- Email: [email protected]
Facilitated percutaneous coronary intervention (PCI) in patients with acute ST-elevation myocardial infarction: Comparison of prehospital tirofiban versus fibrinolysis before direct PCI
International Journal of Cardiology 103 (2005) 193 – 200 www.elsevier.com/locate/ijcard
Facilitated percutaneous coronary intervention (PCI) in patients with acute ST-elevation myocardial infarction: Comparison of prehospital tirofiban versus fibrinolysis before direct PCI Martin Mfckela,T, Wolfgang Bockscha, Sebastian Strohma, York Kqhnlea, Jfrn Vollerta, Lutz Nibbeb, Rainer Dietza a
Department of Cardiology, Charite´-University Medicine Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany b Department of Nephrology/Intensive Care Medicine, Charite´-University Medicine Berlin, Campus Virchow-Klinikum, Germany Received 18 May 2004; received in revised form 11 August 2004; accepted 4 October 2004 Available online 11 March 2005
Abstract Aims: Early start of treatment including coronary revascularization has been recognized as crucial variable in the outcome of acute STsegment elevation myocardial infarction (STEMI). The lack of availability and the realisation that an optimum reperfusion strategy will need to incorporate mechanical reperfusion as part of that strategy has led to a great deal of interest in pharmacologic reperfusion combined with mechanical reperfusion or facilitated PCI. It is not clear whether GPIIb/IIIa-blockade or fibrinolysis better facilitates PCI. Methods: We identified 138 patients who have been primarily treated by our mobile emergency care mobile from July 2001 until February 2003 with tirofiban or fibrinolysis. Seventy-nine patients had ST-elevation myocardial infarction (STEMI) and available angiograms within 24 h. Results: Forty-four patients had tirofiban (TIRO; 60.6 S.D. 11.4 years, 64% male) and 35 patients underwent fibrinolysis (FIB; 31.4% tenecteplase, 54.3% reteplase, 11.4% alteplase, 2.9% streptokinase; 58.8 S.D. 12.2 years, 80% male). Data were analyzed with respect to TIMI-flow and corrected frame count (cTFC) before and after PCI, bleeding complications at 30 days and long-term follow up for major adverse events (median 288 days; MACE: Death, hospitalized re-infarction, intracranial hemorrhage). Catheter films were re-analyzed by an investigator blinded to the prehospital therapy. Time from onset of symptoms to first medical contact was 1.98 h in TIRO compared to 0.5 h in FIB ( pb0.001) and time from first prehospital medical contact to catheter was 1.46 h in the TIRO compared to 2.85 h in the FIB group ( pb0.001). TIMI 3-flow before PCI was observed in 20.5% of TIRO and 62.9% in FIB ( pb0.001). After PCI TIMI 3-flow was achieved in 90.5% and 90.0%, respectively ( p=n.s.). Final cTFC was 24 in TIRO and 29 in FIB ( p=n.s.). Visible thrombi were detected in 30.2% in TIRO and 23.5% in FIB ( p=n.s.). Major bleeding occurred in one TIRO patient (fatal lung bleeding after ultima ratio abciximab on top of tirofiban), 2 patients (4.5%) received transfusions. In FIB 2 intracerebral hemorrhages, 5 transfusions (14.3%) and 3 pulmonary bleedings during mandatory ventilation were observed. After 30 days 4.5% in TIRO and 22.9% in FIB had MACE ( p=0.015). During long-term follow up the primary endpoint was observed in 4.5% of TIRO and 28.6% ( p=0.003) of FIB. Two patients died in TIRO and 9 patients in FIB. Conclusions: We conclude that (1) prehospital start of tirofiban for facilitated PCI is safe and effective if administered by experienced emergency physicians; (2) routine fibrinolysis should be limited to areas where catheter based therapy is not available within 90 min and (3) fibrinolysis should be given for facilitated PCI in randomized trials only at the moment. D 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: ST-elevation myocardial infarction; Facilitated direct PCI; Tirofiban; Fibrinolysis
T Corresponding author. Tel.: +49 30 450 553203; fax: +49 30 450 553927. E-mail address: [email protected] (M. Mfckel). 0167-5273/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.10.015
194
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
1. Introduction
2. Patients and methods
Early start of treatment including coronary revascularization has been recognized as crucial variable in the outcome of acute ST-segment elevation myocardial infarction (STEMI). Primary angioplasty has been recommended by actual guidelines [1] as preferred reperfusion strategy in the treatment of STEMI if reperfusion can be achieved within 90 min (first medical contact to balloon inflation). These new recommendations base on recent studies which could show advantages of mechanical reperfusion over fibrinolytic therapy even if transport of the patient to a catheter facility is necessary (DANAMI-2 [2], PRAGUE-2 [3]). The lack of availability and the realisation that an optimum reperfusion strategy will need to incorporate mechanical reperfusion as part of that strategy has led to a great deal of interest in pharmacologic reperfusion combined with mechanical reperfusion or facilitated PCI. The latter means the use of pharmacologic reperfusion to establish TIMI grade 3 flow as soon as possible followed by immediate PCI to maximize TIMI 3 flow and to stabilize the ruptured plaque. It has been shown from the Primary Angioplasty in Myocardial Infarction (PAMI) investigators [4] that patients undergoing primary PCI for AMI who arrive at the catheterization laboratory with an open versus a closed infarct artery have higher procedural success rates, smaller infarct size, better recovery of left ventricular function, and lower early and late mortality. On the other hand it has been proposed earlier that time to reperfusion seems to be less important with primary PCI compared with fibrinolytic treatment [5] although time to balloon should not exceed 2 h [6]. Additionally, in GUSTO V it has been shown that bleeding risk was increased with the combination of half dose fibrinolytics and abciximab and has the same risk of intracranial hemorrhage as standard fibrinolytic therapy [7]. From these two points it seems to be attractive to facilitate PCI by using early administration of GP IIb/IIIa-blockers alone without fibrinolytic agents. In the placebo controlled ADMIRAL-study [8] it could be shown that early administration of the GP IIb/ IIIa-inhibitor abciximab improved the pre PCI TIMI 3 flow rate of the infarct vessel (16.8% versus 5.4%) and reduced the incidence of the combined endpoint of death, reinfarction and urgent target vessel revascularization. In the TIGER-PA pilot trial [9] randomized early versus periprocedural administration of tirofiban in STEMI patients resulted in improved pre PCI TIMI grade flow (32% versus 10%) but the same final procedural results (TIMI grade flow, TIMI frame count and TIMI myocardial blush grade). We report on our experience with the prehospital administration of the GP IIb/IIIA-inhibitor tirofiban and direct PCI compared to a control group of patients from the same time period and setting who underwent prehospital fibrinolysis and early coronary angiography.
In summer 2001 we established a recommendation for the treatment of STEMI by prehospital tirofiban plus direct PCI on our mobile intensive care unit. As emergency care physicians in our institution are not bound to internal cardiological standards, some preferred early prehospital fibrinolysis in patients with STEMI. We identified 138 patients who have been primarily treated by our mobile emergency care mobile from July 2001 until February 2003 with tirofiban or fibrinolysis. Fig. 1 shows the patient selection of primary care patients with STEMI and early coronary angiography and intervention. The institutional review board approved the collection and analysis of data. STEMI was defined as (1) documented clinical symptoms of acute myocardial infarction (AMI) with the initial onset of chest pain or equivalent symptoms within the past 12 h; (2) ST-elevations of z0.1 mV in 2 or more contiguous leads or documented new left bundle-branch block (LBB). As shown in Fig. 1 no patients were excluded from the analysis. The dose of tirofiban was recommended according to the Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis (RESTORE) protocol (10 Ag/kg as bolus, followed by 0.15 Ag*kg 1*min 1*24 h). Most recently Danzi et al. published that a bolus of 25 Ag/kg is safe. These data were not available during the study [10]. Unfractionated heparin was given as initial bolus of 70 U/ kg followed by 5 U/kg/h. All patients received intravenous aspirin and metoprolol if applicable. Other medications, including nitrates and morphine, were administered on the discretion of the attending emergency care physicians. The 78 patients with initial fibrinolysis received lytic agents at standard doses recommended by the manufactures on the discretion of the attending emergency physicians including tenecteplase, reteplase, alteplase and streptokinase; in the 35 patients finally compared to patients receiving tirofiban, 31.4% had tenecteplase, 54.3% reteplase, 11.4% alteplase and 2.9% (one patient) got streptokinase. Diagnostic angiography was performed in all patients before planned revascularization; Fig. 1 shows the final selection of the two compared groups with STEMI and availability of an early coronary angiography. Revascularization was performed on the discretion of the attending experienced interventional cardiologist. All patients underwent coronary stenting, no drug eluting stents were used in these patients. All patients received a loading dose of 300 mg clopidogrel after the procedure and at least 28 days therapy with ASS 100 mg and clopidogrel 75 mg once daily was mandated. All patients compared received vascular closure devices. Laboratory screening included peak creatine phosphokinase (CPK) and time to peak CPK. Important recorded time intervals included time of first medical contact and time of angiography.
