Critical Review Form Therapy Objectives: Methods:

Transcription

1 Critical Review Form Therapy Clinical Trial Comparing Primary Coronary Angioplasty with Tissue-Plasminogen Activator for Acute Myocardial Infarction (GUSTO-IIb), NEJM 1997; 336: Objectives: To compare primary angioplasty with thrombolytic therapy (and hirudin with heparin) in the initial management of acute myocardial infarction (p 1622) with the hypothesis that immediate angioplasty would result in a lower incidence of death, non-fatal reinfarction, and non-fatal disabling stroke at 30-days than thrombolytic therapy. Methods: International, multi-center (57 hospitals in 9 countries, all experienced cardiac catheterization centers with surgical back-up) comparing angioplasty to tissueplasminogen activator (t-pa) thrombolysis in 1138 patients between July 1994 and January 1996 presenting within 12-hours of symptom onset with at least 20-minutes of chest pain with 0.2 mv ST-segment elevation in two or more contiguous leads or new LBBB. Exclusion criteria (published elsewhere in NEJM 1996; 335: 775) included warfarin therapy, active bleeding or other contraindication to heparin, history of stroke or renal insufficiency, systolic BP > 200 or diastolic BP > 110. After enrollment of a subset intended for a 2x2 factorial design analysis comparing heparin and herudin, all subjects received heparin anti-coagulation. Subjects were randomized via telephone by study coordinators, but subjects were not blinded to the study-arm. In the angioplasty arm, the infarct related artery was the only target, except in patients whose hemodynamic values deteriorated. (p 1622). Angiographic analysis of success was determined at a central angiogram lab and defined as residual stenosis < 50% and final TIMI flow grade of 2 or 3. In the t-pa arm, angiography within three days of lysis was discouraged in the absence of refractory ischemia or hemodynamic deterioration.

2 The primary endpoint was a composite of death, non-fatal re-infarction, and nonfatal disabling stroke at 30-days as confirmed by the clinical events committee, an independent adjudicating body. Secondary endpoints included 30-day all-cause mortality, non-fatal reinfarction at 30-days, recurrent/medically refractory ischemia, disabling stroke, and CHF at 30-days. Follow-up occurred at 30-days and 6-months by self-administered questionnaire, telephone interview, or follow-up physician visit. Study size was selected to detect a 40% reduction in the primary endpoint between the two groups given α = 0.05 and β = Several pre-specified subgroups were analyzed in relation to these outcomes including anterior versus non-anterior AMI and hospital experience with angioplasty. Guide Comments I. Are the results valid? A. Did experimental and control groups begin the study with a similar prognosis (answer the questions posed below)? 1. Were patients randomized? Yes, investigators and study coordinators telephoned or FAXED a 24-hour per day, 7-day per week randomization center (p 1622) 2. Was randomization concealed (blinded)? No, patients and clinicians were not blinded to group allocation.

3 3. Were patients analyzed in the groups to which they were randomized? 4. Were patients in the treatment and control groups similar with respect to known prognostic factors? B. Did experimental and control groups retain a similar prognosis after the study started (answer the questions posed below)? Yes, treatments were compared according to the intention-to-treat principle. (p 1623). In fact, 94% of those randomized to PTCA actually had angioplasty (although only 5% received stents) and 98% of those randomized to t- PA received t-pa with only 1.4% of the t- PA cohort receiving primary angioplasty. Table 1 (p 1623) displays that in fact most prognostic factors (age, tobacco abuse, diabetes) favored t-pa in non-statistically significant fashion. 1. Were patients aware of group allocation? Yes. Failing to blind subjects leaves open the possibility of information bias (more/less likely to report mild adverse events, clinical improvement/deterioration) as well as contamination bias (may give/take treatment outside the study setting). Both of these types of bias are less likely in the current study and blinding (which would involve sham catheterizations) would likely be deemed unethical by many IRB s. For more information on various forms of bias read J Chron Dis 1979; 32: 51-63). 2. Were clinicians aware of group allocation? Yes. Failing to blind treating clinicians leaves open the possibility of cointervention bias (additional, confounding care provided selectively to one group or the other) and ascertainment bias. 3. Were outcome assessors aware of group allocation? No. An independent Clinical Events Committee adjudicated decisions on all possible primary-end-point events. The members remained unaware of the initial treatment assignments (p 1622). 4. Was follow-up complete? Yes. At 6-months 96.9% of eligible patients had been followed up (p 1624).

4 II. What are the results (answer the questions posed below)? 1. How large was the treatment effect? At 30-days angioplasty resulted in 13 fewer deaths (95% CI -15 to 41, p = 0.37) and 41 fewer deaths/infarctions/disabling strokes (95% CI 3-78, p = 0.033) per 1000 patients with most of the benefit noted between Days 5-10 (Figure 1, p 1625). The ARR of the primary outcome (composite of death, re-infarction, or nonfatal disabling stroke at 30-days) was 4.1% (9.6% to 13.7%) producing a NNT 24 to prevent one death or reinfarction or disabling stroke). Six-month follow-up revealed no significant differences in the composite endpoint (15.7% t-pa versus 13.3% PTCA). Surprisingly, with the exception of intracranial hemorrhage, angioplasty was associated with more bleeding events than t-pa. TIMI flow grade correlated with 30-day mortality as follows: %, %, %, 3 1.6% (p < 0.001). Subgroup analysis demonstrated angioplasty superiority in all groups except very early (< 4-hours) AMI, although wide Confidence Intervals all crossed unity. 2. How precise was the estimate of the treatment effect? Fairly broad Confidence Intervals (see above), so not very precise. III. How can I apply the results to patient care (answer the questions posed below)?

5 1. Were the study patients similar to my patient? At a Level I tertiary medical center with cardiothoracic surgery back-up and an experienced cardiac catheterization lab, most definitely. These results should be extrapolated cautiously to less experienced angioplasty centers and to (the underrepresented) Killip 3-4 heart failure patients. 2. Were all clinically important outcomes considered? 3. Are the likely treatment benefits worth the potential harm and costs? Limitations 1) Not easily extrapolated to angioplasty centers with less experience (<200 angioplasties per year) or without surgical back-up. 2) Dated study with perhaps greater efficacy of PTCA to be expected in 2006 with standard use of drug-eluting stents and glycoprotein IIbIIIa inhibitors, neither of which was available in In fact, only 5% of the angioplasty subjects received a stent. 3) Under-representation of high-risk subsets like Killip 3 or 4 heart failure patients and women. Bottom Line Yes, although most clinicians would likely prefer isolated (not composite) outcomes measures. Yes, if cost-benefit analysis of 24-hour capable PTCA can be assessed in comparison with the readily available t-pa at hospitals with varying AMI volumes and cardiac catheterization experience. Among experienced centers, angioplasty without stents appears superior to t-pa for AMI presenting with symptom duration less than 12-hours with an ARR of 4.1% (NNT 24) to prevent one death, non-fatal reinfarction, or disabling stroke at 30-days. Among patients at higher risk for ICH (including those with severe HTN, advanced age, and symptomatic cerebrovascular disease), angioplasty is likely preferable to t-pa. The goal of EM AMI management is rapid restoration of brisk anterograde coronary flow to reduce mortality. In the presence of an experienced interventional Cardiologist, rapid angioplasty (< 2-hours) may be preferable to thrombolytic therapy. However, realizing that not all medical centers are equipped and experienced, thrombolytic therapy should still be regarded as an excellent strategy of reperfusion. (p 1627)