INTRODUCTION — Coronary reperfusion with fibrinolysis or primary percutaneous coronary intervention (PCI) substantially improves survival in patients with an acute ST-elevation myocardial infarction (STEMI) compared with no reperfusion therapy. PCI is preferred for most patients if it can be performed by an experienced operator with less than a 90-minute delay from presentation to the emergency department. However, fibrinolysis remains an important therapeutic modality, due in part to limited availability of primary PCI in some locations. (See "Acute ST-elevation myocardial infarction: Selecting a reperfusion strategy" and "Primary percutaneous coronary intervention in acute ST elevation myocardial infarction: Determinants of outcome" and "Acute ST-elevation myocardial infarction: The use of fibrinolytic therapy".)
The principle reasons for preference of PCI over fibrinolysis is the relatively high frequency of failure of fibrinolysis to establish reperfusion, early reocclusion after apparently successful fibrinolysis, and increased bleeding complications with fibrinolysis. (See 'Primary failure' below and 'Threatened reocclusion or reinfarction' below.)
This topic will review the roles of angiography followed by PCI (or coronary artery bypass graft surgery [CABG] if indicated) for failed fibrinolysis or threatened reocclusion and of retreatment with fibrinolytic drugs for threatened reocclusion. The use of routine angiography after fibrinolysis is discussed separately. (See "Acute ST-elevation myocardial infarction: Selecting a reperfusion strategy".)
IMPORTANCE OF RESTORATION OF NORMAL FLOW — The mortality benefit after fibrinolytic therapy in patients with STEMI is related to the degree to which flow has been restored in the infarct-related artery. In the cardiac catheterization laboratory, patency of the infarct-related artery is usually graded according to the Thrombosis in Myocardial Infarction (TIMI) classification system. In clinical trials of fibrinolysis, patency is usually measured at 60 to 90 minutes after the administration of fibrinolytic therapy:
●TIMI 0 refers to the absence of any antegrade flow beyond a coronary occlusion.
●TIMI 1 flow is faint antegrade coronary flow beyond the occlusion, although filling of the distal coronary bed is incomplete.
●TIMI 2 flow is delayed or sluggish antegrade flow with complete filling of the distal territory.
●TIMI 3 flow is normal flow that fills the distal coronary bed completely.
Only TIMI 3 (normal) flow is associated with a mortality benefit after fibrinolysis (figure 1 and figure 2) [1-6]. The magnitude of this effect was illustrated in a meta-analysis of five trials of almost 4000 patients [2]. Short-term mortality (ranging from in-hospital to 30 days in the different trials) was 3.7, 7, and 8.8 percent among patients with TIMI grade 3, 2, and 0/1, respectively. (See 'Primary failure' below.)
Additionally, in patients who undergo diagnostic angiography after fibrinolysis, both TIMI 0/1 and 2 flow are associated with increased mortality compared with those with TIMI 3 (normal) flow [1]. (See 'Diagnosis' below.)
PRIMARY FAILURE — Primary failure of fibrinolysis is best treated with immediate percutaneous coronary intervention (PCI). Patients should be referred to a facility with PCI capability immediately after fibrinolysis in most cases, so that failed fibrinolysis can be managed expeditiously. This strategy will avoid the need to consider repeat fibrinolysis. (See "Percutaneous coronary intervention after fibrinolysis for acute ST-elevation myocardial infarction", section on 'Patients treated at non-PCI hospitals or in an ambulance'.)
Primary failure based on clinical criteria is common. Using angiographic criteria of TIMI 3 flow within 90 minutes, failure is seen in 40 to 45 percent of patients [7].
Clinical manifestations — Primary failure of fibrinolysis is often clinically suspected by persistent or worsening chest pain (particularly if associated with other symptoms such as dyspnea and diaphoresis), persistent or worsening ST-segment elevation on the electrocardiogram (ECG), hemodynamic instability, and/or heart failure [8].
However, these parameters do not predict failed fibrinolysis in all patients [9-11]. Chest pain alone is subjective, especially if improved, and many patients do not have complete ST-segment elevation resolution [8,12]. In addition, ST-segment elevation resolution and patency of the infarct-related artery are not always correlated. As a result, in the absence of clear indications of reperfusion, the clinician must maintain a high index of suspicion for primary failure [13]. Persistent chest pain or the absence of at least 50 percent of the initial ST-segment elevation on a follow-up ECG 60 to 90 minutes after fibrinolytic therapy are suggestive of lack of adequate reperfusion and warrant early referral for coronary angiography [14-16]. (See 'Primary failure' above.)
