QJM Advance Access originally published online on February 19, 2008
QJM 2008 101(5):387-395; doi:10.1093/qjmed/hcn018
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Repeat thrombolysis or conservative therapy vs. rescue percutaneous coronary intervention for failed thrombolysis: systematic review and meta-analysis
From the 1Institute of Cardiology, John Radcliffe Hospital, Oxford, UK, 2Division of Cardiology, University of Turin, Turin, Italy, 3Department of Medicine, Virginia Commonwealth University, Richmond, VA, USA, and 4Antwerp Cardiovascular Institute Middelheim, AZ Middelheim, Antwerp, Belgium
Address correspondence to Dr Luca Testa, MD, Interventional Cardiology Fellow at John Radcliffe Hospital, Oxford, UK, Fellow of the European Association of Percutaneous Coronary Intervention Research Fellow in Molecular Biology at Policlinico Gemelli, Institute of Cardiology, Rome, Italy. Work Address: Cardiology Ward, level 2, John Radcliffe Hospital, Headley Way, Hoxford, OX3 9DU, UK. email: luctes{at}gmail.com
Received 9 December 2007 and in revised form 18 January 2008
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Background: Despite proven advantages of primary percutaneous coronary intervention (PCI), thrombolysis remains the first line treatment for ST-elevation myocardial infarction (STEMI) worldwide. Management of patients with failed thrombolysis is still debated, and data from existing randomized controlled trials are conflicting.
Aim: To compare the risk/benefit profile of repeat thrombolysis (RT) vs. rescue PCI in patients with failed thrombolysis.
Methods: Search of BioMedCentral, CENTRAL, mRCT and PubMed for randomized controlled trials comparing rescue PCI vs. conservative therapy and/or RT vs. conservative therapy. Outcomes of interest assessed by adjusted indirect meta-analysis: major adverse events (MAE, defined as the composite of overall mortality and re-infarction), stroke, congestive heart failure (CHF), major bleeds (MB), and minor bleeds. Overall mortality and re-infarction have been also analysed individually.
Results: Eight trials were included (1318 patients). Follow-up ranged from ‘in-hospital’ to 6 months. No significant difference was found for the risk of MAE [OR 0.93(0.26–3.35), P = 0.4], overall mortality [OR 1.01(0.52–1.95), P = 0.15], stroke [OR 5.03(0.64–39.1), P = 0.58] and CHF [OR 0.74(0.28–1.96), P = 0.6]. Compared with conservative therapy, rescue PCI was associated with a 70% reduction in the risk of re-infarction [OR 0.32(0.14–0.74), P = 0.008], number needed to treat 17. No difference in terms of MB was found [OR 0.5(0.1–2.5), P = 0.09], while a greater risk of minor bleeds was observed with rescue PCI [OR 2.48(1.08–5.7), P = 0.04], number needed to harm 50.
Conclusion: Although the observed benefit is modest, these data support the use of PCI after failed thrombolysis.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
ST-elevation myocardial infarction (STEMI) continues to be a major health problem in industrialized countries and is becoming increasingly common in developing countries.1 The benefits of an open infarct related artery are well recognized,2 but despite proven advantages of primary percutaneous coronary intervention (PCI), thrombolysis remains the most common reperfusion strategy worldwide.3 Unfortunately, the latter is unable to restore normal coronary flow in 30–40% of treated patients.4 Perhaps surprisingly, management after failed thrombolysis is still debated. Recent data assessing the best strategy for failed thrombolysis are conflicting, and further evaluation using a direct meta-analytical approach is unsuitable as only one trial compared rescue PCI directly with repeat thrombolysis (RT).5 Recent guidelines recommend rescue PCI after failed thrombolysis with a class I recommendation based on current evidence and expert consensus. The evidence on which this is based is currently level B (data derived from a single randomized trial, or non-randomized studies), with level A requiring meta-analysis or multiple head-to-head randomized controlled trials.1 By means of indirect meta-analysis, we sought to improve the level of evidence for which this recommendation is based, by comparing rescue PCI with RT using conservative therapy as the common reference standard.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Data sources and searches
BioMedCentral, CENTRAL, mRCT and PubMed were searched without language restrictions (updated to September 2007), according to an established method.6 Pertinent trials were also searched in major recent international cardiology meetings. References of original and review articles were cross-checked.
