Family Practice, 2015, Vol. 32, No. 2, 192–197 doi:10.1093/fampra/cmv006 Advance Access publication 24 February 2015
Long-term survival benefits of thrombolysis: the Royal College of General Practitioners’ myocardial infarction study Kenneth M Gilmoura, Lisa Iversenb,* and Philip C Hannafordb School of Medicine & Dentistry and bAcademic Primary Care, Division of Applied Health Sciences, University of Aberdeen, Aberdeen, UK. a
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
*Correspondence to Lisa Iversen, Academic Primary Care, Division of Applied Health Sciences, University of Aberdeen, Polwarth Building, Foresterhill, Aberdeen AB25 2ZD, UK; E-mail: [email protected]
Abstract Objective. To investigate whether there is a long-term survival benefit from receipt of thrombolysis in routine care particularly pre-hospital thrombolysis, using 20 year mortality data from the RCGP myocardial infarction (MI) cohort study. Methods. During 1991–92 the RCGP MI study assessed GP delivery of thrombolysis. Participants who received pre-hospital thrombolysis (n = 290), thrombolysis in hospital (n = 781) or no thrombolysis (n = 2021) were followed and mortality data collected to June 2012. The relationship between thrombolysis and survival time was analysed using Cox regression at 28 days, 1, 5, 10, 15 years post-AMI, and at end of follow-up (~20 years post-AMI). Results. Compared to those who did not receive it, participants who received thrombolysis had a significant survival benefit at 28 days [adjusted hazard ratio (HR) 0.72, 95% confidence interval (CI): 0.58–0.90]; 1 year (adjusted HR 0.69, 95% CI: 0.57–0.83); 5 years (adjusted HR 0.76, 95% CI: 0.66–0.86); 10 years (adjusted HR 0.85, 95% CI: 0.77–0.95) and 15 years (adjusted HR 0.88, 95% CI: 0.80–0.96) post-AMI until end of follow-up (adjusted HR 0.92, 95% CI: 0.84–1.00). Pre versus in-hospital thrombolysis did not appear beneficial, although there was evidence among the prehospital group that short symptom onset-to-needle times conferred greater benefit. Conclusions. We found substantial long-term survival benefits associated with thrombolysis when used in routine care. Although primary percutaneous coronary intervention (pPCI) is now the choice treatment, thrombolysis remains an important option when pPCI cannot be delivered within 120 minutes of diagnosis. Key words: Follow-up studies, general practice, myocardial infarction, pre-hospital care, thrombolysis.
Introduction Acute myocardial infarction (AMI) remains a common medical emergency in the UK, requiring prompt intervention to minimize death or morbidity. In many parts of the country first line therapy for ST-elevation AMI has changed in recent years, from thrombolysis to primary percutaneous coronary intervention (pPCI) which has been shown to be a more effective treatment (1). Immediate thrombolysis, however, remains an important therapeutic option when pPCI cannot be provided within 120 minutes of ECG diagnosis (2).
© The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected].
Thrombolysis is a safe and effective treatment for AMI (3), which is most effective when administered as soon as possible after the onset of symptoms (4). Pre-hospital thrombolysis facilitates rapid delivery (5). A meta-analysis of six randomized control trials comparing pre-hospital with in-hospital thrombolysis found a mortality benefit [odds ratio (OR) 0.83, 95% confidence interval (CI) 0.70–0.98; P = 0.03] from pre-hospital administration, with an average time saving of ~60 minutes (P = 0.007) (6). In a 10 year follow-up of the Grampian Region Early Anistreplase Trial (GREAT), pre-hospital thrombolysis was found to maintain its survival benefit (7).
192
Thrombolysis: long-term survival benefits
Methods RCGP MI study The RCGP MI study has been described elsewhere (9). In brief, 344 GPs from 253 practices throughout the UK agreed to be in a ‘user’ group of GPs willing to administer the thrombolytic anistreplase as part of their care of patients with suspected AMI. Another 776 doctors from 552 practices agreed to be in a comparison group of GPs not willing to administer thrombolysis but willing to provide information about their cases of AMI for the study. In total 3596 patients with chest pain were recruited, of whom 3383 patients were managed as an AMI (diagnosed clinically usually based on history and when available electrocardiograph). For these patients, the GP recorded details of: the patient’s clinical condition; the pre-hospital management of the patient, including any contraindications to thrombolysis, whether thrombolysis was administered, and if so when; details of when an ambulance arrived to transport the patient to hospital; other treatments provided. Other information collected for all patients included: demographic details; smoking status; history of previous AMI. Each patient had a unique study number, the key to which was held by the GP. This number was used when communicating between the GP and study centre, removing the need for identifying information and thereby ensuring patient confidentiality. Recruited patients were followed up via the GP for 12 months. Most patients were ‘flagged’ at the NHS central registries upon recruitment so that the study can be informed of the date and cause of subsequent deaths occurring in the cohort. The flagging procedure involved the recruiting GPs providing the central registries with identifying information for each patient recruited together with their unique study number. Death information supplied by the central registries to the study only includes the patient’s study number so that patient confidentiality is maintained. From the 3603 myocardial infarctions on the study database at end of follow-up, useful information was available for 3092 unique patients (Fig. 1). A total of 511 participants were excluded because of: inaccurate information or flagging not possible (n = 294); participant was from Northern Ireland or armed forces (n = 3); missing or inconsistent date of death/birth/AMI (n = 31); not managed as
an AMI (n = 119); not known whether or not participant received thrombolysis (n = 64). Follow up was to 30 June 2012.
