Refinement of intraperitoneal injection of sodium pentobarbital for euthanasia in laboratory rats (Rattus norvegicus)

BACKGROUND The Canadian Council on Animal Care and American Veterinary Medical Association classify intraperitoneal (IP) pentobarbital as an acceptable euthanasia method in rats. However, national guidelines do not exist for a recommended dose or volume and IP euthanasia has been described as unreliable, with misinjections leading to variable success in ensuring a timely death. The aims of this study were to assess and improve efficacy and consistency of IP euthanasia. In a randomized, blinded study, 51 adult female Sprague-Dawley rats (170-495 g) received one of four treatments: low-dose low-volume (LL) IP pentobarbital (n = 13, 200 mg/kg pentobarbital), low-dose high-volume (LH) IP pentobarbital (n = 14, 200 mg/kg diluted 1:3 with phosphate buffered saline), high-dose high-volume (HH, n = 14, 800 mg/kg pentobarbital), or saline. Times to loss of righting reflex (LORR) and cessation of heartbeat (CHB) were recorded. To identify misinjections, necropsy examinations were performed on all rats. Video recordings of LL and HH groups were analyzed for pain-associated behaviors. Between-group comparisons were performed with 1-way ANOVA and Games-Howell post hoc tests. Variability in CHB was assessed by calculating the coefficient of variation (CV). RESULTS The fastest euthanasia method (CHB) was HH (283.7 ± 38.0 s), compared with LL (485.8 ± 140.7 s, p = 0.002) and LH (347.7 ± 72.0 s, p = 0.039). Values for CV were: HH, 13.4%; LH, 20.7%; LL, 29.0%. LORR time was longest in LL (139.5 ± 29.6 s), compared with HH (111.6 ± 19.7 s, p = 0.046) and LH (104.2 ± 19.3 s, p = 0.01). Misinjections occurred in 17.0% (7/41) of euthanasia attempts. Pain-associated behavior incidence ranged from 36% (4/11, LL) to 46% (5/11, HH). CONCLUSIONS These data illustrate refinement of the IP pentobarbital euthanasia technique. Both dose and volume contribute to speed of death, with a dose of 800 mg/kg (HH) being the most effective method. An increase in volume alone does not significantly reduce variability. The proportion of misinjections was similar to that of previous studies.