Comparison of pain on injection during induction of anaesthesia with alfaxalone and two formulations of propofol in dogs

Objective To compare the incidence of pain during injection of three intravenous induction agents in dogs. Study design Prospective, crossover, randomized, blinded, clinical study. Animals Thirty dogs requiring anaesthesia for radiotherapy. Methods Dogs were anaesthetized on three occasions at weekly intervals. An IV cephalic catheter was placed, flushed with saline and alfentanil 0.01 mg kg-1 and atropine 0.02 mg kg-1 administered. After 30 seconds either: propofol lipid macroemulsion (DrugP), propofol lipid-free microemulsion (DrugPC) or alfaxalone (DrugA) was administered over 60 seconds. Each induction agent was administered once to each dog. Induction was recorded by video and reviewed by an assessor, unaware of treatment. Catheter placement (number of attempts, site, size and recent vein use) were recorded. Behavioural changes associated with pain or excitation, were recorded. Severity of pain on injection was recorded (mild, moderate or severe pain). Incidence of pain was analysed using logistic regression, excitation using McNemars test (p < 0.05) and association of pain with induction agent and catheter placement using the Akaike Information Criterion (AIC). Results No dogs reacted to saline or DrugA, thus DrugA was excluded from analysis. Pain on injection occurred in six dogs (20%) with DrugPC and one dog (3.3%) with DrugP. Pain was severe in four dogs with DrugPC. DrugP resulted in a trend for reduced risk of pain compared to DrugPC (p = 0.076, odds ratio [confidence intervals] 0.14 [0.0270.86]). Both propofol formulations resulted in greater risk of excitation than DrugA (p = 0.0003, odds ratio 4.5 [1.8610.90]). Induction agent was associated with pain, whilst catheter placement was not. One dog developed facial oedema and one other dog skin necrosis adjacent to the catheter site following DrugPC. The study was terminated early due to ethical concerns about the severity of reactions with DrugPC. Conclusions and Clinical relevance DrugPC was associated with clinically relevant moderate to severe pain behaviour whilst DrugA and DrugP were not.