Statistical analysis Data were analysed using IBM SPSS Statistics 20. The relationship between the comparison groups (received thrombolysis regardless of location versus thrombolysis not received; received pre-hospital thrombolysis versus in-hospital thrombolysis) and probability of survival to end of follow-up was examined by generating Kaplan– Meier survival curves and applying the log rank test. Kaplan–Meier survival curves were also plotted for the relationship between each of the potential confounding factors and survival time, with visual inspection for deviation from the proportional hazards assumption (10). An unadjusted Cox regression model was fitted from time of AMI to date of death or end of follow-up. The relationship between thrombolysis group and survival time after adjustment for potential confounding was then examined. Sex (men, women), age at AMI (≤49, 50–69, ≥70 years), smoking status (non-smoker, smoker, missing), occupation (non-manual, manual, missing) and history of previous AMI (no, yes, missing) were all entered into the adjusted model. Separate analyses assessed survival to 28 days, 1, 5, 10 and 15 years post-AMI, and end of follow-up (approximately 20 years post-AMI). We repeated the Cox regression models excluding all patients who had a recorded contraindication to thrombolysis (any of: a known bleeding diathesis; active gastro-intestinal bleeding or other internal bleeding in the previous 6 months; prior cerebrovascular accident; neurological procedure in previous 2 months; known intracranial neoplasm or aneurysm; surgery or major trauma in previous 10 days; recent traumatic cardio-pulmonary resuscitation (but not defibrillation); severe uncontrolled hypertension; heavy vaginal bleeding or currently pregnant; received streptokinase or anistreplase within previous 12 months; or chest pain of duration 12 hours). In a separate analysis of participants who received pre-hospital thrombolysis, we investigated whether symptom onset to needle time was associated with mortality risk.
Results By the end of follow-up there were 2320 deaths reported (75%). About 67.4% of deaths were attributed to circulatory diseases, 13.4% cancer and 19.2% other causes. The median duration of follow-up among the thrombolysed group was 11.4 years, (interquartile range, IQR = 3.9–20.0). The median duration of followup among the group that was not thrombolysed was 8.3 years (IQR = 1.4–19.6). Participants who received thrombolysis tended to be men, were aged 50–69 years, were employed in non-manual occupations, had no previous AMI and did not have a contraindication to thrombolysis (Supplementary Table 1). Table 1 shows the association between receipt of thrombolysis, patient characteristics and all-cause mortality at 28 days post-AMI. There was a statistically significant survival benefit from receipt of thrombolysis [adjusted hazard ratio (HR) 0.72, 95% CI 0.58–0.90] after adjustment for sex, age, smoking status, occupation and history of previous MI. Mortality was higher among older patients and those who had a manual occupation. A statistically significant survival benefit from receipt of thrombolysis was also seen at 1, 5, 10 and 15 years post-AMI (Table 2). The risk estimate for end of follow-up (~20 years) was of borderline statistical significance (HR 0.92; 95% CI 0.84–1.00). Figure 2 shows the Kaplan–Meir curves for 1 year, 10 years and end of follow-up.
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Comparatively little is known about whether the benefits demonstrated under clinical trial conditions (where more restrictive inclusion criteria usually apply) are replicated when thrombolysis is delivered as part of routine care. One large observational study in Sweden in 2006 found that pre-hospital thrombolysis was associated with a 29% survival benefit during the first year after use (8). No observational study of pre-hospital thrombolysis has reported results from longer term follow-up. The Royal College of General Practitioners’ (RCGP) MI study was a prospective observational post-marketing surveillance study conducted between March 1991 and September 1992 to assess the safety and practicality of GP administration of thrombolysis to patients thought to be having an AMI in the community (9). The study found that GPs could administer thrombolysis safely and appropriately. Recruited patients were “flagged” at the NHS Central Registries so that the study could be informed of all deaths subsequently occurring among the cohort. Using mortality data accumulated over the last 20 years, we evaluated whether there is a long-term survival benefit to participants who received thrombolysis compared with those who did not receive thrombolysis, and whether there was an additional benefit from pre-hospital compared with hospital administered thrombolysis.
193
Family Practice, 2015, Vol. 32, No. 2
194
The most common contraindication to thrombolysis was chest pain of 12 hours duration (399/775, 51.5%) followed by previous cerebrovascular accident (234/775, 30.2%: data not shown). When we repeated the analysis excluding 775 participants with a recorded contraindication to thrombolysis, the pattern of adjusted HRs remained similar, with the statistically significant survival benefit persisting beyond 5 years post-AMI (Supplementary Table 2). At 28 days and 1 year post-AMI there was no statistically significant difference in survival between men and women (Supplementary Table 3). However, by 5 years post-AMI women had a statistically significant lower mortality than men; an effect which persisted to end of follow-up. The same pattern was found when the patients with any contraindication were removed, with the significant survival benefit for women emerging by 10 years post-AMI (data not shown). At 28 days and 1 year post-AMI there was no statistically significant difference in survival between participants with and without a history of previous MI. By 5 years, however, patients with a history of previous AMI had a statistically significant worse survival than those without such a history; an effect which persisted to end of follow-up (Supplementary Table 4). The effect was also found when the participants with contraindications were removed from the analysis (data not shown). In our study, pre-hospital receipt of thrombolysis was not associated with a statistically significant survival benefit, compared with receipt in hospital (Table 3). We also examined the relationship between time from symptom onset to thrombolysis and all-cause mortality in individuals given thrombolysis before hospitalization.
The amount of data available for analysis at each time period was small, so each risk estimate had a wide confidence interval. The results, however, suggested that people who received thrombolysis within an hour of symptom onset did better than those receiving it later (Supplementary Table 5).
Discussion Summary In this large observational study, patients who received thrombolysis had a significantly reduced risk of dying compared with those who did not receive this treatment; an effect that persisted for many years after the event. We were unable to demonstrate a significantly different mortality risk among those receiving thrombolysis before or after hospitalization. Among those receiving thrombolysis before hospitalization, there was a suggestion that receipt of thrombolysis within an hour of symptom onset was associated with greater benefit than receipt after a longer interval. Women appeared to have better long term survival than males, and those with a history of previous AMI worse long-term survival.