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
195
All 138 patients with prehospital Tirofiban or fibrinolysis in AMI July 2001 - February 2003 61.1 SD 12.9 years, 75% male
Tirofiban n = 60 60.5 SD 11.8 years, 72% male
Lysis n = 78 61.8 SD 13.8 years, 77% male
Patients with NSTEMI n = 13 No death on scene
Patients with STEMI or LBB n = 47 60.7 SD 11.1 years, 66.0% male
Ultima ratio fibrinolysis n = 24 (20/24 [83%] died on scene)
Patients with STEMI or LBB n = 54 61.1 SD 13.5 years, 70.4% male
No cardiac catheter available < 24h n=3 (2 patients refused, n = 1 > 24 h)
Patients with STEMI Cardiac catheter done < 24h n = 44 60.6 SD 11.4 years, 64.0% male
No cardiac catheter < 24h available n = 19 (8 patients had no catheter during index hospital stay, n = 11 > 24h)
Patients with STEMI Cardiac catheter done < 24h n = 35 58.8 SD 12.2 years, 80.0% male
Fig. 1. Patient selection flow chart. (N)STEMI, (non)ST-elevation myocardial infarction; LBB, left bundle branch block.
Bleeding complications, need of transfusion and local complications at the puncture site were recorded. Definition of minor and major bleeding was done according to TIMIgroup criteria [11]. Post-hospital follow-up included at least one telephone conversation and if applicable the review of hospital charts of further admissions. The follow up was started at 6 months after the index event. Median follow up time was 288 (116/467) days for the whole study population. The primary clinical endpoint was the combination of death of all causes, non-fatal myocardial reinfarction and intracranial hemorrhage (ICH). Further endpoints included the components of the primary endpoint, further PCI, coronary artery bypass grafting (CABG) and rehospitalization. Angiographic analysis was done by an investigator who was blinded to the pharmacologic treatment of the patient and not involved in the revascularization procedure. Initial and final TIMI grade flow, corrected TIMI frame count [12], visible thrombi (persisting contrast filling defects), no reflow phenomenon and peripheral embolization (peripheral coronary vessel break off) were recorded. As the original description of the TIMI frame count based on frame rates of 30/s, we calculated comparable values from our frame counts, which base on digital recordings with 12.5 frames/s, by multiplying the results by a constant of 2.4. Occluded vessels (TIMI 0) were not analysed by frame counts. Statistical analysis was performed by use statistical software (SPSSR V 11.0) using a Chi square analysis or Wilcoxon test for unpaired samples comparing the two
treatment groups. Accordingly data are presented as median and (25%-/75%-) percentiles. A result was considered significant if pb0.05. Non-significant p-values below 0.1 are listed. A multiple logistic regression analysis was performed to identify predictors of the choice of therapy including the attending emergency physician as control variable.
3. Results A total of 138 patients were identified in the period of investigation as shown in Fig. 1 and mentioned above. Finally n=44 patients with tirofiban and n=35 patients with fibrinolytic therapy were compared. Patient demographics and presentation are listed in Table 1. No patients were excluded from the analysis. In the fibrinolysis group slightly more patients were smokers (71.4% versus 63.6%, p=0.012) and more patients had previous PCI (37.1% versus 9.1%, p=0.002). The time from first medical contact to PCI was significantly longer in the fibrinolysis group (2.85 versus 1.46 h, pb0.001) but time from onset of symptoms to first medical contact was shorter (0.5 versus 1.98 h, pb0.001). More patients in the fibrinolysis group required prehospital resuscitation (34.3% versus 4.5%, p=0.001). Angiographic analysis (Table 2) showed a significant difference in initial TIMI grade flow (Fig. 2) but not in corrected TIMI frame count of not occluded vessels. The percentage of patients with initial TIMI grade 3 flow before angioplasty was 62.9% in the fibrinolysis group compared
196
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
Initial TIMI-Grade Flow
Table 1 Patient demographics and presentation Fibrinolysis
p
60.6F11.4 63.6 52.3 56.8 63.6 22.8 27.3 22.7 9.1 11.4
58.8F12.2 80.0 54.3 45.7 71.4 22.9 40.0 28.6 37.1 2.9
n.s. n.s. n.s. n.s. 0.012 n.s. n.s. n.s. 0.002 n.s.
30.2 32.6 37.2 1.96 (0.5/5.4)
28.6 54.3 17.1 0.5 (0.25/1.0)
n.s.
b0.001
1.46 (0.99/2.01) 2.85 (1.60/4.90) b0.001 46 (42/63) 6.8 4.5
56 (43/64) 2.9 34.3
n.s. n.s. 0.001
Data expressed as meanFS.D., percent or median (25%/75%)-percentiles; n.s. indicates not statistically significant; *, left ventricular angiography was performed in 26/35 patients of the lysis group and 39/44 patients in the tirofiban group; LVEF, left ventricular ejection fraction; CAD, coronary artery disease; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting.
to 20.9% in the tirofiban group ( pb0.001; Fig. 2). Corrected initial TIMI frame counts were not significantly lower in the fibrinolysis group (47 (32/67) versus 55 (35/120), p=n.s., Table 2). Ninety percent of patients in both groups had postprocedural TIMI grade 3 flow. This was also reflected by the
Table 2 Procedural findings Tirofiban Culprit vessel [%] LAD LCX RCA Initial TGF [%] 3 2 1 0 Initial cTFC Visible thrombi [%] No reflow [%] Peripheral emboli [%] Final TGF-3 [%] Final cTFC
Fibrinolysis
p n.s.
38.6 9.1 45.5
42.9 14.3 42.9
20.5 9.1 22.7 47.7 55 (35/120) 30.2 4.7 20.9 90.5 24 (17/35)
62.9 14.3 11.4 11.4 47 (32/67) 23.5 8.8 8.8 90.0 29 (19/41)
20,5 80% 60%
9,1
TIMI III TIMI II TIMI I TIMI 0
62,9
22,7
40% 20%
47,7
0% Tirofiban
14,3 11,4 11,4 Fibrinolysis
Fig. 2. Initial TIMI grade flow in both treatment groups.
corrected TIMI frame counts (24 (17/35) and 29 (19/41), respectively; p=n.s., Table 2 and Fig. 3). The peak CPK was measured in all patients while in the hospital (Table 3). The peak CPK was not different in both groups. The mean time to peak CPK was significantly lower in the tirofiban group (9.7 (7.3/14.5) h) than in the fibrinolyzed group (12.9 (9.2/23.1) h; p=0.036). No differences were noted between the two groups with regard to bleeding complications (Table 3). Minor bleeding appeared to be more frequent in the fibrinolysed group (8.6%) but the difference was only borderline statistically significant ( p=0.051). The outcomes at 9 month follow up clearly showed a higher event rate in the fibrinolysed group (28.6% versus 4.5%, p=0.003) due to more deaths (25.7% versus 4.5%, p=0.006) and intracerebral hemorrhage (5.7 versus none) in the fibrinolysed group (Table 3). One fatal bleeding occurred in one tirofiban patient (lung bleeding after ultima ratio abciximab on top of tirofiban after massive occurrence of coronary thrombi). There were very few local complications at puncture site and no relevant local bleeding complications occurred which is maybe due to the use of vascular closure devices in all patients (Table 3). No significant differences were seen with respect to reinfarction and further PCI (Table 3). In the fibrinolysed group significantly more patients were rehospitalized (31.4% versus 6.8%, p=0.001) for various causes. Logistic regression resulted in significant odds ratios (OR) for previous PCI (5.0, p=0.023) and delay of the Final Corrected TIMI Frame Count
b0.001 60
n.s. n.s. n.s. n.s. n.s. n.s.