The potential lack of correlation between ST-segment elevation and TIMI flow grade was illustrated in the TIMI 14 trial of 444 patients with an interpretable ECG who underwent angiography at 90 minutes after fibrinolytic therapy [10,17]. TIMI 3 flow was present in only 79 percent of patients with complete (≥70 percent) ST-segment resolution and was also present in 44 to 50 percent of patients with partial or no ST-segment resolution. The latter patients were at increased risk for mortality, as the lack of complete ST-segment resolution was presumably due to extensive microvascular dysfunction [17].
Diagnosis — Primary failure can be suspected based on the patient's symptoms and ECG (see 'Clinical manifestations' above). The diagnosis is confirmed with coronary angiography.
Primary failure is defined as persistent occlusion of the infarct-related artery (TIMI grade 0/1 epicardial flow) at the time of coronary angiography. TIMI grade 0/1 occurred in 27 percent and was associated with a higher short-term mortality rate than TIMI 3 flow (8.8 versus 3.7 percent) [2]. Since mortality is also increased in patients with TIMI 2 flow, some have also considered TIMI 2 flow a sign of failed therapy (figure 1 and figure 2). (See 'TIMI grade 2 flow' below.)
In addition to patency of the epicardial coronary arteries, adequacy of microvascular flow, as determined by the myocardial perfusion grade, can also be assessed during angiography. Myocardial perfusion grade may also be an important predictor of long-term benefit [18]. (See 'Primary failure' above.)
Management — Patients who are diagnosed with or strongly suspected of a primary failure of fibrinolysis should receive an additional attempt at reperfusion. We prefer rescue PCI over repeat fibrinolysis.
Rescue PCI — Rescue or salvage PCI is defined as PCI performed within 12 hours of failed fibrinolysis (primary failure) [19]. If rescue PCI (usually with stenting) is performed, it appears to be important that it be undertaken quickly. In general, we begin to mobilize patients with apparent failure to the cardiac catheterization laboratory at approximately 45 to 60 minutes following fibrinolysis such that, when necessary, rescue PCI may be performed as close to 90 minutes after fibrinolysis as possible. Accordingly, because 40 to 45 percent of patients who receive fibrinolytic therapy will experience primary failure, we generally recommend that all who present at non-PCI-capable hospitals are transferred to a PCI-capable hospital as soon as possible after administration of the fibrinolytic agent. We do not recommend that a decision to transfer be deferred until it is determined whether primary failure has occurred or not. PCI or, if appropriate, bypass surgery is recommended in patients with less than TIMI grade 3 flow.
Revascularization may be deferred in patients with TIMI 3 flow who are hemodynamically stable [8]. However, observational data suggest that even patients with TIMI 3 flow have reduced mortality if PCI of the infarct-related artery is performed. (See "Percutaneous coronary intervention after fibrinolysis for acute ST-elevation myocardial infarction".)
Early randomized trials comparing rescue PCI (most patients underwent balloon angioplasty rather than stenting) with conservative therapy in patients with failed fibrinolysis found a trend toward lower mortality and lower risks of heart failure and reinfarction with rescue PCI [20]. Subsequently, stenting was shown to be superior to balloon angioplasty [13].
The REACT trial randomly assigned 427 patients with STEMI and failed fibrinolysis to conservative medical therapy, repeat fibrinolysis with a fibrin-specific agent, or rescue PCI (69 percent with stenting). Failed fibrinolysis was defined as less than 50 percent ST-segment elevation resolution within 90 minutes after therapy [21,22]. PCI was performed within 12 hours of the onset of pain if angiography revealed less than TIMI grade 3 flow and more than 50 percent stenosis in the infarct-related artery. (See 'Procedural issues' below and 'TIMI grade 2 flow' below.)