Strengths and limitations Strength of this article is its long duration of follow-up. We have been unable to identify any previous research, from either a clinical trial or observational study, which has specifically investigated very long-term mortality outcome among people receiving thrombolysis. Our findings suggest that thrombolysis given in routine clinical settings is associated with important long-term mortality benefits for
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 1. Flowchart showing exclusions from analysis
Thrombolysis: long-term survival benefits
195
Table 1. Association between receipt of thrombolysis, other patient characteristics and all-cause mortality at 28 days post-AMI among 3092 participants in the RCGP MI study Characteristic
N, alive:dead
Unadjusted HR (95% CI)
Adjusteda HR (95% CI)
2021 (65.4) 1071 (34.6)
1721:300 966:105
1.00 0.64 (0.51–0.80)
1.00 0.72 (0.58–0.90)
2036 (65.8) 1056 (34.2)
1791:245 896:160
1.00 1.27 (1.04–1.55)
1.00 1.00 (0.81–1.23)
249 (8.1) 1528 (49.4) 1315 (42.5)
242:7 1412:116 1033:282
1.00 2.76 (1.29–5.93) 8.37 (3.95–17.71)
1.00 2.73 (1.27–5.87) 8.08 (3.78–17.28)
1962 (63.5) 825 (26.7) 305 (9.9)
1694:268 740:85 253:52
1.00 0.75 (0.59–0.95) 1.28 (0.95–1.72)
1.00 1.11 (0.86–1.43) 1.19 (0.88–1.61)
1136 (36.7) 1085 (35.1) 871 (28.2)
1018:118 952:133 717:154
1.00 1.20 (0.94–1.54) 1.75 (1.37–2.22)
1.00 1.31 (1.01–1.68) 1.40 (1.10–1.79)
2290 (74.1) 765 (24.7) 37 (1.2)
1999:291 656:109 32:5
1.00 1.14 (0.92–1.42) 1.05 (0.43–2.53)
1.00 0.99 (0.79–1.23) 0.90 (0.37–2.18)
Adjusted for covariates: sex, age, smoking status, occupation, and previous AMI.
a
Table 2. Association between receipt of thrombolysis and all-cause mortality at different follow-up time points post-AMI Duration of follow-up post-MI
1 year 5 years 10 years 15 years End of follow-up
Thrombolysis not received, n
Thrombolysis received, n
Alive:dead
Adjusted HRa (95% CI)
1568:453 1218:803 911:1110 663:1358 481:1540
1.00b 1.00 1.00 1.00 1.00
Alive:dead 917:154 760:311 571:500 425:646 291:780
Adjusted HRa (95% CI) 0.69 (0.57–0.83) 0.76 (0.66–0.86) 0.85 (0.77–0.95) 0.88 (0.80–0.96) 0.92 (0.84–1.00)
Adjusted for thrombolysis status, sex, age, smoking status, occupation, and previous AMI. 1.00 represents the reference group.
a
b
patients with AMI. The observational nature of the study, however, means that survival might have been influenced by the greater use of beneficial cardiac interventions and secondary prevention treatments among patients receiving thrombolysis. This said, our findings are compatible with evidence from more selective clinical trial settings, and provide a powerful reminder that thrombolysis remains an important treatment option for patients with ST-elevation AMI when pPCI is not available. The flagging of study participants for death notification meant that we were able to follow-up almost all of the cohort efficiently for prolonged periods, with minimal loss to follow-up. The doctors certifying the cause of death, especially for events occurring many years after the recruitment AMI, were highly unlikely to be aware of treatments given at recruitment, so recall bias is unlikely to account for our findings. One limitation of the study was the small number of individuals who received thrombolysis before hospitalization, reflecting the cautious use of thrombolysis by GPs at the time of the study. The small number will have reduced the power of the study when comparing the effects of pre- and post-hospitalization administration of thrombolysis. Furthermore, the study was unable to collect
accurate symptom onset to thrombolysis times for people receiving thrombolysis in hospital, so we were only able to examine this in those receiving thrombolysis in the community, with few data in most strata. The majority of participants with a recorded contraindication did not receive thrombolysis. Our examination of the data excluding patients who had contraindications showed similar survival benefits associated with thrombolysis. The loss of statistical significance by 10 years post-AMI probably reflects loss of statistical power resulting from the exclusion of 25% of individuals from the cohort in these analyses.
Comparison with existing literature As an observational study, confounding and bias must always be considered when interpreting results found. This said, the long-term survival benefit found for thrombolysis is consistent with short-term results from clinical trials of thrombolysis (11,12). A 20-year follow-up of participants in a trial of reperfusion (which included both intra-coronary thrombolysis and coronary angioplasty) also found statistically significant mortality benefit from active therapy, which lasted for up to 20 years (13).
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Thrombolysis No Yes Sex Men Women Age at AMI ≤49 50–69 ≥70 Smoking status Non-smoker Smoker Missing Occupation Non-manual Manual Missing Previous AMI No Yes Missing
n (%)
Family Practice, 2015, Vol. 32, No. 2
196 The GREAT trial contained a smaller number of patients who received pre-hospital thrombolysis (163) than in our study (290), yet was able to find a 52% relative reduction (95% CI 14–89%;
Clinical implications
Figure 2. Survival at different time points during follow-up of participants who received and did not receive thrombolysis in the RCGP MI study
When the RCGP MI study started in 1991 there was clear evidence that thrombolysis saves lives after AMI, and that the benefits could be maximized by minimizing the ‘call to needle’ time. This lead to various initiatives in the UK to optimize delivery times, for example through the provision of thrombolysis in the community by GPs or ambulance paramedic crews, or in accident and emergency departments of hospitals by specialist nurses. Our study provided important evidence that GPs could administer thrombolysis safely and appropriately (9), although there were a number of practical issues which were likely to limit its use in this way. In subsequent years, evidence has emerged that pPCI (coronary angioplasty and stenting) confers greater benefits to patients with ST-elevated AMI than thrombolysis. In 2003, a meta-analysis showed that pPCI results in fewer shortand longer-term deaths, non-fatal re-infarctions and strokes than thrombolysis (1). The National Health Service has responded to this evidence by establishing networks of Heart Attack Centres where pPCI can be provided promptly, so that in 2012/13 virtually all (95%) of the 22 016 ST-elevation AMI patients receiving reperfusion treatments in England, Wales and Northern Ireland received pPCI, and only 5% thrombolysis (17). Many ambulance services have now shifted their focus from the provision of pre-hospital thrombolysis
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
P = 0.007) in mortality in the pre-hospital thrombolysis group at 1 year follow-up (5). However, the GREAT trial only included patients from GP practices in a rural setting at least 16 miles from the local hospital, resulting in a substantial time saved in delivering pre-hospital thrombolysis, which may account for the differences in study results. GP practices in the RCGP study were not exclusively rural so less time was saved by delivering pre-hospital thrombolysis. Thus, the median time delay in GREAT between symptom onset and injection of thrombolysis (or placebo) was 105 minutes (IQR not given), compared with 120 minutes (IQR = 60–215) in the RCGP MI study. In our study, GPs who were willing to use thrombolysis appeared to be cautious in its use (9), partly because community-based thrombolysis was highly controversial at the time of the study. Confounding by indication could have occurred if the GPs tended to use thrombolysis in the community in patients with more serious events, or with risk factors predictive of worse outcome. Although we were able to adjust for gender, age, occupation and previous AMI, there may have been other unmeasured factors that reduced our ability to find a benefit from pre-hospital thrombolysis. Although based on limited data, the symptom onset to thrombolysis analysis (Supplementary Table 5) provides some evidence to support the delivery of thrombolysis as quickly as possible. Furthermore, an observational study in Sweden in 2006 which had accurate and complete information about time from symptom onset to thrombolysis for 1690 patients given treatment pre-hospital and 3685 in-hospital, found a 29% one year mortality benefit in the pre-hospital group (8). Better long-term survival among women post-AMI has been found previously in the Framingham study, which reported an adjusted HR among women compared with men of 0.49 (95% CI 0.31–0.77) at 32-year follow-up (14). Other research has also found poorer long-term survival among those with a history of previous AMI (15). It was noteworthy that the gender and history of previous AMI differences appeared several years after the recruitment AMI, suggesting influences unrelated to initial therapies or interventions. Table 1 highlights that, of the variables measured, age has the greatest influence on prognosis post-AMI. This finding is supported by other studies (15,16).
