1. Myocardial ischaemia-reperfusion in humans is associated with increased formation of 15-F(2t)-isoprostane and 5-F(2t)-isoprostane (15-F(2t)-IsoP and 5-F(2t)-IsoP, respectively). Whether this formation is relevant clinically remains controversial. The present study was performed in order to evaluate the ability of the isoprostanes 15-F(2t)-IsoP and 5-F(2t)-IsoP to reduce myocardial ischaemic injury in rat isolated heart. 2. Rats were divided into six groups. Hearts were excised, perfused retrogradely and pretreated with vehicle (ethanol 5.10(-7) and 2.10(-9) mol/L; n = 6), subjected to ischaemic preconditioning (n = 8) or pretreated with the isoprostanes 15-F(2t)-IsoP (3.10(-10) and 3.10(-7) mol/L; n = 8) or 5-F(2t)-IsoP (10(-9) mol/L; n = 8). After a 5 min treatment-5 min washout period, hearts were submitted to 30 min global ischaemia, followed by a 120 min reperfusion period. 3. The infarct-to-ventricle zone ratio was significantly reduced in ischaemic preconditioned (20.6 +/- 2.6%) compared with vehicle groups (44.5 +/- 4.3 and 51.3 +/- 2.5% in groups pretreated with 5.10(-7) or 2.10(-9) mol/L ethanol, respectively). Pretreatment with either isoprostane had no cardioprotective effect; the infarct-to-ventricle ratios were 43.1 +/- 2.2, 49.4 +/- 5.9 and 44.5 +/- 5.0% for groups treated with 3.10(-10) mol/L 15-F(2t)-IsoP, 3.10(-7) mol/L 15-F(2t)-IsoP or 10(-9) mol/L 5-F(2t)-IsoP, respectively. 4. These data provide evidence that the isoprostanes 15-F(2t)-IsoP and 5-F(2t)-IsoP are not implicated in early myocardial preconditioning at concentrations similar to those found in the human coronary sinus following coronary angioplasty.