The pharmacokinetics and bioavailability of nitroglycerin (GTN) and its metabolites, 1,2-glyceryl dinitrate (1,2-GDN) and 1,3-glyceryl dinitrate (1,3-GDN), were compared after a single 14-hour application of Transderm-Nitro (Ciba-Geigy, Summit, NJ), Nitrodisc (GD Searle, Chicago, IL), and Nitro-Dur II (Key Pharmaceuticals, Kenilworth, NJ) systems to 18 healthy male subjects on 3 separate occasions. A 14-hour intravenous infusion of 15N-labeled GTN was given simultaneously to correct for changes in systemic clearance during the application of each system. Plasma concentrations of 15N-labeled GTN, unlabeled GTN, and their corresponding dinitrate metabolites were measured using a gas chromatography/mass spectrometry method. Results showed that the plasma concentration profiles of nitroglycerin and its metabolites for the three systems were similar during and after system removal. Mean (SD) total amounts (AUCp x CLiv) of GTN transdermally available after adjustment for 15N-labeled GTN clearance were 5.3 (2.1), 5.3 (2.0), and 5.4 (2.6) mg for Transderm-Nitro, Nitrodisc, and Nitro-Dur II, respectively. Mean (SD) AUC values for 1,2-GDN were 44.6 (15.8), 44.3 (16.1), and 42.8 (19.3) ng.h/mL for the 3 systems. Corresponding AUC values for 1,3-GDN were 9.3 (2.9), 9.7 (2.9), and 8.7 (3.0) ng.h/mL. Statistical analysis of the log-transformed data based on 90% conventional confidence interval showed that all 3 systems delivered equivalent amounts of nitroglycerin into the systemic circulation. The AUC ratios for 1,3-GDN to GTN, but not 1,2-GDN to GTN, were statistically different for the intravenous and transdermal routes during all 3 system applications, indicating that the formation and metabolism of 1,3-GDN was dependent on route of administration.