Thrombolysis: long-term survival benefits
197
Table 3. Association between receipt of thrombolysis before (n = 290) or after hospitalization (n = 781) and all-cause mortality Duration of follow-up post-MI
28 days 1 year 5 years 10 years 15 years End of follow-up
Pre-hospital thrombolysis, N
In-hospital thrombolysis, N
Alive:dead
Alive:dead
255:35 239:51 192:98 136:154 108:182 76:214
Adjusted HR (95% CI) 1.00 1.00 1.00 1.00 1.00 1.00
711:70 678:103 568:213 435:346 317:464 215:566
Adjusted HRa (95% CI) 0.75 (0.50–1.14) 0.76 (0.54–1.08) 0.86 (0.67–1.10 0.87 (0.71–1.05) 0.96 (0.81–1.15) 1.00 (0.85–1.17)
Adjusted for thrombolysis status, sex, age, smoking status, occupation, and previous AMI.
a
Supplementary material Supplementary material is available at Family Practice online.
Declaration Funding: The original RCGP MI study was funded by SmithKline Beecham through an unconditional research grant. The company have not had access to the data or been involved in its analysis or interpretation. The University of Aberdeen has provided support for maintenance of the database and death notifications. Ethics approval: The study received ethical approval from the Independent Clinical Research Ethics Committee of the RCGP, and local ethics committees of participating doctors were informed of the study. Conflict of interest: none.
Acknowledgements We thank the GPs who recruited the patients for us, and Dr Clifford Kay who was Principal Investigator for the study at its inception.
References 1. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003; 361: 13–20. 2. Scottish Intercollegiate Guidelines Network (SIGN). Acute Coronary Syndromes. Edinburgh, UK: SIGN, 2013.
3. Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 343: 311–22. 4. Boersma E, Maas AC, Deckers JW et al. Early thrombolytic treatment in acute myocardial infarction: reappraisal of the golden hour. Lancet 1996; 348: 771–5. 5. Rawles JM. Feasibility, safety, and efficacy of domiciliary thrombolysis by general practitioners: Grampian region early anistreplase trial. GREAT Group. BMJ 1992; 305: 548–53. 6. Morrison LJ, Verbeek PR, McDonald AC et al. Mortality and prehospital thrombolysis for acute myocardial infarction: A meta-analysis. JAMA 2000; 283: 2686–92. 7. Rawles J; GREAT. GREAT: 10 year survival of patients with suspected acute myocardial infarction in a randomised comparison of prehospital and hospital thrombolysis. Heart 2003; 89: 563–4. 8. Björklund E, Stenestrand U, Lindbäck J et al. Pre-hospital thrombolysis delivered by paramedics is associated with reduced time delay and mortality in ambulance-transported real-life patients with ST-elevation myocardial infarction. Eur Heart J 2006; 27: 1146–52. 9. Hannaford P, Vincent R, Ferry S et al. Assessment of the practicality and safety of thrombolysis with anistreplase given by general practitioners. Br J Gen Pract 1995; 45: 175–9. 1 0. Katz MH. Multivariable Analysis: A Practical Guide for Clinicians. Cambridge, UK: Cambridge University Press, 2002, pp. 141–58. 11. Franzosi MG, Santoro E, De Vita C et al. Ten-year follow-up of the first megatrial testing thrombolytic therapy in patients with acute myocardial infarction: results of the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto-1 Study. Circulation 1998; 98: 2659–65. 12. Second International Study of Infarct Survival Collaborative Group (ISIS2). Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17 187 cases of suspected acute myocardial infarction: ISIS 2. Lancet 1988; 332: 349–60. 13. van Domburg RT, Sonnenschein K, Nieuwlaat R et al. Sustained benefit 20 years after reperfusion therapy in acute myocardial infarction. J Am Coll Cardiol 2005; 46: 15–20. 14. Wong ND, Cupples A, Ostfeld AM et al. Risk factors for long-term coronory prognosis after initial myocardial infarction: the Framingham Study. Am J Epidemiol 1989; 130: 469–80. 15. Lee KL, Woodlief LH, Topol EJ et al. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction: results from an International Trial of 41 021 patients. Circulation 1995; 91: 1659–68. 16. White HD, Van de Werf FJ. Thrombolysis for acute myocardial infarction. Circulation 1998; 97: 1632–46. 17. Myocardial Ischaemia National Audit Project. How the NHS cares for patients with heart attack. Annual report April 2012–March 2013. https:// www.ucl.ac.uk/nicor/audits/minap/documents/annual_reports/minappublic-report-2014 (accessed on 31 December 2014).
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
to ensuring the rapid admission of patients to a Heart Attack Centre, often bypassing smaller hospitals or accident and emergency departments. Although this focus is appropriate for most patients with ST-elevation AMI, it is important to remember that immediate thrombolysis remains an important treatment for patients living in areas where pPCI cannot be delivered within 120 minutes of initial diagnosis (2). Our results contribute to the evidence base showing that thrombolysis is an effective treatment associated with substantial long term benefits. Although we were unable to corroborate clinical trial data showing important benefits from the pre-hospital administration of thrombolysis, our symptom-to needle time data highlight the importance of providing thrombolysis as quickly as possible, when it is used.
