comparisons of survival between PCI strategies: scheduled, rescue, timely primary and late primary percutaneous intervention (B) Two way comparisons of event free survival between PCI strategies: scheduled, rescue, timely primary and late primary percutaneous intervention.
Hazard ratios for mortality and composite endpoints among patients who underwent timely primary percutaneous coronary intervention, late primary percutaneous coronary intervention, or a pharmaco-invasive strategy.
Discussion
Our study of unselected STEMI patients undergoing PCI during their initial hospitalization provides new information regarding the late outcomes of PI-PCI compared with timely and late pPCI. Despite the appropriately high rates of rescue PCI (33%), patients who received PI-PCI had better late survival compared with all patients who received pPCI. This was predominantly due to the much higher mortality (20%) when pPCI occurred when FMC to FDTs were ≥120 min (Structured Graphical Abstract). The rates of MI, stroke, intracranial haemorrhage, and major bleeding were similar with each reperfusion strategy.
The popularity of pPCI for the treatment of STEMI increased following the landmark meta-analysis conducted by Keeley et al.9 reporting its superiority to fibrinolytic therapy, albeit followed by highly variable rates of rescue PCI.9 The extent to which time delays in routine practice diminish the advantages of pPCI over fibrinolytic therapy has not been well characterized.3 Thus, a pivotal clinical consideration in emergency decision-making about patients with STEMI identified either by paramedics or at non-PCI centres, is whether ‘timely’ pPCI can realistically be performed. In circumstances where timely pPCI is not feasible, a pharmaco-invasive strategy is guideline-recommended for those without contraindications, largely due to its ability to be administered promptly (including in pre-hospital settings).1,3 This is reflected in the increased use of fibrinolytics for STEMI in Australia following the introduction of pre-hospital thrombolytic therapy.7
The Australian and American STEMI guidelines, while nuanced, recommend a FMC to FDT of <90 min for pPCI for those with symptom onset at FMC times of <120 min, whereas European guidelines reflecting systems-of-care that higher population densities can deliver, consider timely pPCI as FMC to FDT of <120 min.,1,3,4,30 We found that mortality rates were similar if FMC and FDT were <90 min or 90–119 min, but mortality was markedly higher when these times were ≥120 min. While our cohort study cannot address the relative merits of guideline FMC to FDT times for timely pPCI of <90 or <120 min, pPCI performed after 120 min had significantly worse outcomes than PI-PCI after propensity matching. Recent series from Alberta (Canada), the French registry on Acute ST-elevation and non-ST-elevation Myocardial Infarction (FAST-MI), Mexico, and the Norwegian Myocardial Infarction Registry (NORMI) have also demonstrated better outcomes among patients who underwent PI-PCI as opposed to late pPCI.14–16,31 FAST-MI defined timely pPCI as at <120 min from the first ECG, whereas in NORMI this was from the FMC.
Also, in our study PI-PCI did not have higher rates of stroke or bleeding (particularly intracranial haemorrhage), which has led to clinician concerns that have resulted in transfers for late pPCI.1 Our results are congruent with findings from the Korean and French MI registries, despite the low rates of radial vascular access (6.2%) during this earlier era at our institution, which has been an important determinant of reduced bleeding after rescue PCI.14,18,32,33 Our STEMI cohort was treated prior to the local introduction of half dose tenecteplase for patients aged ≥75 years, following results from the STREAM trial. This trial reported lower rates of intracranial haemorrhage when half dose tenecteplase was used in patients ≥75 years.12 Half dose tenecteplase is being compared with pPCI in those aged ≥60 years in the STREAM-2 trial.34
The timing of fibrinolytic therapy is critical amongst patients who receive PI-PCI, as older thrombi are more difficult to lyse.35 This is reflected in the decreased administration of fibrinolytics among patients who presented 6–12 h after symptom onset (40% as opposed to 48% among the remainder of the cohort). While more of these later presenting patients underwent late pPCI, as they were a small proportion of our study, they are unlikely to significantly impact our overall results.
