Background and purposeAmbulance trials assessing interventions in suspected stroke patients will recruit patients with currently active symptoms that will resolve into transient ischaemic attack (TIA). The safety and efficacy of glyceryl trinitrate (GTN) in the pre‐specified subgroup of patients with TIA in the Rapid Intervention with Glyceryl Trinitrate in Hypertensive Stroke Trial 2 (RIGHT‐2) was assessed.MethodsRIGHT‐2 was a pre‐hospital‐initiated multicentre randomized sham‐controlled blinded‐endpoint trial that randomized patients with presumed ultra‐acute stroke within 4 h of symptom onset to transdermal GTN or sham. Final diagnosis was determined by site investigators. The primary outcome was a shift in modified Rankin Scale (mRS) scores at 90 days analysed using ordinal logistic regression reported as adjusted common odds ratio with 95% confidence intervals (CIs). Secondary outcomes included death or dependence (mRS >2).ResultsIn all, 109 of 1149 (9.5%) patients had a final diagnosis of TIA (GTN 57, sham 52) with mean age 73 (SD 13) years, 19 (17.4%) had pre‐morbid mRS >2, and onset to randomization was 80 min (interquartile range 49, 105). GTN lowered blood pressure by 7.4/5.2 mmHg compared with sham by hospital arrival. At day 90, GTN had no effect on shift in mRS scores (common odds ratio for increased dependence 1.47, 95% CI 0.70–3.11) but was associated with increased death or dependence (mRS >2): GTN 29 (51.8%) versus sham 23 (46.9%), odds ratio 3.86 (95% CI 1.09–13.59).ConclusionsPre‐hospital ultra‐acute transdermal GTN did not improve overall functional outcome in patients with investigator‐diagnosed TIA compared with sham treatment.