Data expressed as meanFS.D., percent or median (25%/75%)-percentiles; n.s. indicates not statistically significant; TGF, TIMI-grade flow; cTFC, corrected TIMI-frame count.
cTFC
Age [years] Gender, [%] male Hypertension [%] Hyperlipidemia [%] Smoking [%] Diabetes [%] Previous CAD [%] Previous MI [%] Previous PCI [%] Previous CABG [%] Severity of CAD [%] Single-vessel disease Double-vessel disease Triple-vessel disease Symptom onset to first medical contact [h] First medical contact to PCI [h] Initial angiographic LVEF [%]* Shock [%] Prehospital resuscitation [%]
100%
Tirofiban
40
20
0 Tirofiban
Fibrinolysis
Fig. 3. Final corrected TIMI frame count (cTFC). Median and (25%/75%)percentiles. Differences are not significant.
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200 Table 3 Complications and outcomes Tirofiban Bleeding [%] Minor# Major# Loss## Transfusions [%] Local complications at puncture site### [%] Outcomes at 30 days follow Peak CPK [U/L] Time to peak CPK [h] Composite [%] Death Re-infarction ICH PCI [%] CABG (within 24 h) [%] CABG (after 24 h) [%]
2.3 0 2.3 2.3 4.5 2.3
up 557 (200/1483) 9.7 (7.3/14.5) 4.5 4.5 0 0 2.3 0 0
Cumulative outcomes at long-term follow up Follow up time [days] 240 (103/359) Composite [%] 4.5 Death 4.5 Re-infarction 0 ICH [%] 0 PCI [%] 22.8 CABG [%] 2.3 Composite-2* [%] 27.3 Rehospitalization [%] 6.8
Fibrinolysis
p
11.4 8.6 2.9 0 14.3 2.9
n.s. 0.051 n.s. n.s. n.s. n.s.
639 (161/2087) 12.9 (9.2/23.1) 22.9 20.9 2.9 5.7 0 2.9 2.9
n.s. 0.036 0.015 0.028 n.s. n.s. n.s. n.s. n.s.
379 (105/471) 28.6 25.7 2.9 5.7 14.3 8.6 40.0 31.4
n.s. 0.003 0.006 n.s. n.s. n.s. n.s. n.s. 0.001
# As defined in the methods section according to the TIMI-criteria; ##Loss denotes a decrease in hemoglobin of more than 4 g/dL but not exceeding 5 g/dL with no identified bleeding site. ###Aneurysma spurium and/or a-vfistula. Severe local bleeding/hematoma were not observed. PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; CPK, creatinphosphokinase. *Composite-2 denotes the second composite endpoint containing death, reinfarction, PCI or CABG for comparison with the MITRA-registry [19].
patient below 3 h (5.1, p=0.036) for the choice of prehospital fibrinolysis instead of tirofiban and direct PCI.
4. Discussion Our study presents real life data on the prehospital application of the GP IIb/IIIa-inhibitor tirofiban in patients with ST-elevation myocardial infarction and intended percutaneous revascularization. Patients are compared to a group with prehospital fibrinolysis and early angiography from the same emergency system and the same time period. Mechanical reperfusion has gained increasing acceptance as the preferred reperfusion strategy for acute myocardial infarction. Most recently this has been included in actual guidelines for treatment of ST-elevation acute myocardial infarction (STEMI) [1]. The superiority of percutaneous coronary intervention (PCI) compared to fibrinolysis is due to more than 90% open arteries with TIMI III flow after the procedure while fibrinolysis with
197
modern agents as tenecteplase reaches only around 60% (e.g. 62.8% in TIMI 10b [13]). The rationale behind facilitated PCI has been outlined above. In our study a full dose fibrinolyis followed by early angiography and PCI was compared to regular dose tirofiban and PCI. Up to now there are no randomized trials published with clinical endpoints comparing both strategies. The angiographic findings of this real-life observational study suggest that prehospital administration of tirofiban in the emergency care mobile improves initial TIMI III flow grade. Compared to the TIGER-PA study [9] TIMI III flow rate in our study was in between the early and late group of that study (20.5% versus 32%/10%) and nearly matches the presence of 19% TIMI 3 flow with early tirofiban in the larger On-TIME study [14]. Additionally our results fit very well in previous data achieved with abciximab (ADMIRAL, 16.8% TIMI III flow) and are clearly higher than TIMI flow rates reported from the placebo group of that study (5.4%) [8]. Therefore, in real-life emergency care for STEMI patients, we achieved the same pre PCI TIMI III flow rate as in randomized studies. On the other hand there were significantly more open arteries in the fibrinolysed patients (62.9%), a result which fits very well in the previous published data [13]. Although facilitation of PCI by fibrinolysis leads to significantly more open arteries prior to stent placement the crucial questions are (1) whether more open arteries lead to a higher procedural success and (2) improved outcomes. Our study was designed to answer the first question and give some idea of the second. We found that procedural success was the same in both treatment groups. Especially final TIMI III flow and final TIMI frame count (Table 2, Fig. 3) were not different. These results are in line with others who described excellent procedural success with early pretreatment with tirofiban [9]. It is of interest that longer time to peak CPK was detected in fibrinolysed patients (Table 3). This underlines that the initial TIMI-3 flow is not necessarily identical with myocardial tissue reperfusion which is probably better assessed by myocardial blush grades [15]. It has been taken into account that (I) the vessels need a certain time to open under fibrinolysis and the net benefit compared to direct PCI is most probably short (II) that indeed due to the later catheter time in our lysis group less patients would have had TIMI 3 flow at an earlier time point. Nevertheless it can be proposed that every fibrinolysed patient should undergo immediate cardiac catheterization for rescue PCI which must not be delayed as it has happened in our real-life szenario. Immediate PCI after lysis has been shown earlier to be effective and safe if short acting fibrinolytics are used [16]. The outcomes data of our study are potentially biased as discussed in detail below and limited due to the small number of patients. Nevertheless, we observed two patients with intracranial hemorrhage (ICH) after fibrinolyis which is
198
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
the well known major complication of fibrinolytic treatment. This confirms even recent reports of fibrinolytic trials with 1–2% ICH [17,18] depending on the co-therapy. Dose reduction of the fibrinolytic agent and combination with a GP IIb/IIIa-inhibitor does not reduce ICH (e.g. GUSTO V [7]). In our study the secondary endpoint showed very good outcomes of patients with tirofiban and PCI but high event rates for patients with fibrinolysis. As it is known from the MITRA registry [19], mortality and event rates are substantially higher in patients outside of controlled randomized trials. Our 12 months mortality rate in the fibrinolysis group seems to be higher compared to those in the registry (25.7% versus 13.9% [19]). In the MITRA registry resuscitation was 4.9% in the fibrinolysis group compared to 34.3% in our study indicating high risk, which clearly explains the difference in mortality. On the other hand the composite 12 months endpoint of the MITRAregistry (see Table 3 for definition bcomposite-2Q) has similar results in the PCI group compared to our tirofibangroup (25.6% versus 27.3%; Table 3). The major component of this endpoint is further PCI (not only of the target vessel). The fibrinolyis group in MITRA has a 32.3% event rate compared to 40.0% in our high risk fibrinolysis group. This shows that the overall long-term risk of fibrinolysed patients in our study is comparable to previously published data. The high early mortality is due to a selection bias by including many patients with prehospital resuscitation. At many hospitals, primary PCI is the preferred reperfusion strategy for STEMI patients. According to actual guidelines and recent studies this can be recommended in general if time from first medical contact until balloon inflation by an experienced investigator is below 90 min [1]. Mortality seems to be excessively increased if doorto-balloon time exceeds 2 h [6]. This fits well with the DANAMI-2 study which shows that patients profit from primary PCI even if transfer is necessary. In this study the patients from referral hospitals had a median delay from randomization to start of treatment of 90 min [2]. The present study suggests that, in centers where primary PCI is the preferred reperfusion strategy, the prehospital administration of tirofiban by experienced emergency physicians is safe and effective. The study was underpowered and biased by selection to prove superiority of tirofiban in clinical endpoints. Larger randomized studies are necessary and justified to compare different measures to facilitate direct PCI for STEMI. 4.1. Limitations The present study presents prospective data on the safety and efficacy of the prehospital application of tirofiban before percutaneous intervention. The control group has been retrospectively taken from the same period of time and the same emergency system. All applicable patients in the inclusion period have been identified and analysed as displayed in Fig. 1. As the study was not randomized some
limitations has to be addressed with respect to the discussion of the results. (1) The tirofiban patients were all scheduled for PCI. Fibrinolysed patients are usually scheduled for coronary angiography but as reperfusion therapy is already started, the priority of angiography in these patients is sometimes judged lower in clinical routine. This explains the significantly longer time from first medical contact to PCI in the fibrinolysed group (Table 1). On the other hand the time from onset of symptoms to first medical contact was significantly shorter in the lysis group reflecting that the maximum benefit of fibrinolysis in the first 3 h was taken into account by the emergency physicians. With respect to the interpretation of the results this means that TIMI 3 flow is maybe overestimated in the fibrinolysed group compared to tirofiban patients who were earlier in the catheterization laboratory. (2) The major problem of non-randomized studies is the bias of selection. We aimed at minimizing this factor by analysing all patients of the inclusion period. Table 1 shows only few significant differences of clinical variables in the both groups, including delay of the patient (time from onset of symptoms to first medical contact, Table 1), smoking, previous PCI and prehospital resuscitation. Another potential confounder is the attending emergency physician who was responsible for the initial choice of therapy. We performed a multiple logistic regression analysis with respect to the choice of therapy including these critical variables. The analysis clearly resulted in a significant influence of previous PCI and delay below 3 h as variables in favor for prehospital fibrinolysis (OR 5.0, p=0.023, OR 5.1, p=0.036, respectively). Prehospital resuscitation was not longer significant when the variable delay was included in the model. The variable battending Table 4 Complications and outcomes of the fibrinolysed group with late or no PCI n=19 Bleeding [%] Minor Major Transfusions Cardiac catheter during index admission Local complications at puncture site [n] Outcomes at 9 month follow up Composite [%] Death Re-infarction Stroke PCI Rehospitalization Peak CPK [U/L] Time to peak CPK [h]
5.3 5.3 0 0 57.9 0 21.1 21.1 0 0 21.1 42.1 360 (110/860) 27.7 (10.1/46.5)
Results were not significantly different form those in group with early angiography (Tables 1–3).