Study selection
Inclusion criteria were: (i) randomized allocation; (ii) controlled comparison of rescue PCI or RT vs. conservative treatment, in the setting of STEMI patients with failed thrombolysis and (iii) intention-to-treat analysis. Exclusion criteria were: (a) equivocal treatment allocation process; (b) significant imbalances in major baseline characteristics among study groups and (c) incomplete (<80%) follow-up.
Data extraction and quality assessment
Four trained and independent reviewers (L.T., W.v.G., G.B.Z. and P.A.) performed data abstraction blindly. Patient-level data were not available. Frequency of independent reviewer disagreement was low and divergences were resolved by consensus. The endpoints of interest were: (i) the combined rate of major adverse events (MAE), defined as all cause death and non-fatal re-infarction, (ii) the rate of stroke, (iii) the rate of congestive heart failure (CHF), (iv) major bleeds (MB) defined as clinically overt bleedings in a critical anatomical site (intra-cranial, intra-spinal, intra-ocular, retroperitoneal, pericardial, intra-articular), bleeds requiring transfusions, or those associated with a 
2 g/dl drop in haemoglobin and (v) the rate of minor bleeds, defined as bleedings not fitting with the features listed above. Additional analyses were carried out for single end points. Original study definitions for re-infarction, stroke and CHF were used.
The internal validity of the included trials was appraised according to the Cochrane Collaboration criteria, i.e. judging the risk of selection, performance, attrition and adjudication biases. The risk of bias was expressed as low (A), moderate (B), or high (C) with incomplete reporting leading to inability to ascertain the underlying risk of bias scored ‘D’. Allocation concealment was distinguished as adequate (A), unclear (B), inadequate (C) or not used (D).7 The Jadad score8 was also applied to evaluate the quality of the included reports. Sensitivity analysis was performed by excluding trials one at time, from those with the lowest to those with the highest quality score, in order to assess the contribution of each study to the pooled estimates.7 The likelihood of publication bias was assessed graphically by generating a funnel plot for the combined endpoint of MAE and mathematically by means of Egger's9 test (P for significant asymmetry <0.1).
This study is inspired by good practice guidelines, including those from the Cochrane Collaboration, and the Quality of Reporting of Meta-analyses (QUOROM) statement.7
Data synthesis and analysis
OpenOffice.org, SPSS 11.0 for Windows®, and the Review Manager freeware package (RevMan 4.2)10 were used for analysis. Review Manager is a comprehensive statistical and reviewing programme, developed and maintained by The Cochrane Collaboration (Oxford, UK), which includes ad hoc statistical tools for pooled estimate calculations, according to several methods. At first we calculated odds ratios (OR) for the comparisons of rescue PCI vs. conservative therapy and RT vs. conservative therapy, with 95% confidence intervals (95%CI), as summary statistics. Binary outcomes from individual studies were combined with both Der Simonian and Laird random-effect model and fixed-effect model, according to an intention-to-treat analysis. We also carried out the ‘z’- test with z = estimated effect size/standard error of the estimated effect size, and the OR considered on the log scale. As log(OR) has a unimodal distribution, the reported z-values were analysed to obtain a two-tailed ‘P ’, and hypothesis testing results were considered statistically significant at the 0.05 level.7
We computed Cochrane Q heterogeneity test (H) by summing the squared deviations of each study's estimate from the overall meta-analytic estimate, weighting each study's contribution in the same manner. Heterogeneity was considered significant at ‘P for H’ <0.10.7 According to Higgins et al.,11 we used the Q together with the resulting degrees of freedom (df) to calculate the proportion of variation due to heterogeneity [Inconsistency: (I2) = (Q–df)/Q]. The degree of inconsistency among studies (I2) was estimated with scores of <25%, between 25 and 75%, and >75% representing, respectively, low, moderate or high inconsistency.
