graphic evidence of intraluminal thrombus were random-ized to either intracoronary thrombectomy followed bystenting or to a conventional strategy of stenting. Throm-bectomy was performed using the X-Sizer catheter (Endi-COR Inc., San Clemente, California). Myocardial reperfu-sion was assessed by myocardial blush and ST resolution.Results: Postprocedure Thrombolysis in Myocardial Infarc-tion-3 flow was not different between groups (93.5% vs.95.7%, p�0.39). Myocardial blush-3 was observed in71.7% of patients undergoing thrombectomy and in 36.9%of patients undergoing conventional strategy (p�0.006).ST-segment resolution �50% occurred more often in pa-tients undergoing thrombectomy (82.6% vs. 52.2%,p�0.001). By multivariate analysis, adjunctive thrombec-tomy was an independent predictor of blush-3 (odds ratio,3.27; 95% confidence interval, 1.06–10.05; p�0.039).Conclusions: The authors concluded that intracoronarythrombectomy as adjunct to stenting during direct angio-plasty for AMI improves myocardial reperfusion as assessedby myocardial blush and ST resolution.Perspective: This single-center experience in a small num-ber of patients demonstrated that intracoronary thrombec-tomy with the X-Sizer catheter during catheter-based treat-ment of AMI improved myocardial reperfusion as assessedby myocardial blush and ST-segment resolution. Whetherthese surrogate end points translate into meaningful im-provements in clinical outcomes will need to be tested inlarger appropriately designed clinical trials. DM

Transfer for Primary Angioplasty vs. ImmediateThrombolysis in Acute Myocardial Infarction: AMeta-AnalysisDalby M, Bouzamondo A, Lechat P, Montalescot G. Circulation2003;108:1809 –14.

Study Question: The objective was to evaluate the best ther-apeutic strategy for a patient with AMI presenting to acutecare services without catheterization facilities. The investi-gators used all available information from clinical trialscomparing transfer of patients experiencing AMI for angio-plasty vs. immediate thrombolysis.Methods: The study was a meta-analysis of six clinical trialscomparing the two strategies. The primary end point wasthe combined criteria (CC) of death/reinfarction/stroke asdefined in each trial. Relative risk (RR) evaluated the treat-ment effect. The investigators identified six clinical trialsincluding 3750 patients.Results: Transfer time was always �3 hours. The CC wassignificantly reduced by 42% (95% confidence interval [CI]29–53%, p�0.001) in the group transferred for primaryPCI compared with the group receiving on-site thrombol-ysis. When CC parameters were considered separately,reinfarction was significantly reduced by 68% (95% CI,34–84%; p�0.001) and stroke by 56% (95% CI, �15–77%; p�0.015). There was a trend toward reduction in

all-cause mortality of 19% (95% CI, �3–36%; p�0.08)with transfer for PCI.Conclusions: The authors concluded that even when transferto an angioplasty center is necessary, primary PCI remainssuperior to immediate thrombolysis. They state that orga-nization of ambulance systems, prehospital managementand adequate PCI capacity appears to be the key issues inproviding reperfusion therapy for AMI.Perspective: This meta-analysis of six contemporary clinicaltrials indicates that even when transport is required, pri-mary PCI remains a superior strategy to local thrombolysisas long as transfer time is �3 hours. To benefit fromprimary angioplasty we need systems in place to organizerapid and safe transfer of patients with AMI to primaryangioplasty facilities. The goal of treatment in AMI shouldbe early and complete reperfusion, which is best achievedwith primary angioplasty. However, the relevance of thisanalysis to rural communities where cath facilities are manyhours away is minimal. DM

Clinical and Angiographic Correlates and Outcomesof Suboptimal Coronary Flow in Patients With AcuteMyocardial Infarction Undergoing PrimaryPercutaneous Coronary Intervention

Mehta RH, Harjai KJ, Cox D, et al., and Primary Angioplasty inMyocardial Infarction (PAMI) Investigators. J Am Coll Cardiol2003;42:1739 – 46.

Study Question: The purpose of this study was to determinethe clinical and angiographic correlates and outcomes ofpatients with suboptimal coronary flow after primary per-cutaneous coronary interventions (PCI).Methods: The Primary Angioplasty in Myocardial Infarction(PAMI) studies prospectively enrolled patients with STEMI.The current study is post hoc analysis of patients withsuboptimal flow after primary PCI in the PAMI studies. Theprincipal outcomes were the difference in the hospital and1-year mortality and hospital and 1-year incidence of majoradverse cardiovascular events (MACE, defined as death, orreinfarction or ischemia-driven target vessel revasculariza-tion) in patients with normal and suboptimal flow.Results: Post-procedural final TIMI �2 flow occurred in232 (6.9%) patients. Multivariate analysis identified age�70 years (odds ratio [OR], 1.6; 95% confidence interval[CI], 1.1–2.2), diabetes (OR 1.9; 95% CI, 1.3–2.7), symp-tom onset to emergency room presentation (OR 1.1; 95%CI, 1.1–1.2); initial TIMI �1 flow (OR 3.2; 95% CI, 1.9–5.5) and left ventricular ejection fraction �50% (OR 1.7;95% CI, 1.2–2.4) as independent correlates of final TIMI�2 flow. In-hospital (composite of reinfarction, ischemictarget vessel revascularization or death, as well as theseevents individually) and 1-year (reinfarction and/or death)events occurred more frequently in patients with TIMI �2flow. The Cox proportional hazards model identified TIMI

ACC CURRENT JOURNAL REVIEW Feb 2004

43