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
emergency physicianQ did not significantly influence the choice of reperfusion therapy. With respect to the interpretation of the results this means that the emergency physicians maybe selected more patients with definite coronary artery disease who present early for prehospital fibrinolysis which has to be taken into account when outcome data are interpreted. (3) In the whole study population 19 patients with prehospital fibrinolysis had no coronary angiography within 24 h. One could ask if these patients are more or less severe ill and therefore could have biased the results. The Table 4 summarizes the complications and outcomes data of this group. It is clearly evident that this group is not significantly different from the fibrinolysis group with early angiography and therefore an additional bias is most probably excluded. (4) Finally it has to be pointed out that the comparison of the procedural findings was the primary goal of this study and with respect to the very similar patients characteristics (Table 1) it can be assumed that herein valid results could be obtained. The advantage of these real life data is that no selection bias due to exclusion criteria (e.g. all resuscitated patients would have not been included in a randomized study), non-consent of patients or missing inclusion due to organisational or administrative causes can occur which usually limits the applicability of randomized trials to daily practice.
5. Conclusions We conclude that (1) prehospital start of tirofiban for facilitated PCI is safe and effective if administered by experienced emergency physicians; (2) routine fibrinolysis should be limited to areas where catheter based therapy is not available within 90 min and (3) fibrinolysis should be given for facilitated PCI in randomized trials only at the moment.
Acknowledgement The preparation of the manuscript was supported by a grant from MSD. This work is part of the doctoral thesis of Lutz Nibbe.
References [1] Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KA, et al. Management of acute myocardial infarction in patients presenting with ST-segment elevation. The task force on the management of acute myocardial infarction of the European Society of Cardiology. Eur Heart J 2003;24(1):28 – 66. [2] Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H, Thayssen P, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med 2003;349(8):733 – 42.
199
[3] Widimsky P, Budesinsky T, Vorac D, Groch L, Zelizko M, Aschermann M, et al. Long distance transport for primary angioplasty vs immediate thrombolysis in acute myocardial infarction. Final results of the randomized national multicentre trial-PRAGUE-2. Eur Heart J 2003;24(1):94 – 104. [4] Stone GW, Cox D, Garcia E, Brodie BR, Morice MC, Griffin J, et al. Normal flow (TIMI-3) before mechanical reperfusion therapy is an independent determinant of survival in acute myocardial infarction: analysis from the primary angioplasty in myocardial infarction trials. Circulation 2001;104(6):636 – 41. [5] Brodie BR, Stuckey TD, Wall TC, Kissling G, Hansen CJ, Muncy DB, et al. Importance of time to reperfusion for 30-day and late survival and recovery of left ventricular function after primary angioplasty for acute myocardial infarction. J Am Coll Cardiol 1998;32(5):1312 – 9. [6] Cannon CP, Gibson CM, Lambrew CT, Shoultz DA, Levy D, French WJ, et al. Relationship of symptom-onset-to-balloon time and door-toballoon time with mortality in patients undergoing angioplasty for acute myocardial infarction. JAMA 2000;283(22):2941 – 7. [7] Topol EJ. Reperfusion therapy for acute myocardial infarction with fibrinolytic therapy or combination reduced fibrinolytic therapy and platelet glycoprotein IIb/IIIa inhibition: the GUSTO V randomised trial. Lancet 2001;357(9272):1905 – 14. [8] Montalescot G, Barragan P, Wittenberg O, Ecollan P, Elhadad S, Villain P, et al. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction. N Engl J Med 2001;344(25):1895 – 903. [9] Lee DP, Herity NA, Hiatt BL, Fearon WF, Rezaee M, Carter AJ, et al. Adjunctive platelet glycoprotein IIb/IIIa receptor inhibition with tirofiban before primary angioplasty improves angiographic outcomes: results of the TIrofiban Given in the Emergency Room before Primary Angioplasty (TIGER-PA) pilot trial. Circulation 2003;107(11):1497 – 501. [10] Danzi GB, Capuano C, Sesana M, Baglini R. Safety of a high bolus dose of tirofiban in patients undergoing coronary stent placement. Catheter Cardiovasc Interv 2004;61(2):179 – 84. [11] Bovill EG, Terrin ML, Stump DC, Berke AD, Frederick M, Collen D, et al. Hemorrhagic events during therapy with recombinant tissue-type plasminogen activator, heparin, and aspirin for acute myocardial infarction. Results of the Thrombolysis in Myocardial Infarction (TIMI), Phase II Trial. Ann Intern Med 1991;115(4):256 – 65. [12] Gibson CM, Cannon CP, Daley WL, Dodge Jr JT, Alexander B, Marble SJ, et al. TIMI frame count: a quantitative method of assessing coronary artery flow. Circulation 1996;93(5):879 – 88. [13] Cannon CP, Gibson CM, McCabe CH, Adgey AAJ, Schweiger MJ, Sequeira RF, et al. TNK-tissue plasminogen activator compared with front-loaded alteplase in acute myocardial infarction: results of the TIMI 10B trial. Circulation 1998;98(25):2805 – 14. [14] van’t Hof AWJ, Ernst N, de Boer MJ, de Winter R, Boersma E, Bunt T, et al. Facilitation of primary coronary angioplasty by early start of a glycoprotein 2b/3a inhibitor: results of the ongoing tirofiban in myocardial infarction evaluation (On-TIME) trial. Eur Heart J 2004;25(10):837 – 46. [15] Gibson CM, Cannon CP, Murphy SA, Marble SJ, Barron HV, Braunwald E, et al. Relationship of the TIMI myocardial perfusion grades, flow grades, frame count, and percutaneous coronary intervention to long-term outcomes after thrombolytic administration in acute myocardial infarction. Circulation 2002;105(16):1909 – 13. [16] Ross AM, Coyne KS, Reiner JS, Greenhouse SW, Fink C, Frey A, et al. A randomized trial comparing primary angioplasty with a strategy of short-acting thrombolysis and immediate planned rescue angioplasty in acute myocardial infarction: the PACT trial. J Am Coll Cardiol 1999;34(7):1954 – 62. [17] Wallentin L, Goldstein P, Armstrong PW, Granger CB, Adgey AAJ, Arntz HR, et al. Efficacy and safety of tenecteplase in combination with the low-molecular-weight heparin enoxaparin or unfractionated heparin in the prehospital setting: the Assessment of the Safety and
200
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
Efficacy of a New Thrombolytic Regimen (ASSENT)-3 PLUS randomized trial in acute myocardial infarction. Circulation 2003;108(2):135 – 42. [18] ASSENT-2 Investigators. Single-bolus tenecteplase compared with front-loaded alteplase in acute myocardial infarction: the ASSENT-2 double-blind randomised trial. Lancet 1999;354(9180):716 – 22.
[19] Zahn R, Schiele R, Schneider S, Gitt AK, Heer T, Wienbergen H, et al. Long-term follow-up of patients with acute myocardial infarction treated with primary angioplasty or thrombolysis. Results of the MITRA trial. Z Kardiol 2002;91(1):49 – 57.