The following significantly improved outcomes were noted with rescue PCI:
●A higher rate of event-free survival (primary endpoint of death, reinfarction, stroke, or severe heart failure) at six months (85 versus 69 and 70 percent compared with repeat fibrinolysis or conservative therapy, adjusted hazard ratio [HR] 0.43 [95% CI 0.26-0.72] and 0.47 [95% CI 0.28-0.79]) and at one year (82 versus 64 and 68 percent, adjusted HR 0.44 [95% CI 0.28-0.71] and 0.51 [95% CI 0.32-0.83], respectively).
●A lower rate of all-cause mortality at a median follow-up of 4.4 years (6.2 versus 12.7 and 12.8 percent compared with repeat fibrinolysis or conservative therapy).
●A higher rate of freedom from revascularization at one year (85 versus 67 percent compared with repeat fibrinolysis or conservative therapy).
TIMI grade 2 flow — The above studies largely evaluated the efficacy of rescue PCI in patients, most of whom had failed fibrinolysis (TIMI grade 0/1 flow), although some patients with TIMI grade 2 flow were enrolled. Thus, the optimum treatment in patients with TIMI grade 2 flow is less well studied [18,23,24]. As noted above, patients with TIMI grade 2 flow have higher mortality rates than those with TIMI grade 3 (normal) flow: 7 versus 3.7 percent in a meta-analysis (figure 1 and figure 2) [2]. (See 'Importance of restoration of normal flow' above.)
Additional evidence to support rescue PCI for patients with TIMI grade 2 flow comes from an analysis of 668 patients (105 of whom were treated with rescue PCI) with TIMI grade 2 or 3 flow at angiography in the TIMI 10B trial [18]. Rescue PCI was associated with a trend toward reduction in mortality at two years (4 versus 11 percent).
Although the evidence to support rescue PCI for patients with TIMI grade 2 flow is less robust than that for TIMI grade 0/1, we suggest this approach. This is based on the inclusion of such patients in REACT and the increasing data suggesting benefit from routine angiography and PCI after fibrinolysis when appropriately timed. (See 'Rescue PCI' above and "Percutaneous coronary intervention after fibrinolysis for acute ST-elevation myocardial infarction".)
Procedural issues — Depending on the fibrinolytic agent chosen, the half-life of fibrinolytic activity may range from 50 to 90 minutes. Consequently, there is likely to be at least some fibrinolytic still on board when proceeding with rescue PCI. This may have an impact on decisions made at the time of PCI regarding the use of adjunctive antiplatelet and antithrombin therapy. Although large randomized clinical trials have not been performed to guide decisions, some reasonable options for adjunctive therapy are as follows:
●Use of adjunctive IIb/IIIa glycoprotein inhibitors – Because of the potential increased risk of intracranial hemorrhage associated with the combined use of IIb/IIIa inhibitors and fibrinolytic agents, it is wise to use caution when using it during rescue PCI. Factors to consider include the thrombus burden identified during coronary angiography, the age of the patient, the type of fibrinolytic used, and the time passed since the fibrinolytic agent was administered. (See "Acute ST-elevation myocardial infarction: Antiplatelet therapy", section on 'Intravenous agents'.)
●Choice of oral P2Y12 inhibitor – Because most guidelines recommend that clopidogrel 300 mg be administered at the time of administration of a fibrinolytic agent, in most circumstances, clopidogrel will already be on board. It is our practice at the time of rescue PCI to complete a standard 600 mg load of clopidogrel by administering another 300 mg. If no clopidogrel has yet been given, it would be the interventionalist's choice regarding which P2Y12 agent to administer. (See "Acute ST-elevation myocardial infarction: Antiplatelet therapy", section on 'Our approach to early DAPT'.)
●Choice of antithrombin agent – Because heparin is often started at the time of fibrinolytic agent administration, it seems reasonable to continue with heparin as the antithrombin agent during rescue PCI. Adjusting the dose according to the measured activated clotting time (ACT) may be accomplished in a similar manner as during standard PCI. Although not much data exist regarding the use of bivalirudin, in some circumstances, such as the presence of heparin-induced thrombocytopenia, its use may be considered. (See "Acute ST-elevation myocardial infarction: Management of anticoagulation", section on 'Primary percutaneous coronary intervention'.)
●Radial or femoral access – In most patients, we prefer radial rather than femoral artery access in order to decrease the likelihood of significant bleeding after fibrinolysis. (See "Periprocedural complications of percutaneous coronary intervention", section on 'Radial artery access'.)
