How do Smokers Differ From Nonsmokers in Their Response to Thrombolysis? (The TIMI-4 Trial) Doron Zahger, MD, Bojan Cercek, MD, Christopher P. Cannon, MD, Matthew Jordan, MS, Vicki Davis, DrPH, Eugene Braunwald, MD, and Prediman K. Shah, MD, for the TIMI-4 Investigators Smokers with acute myocardial infarction appear to have a better outcome after thrombolysis than do non- smokers. To identify factors that could contribute to this curious finding, we analyzed data from the Thrombol- ysis in Myocardial Infarction (TIMI-4) trial, in which 382 patients with acute myocardial infarction were ran- domized to tissue plasminogen activator, anistreplase, or both. Coronary angiography was performed 90 min- utes and 18 to 36 hours after randomization, a myo- cardial perfusion scan was performed at 18 to 36 hours and before discharge, and o radionuclide venMculo- gram was obtained before discharge. Angiographic and clinical outcome variables were determinearin cur- rent smokers, ex-smokers, and nonsmokers, and re- analysis was used to correct far differences in seline characteristics. The in-hospital mortality of cur- rent smokers was lower than that of ex-smokers and nonsmokers: 2.3% versus 5.2% versus 7.0%, respec- tively (p = 0.04 by paired comparison, current vs non- smokers). Ninety minutes after randomization, the inci- dence of TIMI grade 3 flow was significantly higher in smokers than in ex-smokers and nonsmokers (55% vs 43% and 45%, p = 0.02); this difference was no longer observed at the second angiogram, nor did smokers differ from nonsmokers with respect to residual steno- sis, thrombus grade, infarct size, ejection fraction, or recurrent ischemia. Because a slmng inverse relation exists between TIMI grade 3 flow at 90 minutes and mortality, our findings suggest that the lower mortali- ty of current smokers after thrombolytlc therapy may be related to a higher incidence of early, complete reperfusion. (Am J Cardiol 1995;75:232-236) C lgarette smoking ~s associated with increased coro- nary morbidity and mortality, 1-7 which is mediated by multiple deleterious effects of smoking on atherogen- esis, thrombosis, vasomotlon, and arrhythmogenesIs.8-14 Paradoxically, smoking is associated with better outcome during the hospital phase of acute myocardml infarction (AMI), both with and without thrombolytlc therapy 15-20 Because AMI occurs at a younger age in smokers and because of the prothrombotlc effect of smoking, smok- ers with AMI may have a relatively smaller atheroscle- rotic burden and more thrombus at the site of coronary occlusion, el and consequently respond better to throm- bolytic therapy. To examine this and other potential mechanisms, we analyzed data from the Thrombolysis in Myocardial Infarction (T1MI)-4 trial METHODS TIMI-4 was a randomized, double-blind, multicenter trial companng anlstreplase, front-loaded ussue plas- minogen actwator (t-PA), and a combinaUon of the 2 in From the Division of Cardiology, Departmentof Medlane, Cedars- Sinai Me&cal Centerand the Unrverslty of Cahfornla School of Med- ~ane, Los Angeles, Cahfornla, the Cardtovascular DIwslon, Depart- ment of Medlane, Bngham & Women's Hospital and Harvard Medical School, Boston, Massachusetts, and the ResearchTnangle Institute, Research Tnangle Park, North Carohna Dr Zahger was supported by the Wdham Ganz ResearchFellowship of the Save A Heart Foundation and the study was supported m part by a grant from SmlthKhne Beecham, Phdadelph~a,Pennsylvania Manuscnpt receivedJune 29, 1994, revised manuscriptreceived October 20, 1994, and accepted October 22 Address for repnnts Predlman K Shah, MD, Division of Cardi- ology, #5314, Cedars-Stoat Me&cal Center, 8700 Beverly Blvd, Los Angeles, Cahfornla 90048 382 patients with evolvmg AMI An open-label phase using the combination was first conducted In 34 paaents who are included in this analysis The study was con- ducted in 18 participating centers (see Appendix). Patients were eligible for inclusion if they were <80 years, had typical lschemlc chest pain for >30 nunutes, presented within 6 hours, and had either new ST-segment elevation of 0.1 mV m 2 contiguous leads or new left bundle branch block. Patients were excluded if they had contramdlcatlons to thrombolysls, as well as recent (with- in 2 months) coronary bypass surgery, previous use of streptokmase or anlstreplase, chronic left bundle branch block, were women of chlldbeanng potential, were re- ceiving oral antlcoagulatlon, had any other serious illness, or were not wllhng or unable to give reformed consent. Patients were randomized to 1 of 3 thrombolyttc reg- imens" (1) front-loaded t-PA (Actlvase TM, Genentech, South San Francisco, California) given as an initial bolus of 15 mg, followed by an infusion of 0 75 mg/kg over 30 minutes and then 0.5 mg/kg over 60 minutes, the total dose not to exceed 100 mg; (2) anistreplase (amsoylat- ed plasminogen streptolonase activator complex, Eml- nase TM, SmlthKlme Beecham, Philadelphia, Pennsylva- nia) 30 U gwen over 2 to 5 minutes; (3) a combination of t-PA (15 mg bolus and 075 mg/kg over 30 minutes, maximal dose 65 mg) and anistreplase 20 U. Drugs were administered in a double-blmd manner. All patients received 325 mg/day of enteric-coated as- pirin (Ecotrln TM, SmithKhne Beecham); patients who had not been taking aspirin were asked to chew the tablet on admission. Hepann was administered as a 5,000 U bolus dose followed by an infusion of 1,000 U/hour, and adjusted to maintain the activated partial thromboplas- 232 THE AMERICAN JOURNAL OF CARDIOLOGY ® VOL 75 FEB 1, 1995