Facilitated percutaneous coronary intervention (PCI) in patients with acute ST-elevation myocardial infarction: Comparison of prehospital tirofiban versus fibrinolysis before direct PCI Martin Mfckela,T, Wolfgang Bockscha, Sebastian Strohma, York Kqhnlea, Jfrn Vollerta, Lutz Nibbeb, Rainer Dietza a
Department of Cardiology, Charite´-University Medicine Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany b Department of Nephrology/Intensive Care Medicine, Charite´-University Medicine Berlin, Campus Virchow-Klinikum, Germany Received 18 May 2004; received in revised form 11 August 2004; accepted 4 October 2004 Available online 11 March 2005
Abstract Aims: Early start of treatment including coronary revascularization has been recognized as crucial variable in the outcome of acute STsegment elevation myocardial infarction (STEMI). The lack of availability and the realisation that an optimum reperfusion strategy will need to incorporate mechanical reperfusion as part of that strategy has led to a great deal of interest in pharmacologic reperfusion combined with mechanical reperfusion or facilitated PCI. It is not clear whether GPIIb/IIIa-blockade or fibrinolysis better facilitates PCI. Methods: We identified 138 patients who have been primarily treated by our mobile emergency care mobile from July 2001 until February 2003 with tirofiban or fibrinolysis. Seventy-nine patients had ST-elevation myocardial infarction (STEMI) and available angiograms within 24 h. Results: Forty-four patients had tirofiban (TIRO; 60.6 S.D. 11.4 years, 64% male) and 35 patients underwent fibrinolysis (FIB; 31.4% tenecteplase, 54.3% reteplase, 11.4% alteplase, 2.9% streptokinase; 58.8 S.D. 12.2 years, 80% male). Data were analyzed with respect to TIMI-flow and corrected frame count (cTFC) before and after PCI, bleeding complications at 30 days and long-term follow up for major adverse events (median 288 days; MACE: Death, hospitalized re-infarction, intracranial hemorrhage). Catheter films were re-analyzed by an investigator blinded to the prehospital therapy. Time from onset of symptoms to first medical contact was 1.98 h in TIRO compared to 0.5 h in FIB ( pb0.001) and time from first prehospital medical contact to catheter was 1.46 h in the TIRO compared to 2.85 h in the FIB group ( pb0.001). TIMI 3-flow before PCI was observed in 20.5% of TIRO and 62.9% in FIB ( pb0.001). After PCI TIMI 3-flow was achieved in 90.5% and 90.0%, respectively ( p=n.s.). Final cTFC was 24 in TIRO and 29 in FIB ( p=n.s.). Visible thrombi were detected in 30.2% in TIRO and 23.5% in FIB ( p=n.s.). Major bleeding occurred in one TIRO patient (fatal lung bleeding after ultima ratio abciximab on top of tirofiban), 2 patients (4.5%) received transfusions. In FIB 2 intracerebral hemorrhages, 5 transfusions (14.3%) and 3 pulmonary bleedings during mandatory ventilation were observed. After 30 days 4.5% in TIRO and 22.9% in FIB had MACE ( p=0.015). During long-term follow up the primary endpoint was observed in 4.5% of TIRO and 28.6% ( p=0.003) of FIB. Two patients died in TIRO and 9 patients in FIB. Conclusions: We conclude that (1) prehospital start of tirofiban for facilitated PCI is safe and effective if administered by experienced emergency physicians; (2) routine fibrinolysis should be limited to areas where catheter based therapy is not available within 90 min and (3) fibrinolysis should be given for facilitated PCI in randomized trials only at the moment. D 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: ST-elevation myocardial infarction; Facilitated direct PCI; Tirofiban; Fibrinolysis
T Corresponding author. Tel.: +49 30 450 553203; fax: +49 30 450 553927. E-mail address: [email protected] (M. Mfckel). 0167-5273/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.10.015
194
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
1. Introduction
2. Patients and methods
Early start of treatment including coronary revascularization has been recognized as crucial variable in the outcome of acute ST-segment elevation myocardial infarction (STEMI). Primary angioplasty has been recommended by actual guidelines [1] as preferred reperfusion strategy in the treatment of STEMI if reperfusion can be achieved within 90 min (first medical contact to balloon inflation). These new recommendations base on recent studies which could show advantages of mechanical reperfusion over fibrinolytic therapy even if transport of the patient to a catheter facility is necessary (DANAMI-2 [2], PRAGUE-2 [3]). The lack of availability and the realisation that an optimum reperfusion strategy will need to incorporate mechanical reperfusion as part of that strategy has led to a great deal of interest in pharmacologic reperfusion combined with mechanical reperfusion or facilitated PCI. The latter means the use of pharmacologic reperfusion to establish TIMI grade 3 flow as soon as possible followed by immediate PCI to maximize TIMI 3 flow and to stabilize the ruptured plaque. It has been shown from the Primary Angioplasty in Myocardial Infarction (PAMI) investigators [4] that patients undergoing primary PCI for AMI who arrive at the catheterization laboratory with an open versus a closed infarct artery have higher procedural success rates, smaller infarct size, better recovery of left ventricular function, and lower early and late mortality. On the other hand it has been proposed earlier that time to reperfusion seems to be less important with primary PCI compared with fibrinolytic treatment [5] although time to balloon should not exceed 2 h [6]. Additionally, in GUSTO V it has been shown that bleeding risk was increased with the combination of half dose fibrinolytics and abciximab and has the same risk of intracranial hemorrhage as standard fibrinolytic therapy [7]. From these two points it seems to be attractive to facilitate PCI by using early administration of GP IIb/IIIa-blockers alone without fibrinolytic agents. In the placebo controlled ADMIRAL-study [8] it could be shown that early administration of the GP IIb/ IIIa-inhibitor abciximab improved the pre PCI TIMI 3 flow rate of the infarct vessel (16.8% versus 5.4%) and reduced the incidence of the combined endpoint of death, reinfarction and urgent target vessel revascularization. In the TIGER-PA pilot trial [9] randomized early versus periprocedural administration of tirofiban in STEMI patients resulted in improved pre PCI TIMI grade flow (32% versus 10%) but the same final procedural results (TIMI grade flow, TIMI frame count and TIMI myocardial blush grade). We report on our experience with the prehospital administration of the GP IIb/IIIA-inhibitor tirofiban and direct PCI compared to a control group of patients from the same time period and setting who underwent prehospital fibrinolysis and early coronary angiography.
In summer 2001 we established a recommendation for the treatment of STEMI by prehospital tirofiban plus direct PCI on our mobile intensive care unit. As emergency care physicians in our institution are not bound to internal cardiological standards, some preferred early prehospital fibrinolysis in patients with STEMI. We identified 138 patients who have been primarily treated by our mobile emergency care mobile from July 2001 until February 2003 with tirofiban or fibrinolysis. Fig. 1 shows the patient selection of primary care patients with STEMI and early coronary angiography and intervention. The institutional review board approved the collection and analysis of data. STEMI was defined as (1) documented clinical symptoms of acute myocardial infarction (AMI) with the initial onset of chest pain or equivalent symptoms within the past 12 h; (2) ST-elevations of z0.1 mV in 2 or more contiguous leads or documented new left bundle-branch block (LBB). As shown in Fig. 1 no patients were excluded from the analysis. The dose of tirofiban was recommended according to the Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis (RESTORE) protocol (10 Ag/kg as bolus, followed by 0.15 Ag*kg 1*min 1*24 h). Most recently Danzi et al. published that a bolus of 25 Ag/kg is safe. These data were not available during the study [10]. Unfractionated heparin was given as initial bolus of 70 U/ kg followed by 5 U/kg/h. All patients received intravenous aspirin and metoprolol if applicable. Other medications, including nitrates and morphine, were administered on the discretion of the attending emergency care physicians. The 78 patients with initial fibrinolysis received lytic agents at standard doses recommended by the manufactures on the discretion of the attending emergency physicians including tenecteplase, reteplase, alteplase and streptokinase; in the 35 patients finally compared to patients receiving tirofiban, 31.4% had tenecteplase, 54.3% reteplase, 11.4% alteplase and 2.9% (one patient) got streptokinase. Diagnostic angiography was performed in all patients before planned revascularization; Fig. 1 shows the final selection of the two compared groups with STEMI and availability of an early coronary angiography. Revascularization was performed on the discretion of the attending experienced interventional cardiologist. All patients underwent coronary stenting, no drug eluting stents were used in these patients. All patients received a loading dose of 300 mg clopidogrel after the procedure and at least 28 days therapy with ASS 100 mg and clopidogrel 75 mg once daily was mandated. All patients compared received vascular closure devices. Laboratory screening included peak creatine phosphokinase (CPK) and time to peak CPK. Important recorded time intervals included time of first medical contact and time of angiography.