●Direct stenting and predilation – Direct stenting (stenting without predilation of the coronary lesion) may improve myocardial perfusion and the incidence of death, MI, or heart failure compared with conventional stenting in patients who undergo PCI after fibrinolysis. This was illustrated in an observational study in which the choice of procedure was at the discretion of the treating physician [25]. Among patients with occluded infarct-related arteries, the likelihood of achieving TIMI myocardial perfusion grade 3 was significantly greater with direct stenting (54 versus 25 percent with conventional stenting).
Repeat fibrinolysis — Repeat fibrinolysis for failed primary fibrinolysis is usually considered when access to PCI is not available. However, it is not well studied. A 2007 meta-analysis included three trials that randomly assigned 410 patients to repeat fibrinolysis (all with alteplase) or conservative therapy with follow-up between in-hospital discharge and six months [20]. There was no significant difference in all-cause mortality, reinfarction, or stroke between the two groups, and both treatments were inferior to rescue PCI. (See 'Rescue PCI' above.)
THREATENED REOCCLUSION OR REINFARCTION — Threatened reocclusion is best treated with immediate percutaneous coronary intervention (PCI) with stenting (rather than repeat fibrinolysis). Patients should be referred to a facility with PCI capability soon after fibrinolysis in most cases. This strategy will avoid the need to consider repeat lysis, except in rare cases.
After apparently successful fibrinolysis by clinical criteria, early recurrence of ischemia or ST segment shifts on the ECG has been observed in 20 to 30 percent of patients [26,27] and thrombotic coronary reocclusion at the time of angiography in 5 to 15 percent [4,28-31]. Reinfarction occurs in 3 to 5 percent and is clinically silent in over 50 percent of cases [31-34].
Demographic and clinical variables have not proven very helpful in predicting which patients will develop reocclusion [28,29,31,32]. One early fibrinolytic trial attempted to identify angiographic predictors of reocclusion [35]. The following angiographic variables, identified at 90 minutes after thrombolytic therapy, were associated with a significantly higher rate of infarct-related artery reocclusion (in the absence of PCI): TIMI grade 2 flow, lesion ulceration, the presence of collaterals, high-grade residual stenosis, and lesions that were eccentric or thrombotic. Other studies did not find angiographic features at a 90-minute baseline study to be helpful for predicting reocclusion [36,37]. Differences among these trials could explain the different findings [35].
Outcomes — Reocclusion of an infarct-related artery after reperfusion therapy is associated with a significant increase in mortality [28,31-34]. Patients with coronary reocclusion have a worse 30-day and one-year mortality compared with those without [27,28,31-33]. The data are conflicting as to whether 30-day survivors of reinfarction do [32] or do not [31] have a modest increase in risk from 30 days to one year.
Diagnosis — The following are accepted criteria to diagnose, or at a minimum strongly suspect, threatened reocclusion [15,16]:
●Within the first 18 hours after initial fibrinolysis, a recurrent elevation in cardiac biomarkers accompanied by recurrent ST-segment elevation on ECG and at least one other supporting criterion (such as recurrent chest pain or hemodynamic decompensation).
●After 18 hours, a biomarker rise of at least 50 percent and at least one additional criterion are sufficient for the diagnosis.
Threatened reocclusion is characterized by the development of early recurrent ischemia after apparently successful fibrinolysis.
Objective evidence of thrombotic reocclusion includes persistent or recurrent chest pain associated with new or worsening ST-segment elevation, the development of new ischemic ECG changes after initial resolution, or a second elevation in cardiac biomarkers (usually troponin).
Management — Patients with evidence of recurrent ischemia or infarction should be rapidly managed with angiography followed by PCI (or coronary artery bypass graft surgery [CABG] if appropriate), if possible. If urgent angiography is not available, repeat fibrinolysis is a reasonable alternative if ST-segment elevation on an ECG is present.
We prefer angiography and PCI in these patients, assuming that catheterization can be performed within two hours. If catheterization will be delayed, repeat fibrinolysis with tenecteplase, alteplase, or reteplase may be preferred in patients with ST elevation. Patients receiving repeat fibrinolytic therapy should be observed for stability as other medical therapies are continued or augmented.
