*2.3. Materials*

In this study, a 12 μg/h transdermal fentanyl patch (Fentanyl-Mepha Matrixpfl, Mepha Pharma AG, Switzerland) was used. Due to the recommendation of a dosage of 2 μg/kg/h in sheep and up to 5 μg/kg/h in small animals (dogs, cats) described in several former studies, the 12 μg/h patch size was chosen [5,7,22]. The mean administered dose rate of fentanyl was 3.60 μg/kg/h (Dmax = 3.93 μg/kg/h, Dmin = 3.16 μg/kg/h).

### *2.4. Fentanyl Patch Application Process*

Rabbits were sedated with Medetomidin (200 μg/kg) and Midazolam (0.5 mg/kg) both mixed in a syringe and given intramuscularly approximately 15 min before starting skin preparation. Thereafter, the rabbits were placed in sternal recumbency, eye ointment (Vitamin A Blache Augensalbe, Bausch & Lomb Swiss AG, Switzerland) was applied and oxygen (flow rate 1 L/min) was provided via face mask. The required site was carefully clipped using two different clippers depending on the location. For the location on the ear, the clipper Isis (GT420 Aesculap, Germany; cutting length 0.5 mm) was utilized. For the neck, another mechanical fur clipper (model GH703/10 (Favorita II GT104), Aesculap, Germany; cutting length 0.1 mm) was used in the interscapular region due to the increased density of hair at this site. During clipping extreme precautions were taken not to traumatize the application site to avoid influencing fentanyl absorption.

Afterwards, skin was degreased with swabs (Mesoft, Moelnlycke, Sweden) soaked in alcohol (Softasept, B. Braun Vet Care GmbH, Germany) to ensure a good contact of the patch with the skin. While waiting for 5 min to air dry, the ear not used for patch application was clipped and prepared aseptically with alcohol for the placement of a 22 G catheter (Vasofix Safety, B.Braun Meisungen AG, Germany) in the marginal ear vein. The catheter, inserted to facilitate blood withdrawal, was wrapped with tape (Durapore, 3 M (Schweiz) GmbH, Switzerland) and a roll of gauze (Mesoft) to hold it in place. After each blood withdrawal, the catheter was flushed with approximately 0.5 mL of heparinized 0.9% NaCl (B.Braun Meisungen AG, Germany) and a mandrin (Mandrin Vasofix, B.Braun Meisungen AG, Germany) was inserted. On the dry application site, the sticky side of the patch was then pressed on the skin with the palm of the hand and kept there for 1 min to ensure a good patch-skin contact. Finally, the patch was fixed with tape (Leukoplast) to prevent the loosening of the patch detachment during animal handling.

### *2.5. Sample Collection and Plasma Fentanyl Analysis via ELISA*

Venous blood sampling was performed immediately prior to the application of the patch and 3, 6, 9, 12, 18, 24, 36, 48, 72, 96, 120 h thereafter. In a few cases arterial blood was sampled, if no venous sample could be obtained. At each sampling time point, approximately 1 mL of blood was collected in a 1.3 mL EDTA-covered tube (1.3 mL K3E, Sarstedt, AG & Co, Nümbrecht, Germany). A venous catheter was always used to withdraw blood samples. For the first 24 h the catheter was left in the marginal ear vein. Thereafter, a new catheter was inserted for each blood sampling. EDTA-tubes were then centrifuged for 15 min, at 23 ◦C and 1000 rcf (Centrifuge 5810R, Vaudaux-Eppendorf AG, Switzerland) within 2 h of collection.

Subsequently, the obtained plasma was transferred into Eppendorf tubes (Vaudaux-Eppendorf AG, Switzerland) and stored at −80 ◦C until analysis. Plasma fentanyl concentrations were measured by using a commercially available human enzyme-linked immunoabsorbent assay (ELISA). Fentanyl Kits (Adnova, Fentanyl (Human) ELISA Kit, Taiwan) were used according to the manufacturer's instructions. Additionally, serial dilutions of fentanyl standard (Fentanyl Sintetica 0.5 mg/10 mL) in fentanyl negative rabbit plasma were measured on the ELISA-plate, as a standard curve. The absorbance was read at 450 nm using a Mithras microplate reader (Berthold Technologies, Bad Wildbad, Germany). Each sample was tested in duplicates.

### *2.6. Scoring and Practicability*

During the study, the rabbits were scored twice a day. The scoring system included observation of food and water intake, assessment of the general conditions, body weight and rectal temperature as well as defecation and coprophagy. Additionally, practicability was assessed based on the following criteria: ease of preparation, quality of patch adhesiveness, ease of daily checks, occurrence of undesired patch detachment before the end of the study, ease of patch removal and skin condition after patch removal. Any findings in respect to these criteria were noted and at the end a subjective three level scoring system (positive, neutral and negative assessment) comparing the three groups was applied by one veterinarian.