From random effect (der Simonian and Laird method) OR comparing rescue PCI vs. conservative therapy and RT vs. conservative therapy, we generated an interaction OR with 95% confidence interval for the different combined treatments and z-scores for two-tailed hypothesis testing (P is significant if <0.05). Specifically, interaction OR with respective 95% confidence intervals and inconsistency among trials are calculated according to the following formulas:
ln (ORrescuePCI vs. RT) = ln (ORrescuePCI vs. CT) – ln (ORRT vs. CT);
var [ln (ORrescuePCI vs. RT)] = var [ln (ORrescuePCI vs. CT)] + var [ln (ORRT vs. CT)]; and
I2(rescuePCI vs. RT) = (
2(rescuePCI) + 2(RT)) x 10/degress of freedom(rescuePCI) + degrees of freedom(RT) (Appendix).12
As per protocol, we calculated the number needed to treat (NNT) and the number needed to harm (NNH) as the inverse of absolute risk difference, with relative confidence intervals. The degree of inconsistency among studies (I2) was also estimated according to Higgins et al.11
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Search results and study characteristics
Out of 40 potentially relevant citations retrieved (Figure 1), 8 studies5,13–19 comparing rescue PCI or RT vs. conservative therapy and totalling 1318 patients, were finally included. Main characteristics of included studies are listed in Table 1. Some discrepancies were found between random effect and fixed effect model, so all the results according to both models are listed in Table 2.
|
|
|
Repeat thrombolysis vs. conservative therapy
At a follow-up ranging from ‘in hospital’ to 6 months, no differences between RT and conservative therapy were found for the risk of MAE, overall death, re-infarction or MB. Stroke and CHF risks were reported only in the REACT trial. In presence of moderate inconsistency (I2 = 55.5%), RT was associated with a significantly increased risk of minor bleeds, with an NNH of 14 (95% CI: 7.5–100), according to the fixed effect but not the random effect model (Figure 2).
|
Rescue PCI vs. conservative therapy
At a follow-up ranging from ‘in hospital’ to 6 months, no differences were found between rescue PCI and conservative therapy in the risk of death or MB with either model. Rescue PCI was associated with a significant increase in the risk of minor bleeds with an NNH of nine (6–18), in the absence of inconsistency (I2 = 0%). Notably, with a fixed effect model rescue PCI was associated with a significant reduction in the risk of MAE and re-infarction, with an NNT of 14 (9–50) and 25 (11–100), respectively. Such differences were confirmed for MAE but not re-infarction using a random effect model. Rescue PCI was associated, according to the fixed effect model, with a significant lower risk of CHF with an NNT of 20 (11–100). Rescue PCI was consistently associated with an increased risk of stroke, with an NNH of 33 (20–100) (Figure 2).
Analysis of heterogeneity/inconsistency
Moderate inconsistency was found in the majority of the comparisons probably reflecting differences in sample size, setting and length of follow-up between trials.
Sensitivity analysis and assessment of possible biases
On sensitivity analysis based on study quality, for the end point of death at the longest available follow-up, the sequential exclusion of each trial in the setting of rescue-PCI vs. conservative therapy did not significantly change the result. On the other hand, for RT vs. conservative therapy, the exclusion of Sarullo et al.18 slightly reduced the magnitude of the effect, possibly reflecting the comparatively shorter follow-up in this study.
The overall quality of the included trials was adjudicated at low risk of bias according to the Cochrane Collaboration Criteria, and the average Jadad score was estimated as 3.2 (2.8 for trials evaluating rescue-PCI vs. conservative therapy and 4.0 for those evaluating RT vs. conservative therapy).
Funnel plots for MAE showed an overall symmetry, further confirmed by results of Egger's test with P for significant asymmetry of 0.65 for rescue PCI vs. conservative therapy and 0.96 for RT vs. conservative therapy, thus excluding possible publication or ‘small study’ bias. Our decision to exclude trials with incomplete follow-up could have introduced a ‘selection’ bias, however, none of the possible hits found with either the electronic or manual search was excluded for this reason (Figure 3).
