Objective

To compare injectate distribution and likelihood of regional anesthesia to the orbit following retrobulbar (RB) or peribulbar (PB) injections in dog cadavers.

Study design

Randomized, masked study.

Animals

Twenty-four dog cadavers (aged 5.5-17 years, 2.0-36.3 kg).

Methods

Orbits underwent one of three injection techniques with bupivacaine 0.5% and iohexol (1:1): ventrolateral RB injection (1-2 mL; 15 orbits), medial canthal PB injection (2-8 mL; PB-1; 16 orbits), or dorsomedial and ventrolateral PB injections (each 1-4 mL; PB-2; 16 orbits). The likelihood of successful regional anesthesia was estimated based on computed tomographic images scored for injectate volume of distribution at the base and within the extraocular muscle cone (EOMC), and injectate distribution around the optic nerve. Intraocular pressure (IOP) was measured before and after injections. Mixed-effects linear regression with post hoc Bonferroni contrast adjustments was performed. Significance was set at 0.05.

Results

A difference in injectate volume of distribution within or at the base of the EOMC was not detected among groups. The median optic nerve circumference of injectate distribution was significantly higher in the RB injected group than in the PB-2 group. Injectate distribution following RB, PB-1 and PB-2 injections was graded as likely to provide regional anesthesia within the EOMC in 40%, 19% and 31% of eyes, and at the EOMC base in 60%, 63% and 50% of eyes, respectively. The probability of likelihood to provide regional anesthesia was lower in dogs of higher body weights. The IOP was significantly higher than baseline following PB-1 (18 ± 14 mmHg) and in comparison with RB (2 ± 3 mmHg), but not different from PB-2 injection (10 ± 11 mmHg).

Conclusions and clinical relevance

None of the techniques reliably produced 'successful' injectate distribution based on this study's definitions; however, clinical assessment of anesthetic success is required.