Long-term survival benefits of thrombolysis: the Royal College of General Practitioners’ myocardial infarction study Kenneth M Gilmoura, Lisa Iversenb,* and Philip C Hannafordb School of Medicine & Dentistry and bAcademic Primary Care, Division of Applied Health Sciences, University of Aberdeen, Aberdeen, UK. a
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
*Correspondence to Lisa Iversen, Academic Primary Care, Division of Applied Health Sciences, University of Aberdeen, Polwarth Building, Foresterhill, Aberdeen AB25 2ZD, UK; E-mail: [email protected]
Abstract Objective. To investigate whether there is a long-term survival benefit from receipt of thrombolysis in routine care particularly pre-hospital thrombolysis, using 20 year mortality data from the RCGP myocardial infarction (MI) cohort study. Methods. During 1991–92 the RCGP MI study assessed GP delivery of thrombolysis. Participants who received pre-hospital thrombolysis (n = 290), thrombolysis in hospital (n = 781) or no thrombolysis (n = 2021) were followed and mortality data collected to June 2012. The relationship between thrombolysis and survival time was analysed using Cox regression at 28 days, 1, 5, 10, 15 years post-AMI, and at end of follow-up (~20 years post-AMI). Results. Compared to those who did not receive it, participants who received thrombolysis had a significant survival benefit at 28 days [adjusted hazard ratio (HR) 0.72, 95% confidence interval (CI): 0.58–0.90]; 1 year (adjusted HR 0.69, 95% CI: 0.57–0.83); 5 years (adjusted HR 0.76, 95% CI: 0.66–0.86); 10 years (adjusted HR 0.85, 95% CI: 0.77–0.95) and 15 years (adjusted HR 0.88, 95% CI: 0.80–0.96) post-AMI until end of follow-up (adjusted HR 0.92, 95% CI: 0.84–1.00). Pre versus in-hospital thrombolysis did not appear beneficial, although there was evidence among the prehospital group that short symptom onset-to-needle times conferred greater benefit. Conclusions. We found substantial long-term survival benefits associated with thrombolysis when used in routine care. Although primary percutaneous coronary intervention (pPCI) is now the choice treatment, thrombolysis remains an important option when pPCI cannot be delivered within 120 minutes of diagnosis. Key words: Follow-up studies, general practice, myocardial infarction, pre-hospital care, thrombolysis.
Introduction Acute myocardial infarction (AMI) remains a common medical emergency in the UK, requiring prompt intervention to minimize death or morbidity. In many parts of the country first line therapy for ST-elevation AMI has changed in recent years, from thrombolysis to primary percutaneous coronary intervention (pPCI) which has been shown to be a more effective treatment (1). Immediate thrombolysis, however, remains an important therapeutic option when pPCI cannot be provided within 120 minutes of ECG diagnosis (2).
© The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected].
Thrombolysis is a safe and effective treatment for AMI (3), which is most effective when administered as soon as possible after the onset of symptoms (4). Pre-hospital thrombolysis facilitates rapid delivery (5). A meta-analysis of six randomized control trials comparing pre-hospital with in-hospital thrombolysis found a mortality benefit [odds ratio (OR) 0.83, 95% confidence interval (CI) 0.70–0.98; P = 0.03] from pre-hospital administration, with an average time saving of ~60 minutes (P = 0.007) (6). In a 10 year follow-up of the Grampian Region Early Anistreplase Trial (GREAT), pre-hospital thrombolysis was found to maintain its survival benefit (7).
192
Thrombolysis: long-term survival benefits
Methods RCGP MI study The RCGP MI study has been described elsewhere (9). In brief, 344 GPs from 253 practices throughout the UK agreed to be in a ‘user’ group of GPs willing to administer the thrombolytic anistreplase as part of their care of patients with suspected AMI. Another 776 doctors from 552 practices agreed to be in a comparison group of GPs not willing to administer thrombolysis but willing to provide information about their cases of AMI for the study. In total 3596 patients with chest pain were recruited, of whom 3383 patients were managed as an AMI (diagnosed clinically usually based on history and when available electrocardiograph). For these patients, the GP recorded details of: the patient’s clinical condition; the pre-hospital management of the patient, including any contraindications to thrombolysis, whether thrombolysis was administered, and if so when; details of when an ambulance arrived to transport the patient to hospital; other treatments provided. Other information collected for all patients included: demographic details; smoking status; history of previous AMI. Each patient had a unique study number, the key to which was held by the GP. This number was used when communicating between the GP and study centre, removing the need for identifying information and thereby ensuring patient confidentiality. Recruited patients were followed up via the GP for 12 months. Most patients were ‘flagged’ at the NHS central registries upon recruitment so that the study can be informed of the date and cause of subsequent deaths occurring in the cohort. The flagging procedure involved the recruiting GPs providing the central registries with identifying information for each patient recruited together with their unique study number. Death information supplied by the central registries to the study only includes the patient’s study number so that patient confidentiality is maintained. From the 3603 myocardial infarctions on the study database at end of follow-up, useful information was available for 3092 unique patients (Fig. 1). A total of 511 participants were excluded because of: inaccurate information or flagging not possible (n = 294); participant was from Northern Ireland or armed forces (n = 3); missing or inconsistent date of death/birth/AMI (n = 31); not managed as
an AMI (n = 119); not known whether or not participant received thrombolysis (n = 64). Follow up was to 30 June 2012.
Statistical analysis Data were analysed using IBM SPSS Statistics 20. The relationship between the comparison groups (received thrombolysis regardless of location versus thrombolysis not received; received pre-hospital thrombolysis versus in-hospital thrombolysis) and probability of survival to end of follow-up was examined by generating Kaplan– Meier survival curves and applying the log rank test. Kaplan–Meier survival curves were also plotted for the relationship between each of the potential confounding factors and survival time, with visual inspection for deviation from the proportional hazards assumption (10). An unadjusted Cox regression model was fitted from time of AMI to date of death or end of follow-up. The relationship between thrombolysis group and survival time after adjustment for potential confounding was then examined. Sex (men, women), age at AMI (≤49, 50–69, ≥70 years), smoking status (non-smoker, smoker, missing), occupation (non-manual, manual, missing) and history of previous AMI (no, yes, missing) were all entered into the adjusted model. Separate analyses assessed survival to 28 days, 1, 5, 10 and 15 years post-AMI, and end of follow-up (approximately 20 years post-AMI). We repeated the Cox regression models excluding all patients who had a recorded contraindication to thrombolysis (any of: a known bleeding diathesis; active gastro-intestinal bleeding or other internal bleeding in the previous 6 months; prior cerebrovascular accident; neurological procedure in previous 2 months; known intracranial neoplasm or aneurysm; surgery or major trauma in previous 10 days; recent traumatic cardio-pulmonary resuscitation (but not defibrillation); severe uncontrolled hypertension; heavy vaginal bleeding or currently pregnant; received streptokinase or anistreplase within previous 12 months; or chest pain of duration 12 hours). In a separate analysis of participants who received pre-hospital thrombolysis, we investigated whether symptom onset to needle time was associated with mortality risk.