|
Repeat thrombolysis vs. rescue PCI: adjusted indirect comparison
At a follow-up ranging from ‘in hospital’ to 6 months, no significant difference was found for the risk of MAE [OR 0.93 (0.26–3.35), P = 0.4], nor for the risk of overall mortality [OR 1.01 (0.52–1.95), P = 0.15] or CHF [OR 0.74 (0.28–1.96), P = 0.6]. Of note, rescue PCI was associated with a 70% reduction in the risk of re-infarction [OR 0.32 (0.14–0.74), P = 0.008], with an NNT of 17 (8–55). No statistically significant difference was found, despite a slight trend in favour of RT, for the risk of stroke [OR 5.03 (0.64–39.1), P = 0.58]. No difference in terms of MB was found [OR 0.5 (0.1–2.5), P = 0.1], while a greater risk of minor bleeds was observed with rescue PCI compare with RT [OR 2.48 (1.08–5.7), P = 0.04], with an NNH of 50 (20–100) (Figure 4).
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Contribution of the present study to the current context
Our independent systematic review and meta-analysis excludes any clinical benefit for RT over conservative therapy in patients with failed thrombolysis for STEMI. It shows benefit from rescue PCI over conservative therapy with respect to the risks of MAE (NNT = 14) and death (NNT = 25). Allocating 100 patients with failed thrombolysis to rescue PCI would prevent seven MAE and four fatalities compared to conservative therapy. There was also a strong trend in favour of rescue PCI over conservative therapy with respect to the risk of heart failure.
Compared to repeat thrombolysis, rescue PCI reduced the risk of re-infarction by 70% (NNT = 17), with only a slight increase in the risk of minor bleeds (NNH = 50). Therefore, allocating 100 patients with failed thrombolysis to rescue PCI would prevent six re-infarctions while causing two minor bleeds compared with repeat thrombolysis.
These data show no major benefit in terms of MAE as there is no impact on overall mortality. This could reflect selection as patients who reach hospital and receive thrombolysis have already proved themselves as ‘survivors’ and as a consequence demonstrating a survival benefit of further treatment in this population is difficult. The lack of survival benefit might also relate to several features of the included studies like the relatively short follow-up and the small sample sizes, limiting the power to show any difference in terms of risk of death. Similarly no benefit has been shown in lowering the risk of CHF which probably reflects the limited ability of rescue PCI to salvage myocardium after prolonged ischaemia.
Rescue PCI does appear to reduce re-infarction. Pharmacological management of STEMI only lyses thrombus, and residual unstable plaque can act as a substrate for further events. The large reduction in risk of re-infarction may be due to a ‘plaque sealing’ effect obtained by means of stent placement during PCI.20 As for the risk of stroke, the small number of events possibly precluded the chance of highlighting any specific effect of the treatments. The higher risk of minor bleeds with rescue PCI is related to arterial access site complications during the procedure, but it could also be associated to the exclusion of patients at high bleeding risk from the trials evaluating repeat thrombolysis. The latter, together with the usage of IIb/IIIa in the rescue PCI approach could also explain the lack of advantage in terms of MB for the rescue PCI approach.
Rescue PCI is defined as PCI within 12 h after failed fibrinolysis for patients with continuing or recurrent myocardial ischaemia.1 A major limitation in adopting a strategy of rescue PCI is the difficulty in identifying patients for whom fibrinolytic therapy has not restored antegrade coronary flow.1 Unless unsuccessful fibrinolysis is recognized and treated quickly (within 3–6 h of onset of symptoms), salvage of ischaemic myocardium is unlikely. Unfortunately, clinical markers of reperfusion, such as relief of ischaemic-type chest discomfort, partial resolution of ST-segment elevation (50%) and reperfusion arrhythmias, have limited predictive value in identifying successful reperfusion.21 In a prior era in which the practice of PCI was less mature, immediate catheterization of all patients after fibrinolytic therapy to identify those with an occluded infarct artery was found to be impractical, costly and often associated with bleeding complications.22,23 Irreversible myocardial damage will occur once coronary occlusion has been present for more than 3 h.22 Thus if mandated rescue PCI is to be effective, it must be performed in a timely manner, and needs a similar network to primary PCI. Currently, there are too many delays associated with this strategy including the presentation of the patient to hospital, infusion of the fibrinolytic agent, recognition of failed fibrinolysis and subsequent initiation of PCI.