PCI — PCI is the treatment of choice for patients with threatened reocclusion or recurrent MI. The magnitude of the benefit was illustrated in a retrospective review of 20,100 patients from trials of fibrinolytic therapy [31]. Among the 4.2 percent with a recurrent MI, the in-hospital mortality was 23.6 percent in those who were treated medically and did not undergo revascularization compared with 5.2 percent with PCI; an equivalent benefit was seen with the smaller number of patients who underwent CABG.
A similar dramatic reduction in mortality with revascularization was noted in a review from two fibrinolytic trials [33]. Among the 4 percent of patients with reinfarction, 30-day mortality was significantly lower in patients treated with revascularization compared with medical therapy (11 versus 28 percent). The revascularization procedure was PCI or surgery (performed in 35 percent in United States centers and 15 percent in centers in other countries) or repeat thrombolysis (performed in 28 percent in United States centers and 47 percent in centers in other countries). Thirty-day mortality (11 percent) was not affected by the type of revascularization.
Repeat fibrinolytic therapy — The efficacy of retreatment with a fibrinolytic agent has been evaluated primarily in patients with threatened reocclusion and ST-segment elevation on an ECG. Most retreatment studies have involved alteplase and, to a lesser degree, other nonantigenic agents such as tenecteplase; streptokinase should be avoided because of antigenicity and relative resistance. After repeat fibrinolysis, patients should be referred to a facility with PCI capability.
Choice of agent — In patients who will be retreated with fibrinolytic therapy, we use alteplase, tenecteplase, or reteplase. Alteplase is the best studied of the three. Retreatment with streptokinase or anistreplase should be avoided due to the potential for neutralizing antibodies, which might render such therapy ineffective [38,39].
There are two major concerns with repeat fibrinolysis:
●Excess bleeding risk, particularly with early reuse
●Decreased efficacy, in view of initial treatment failure
A number of reports have demonstrated the feasibility and efficacy of repeat fibrinolytic therapy with alteplase [40-43]. As an example, a retrospective study assessed the outcome of repeat infusions of alteplase in 52 patients with an acute MI who developed early recurrent myocardial ischemia with threatened reinfarction after initial therapy with either alteplase (46 patients) or streptokinase (six patients) [42]. Retreatment was given within one hour of the first infusion in 67 percent and within 72 hours in 92 percent. The following findings were noted:
●Complete resolution of acute ischemia within one hour of the second infusion was achieved in 85 percent; one-half of these patients had a sustained response and avoided further coronary intervention.
●Bleeding complications occurred in 19 percent, but only 4 percent required transfusion.
●Fibrinogen levels fell by 25 percent and plasminogen levels by 63 percent; these changes were only slightly greater than those in patients who had received only one alteplase infusion and were not additive to those observed after the first infusion.
A later study evaluated the effects of retreatment with alteplase given for early signs of reocclusion after fibrinolysis, as manifested by recurrent chest pain lasting for more than 30 minutes with re-elevation of the ST segment [43]. This complication occurred in 26 of 652 patients (4 percent) treated with alteplase, usually within 24 hours of the first infusion, particularly if intravenous heparin was not used. The second dose of alteplase was 50 mg when threatened reocclusion occurred within 24 hours of initial therapy and 100 mg after 24 hours. The following results were noted:
●All patients had a good clinical response to retreatment; the new ST-segment changes disappeared and pain resolved within 100 minutes (median 50 minutes).
●Angiographic findings, determined at one hour, were less favorable. Coronary artery patency was observed in 73 percent of patients who had received intravenous heparin and only 40 percent of those not receiving heparin.
●Bleeding rates were similar to those after a single dose of alteplase.
Randomized trials have not been performed to assess the comparative efficacy of repeat fibrinolysis and PCI in patients with recurrent ischemia or threatened reinfarction. This issue was assessed in a retrospective review of the 4 percent of patients (n = 2301) who developed early reinfarction after initial fibrinolysis in the GUSTO-I (alteplase) and ASSENT 2 (tenecteplase) trials [33]. Repeat fibrinolysis (performed in 28 percent in United States centers and 47 percent in centers in other countries) and revascularization with PCI or surgery (performed in 35 percent in United States centers and 15 percent in centers in other countries) were equally effective and were associated with a significantly lower 30-day mortality than conservative therapy (11 and 11 versus 28 percent) without any increase in the rate of stroke.