Results By the end of follow-up there were 2320 deaths reported (75%). About 67.4% of deaths were attributed to circulatory diseases, 13.4% cancer and 19.2% other causes. The median duration of follow-up among the thrombolysed group was 11.4 years, (interquartile range, IQR = 3.9–20.0). The median duration of followup among the group that was not thrombolysed was 8.3 years (IQR = 1.4–19.6). Participants who received thrombolysis tended to be men, were aged 50–69 years, were employed in non-manual occupations, had no previous AMI and did not have a contraindication to thrombolysis (Supplementary Table 1). Table 1 shows the association between receipt of thrombolysis, patient characteristics and all-cause mortality at 28 days post-AMI. There was a statistically significant survival benefit from receipt of thrombolysis [adjusted hazard ratio (HR) 0.72, 95% CI 0.58–0.90] after adjustment for sex, age, smoking status, occupation and history of previous MI. Mortality was higher among older patients and those who had a manual occupation. A statistically significant survival benefit from receipt of thrombolysis was also seen at 1, 5, 10 and 15 years post-AMI (Table 2). The risk estimate for end of follow-up (~20 years) was of borderline statistical significance (HR 0.92; 95% CI 0.84–1.00). Figure 2 shows the Kaplan–Meir curves for 1 year, 10 years and end of follow-up.
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Comparatively little is known about whether the benefits demonstrated under clinical trial conditions (where more restrictive inclusion criteria usually apply) are replicated when thrombolysis is delivered as part of routine care. One large observational study in Sweden in 2006 found that pre-hospital thrombolysis was associated with a 29% survival benefit during the first year after use (8). No observational study of pre-hospital thrombolysis has reported results from longer term follow-up. The Royal College of General Practitioners’ (RCGP) MI study was a prospective observational post-marketing surveillance study conducted between March 1991 and September 1992 to assess the safety and practicality of GP administration of thrombolysis to patients thought to be having an AMI in the community (9). The study found that GPs could administer thrombolysis safely and appropriately. Recruited patients were “flagged” at the NHS Central Registries so that the study could be informed of all deaths subsequently occurring among the cohort. Using mortality data accumulated over the last 20 years, we evaluated whether there is a long-term survival benefit to participants who received thrombolysis compared with those who did not receive thrombolysis, and whether there was an additional benefit from pre-hospital compared with hospital administered thrombolysis.
193
Family Practice, 2015, Vol. 32, No. 2
194
The most common contraindication to thrombolysis was chest pain of 12 hours duration (399/775, 51.5%) followed by previous cerebrovascular accident (234/775, 30.2%: data not shown). When we repeated the analysis excluding 775 participants with a recorded contraindication to thrombolysis, the pattern of adjusted HRs remained similar, with the statistically significant survival benefit persisting beyond 5 years post-AMI (Supplementary Table 2). At 28 days and 1 year post-AMI there was no statistically significant difference in survival between men and women (Supplementary Table 3). However, by 5 years post-AMI women had a statistically significant lower mortality than men; an effect which persisted to end of follow-up. The same pattern was found when the patients with any contraindication were removed, with the significant survival benefit for women emerging by 10 years post-AMI (data not shown). At 28 days and 1 year post-AMI there was no statistically significant difference in survival between participants with and without a history of previous MI. By 5 years, however, patients with a history of previous AMI had a statistically significant worse survival than those without such a history; an effect which persisted to end of follow-up (Supplementary Table 4). The effect was also found when the participants with contraindications were removed from the analysis (data not shown). In our study, pre-hospital receipt of thrombolysis was not associated with a statistically significant survival benefit, compared with receipt in hospital (Table 3). We also examined the relationship between time from symptom onset to thrombolysis and all-cause mortality in individuals given thrombolysis before hospitalization.
The amount of data available for analysis at each time period was small, so each risk estimate had a wide confidence interval. The results, however, suggested that people who received thrombolysis within an hour of symptom onset did better than those receiving it later (Supplementary Table 5).
Discussion Summary In this large observational study, patients who received thrombolysis had a significantly reduced risk of dying compared with those who did not receive this treatment; an effect that persisted for many years after the event. We were unable to demonstrate a significantly different mortality risk among those receiving thrombolysis before or after hospitalization. Among those receiving thrombolysis before hospitalization, there was a suggestion that receipt of thrombolysis within an hour of symptom onset was associated with greater benefit than receipt after a longer interval. Women appeared to have better long term survival than males, and those with a history of previous AMI worse long-term survival.
Strengths and limitations Strength of this article is its long duration of follow-up. We have been unable to identify any previous research, from either a clinical trial or observational study, which has specifically investigated very long-term mortality outcome among people receiving thrombolysis. Our findings suggest that thrombolysis given in routine clinical settings is associated with important long-term mortality benefits for
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 1. Flowchart showing exclusions from analysis
Thrombolysis: long-term survival benefits
195
Table 1. Association between receipt of thrombolysis, other patient characteristics and all-cause mortality at 28 days post-AMI among 3092 participants in the RCGP MI study Characteristic
N, alive:dead
Unadjusted HR (95% CI)
Adjusteda HR (95% CI)
2021 (65.4) 1071 (34.6)
1721:300 966:105
1.00 0.64 (0.51–0.80)
1.00 0.72 (0.58–0.90)
2036 (65.8) 1056 (34.2)
1791:245 896:160
1.00 1.27 (1.04–1.55)
1.00 1.00 (0.81–1.23)
249 (8.1) 1528 (49.4) 1315 (42.5)
242:7 1412:116 1033:282
1.00 2.76 (1.29–5.93) 8.37 (3.95–17.71)
1.00 2.73 (1.27–5.87) 8.08 (3.78–17.28)
1962 (63.5) 825 (26.7) 305 (9.9)
1694:268 740:85 253:52
1.00 0.75 (0.59–0.95) 1.28 (0.95–1.72)
1.00 1.11 (0.86–1.43) 1.19 (0.88–1.61)
1136 (36.7) 1085 (35.1) 871 (28.2)
1018:118 952:133 717:154
1.00 1.20 (0.94–1.54) 1.75 (1.37–2.22)
1.00 1.31 (1.01–1.68) 1.40 (1.10–1.79)
2290 (74.1) 765 (24.7) 37 (1.2)
1999:291 656:109 32:5
1.00 1.14 (0.92–1.42) 1.05 (0.43–2.53)
1.00 0.99 (0.79–1.23) 0.90 (0.37–2.18)
Adjusted for covariates: sex, age, smoking status, occupation, and previous AMI.