More recently, routine PCI after the administration of thrombolysis, i.e. not driven by persistent symptoms and/or ST elevation, has been shown to significantly reduce the risk of death and new myocardial infarction compared to a rescue PCI approach according to current guidelines.24 This advantage might be related to a higher percentage of TIMI 3 flow with the routine percutaneous revascularization approach but also to the minimized delay between thrombolysis and PCI.
It has been suggested that ‘facilitated PCI’ might optimize the therapy for STEMI but it is now clear that it has a worse outcome compared to primary PCI.25,26 Sub-optimum antithrombotic treatment in the facilitated PCI approach, in particular the absence of upfront administration of clopidogrel and treatment delays probably underline this result.
Limitation of the present study
A limitation inherent to all meta-analyses is the potential heterogeneity among studies, in terms of protocols, patients, and sample sizes and the unavailability of patient-level data.
However, the primary disagreement that arises in meta-analyses is whether to incorporate between-study variation (heterogeneity and inconsistency) in estimating summaries of effect size. If there is little between-study variation (i.e. non-significant heterogeneity), the choice between random effect and fixed effect models usually make little difference in the results. In presence of significant heterogeneity it may be more appropriate to analyse results using both methods.
A statistically significant result with the fixed effect model indicates that there is an effect in at least one of the studies, and the overall result is an average measure of treatment effect of the studies in the analysis.
On the other hand, the random effects approach relies on assumptions that the studies are a random sample from a hypothetical population of studies and that the heterogeneity between studies can be represented by a single variance. The random effects model tends to give a more conservative estimate (i.e. with wider confidence intervals).
If the test of heterogeneity is statistically significant and the differences in results are of practical importance, then more emphasis should be placed on the random effects model.10
Unlike previous reports,27 that only by means of fixed effect relative risk calculation, stated the superiority of rescue PCI over conservative therapy, we presented and discussed all the results according to both validated models.7 We also evaluated the external validity of the included trials with two validated and widely used models.7,8
Moreover, although being an indirect comparison, adjusted meta-analyses with inverse variance weightings are useful instruments in the assessment of different treatment strategies as they are usually confirmed by properly design randomized trials.28,29 Moreover, such methodology is receiving progressive attention and acceptance.30
Avenues for future research
In our view, a better definition of failed thrombolysis should be sought. Moreover, an exact stratification of the risk of stroke and haemorrhage in patients with failed thrombolysis would be of great utility in order to appraise homogeneous populations which clearly evaluate the risk/benefit ratio of rescue PCI.
|
Conclusion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Our data excludes the clinical utility of RT while suggesting a modest advantage of rescue PCI over conservative therapy for the treatment of patients with failed thrombolysis. These data further support the utility of a primary angioplasty approach rather than either early mandated PCI following thrombolysis or routine facilitated PCI.
|
Appendix |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Model for the calculation of natural logarithm (ln), Variance (var) and Inconsistency (I2) of competing interventions by means of adjusted indirect meta-analysis with inverse variance weighting. RT: repeat thrombolysis; CT: conservative treatment;
2: test for heterogeneity, OR: odd ratio. - ln (ORrescuePCI vs. RT) = ln (ORrescuePCI vs. CT) – ln (ORRT vs. CT);
- var [ln (ORrescuePCI vs. RT)] = var [ln (ORrescuePCI vs. CT)] + var [ln (ORRT vs. CT)]; and
- I2(rescuePCI vs. RT) = (2(rescuePCI) + 2(RT)) x 10/degress of freedom(rescuePCI) + degrees of freedom(RT).
- Once calculated the variance, it is necessary to calculate the standard error (SE), i.e. the square root of the variance.
- Confidence interval (CI) at 95% of the ln is then calculated as follows: ln + 1.96*SE (upper CI) and ln–1.96*SE (lower CI).
- The OR for the comparison of rescue PCI vs. RT can be calculated as the exponent of the ln.
- CI of the OR is finally calculated as the exponent of the CI of the ln.
