We obtained information about cessation of smoking in 310 survivors of out-of-hospital cardiac arrest who had been habitual cigarette smokers at the time of their arrest. Patients with coronary heart disease were stratified according to mortality risk on the basis of recognized criteria. The expected first-year rate of recurrent arrest ranged from 2 to 40 percent among the strata. Life-table analyses showed that reformed smokers had a lower incidence of recurrent arrest than patients who continued to smoke (19 vs. 27 percent at three years; P = 0.038 by one-sided test adjusted across strata). This effect occurred to varying degrees in all but the highest risk stratum. No differences in survival were observed for mortality due to other causes. It is possible that continued smoking in these patients led to acceleration of an ongoing atherosclerotic process, but the differences in early survival suggest that smoking may also act in the short term to enhance vulnerability to cardiac arrest.