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
195
All 138 patients with prehospital Tirofiban or fibrinolysis in AMI July 2001 - February 2003 61.1 SD 12.9 years, 75% male
Tirofiban n = 60 60.5 SD 11.8 years, 72% male
Lysis n = 78 61.8 SD 13.8 years, 77% male
Patients with NSTEMI n = 13 No death on scene
Patients with STEMI or LBB n = 47 60.7 SD 11.1 years, 66.0% male
Ultima ratio fibrinolysis n = 24 (20/24 [83%] died on scene)
Patients with STEMI or LBB n = 54 61.1 SD 13.5 years, 70.4% male
No cardiac catheter available < 24h n=3 (2 patients refused, n = 1 > 24 h)
Patients with STEMI Cardiac catheter done < 24h n = 44 60.6 SD 11.4 years, 64.0% male
No cardiac catheter < 24h available n = 19 (8 patients had no catheter during index hospital stay, n = 11 > 24h)
Patients with STEMI Cardiac catheter done < 24h n = 35 58.8 SD 12.2 years, 80.0% male
Fig. 1. Patient selection flow chart. (N)STEMI, (non)ST-elevation myocardial infarction; LBB, left bundle branch block.
Bleeding complications, need of transfusion and local complications at the puncture site were recorded. Definition of minor and major bleeding was done according to TIMIgroup criteria [11]. Post-hospital follow-up included at least one telephone conversation and if applicable the review of hospital charts of further admissions. The follow up was started at 6 months after the index event. Median follow up time was 288 (116/467) days for the whole study population. The primary clinical endpoint was the combination of death of all causes, non-fatal myocardial reinfarction and intracranial hemorrhage (ICH). Further endpoints included the components of the primary endpoint, further PCI, coronary artery bypass grafting (CABG) and rehospitalization. Angiographic analysis was done by an investigator who was blinded to the pharmacologic treatment of the patient and not involved in the revascularization procedure. Initial and final TIMI grade flow, corrected TIMI frame count [12], visible thrombi (persisting contrast filling defects), no reflow phenomenon and peripheral embolization (peripheral coronary vessel break off) were recorded. As the original description of the TIMI frame count based on frame rates of 30/s, we calculated comparable values from our frame counts, which base on digital recordings with 12.5 frames/s, by multiplying the results by a constant of 2.4. Occluded vessels (TIMI 0) were not analysed by frame counts. Statistical analysis was performed by use statistical software (SPSSR V 11.0) using a Chi square analysis or Wilcoxon test for unpaired samples comparing the two
treatment groups. Accordingly data are presented as median and (25%-/75%-) percentiles. A result was considered significant if pb0.05. Non-significant p-values below 0.1 are listed. A multiple logistic regression analysis was performed to identify predictors of the choice of therapy including the attending emergency physician as control variable.
3. Results A total of 138 patients were identified in the period of investigation as shown in Fig. 1 and mentioned above. Finally n=44 patients with tirofiban and n=35 patients with fibrinolytic therapy were compared. Patient demographics and presentation are listed in Table 1. No patients were excluded from the analysis. In the fibrinolysis group slightly more patients were smokers (71.4% versus 63.6%, p=0.012) and more patients had previous PCI (37.1% versus 9.1%, p=0.002). The time from first medical contact to PCI was significantly longer in the fibrinolysis group (2.85 versus 1.46 h, pb0.001) but time from onset of symptoms to first medical contact was shorter (0.5 versus 1.98 h, pb0.001). More patients in the fibrinolysis group required prehospital resuscitation (34.3% versus 4.5%, p=0.001). Angiographic analysis (Table 2) showed a significant difference in initial TIMI grade flow (Fig. 2) but not in corrected TIMI frame count of not occluded vessels. The percentage of patients with initial TIMI grade 3 flow before angioplasty was 62.9% in the fibrinolysis group compared
196
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
Initial TIMI-Grade Flow
Table 1 Patient demographics and presentation Fibrinolysis
p
60.6F11.4 63.6 52.3 56.8 63.6 22.8 27.3 22.7 9.1 11.4
58.8F12.2 80.0 54.3 45.7 71.4 22.9 40.0 28.6 37.1 2.9
n.s. n.s. n.s. n.s. 0.012 n.s. n.s. n.s. 0.002 n.s.
30.2 32.6 37.2 1.96 (0.5/5.4)
28.6 54.3 17.1 0.5 (0.25/1.0)
n.s.
b0.001
1.46 (0.99/2.01) 2.85 (1.60/4.90) b0.001 46 (42/63) 6.8 4.5
56 (43/64) 2.9 34.3
n.s. n.s. 0.001
Data expressed as meanFS.D., percent or median (25%/75%)-percentiles; n.s. indicates not statistically significant; *, left ventricular angiography was performed in 26/35 patients of the lysis group and 39/44 patients in the tirofiban group; LVEF, left ventricular ejection fraction; CAD, coronary artery disease; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting.
to 20.9% in the tirofiban group ( pb0.001; Fig. 2). Corrected initial TIMI frame counts were not significantly lower in the fibrinolysis group (47 (32/67) versus 55 (35/120), p=n.s., Table 2). Ninety percent of patients in both groups had postprocedural TIMI grade 3 flow. This was also reflected by the
Table 2 Procedural findings Tirofiban Culprit vessel [%] LAD LCX RCA Initial TGF [%] 3 2 1 0 Initial cTFC Visible thrombi [%] No reflow [%] Peripheral emboli [%] Final TGF-3 [%] Final cTFC
Fibrinolysis
p n.s.
38.6 9.1 45.5
42.9 14.3 42.9
20.5 9.1 22.7 47.7 55 (35/120) 30.2 4.7 20.9 90.5 24 (17/35)
62.9 14.3 11.4 11.4 47 (32/67) 23.5 8.8 8.8 90.0 29 (19/41)
20,5 80% 60%
9,1
TIMI III TIMI II TIMI I TIMI 0
62,9
22,7
40% 20%
47,7
0% Tirofiban
14,3 11,4 11,4 Fibrinolysis
Fig. 2. Initial TIMI grade flow in both treatment groups.
corrected TIMI frame counts (24 (17/35) and 29 (19/41), respectively; p=n.s., Table 2 and Fig. 3). The peak CPK was measured in all patients while in the hospital (Table 3). The peak CPK was not different in both groups. The mean time to peak CPK was significantly lower in the tirofiban group (9.7 (7.3/14.5) h) than in the fibrinolyzed group (12.9 (9.2/23.1) h; p=0.036). No differences were noted between the two groups with regard to bleeding complications (Table 3). Minor bleeding appeared to be more frequent in the fibrinolysed group (8.6%) but the difference was only borderline statistically significant ( p=0.051). The outcomes at 9 month follow up clearly showed a higher event rate in the fibrinolysed group (28.6% versus 4.5%, p=0.003) due to more deaths (25.7% versus 4.5%, p=0.006) and intracerebral hemorrhage (5.7 versus none) in the fibrinolysed group (Table 3). One fatal bleeding occurred in one tirofiban patient (lung bleeding after ultima ratio abciximab on top of tirofiban after massive occurrence of coronary thrombi). There were very few local complications at puncture site and no relevant local bleeding complications occurred which is maybe due to the use of vascular closure devices in all patients (Table 3). No significant differences were seen with respect to reinfarction and further PCI (Table 3). In the fibrinolysed group significantly more patients were rehospitalized (31.4% versus 6.8%, p=0.001) for various causes. Logistic regression resulted in significant odds ratios (OR) for previous PCI (5.0, p=0.023) and delay of the Final Corrected TIMI Frame Count
b0.001 60
n.s. n.s. n.s. n.s. n.s. n.s.
Data expressed as meanFS.D., percent or median (25%/75%)-percentiles; n.s. indicates not statistically significant; TGF, TIMI-grade flow; cTFC, corrected TIMI-frame count.
cTFC
Age [years] Gender, [%] male Hypertension [%] Hyperlipidemia [%] Smoking [%] Diabetes [%] Previous CAD [%] Previous MI [%] Previous PCI [%] Previous CABG [%] Severity of CAD [%] Single-vessel disease Double-vessel disease Triple-vessel disease Symptom onset to first medical contact [h] First medical contact to PCI [h] Initial angiographic LVEF [%]* Shock [%] Prehospital resuscitation [%]
100%
Tirofiban
40
20
0 Tirofiban
Fibrinolysis
Fig. 3. Final corrected TIMI frame count (cTFC). Median and (25%/75%)percentiles. Differences are not significant.