- Zeta score for the calculation of the P for significance is finally derived from ln/SE.
|
Acknowledgements |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
Dr Banning is partially funded by the Oxford Partnership Comprehensive Biomedical Research Centre with funding from the Department of Health NIHR BMRC funding scheme. This study is part of a senior investigator project of the Center for Overview, Meta-analysis, and Evidence-based medicine Training (METCARDIO), based in Oxford, UK (http://www.metcardio.org).
Conflict of interest: None declared.
|
REFERENCES |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionConclusionAppendixAcknowledgementsREFERENCES |
|---|
1. Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, et al. American College of Cardiology; American Heart Association; Canadian Cardiovascular Society. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction – executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1999 guidelines for the management of patients with acute myocardial infarction). J Am Coll Cardiol (2004) 44:671–719. Available at www.americanheart.org.
2. The GUSTO Angiographic Investigators. The effects of tissue plasminogen activator, streptokinase, or both on coronaryartery patency, ventricular function, and survival after acute myocardial infarction. N Engl J Med (1993) 329:1615–22. [Erratum, N Engl J Med 1994; 330:516.].
3. Wiviott SD, Morrow DA, Giugliano RP, Gibson CM, McCabe CH, Cannon CP, et al. Performance of the thrombolysis in myocardial infarction risk index for early acute coronary syndrome in the National Registry of Myocardial Infarction: a simple risk index predicts mortality in both ST and non-ST elevation myocardial infarction. J Am Coll Cardiol (2004) 44:783–9.
4. 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.[CrossRef][Web of Science][Medline]
5. Gershlick AH, Stephens-Lloyd A, Hughes S, Abrams KR, Stevens SE, Uren NG, et al. REACT Trial Investigators. Rescue angioplasty after failed thrombolytic therapy for acute myocardial infarction. N Engl J Med (2005) 353:2758–68.
6. Biondi Zoccai GG, Agostoni P, Abbate A, Testa L, Burzotta F. A simple hint to improve Robinson and Dickersin's highly sensitive. PubMed search strategy for controlled clinical trials. Int J Epidemiol (2005) 34:224–5.
7. Clarke M, Oxman AD, eds. Cochrane Reviewers’ Handbook 4.1.3. The Cochrane Collaboration, The Cochrane Library, Issue 3, 2001. [http://www.cochrane.org/resources/handbook/] Accessed 12 February 2008.
8. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials (1996) 17:1–12.[CrossRef][Web of Science][Medline]
9. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. Br Med J (1997) 315:629–34.
10. Review Manager 4.2.2. Oxford, UK: Update Software. Updated quarterly. Available from The Cochrane Collaboration at http://www.cochrane.org/***.
11. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Br Med J (2003) 327:557–60.
12. Song F, Altman D, Glenny A-M, Deeks JJ. Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analysis. Br Med J (2003) 326:472.
13. Sutton AG, Campbell PG, Graham R, Price DJ, Gray JC, Grech ED, et al. A randomized trial of rescue angioplasty versus a conservative approach for failed fibrinolysis in ST-segment elevation myocardial infarction: the Middlesbrough Early Revascularization to Limit INfarction (MERLIN) trial. J Am Coll Cardiol (2004) 44:287–96.
14. Ellis SG, Da Silva ER, Spaulding CM, Nobuyoshi M, Weiner B, Talley JD. Review of immediate angioplasty after fibrinolytic therapy for acute myocardial infarction: insights from the RESCUE I, RESCUE II, and other contemporary clinical experiences. Am Heart J (2000) 139:1046–53.[CrossRef][Web of Science][Medline]
15. Ellis SG, da Silva ER, Heyndrickx G, Talley JD, Cernigliaro C, Steg G, et al. Randomized comparison of rescue angioplasty with conservative management of patients with early failure of thrombolysis for acute anterior myocardial infarction. Circulation (1994) 90:2280–4.
