Objective This study investigated the outcome of patients who received bail-out study medication and evaluated whether high-dose tirofiban (HDT) pretreatment may reduce the need for bail-out study medication.

Design A prespecified analysis of the multicentre, double-blind, placebo controlled, randomised On-TIME 2 trial. Bail-out use of study medication was predefined and part of the combined clinical end point.

Patients 984 patients excluded from many coronary intervention hospitals in different countries were randomly assigned to HDT or placebo. In the subgroup who received blinded bail-out treatment, patients pretreated with placebo who received bail-out HDT were compared with those pretreated with HDT who received bail-out placebo.

Interventions Routine prehospital initiation of HDT versus bail-out use of HDT.

Main Outcome Measures Electrocardiographic and clinical outcome.

Results Blinded bail-out study medication was used in 24% (237/980) of patients, with a higher rate in patients pretreated with placebo: 29% (140/492) versus 20% (97/488), p=0.002. Bail-out versus no bail-out use of study medication was associated with more residual ST deviation (5.5±7.2 vs 3.7±4.8 mm, p=0.005), and worse clinical outcome (major adverse cardiac events (MACE) at 30 days 12.2% vs 5.6%, p<0.001), mainly due to poor outcome in patients who received HDT bail-out. In patients pretreated with HDT who received placebo bail-out study medication, residual ST deviation and clinical outcome did not differ significantly compared with patients who did not receive bail-out medication (4.0±4.6 vs 3.7± 4.8 mm, p=0.703, MACE 7.2% vs 5.6%, p=0.535).

Conclusions Routine prehospital treatment with HDT significantly reduced the use of blinded bail-out study medication. The need for bail-out therapy was associated with a less favourable outcome. This analysis suggests that routine pretreatment is superior to provisional use of HDT in patients with ST-segment elevation myocardial infarction.