Thus, repeat fibrinolytic infusions are feasible, can stabilize a substantial number of patients with threatened early infarction, and, even when relief of ischemia is temporary, can reduce myocardial damage in those proceeding to mechanical or surgical revascularization. However, there is limited information regarding the safety of PCI or surgery in the early hours after a second course of therapy with fibrinolytic drugs.
Reinfarction in a new territory — Studies in patients with an acute coronary syndrome have demonstrated multiple unstable plaques within the coronary circulation. Thus, reinfarction might occur in a new territory rather than reflecting failed fibrinolysis in the index territory.
This issue was addressed in a report from the HERO-2 trial of bivalirudin versus unfractionated heparin prior to streptokinase administration [44]. Confirmed reinfarction occurred in 552 patients (3.2 percent). Among these patients, 67 percent had ST elevation in the index territory, 18 percent had no new ECG changes, 2 percent had new bundle branch block, and 4 percent (0.15 percent of all patients) had ST elevation in a new territory at a mean of 46 hours.
For patients with reinfarction in a new territory, which is usually diagnosed at the time of coronary angiography, we usually perform PCI of the artery supplying this new territory unless the coronary anatomy requires CABG.
PATIENTS REQUIRING CABG — Urgent coronary artery bypass graft surgery (CABG) is needed infrequently after fibrinolysis. Percutaneous coronary intervention (PCI) is preferred in this setting because of increased risks of perioperative mortality and major hemorrhage with CABG, especially if the interval between administration of the fibrinolytic agent and surgery is short. In an analysis from the TIMI II study, 390 patients underwent CABG after fibrinolytic treatment [45]. Patients undergoing surgery within 24 hours of study, compared with those undergoing later surgery, had higher rates of perioperative mortality (17 versus 4 percent) and major hemorrhage (74 versus 51 percent).
The risk of bleeding for patients undergoing CABG after fibrinolysis may vary with the specific fibrinolytic agent. Tenecteplase is associated with significantly less fibrinogen depletion than either alteplase or reteplase [46]. As a result, patients treated with tenecteplase may be able to undergo CABG within 24 hours of fibrinolysis more safely than patients treated with the other agents, although this has not been specifically demonstrated in clinical studies.
While early morbidity and mortality are of concern in this setting, long-term outcome is much better and suggests that surgery may be appropriate for carefully selected patients [31,45].
RECOMMENDATIONS OF OTHERS — Our recommendations are generally in agreement with those of other major societal organizations [15,16,47].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: ST-elevation myocardial infarction (STEMI)".)
SUMMARY AND RECOMMENDATIONS
●Definitions – Among patients treated with fibrinolytic drugs for ST-elevation myocardial infarction (STEMI), there are two types of failed therapy: primary failure and threatened reocclusion.
•Primary failure – Primary failure of fibrinolysis occurs when there is evidence of continuing or worsening myocardial ischemia after fibrinolysis. (See 'Primary failure' above.)
•Threatened occlusion – Threatened reocclusion is characterized by the development of early recurrent ischemia after apparently successful fibrinolysis. (See 'Threatened reocclusion or reinfarction' above.)
●Management
•Primary failure of fibrinolysis – For patients with evidence of primary failure of fibrinolysis, we recommend urgent rescue percutaneous coronary intervention (PCI) as opposed to either repeat fibrinolysis or conservative care (Grade 1B). PCI should be carried out as soon as the diagnosis of primary failure of fibrinolysis is confirmed. If rescue PCI cannot be performed within 12 hours, we suggest conservative care as opposed to repeat fibrinolysis (Grade 2B). (See 'Rescue PCI' above.)
•Threatened occlusion – For patients with evidence of threatened reocclusion after fibrinolysis, we recommend urgent PCI as opposed to either repeat fibrinolysis or conservative care (Grade 1B). If PCI cannot be performed within two hours, we suggest repeat fibrinolysis with a nonantigenic agent as opposed to conservative care (Grade 2B). (See 'Threatened reocclusion or reinfarction' above.)
Do you want to add Medilib to your home screen?