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200 Table 3 Complications and outcomes Tirofiban Bleeding [%] Minor# Major# Loss## Transfusions [%] Local complications at puncture site### [%] Outcomes at 30 days follow Peak CPK [U/L] Time to peak CPK [h] Composite [%] Death Re-infarction ICH PCI [%] CABG (within 24 h) [%] CABG (after 24 h) [%]
2.3 0 2.3 2.3 4.5 2.3
up 557 (200/1483) 9.7 (7.3/14.5) 4.5 4.5 0 0 2.3 0 0
Cumulative outcomes at long-term follow up Follow up time [days] 240 (103/359) Composite [%] 4.5 Death 4.5 Re-infarction 0 ICH [%] 0 PCI [%] 22.8 CABG [%] 2.3 Composite-2* [%] 27.3 Rehospitalization [%] 6.8
Fibrinolysis
p
11.4 8.6 2.9 0 14.3 2.9
n.s. 0.051 n.s. n.s. n.s. n.s.
639 (161/2087) 12.9 (9.2/23.1) 22.9 20.9 2.9 5.7 0 2.9 2.9
n.s. 0.036 0.015 0.028 n.s. n.s. n.s. n.s. n.s.
379 (105/471) 28.6 25.7 2.9 5.7 14.3 8.6 40.0 31.4
n.s. 0.003 0.006 n.s. n.s. n.s. n.s. n.s. 0.001
# As defined in the methods section according to the TIMI-criteria; ##Loss denotes a decrease in hemoglobin of more than 4 g/dL but not exceeding 5 g/dL with no identified bleeding site. ###Aneurysma spurium and/or a-vfistula. Severe local bleeding/hematoma were not observed. PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; CPK, creatinphosphokinase. *Composite-2 denotes the second composite endpoint containing death, reinfarction, PCI or CABG for comparison with the MITRA-registry [19].
patient below 3 h (5.1, p=0.036) for the choice of prehospital fibrinolysis instead of tirofiban and direct PCI.
4. Discussion Our study presents real life data on the prehospital application of the GP IIb/IIIa-inhibitor tirofiban in patients with ST-elevation myocardial infarction and intended percutaneous revascularization. Patients are compared to a group with prehospital fibrinolysis and early angiography from the same emergency system and the same time period. Mechanical reperfusion has gained increasing acceptance as the preferred reperfusion strategy for acute myocardial infarction. Most recently this has been included in actual guidelines for treatment of ST-elevation acute myocardial infarction (STEMI) [1]. The superiority of percutaneous coronary intervention (PCI) compared to fibrinolysis is due to more than 90% open arteries with TIMI III flow after the procedure while fibrinolysis with
197
modern agents as tenecteplase reaches only around 60% (e.g. 62.8% in TIMI 10b [13]). The rationale behind facilitated PCI has been outlined above. In our study a full dose fibrinolyis followed by early angiography and PCI was compared to regular dose tirofiban and PCI. Up to now there are no randomized trials published with clinical endpoints comparing both strategies. The angiographic findings of this real-life observational study suggest that prehospital administration of tirofiban in the emergency care mobile improves initial TIMI III flow grade. Compared to the TIGER-PA study [9] TIMI III flow rate in our study was in between the early and late group of that study (20.5% versus 32%/10%) and nearly matches the presence of 19% TIMI 3 flow with early tirofiban in the larger On-TIME study [14]. Additionally our results fit very well in previous data achieved with abciximab (ADMIRAL, 16.8% TIMI III flow) and are clearly higher than TIMI flow rates reported from the placebo group of that study (5.4%) [8]. Therefore, in real-life emergency care for STEMI patients, we achieved the same pre PCI TIMI III flow rate as in randomized studies. On the other hand there were significantly more open arteries in the fibrinolysed patients (62.9%), a result which fits very well in the previous published data [13]. Although facilitation of PCI by fibrinolysis leads to significantly more open arteries prior to stent placement the crucial questions are (1) whether more open arteries lead to a higher procedural success and (2) improved outcomes. Our study was designed to answer the first question and give some idea of the second. We found that procedural success was the same in both treatment groups. Especially final TIMI III flow and final TIMI frame count (Table 2, Fig. 3) were not different. These results are in line with others who described excellent procedural success with early pretreatment with tirofiban [9]. It is of interest that longer time to peak CPK was detected in fibrinolysed patients (Table 3). This underlines that the initial TIMI-3 flow is not necessarily identical with myocardial tissue reperfusion which is probably better assessed by myocardial blush grades [15]. It has been taken into account that (I) the vessels need a certain time to open under fibrinolysis and the net benefit compared to direct PCI is most probably short (II) that indeed due to the later catheter time in our lysis group less patients would have had TIMI 3 flow at an earlier time point. Nevertheless it can be proposed that every fibrinolysed patient should undergo immediate cardiac catheterization for rescue PCI which must not be delayed as it has happened in our real-life szenario. Immediate PCI after lysis has been shown earlier to be effective and safe if short acting fibrinolytics are used [16]. The outcomes data of our study are potentially biased as discussed in detail below and limited due to the small number of patients. Nevertheless, we observed two patients with intracranial hemorrhage (ICH) after fibrinolyis which is
198
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
the well known major complication of fibrinolytic treatment. This confirms even recent reports of fibrinolytic trials with 1–2% ICH [17,18] depending on the co-therapy. Dose reduction of the fibrinolytic agent and combination with a GP IIb/IIIa-inhibitor does not reduce ICH (e.g. GUSTO V [7]). In our study the secondary endpoint showed very good outcomes of patients with tirofiban and PCI but high event rates for patients with fibrinolysis. As it is known from the MITRA registry [19], mortality and event rates are substantially higher in patients outside of controlled randomized trials. Our 12 months mortality rate in the fibrinolysis group seems to be higher compared to those in the registry (25.7% versus 13.9% [19]). In the MITRA registry resuscitation was 4.9% in the fibrinolysis group compared to 34.3% in our study indicating high risk, which clearly explains the difference in mortality. On the other hand the composite 12 months endpoint of the MITRAregistry (see Table 3 for definition bcomposite-2Q) has similar results in the PCI group compared to our tirofibangroup (25.6% versus 27.3%; Table 3). The major component of this endpoint is further PCI (not only of the target vessel). The fibrinolyis group in MITRA has a 32.3% event rate compared to 40.0% in our high risk fibrinolysis group. This shows that the overall long-term risk of fibrinolysed patients in our study is comparable to previously published data. The high early mortality is due to a selection bias by including many patients with prehospital resuscitation. At many hospitals, primary PCI is the preferred reperfusion strategy for STEMI patients. According to actual guidelines and recent studies this can be recommended in general if time from first medical contact until balloon inflation by an experienced investigator is below 90 min [1]. Mortality seems to be excessively increased if doorto-balloon time exceeds 2 h [6]. This fits well with the DANAMI-2 study which shows that patients profit from primary PCI even if transfer is necessary. In this study the patients from referral hospitals had a median delay from randomization to start of treatment of 90 min [2]. The present study suggests that, in centers where primary PCI is the preferred reperfusion strategy, the prehospital administration of tirofiban by experienced emergency physicians is safe and effective. The study was underpowered and biased by selection to prove superiority of tirofiban in clinical endpoints. Larger randomized studies are necessary and justified to compare different measures to facilitate direct PCI for STEMI. 4.1. Limitations The present study presents prospective data on the safety and efficacy of the prehospital application of tirofiban before percutaneous intervention. The control group has been retrospectively taken from the same period of time and the same emergency system. All applicable patients in the inclusion period have been identified and analysed as displayed in Fig. 1. As the study was not randomized some
limitations has to be addressed with respect to the discussion of the results. (1) The tirofiban patients were all scheduled for PCI. Fibrinolysed patients are usually scheduled for coronary angiography but as reperfusion therapy is already started, the priority of angiography in these patients is sometimes judged lower in clinical routine. This explains the significantly longer time from first medical contact to PCI in the fibrinolysed group (Table 1). On the other hand the time from onset of symptoms to first medical contact was significantly shorter in the lysis group reflecting that the maximum benefit of fibrinolysis in the first 3 h was taken into account by the emergency physicians. With respect to the interpretation of the results this means that TIMI 3 flow is maybe overestimated in the fibrinolysed group compared to tirofiban patients who were earlier in the catheterization laboratory. (2) The major problem of non-randomized studies is the bias of selection. We aimed at minimizing this factor by analysing all patients of the inclusion period. Table 1 shows only few significant differences of clinical variables in the both groups, including delay of the patient (time from onset of symptoms to first medical contact, Table 1), smoking, previous PCI and prehospital resuscitation. Another potential confounder is the attending emergency physician who was responsible for the initial choice of therapy. We performed a multiple logistic regression analysis with respect to the choice of therapy including these critical variables. The analysis clearly resulted in a significant influence of previous PCI and delay below 3 h as variables in favor for prehospital fibrinolysis (OR 5.0, p=0.023, OR 5.1, p=0.036, respectively). Prehospital resuscitation was not longer significant when the variable delay was included in the model. The variable battending Table 4 Complications and outcomes of the fibrinolysed group with late or no PCI n=19 Bleeding [%] Minor Major Transfusions Cardiac catheter during index admission Local complications at puncture site [n] Outcomes at 9 month follow up Composite [%] Death Re-infarction Stroke PCI Rehospitalization Peak CPK [U/L] Time to peak CPK [h]
5.3 5.3 0 0 57.9 0 21.1 21.1 0 0 21.1 42.1 360 (110/860) 27.7 (10.1/46.5)
Results were not significantly different form those in group with early angiography (Tables 1–3).
