Background. For more than 60 years, the synthetic opioid fentanyl has been widely used in anaesthesia and analgesia. While the intravenous formulation is primarily used for general anaesthesia and intensive care settings, the drug’s high lipophilic properties also allow various noninvasive routes of administration. Published data suggest that intranasal administration is also attractive for use as intranasal patient-controlled analgesia (PCA). A newly developed intranasal fentanyl formulation containing 47 μg fentanyl, intravenous fentanyl, and oral transmucosal fentanyl citrate were characterised, and bioavailability was compared to assess the suitability of the intranasal formulation for an intranasal PCA product. Methods. 27 healthy volunteers were enrolled in a single-centre, open-label, randomised (order of treatments), single-dose study in a three-period crossover design. The pharmacokinetics of one intranasal puff of fentanyl formulation (47 μg, 140 mL per puff), one short intravenous infusion of 50 μg fentanyl, and one lozenge with an integrated applicator (200 μg fentanyl) were studied, and bioavailability was calculated. Blood samples were collected over 12 hours, and plasma concentrations of fentanyl were determined by HPLC with MS/MS detection. Results. 24 volunteers completed the study. The geometric mean of AUC0-tlast was the highest with oral transmucosal administration (1106 h  ∗  pg/ml, CV% = 32.86), followed by intravenous (672 h  ∗  pg/ml, CV% = 32.18) and intranasal administration (515 h  ∗  pg/ml, CV% = 30.10). Cmax was 886 pg/ml (CV% = 59.38) for intravenous, 338 pg/ml (CV% = 45.61) for intranasal, and 310 pg/ml (CV% = 29.58) for oral transmucosal administration. tmax was shortest for intravenous administration (0.06 h, SD = 0.056), followed by intranasal (0.21 h, SD = 0.078) and oral transmucosal administration (1.20 h, SD = 0.763). Dose-adjusted absolute bioavailability was determined to be 74.70% for the intranasal formulation and 41.25% for the oral transmucosal product. In total, 38 adverse events (AEs) occurred. Fourteen AEs were potentially related to the investigational items. No serious AE occurred. Conclusion. Pharmacokinetic parameters and bioavailability of the investigated intranasal fentanyl indicated suitability for its intended use as an intranasal PCA option.