a
Table 2. Association between receipt of thrombolysis and all-cause mortality at different follow-up time points post-AMI Duration of follow-up post-MI
1 year 5 years 10 years 15 years End of follow-up
Thrombolysis not received, n
Thrombolysis received, n
Alive:dead
Adjusted HRa (95% CI)
1568:453 1218:803 911:1110 663:1358 481:1540
1.00b 1.00 1.00 1.00 1.00
Alive:dead 917:154 760:311 571:500 425:646 291:780
Adjusted HRa (95% CI) 0.69 (0.57–0.83) 0.76 (0.66–0.86) 0.85 (0.77–0.95) 0.88 (0.80–0.96) 0.92 (0.84–1.00)
Adjusted for thrombolysis status, sex, age, smoking status, occupation, and previous AMI. 1.00 represents the reference group.
a
b
patients with AMI. The observational nature of the study, however, means that survival might have been influenced by the greater use of beneficial cardiac interventions and secondary prevention treatments among patients receiving thrombolysis. This said, our findings are compatible with evidence from more selective clinical trial settings, and provide a powerful reminder that thrombolysis remains an important treatment option for patients with ST-elevation AMI when pPCI is not available. The flagging of study participants for death notification meant that we were able to follow-up almost all of the cohort efficiently for prolonged periods, with minimal loss to follow-up. The doctors certifying the cause of death, especially for events occurring many years after the recruitment AMI, were highly unlikely to be aware of treatments given at recruitment, so recall bias is unlikely to account for our findings. One limitation of the study was the small number of individuals who received thrombolysis before hospitalization, reflecting the cautious use of thrombolysis by GPs at the time of the study. The small number will have reduced the power of the study when comparing the effects of pre- and post-hospitalization administration of thrombolysis. Furthermore, the study was unable to collect
accurate symptom onset to thrombolysis times for people receiving thrombolysis in hospital, so we were only able to examine this in those receiving thrombolysis in the community, with few data in most strata. The majority of participants with a recorded contraindication did not receive thrombolysis. Our examination of the data excluding patients who had contraindications showed similar survival benefits associated with thrombolysis. The loss of statistical significance by 10 years post-AMI probably reflects loss of statistical power resulting from the exclusion of 25% of individuals from the cohort in these analyses.
Comparison with existing literature As an observational study, confounding and bias must always be considered when interpreting results found. This said, the long-term survival benefit found for thrombolysis is consistent with short-term results from clinical trials of thrombolysis (11,12). A 20-year follow-up of participants in a trial of reperfusion (which included both intra-coronary thrombolysis and coronary angioplasty) also found statistically significant mortality benefit from active therapy, which lasted for up to 20 years (13).
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Thrombolysis No Yes Sex Men Women Age at AMI ≤49 50–69 ≥70 Smoking status Non-smoker Smoker Missing Occupation Non-manual Manual Missing Previous AMI No Yes Missing
n (%)
Family Practice, 2015, Vol. 32, No. 2
196 The GREAT trial contained a smaller number of patients who received pre-hospital thrombolysis (163) than in our study (290), yet was able to find a 52% relative reduction (95% CI 14–89%;
Clinical implications
Figure 2. Survival at different time points during follow-up of participants who received and did not receive thrombolysis in the RCGP MI study
When the RCGP MI study started in 1991 there was clear evidence that thrombolysis saves lives after AMI, and that the benefits could be maximized by minimizing the ‘call to needle’ time. This lead to various initiatives in the UK to optimize delivery times, for example through the provision of thrombolysis in the community by GPs or ambulance paramedic crews, or in accident and emergency departments of hospitals by specialist nurses. Our study provided important evidence that GPs could administer thrombolysis safely and appropriately (9), although there were a number of practical issues which were likely to limit its use in this way. In subsequent years, evidence has emerged that pPCI (coronary angioplasty and stenting) confers greater benefits to patients with ST-elevated AMI than thrombolysis. In 2003, a meta-analysis showed that pPCI results in fewer shortand longer-term deaths, non-fatal re-infarctions and strokes than thrombolysis (1). The National Health Service has responded to this evidence by establishing networks of Heart Attack Centres where pPCI can be provided promptly, so that in 2012/13 virtually all (95%) of the 22 016 ST-elevation AMI patients receiving reperfusion treatments in England, Wales and Northern Ireland received pPCI, and only 5% thrombolysis (17). Many ambulance services have now shifted their focus from the provision of pre-hospital thrombolysis
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
P = 0.007) in mortality in the pre-hospital thrombolysis group at 1 year follow-up (5). However, the GREAT trial only included patients from GP practices in a rural setting at least 16 miles from the local hospital, resulting in a substantial time saved in delivering pre-hospital thrombolysis, which may account for the differences in study results. GP practices in the RCGP study were not exclusively rural so less time was saved by delivering pre-hospital thrombolysis. Thus, the median time delay in GREAT between symptom onset and injection of thrombolysis (or placebo) was 105 minutes (IQR not given), compared with 120 minutes (IQR = 60–215) in the RCGP MI study. In our study, GPs who were willing to use thrombolysis appeared to be cautious in its use (9), partly because community-based thrombolysis was highly controversial at the time of the study. Confounding by indication could have occurred if the GPs tended to use thrombolysis in the community in patients with more serious events, or with risk factors predictive of worse outcome. Although we were able to adjust for gender, age, occupation and previous AMI, there may have been other unmeasured factors that reduced our ability to find a benefit from pre-hospital thrombolysis. Although based on limited data, the symptom onset to thrombolysis analysis (Supplementary Table 5) provides some evidence to support the delivery of thrombolysis as quickly as possible. Furthermore, an observational study in Sweden in 2006 which had accurate and complete information about time from symptom onset to thrombolysis for 1690 patients given treatment pre-hospital and 3685 in-hospital, found a 29% one year mortality benefit in the pre-hospital group (8). Better long-term survival among women post-AMI has been found previously in the Framingham study, which reported an adjusted HR among women compared with men of 0.49 (95% CI 0.31–0.77) at 32-year follow-up (14). Other research has also found poorer long-term survival among those with a history of previous AMI (15). It was noteworthy that the gender and history of previous AMI differences appeared several years after the recruitment AMI, suggesting influences unrelated to initial therapies or interventions. Table 1 highlights that, of the variables measured, age has the greatest influence on prognosis post-AMI. This finding is supported by other studies (15,16).