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
emergency physicianQ did not significantly influence the choice of reperfusion therapy. With respect to the interpretation of the results this means that the emergency physicians maybe selected more patients with definite coronary artery disease who present early for prehospital fibrinolysis which has to be taken into account when outcome data are interpreted. (3) In the whole study population 19 patients with prehospital fibrinolysis had no coronary angiography within 24 h. One could ask if these patients are more or less severe ill and therefore could have biased the results. The Table 4 summarizes the complications and outcomes data of this group. It is clearly evident that this group is not significantly different from the fibrinolysis group with early angiography and therefore an additional bias is most probably excluded. (4) Finally it has to be pointed out that the comparison of the procedural findings was the primary goal of this study and with respect to the very similar patients characteristics (Table 1) it can be assumed that herein valid results could be obtained. The advantage of these real life data is that no selection bias due to exclusion criteria (e.g. all resuscitated patients would have not been included in a randomized study), non-consent of patients or missing inclusion due to organisational or administrative causes can occur which usually limits the applicability of randomized trials to daily practice.
5. Conclusions We conclude that (1) prehospital start of tirofiban for facilitated PCI is safe and effective if administered by experienced emergency physicians; (2) routine fibrinolysis should be limited to areas where catheter based therapy is not available within 90 min and (3) fibrinolysis should be given for facilitated PCI in randomized trials only at the moment.
Acknowledgement The preparation of the manuscript was supported by a grant from MSD. This work is part of the doctoral thesis of Lutz Nibbe.
References [1] Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KA, et al. Management of acute myocardial infarction in patients presenting with ST-segment elevation. The task force on the management of acute myocardial infarction of the European Society of Cardiology. Eur Heart J 2003;24(1):28 – 66. [2] Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H, Thayssen P, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med 2003;349(8):733 – 42.
199
[3] Widimsky P, Budesinsky T, Vorac D, Groch L, Zelizko M, Aschermann M, et al. Long distance transport for primary angioplasty vs immediate thrombolysis in acute myocardial infarction. Final results of the randomized national multicentre trial-PRAGUE-2. Eur Heart J 2003;24(1):94 – 104. [4] Stone GW, Cox D, Garcia E, Brodie BR, Morice MC, Griffin J, et al. Normal flow (TIMI-3) before mechanical reperfusion therapy is an independent determinant of survival in acute myocardial infarction: analysis from the primary angioplasty in myocardial infarction trials. Circulation 2001;104(6):636 – 41. [5] Brodie BR, Stuckey TD, Wall TC, Kissling G, Hansen CJ, Muncy DB, et al. Importance of time to reperfusion for 30-day and late survival and recovery of left ventricular function after primary angioplasty for acute myocardial infarction. J Am Coll Cardiol 1998;32(5):1312 – 9. [6] Cannon CP, Gibson CM, Lambrew CT, Shoultz DA, Levy D, French WJ, et al. Relationship of symptom-onset-to-balloon time and door-toballoon time with mortality in patients undergoing angioplasty for acute myocardial infarction. JAMA 2000;283(22):2941 – 7. [7] Topol EJ. Reperfusion therapy for acute myocardial infarction with fibrinolytic therapy or combination reduced fibrinolytic therapy and platelet glycoprotein IIb/IIIa inhibition: the GUSTO V randomised trial. Lancet 2001;357(9272):1905 – 14. [8] Montalescot G, Barragan P, Wittenberg O, Ecollan P, Elhadad S, Villain P, et al. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction. N Engl J Med 2001;344(25):1895 – 903. [9] Lee DP, Herity NA, Hiatt BL, Fearon WF, Rezaee M, Carter AJ, et al. Adjunctive platelet glycoprotein IIb/IIIa receptor inhibition with tirofiban before primary angioplasty improves angiographic outcomes: results of the TIrofiban Given in the Emergency Room before Primary Angioplasty (TIGER-PA) pilot trial. Circulation 2003;107(11):1497 – 501. [10] Danzi GB, Capuano C, Sesana M, Baglini R. Safety of a high bolus dose of tirofiban in patients undergoing coronary stent placement. Catheter Cardiovasc Interv 2004;61(2):179 – 84. [11] Bovill EG, Terrin ML, Stump DC, Berke AD, Frederick M, Collen D, et al. Hemorrhagic events during therapy with recombinant tissue-type plasminogen activator, heparin, and aspirin for acute myocardial infarction. Results of the Thrombolysis in Myocardial Infarction (TIMI), Phase II Trial. Ann Intern Med 1991;115(4):256 – 65. [12] Gibson CM, Cannon CP, Daley WL, Dodge Jr JT, Alexander B, Marble SJ, et al. TIMI frame count: a quantitative method of assessing coronary artery flow. Circulation 1996;93(5):879 – 88. [13] Cannon CP, Gibson CM, McCabe CH, Adgey AAJ, Schweiger MJ, Sequeira RF, et al. TNK-tissue plasminogen activator compared with front-loaded alteplase in acute myocardial infarction: results of the TIMI 10B trial. Circulation 1998;98(25):2805 – 14. [14] van’t Hof AWJ, Ernst N, de Boer MJ, de Winter R, Boersma E, Bunt T, et al. Facilitation of primary coronary angioplasty by early start of a glycoprotein 2b/3a inhibitor: results of the ongoing tirofiban in myocardial infarction evaluation (On-TIME) trial. Eur Heart J 2004;25(10):837 – 46. [15] Gibson CM, Cannon CP, Murphy SA, Marble SJ, Barron HV, Braunwald E, et al. Relationship of the TIMI myocardial perfusion grades, flow grades, frame count, and percutaneous coronary intervention to long-term outcomes after thrombolytic administration in acute myocardial infarction. Circulation 2002;105(16):1909 – 13. [16] Ross AM, Coyne KS, Reiner JS, Greenhouse SW, Fink C, Frey A, et al. A randomized trial comparing primary angioplasty with a strategy of short-acting thrombolysis and immediate planned rescue angioplasty in acute myocardial infarction: the PACT trial. J Am Coll Cardiol 1999;34(7):1954 – 62. [17] Wallentin L, Goldstein P, Armstrong PW, Granger CB, Adgey AAJ, Arntz HR, et al. Efficacy and safety of tenecteplase in combination with the low-molecular-weight heparin enoxaparin or unfractionated heparin in the prehospital setting: the Assessment of the Safety and
200
M. Mo¨ckel et al. / International Journal of Cardiology 103 (2005) 193–200
Efficacy of a New Thrombolytic Regimen (ASSENT)-3 PLUS randomized trial in acute myocardial infarction. Circulation 2003;108(2):135 – 42. [18] ASSENT-2 Investigators. Single-bolus tenecteplase compared with front-loaded alteplase in acute myocardial infarction: the ASSENT-2 double-blind randomised trial. Lancet 1999;354(9180):716 – 22.
[19] Zahn R, Schiele R, Schneider S, Gitt AK, Heer T, Wienbergen H, et al. Long-term follow-up of patients with acute myocardial infarction treated with primary angioplasty or thrombolysis. Results of the MITRA trial. Z Kardiol 2002;91(1):49 – 57.