16. Ellis SG, Lincoff AM, George BS, Kereiakes DJ, Ohman EM, Krucoff MW, et al. Randomized evaluation of coronary angioplasty for early TIMI 2 flow after thrombolytic therapy for the treatment of acute myocardial infarction: a new look at an old study. The Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) Study Group. Coron Artery Dis (1994) 5:611–5.[Web of Science][Medline]
17. Belenkie I, Traboulsi M, Hall CA, Hansen JL, Roth DL, Manyari D, Filipchuck NG, Schnurr LP, Rosenal TW, Smith ER. Rescue angioplasty during myocardial infarction has a beneficial effect on mortality: a tenable hypothesis. Can J Cardiol (1992) 8:357–62.[Web of Science][Medline]
18. Sarullo FM, Americo L, Di Pasquale P, Castello A, Mauri F. Efficacy of rescue thrombolysis in patients with acute myocardial infarction: preliminary findings. Cardiovasc Drugs Ther (2000) 14:83–9.[CrossRef][Web of Science][Medline]
19. Mounsey JP, Skinner JS, Hawkins T, MacDermott AF, Furniss SS, Adams PC, et al. Rescue thrombolysis: alteplase as adjuvant treatment after streptokinase in acute myocardial infarction. Br Heart J (1995) 74:348–53.
20. Colombo A, Iakovou I. Plaque sealing: a concept waiting for support. Int J Cardiovasc Intervent (2005) 7:72–4.[CrossRef][Medline]
21. Goldman LE, Eisenberg MJ. Identification and management of patients with failed thrombolysis after acute myocardial infarction. Ann Intern Med (2000) 132:556–65.
22. The TIMI Research Group. Immediate vs delayed catheterization and angioplasty following thrombolytic therapy for acute myocardial infarction: TIMI II A results. JAMA (1988) 260:2849–58.
23. Topol EJ, Califf RM, George BS, Kereiakes DJ, Abbottsmith CW, Candela RJ, et al. A randomized trial of immediate versus delayed elective angioplasty after intravenous tissue plasminogen activator in acute myocardial infarction. N Engl J Med (1987) 317:581–8.[Abstract]
24. Mario Carlo Di. Randomised evaluation of routine transfer for urgent PCI or local management for patients admitted with STEMI to centres without PCI facilities initially treated with reteplase, heparin and abciximab (CARESS in AMI). In: Hot Line Session II. (2007) September 3. European Society of Cardiology Congress.
25. Assessment of the Safety and Efficacy of a New Treatment Strategy with Percutaneous Coronary Intervention (ASSENT-4 PCI) investigators. Primary versus tenecteplase-facilitated percutaneous coronary intervention in patients with ST-segment elevation acute myocardial infarction (ASSENT-4 PCI): randomised trial. Lancet (2006) 367:569–78.[CrossRef][Web of Science][Medline]
26. Ellis S. Final results of the FINESSE (Facilitated INtervention with Enhanced Reperfusion Speed to Stop Events) trial: evaluation of abciximab + half-dose reteplase, abciximab alone, or placebo for facilitation of primary PCI for ST elevation MI. In: Hot line II session. (2007) September 3. European Society of Cardiology Congress.
27. Wijeysundera HC, Vijayaraghavan R, Nallamothu BK, Foody JM, Krumholz HM, Phillips CO, et al. Rescue angioplasty or repeat fibrinolysis after failed fibrinolytic therapy for ST-segment myocardial infarction: a meta-analysis of randomized trials. J Am Coll Cardiol (2007) 49:422–30.
28. Biondi-Zoccai GGL, Agostoni P, Abbate A, Testa L, Burzotta F, Lotrionte M, et al. Adjusted indirect comparison of intracoronary drug-eluting stents: evidence from a meta analysis of randomized bare-metal-stent-controlled trias. Int J Cardiol (2005) 100:119–23.[CrossRef][Web of Science][Medline]
29. Kastrati A, Mehilli J, von Beckerath N, Dibra A, Hausleiter J, Pache J, et al. ISAR-DESIRE Study Investigators. Sirolimus-eluting stent or paclitaxel-eluting stent vs balloon angioplasty for prevention of recurrences in patients with coronary in-stent restenosis: a randomized controlled trial. JAMA (2005) 293:165–71.
30. Testa L, Biondi Zoccai GGL, Porto I, Andreotti F, Crea F. Aspirin plus warfarin versus aspirin plus clopidogrel after acute coronary syndromes: adjusted indirect meta analysis on 69741 patients. Am J Cardiol (2007) 99:1637–42.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||