As this study occurred over 10.5 years, several aspects of practice evolved. This included the adoption of a 24 h pPCI service, increasing rates of radial access, the introduction of second-generation DES, more potent P2Y12 inhibitors, and improved systems of care for streamlining transport for pPCI. Pre-hospital ECG identification of STEMI with transmission to on-call interventional cardiologists was introduced in May 2010 (Period 3 of this study), which reduced FMC to FDT for pPCI. As such, the increased survival reported during the later years of our study may be attributed to improvements in STEMI systems of care over the study decade.
Interestingly, PI-PCI was associated with significantly lower mortality compared with pPCI in all except Period 3. This is partly because patients previously transported to the closest hospital, often a non-PCI centre, were now identified by paramedics as having STEMI and transported directly to the PCI centre, improving FMC to FDT. Also, our lowest mortality was in those who had scheduled PCI, a finding similar to that reported from STREAM-1 among those undergoing scheduled angiography16
Limitations
This study has limitations inherent in all observational studies, specifically that the reported associations cannot be assumed to be causative. Also, the data are derived from our procedural database, and represent a STEMI cohort that survived and was stable enough to undergo PCI. Whilst this likely represents most of our STEMI population (reported as 84% in FAST-MI), outcomes cannot be generalized to STEMI patients who did not undergo PCI (for whom resources precluded routine follow-up). Despite enrolling consecutive patients, given the study design, a selection bias cannot be excluded. As these data were derived from a cardiology department procedural database, information about a small percentage (10–15%36) of patients receiving fibrinolytic therapy at referral hospitals who were not referred for angiography is not available. Some clinical data that was not available for all patients, included body mass index, ejection fraction, and Killip class. Although multi-vessel disease was identified in ∼50% of our STEMI population and is a known independent predictor of mortality in patients with STEMI undergoing PCI, rates of staged, non-culprit PCI were not routinely recorded.37,38 However, we have previously reported a residual Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery (SYNTAX) score analysis for the 2010–14 STEMI cohort, and only a minority of patients had complete revascularization.39–41 The lack of a Global Registry of Acute Coronary Events (GRACE-ACS) risk score is another limitation in our multivariable model.42 Moreover, adjustments for medications on discharge (eg. β-blockers, statins, and angiotensin-converting enzyme inhibitors) could not be performed due to both a proportion of deaths being in-hospital and unrecorded compliance rates, so a potential positive prognostic impact of these evidence-based therapies could not be determined. Whilst our results are congruent with other recent studies, they may not apply to other healthcare systems.14,15,43
Conclusion
Among unselected patients with STEMI who underwent PCI, those who received PI-PCI and those who received pPCI at <120 min from FMC to FDT times had similar late mortality rates. Markedly higher late mortality occurred amongst the one third of patients undergoing late pPCI. Rates of major bleeding and stroke were comparable among both groups. The efficacy and safety of PI-PCI, including with half dose fibrinolytic therapy, should be further evaluated in large clinical trials among suitable patients.
Acknowledgements
The authors would like to acknowledge the patients who agreed to participate in this study at a very difficult and daunting time.
Funding
This work was supported by the Australian Government (Research Training Program Scholarship) to J.J.
Declaration of Helsinki: The authors declare that the study complies with the Declaration of Helsinki, has been approved as a quality assurance project by the SWSLHD human research ethics committee (QA2008/034) and informed consent has been obtained from the subjects (or their legally authorized representative).
Data availability
The data sources for this manuscript cannot be shared publicly to protect the privacy of the individuals who were included in the study, for which the ethics committee approved data collection as a quality assurance project, and thus informed consent was not required. De-identified data can be shared upon reasonable request to the corresponding author. The authors do hereby declare that all illustrations and figures in the manuscript are entirely original and do not require reprint permission.
References
Author notes
Presented in part at the American College of Cardiology Scientific Sessions, Washington, DC, United States, April 2022.
Conflict of interest: J.K.F., as national co-ordinator for STREAM-2 has an unrestricted research grant in aid from Boehringer Ingelheim; he was also a co-author of the 2016 Australian ACS guidelines.