Thrombolysis: long-term survival benefits
197
Table 3. Association between receipt of thrombolysis before (n = 290) or after hospitalization (n = 781) and all-cause mortality Duration of follow-up post-MI
28 days 1 year 5 years 10 years 15 years End of follow-up
Pre-hospital thrombolysis, N
In-hospital thrombolysis, N
Alive:dead
Alive:dead
255:35 239:51 192:98 136:154 108:182 76:214
Adjusted HR (95% CI) 1.00 1.00 1.00 1.00 1.00 1.00
711:70 678:103 568:213 435:346 317:464 215:566
Adjusted HRa (95% CI) 0.75 (0.50–1.14) 0.76 (0.54–1.08) 0.86 (0.67–1.10 0.87 (0.71–1.05) 0.96 (0.81–1.15) 1.00 (0.85–1.17)
Adjusted for thrombolysis status, sex, age, smoking status, occupation, and previous AMI.
a
Supplementary material Supplementary material is available at Family Practice online.
Declaration Funding: The original RCGP MI study was funded by SmithKline Beecham through an unconditional research grant. The company have not had access to the data or been involved in its analysis or interpretation. The University of Aberdeen has provided support for maintenance of the database and death notifications. Ethics approval: The study received ethical approval from the Independent Clinical Research Ethics Committee of the RCGP, and local ethics committees of participating doctors were informed of the study. Conflict of interest: none.
Acknowledgements We thank the GPs who recruited the patients for us, and Dr Clifford Kay who was Principal Investigator for the study at its inception.
References 1. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003; 361: 13–20. 2. Scottish Intercollegiate Guidelines Network (SIGN). Acute Coronary Syndromes. Edinburgh, UK: SIGN, 2013.
3. Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 343: 311–22. 4. Boersma E, Maas AC, Deckers JW et al. Early thrombolytic treatment in acute myocardial infarction: reappraisal of the golden hour. Lancet 1996; 348: 771–5. 5. Rawles JM. Feasibility, safety, and efficacy of domiciliary thrombolysis by general practitioners: Grampian region early anistreplase trial. GREAT Group. BMJ 1992; 305: 548–53. 6. Morrison LJ, Verbeek PR, McDonald AC et al. Mortality and prehospital thrombolysis for acute myocardial infarction: A meta-analysis. JAMA 2000; 283: 2686–92. 7. Rawles J; GREAT. GREAT: 10 year survival of patients with suspected acute myocardial infarction in a randomised comparison of prehospital and hospital thrombolysis. Heart 2003; 89: 563–4. 8. Björklund E, Stenestrand U, Lindbäck J et al. Pre-hospital thrombolysis delivered by paramedics is associated with reduced time delay and mortality in ambulance-transported real-life patients with ST-elevation myocardial infarction. Eur Heart J 2006; 27: 1146–52. 9. Hannaford P, Vincent R, Ferry S et al. Assessment of the practicality and safety of thrombolysis with anistreplase given by general practitioners. Br J Gen Pract 1995; 45: 175–9. 1 0. Katz MH. Multivariable Analysis: A Practical Guide for Clinicians. Cambridge, UK: Cambridge University Press, 2002, pp. 141–58. 11. Franzosi MG, Santoro E, De Vita C et al. Ten-year follow-up of the first megatrial testing thrombolytic therapy in patients with acute myocardial infarction: results of the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto-1 Study. Circulation 1998; 98: 2659–65. 12. Second International Study of Infarct Survival Collaborative Group (ISIS2). Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17 187 cases of suspected acute myocardial infarction: ISIS 2. Lancet 1988; 332: 349–60. 13. van Domburg RT, Sonnenschein K, Nieuwlaat R et al. Sustained benefit 20 years after reperfusion therapy in acute myocardial infarction. J Am Coll Cardiol 2005; 46: 15–20. 14. Wong ND, Cupples A, Ostfeld AM et al. Risk factors for long-term coronory prognosis after initial myocardial infarction: the Framingham Study. Am J Epidemiol 1989; 130: 469–80. 15. Lee KL, Woodlief LH, Topol EJ et al. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction: results from an International Trial of 41 021 patients. Circulation 1995; 91: 1659–68. 16. White HD, Van de Werf FJ. Thrombolysis for acute myocardial infarction. Circulation 1998; 97: 1632–46. 17. Myocardial Ischaemia National Audit Project. How the NHS cares for patients with heart attack. Annual report April 2012–March 2013. https:// www.ucl.ac.uk/nicor/audits/minap/documents/annual_reports/minappublic-report-2014 (accessed on 31 December 2014).
Downloaded from http://fampra.oxfordjournals.org/ at North Dakota State University on May 26, 2015
to ensuring the rapid admission of patients to a Heart Attack Centre, often bypassing smaller hospitals or accident and emergency departments. Although this focus is appropriate for most patients with ST-elevation AMI, it is important to remember that immediate thrombolysis remains an important treatment for patients living in areas where pPCI cannot be delivered within 120 minutes of initial diagnosis (2). Our results contribute to the evidence base showing that thrombolysis is an effective treatment associated with substantial long term benefits. Although we were unable to corroborate clinical trial data showing important benefits from the pre-hospital administration of thrombolysis, our symptom-to needle time data highlight the importance of providing thrombolysis as quickly as possible, when it is used.
