BACKGROUND

Rescue angioplasty (rPCI) for failed fibrinolysis is associated with a low mortality if successful, but a high mortality if it fails. The latter may reflect a high-risk group or harm in some patients. Predictors of success or failure of rPCI may aid selection of patients to be treated.

METHODS

Unselected patients referred for rPCI from March 1994 to March 2005 were studied to determine the predictors of a failed procedure and 1-year mortality.

RESULTS

Of 440 patients undergoing emergency coronary angiography for failed fibrinolysis (1-year mortality 18%), 101 had thrombolysis in myocardial infarction flow grade (TFG) 3 in the infarct-related vessel. rPCI was attempted in 318 of 339 patients with <TFG 3 flow, but not in 21 patients (angiography-produced TFG 3 [n = 7] or unsuitable anatomy [n = 14]). Of the rPCI cohort, 77% had a successful procedure (no in-lab death or emergency coronary artery bypass grafting and TFG 3 in the infarct-related vessel); rPCI failed in 23%. One-year mortality rates for successful and failed rPCI were 14 and 43%, respectively. Patients with failed rPCI were older and more likely to be diabetic, have anterior MI, be interhospital transfers, be in cardiogenic shock, and less likely to be a current smoker. Shock was the only independent predictor of failed rPCI. Age group >75 years, shock, and final TFG < 3 were independent predictors of 1-year mortality.

CONCLUSIONS

Cardiogenic shock is an independent predictor of a failed rPCI. Age group >75 years and shock were the only independent clinical predictors of 1-year mortality. These clinical variables may help in selecting patients for either a strategy of rescue angioplasty after failed fibrinolysis, or in selecting specific patients who might do better with a policy of primary angioplasty.