Clinical and pharmacokinetic effects of regional or general anaesthesia on intravenous regional limb perfusion with amikacin in horses

Abstract

Introduction: Orthopaedic infection in horses is a common and life-threatening condition, which requires early and aggressive treatment. Apart from systemic antimicrobial therapy, local antimicrobial therapy is often added to the treatment regime with the aim to obtain high levels of antimicrobials in the affected areas. This includes intravenous regional limb perfusion (IVRLP) with antimicrobials. Antimicrobial IVRLP consists of the infusion of an antimicrobial agent under pressure into the vascular system of a portion of a limb isolated from the systemic circulation by the use of a proximally placed tourniquet. The infusion can take place via a peripheral blood vessel or the medullary cavity of a bone. Synovial antimicrobial concentrations after IVRLP are often > 50 times the minimum inhibitory concentration (MIC) for clinically-relevant bacterial organisms. Intravenous regional limb perfusion can be performed on the standing, sedated horse or with the horse under general anaesthesia. In an attempt to increase the comfort of the standing horse and decrease movement due to discomfort, a local anaesthetic agent is often administered as perineural regional anaesthesia or intravenous regional anaesthesia. Some authors regard antimicrobial IVRLP under general anaesthesia to be superior to that on the standing sedated horse due to better efficacy of the tourniquet and absence of movement of the animal. This study aimed to evaluate the effects of regional or general anaesthesia on the animal’s comfort and synovial pharmacokinetic parameters of amikacin administered by IVRLP to horses. Materials and methods: Eight healthy horses received 4 treatments of amikacin IVRLP in a randomised, blinded, cross-over design: under standing sedation without regional anaesthesia (CNT), under standing sedation with intravenous regional anaesthesia (IVA), under standing sedation with perineural regional anaesthesia (PNA) or under general anaesthesia (GA). For all treatments, a tourniquet was applied proximal to the carpus and the cephalic vein was used for injection of the perfusate. Synovial fluid amikacin concentrations in the middle carpal joint were measured over 24 hours and the regional synovial pharmacokinetic parameters (rPkp) calculated. Heart and respiratory rates, visual analogue scale (VAS) of discomfort, number of times the limb was lifted (LIFT) and number of additional sedations administered were recorded. ANOVA cross-over analysis was applied with significance level at P < 0.05. Results: One horse was removed from the analysis due to recumbency during IVA and CNT treatments. Amikacin concentrations and rPkp did not differ significantly among treatments. Maximum amikacin concentrations (mean ± CV; μg/mL) in middle carpal joint synovial fluid were 239 ± 0.97, 172 ± 0.55, 344 ± 1.25 and 503 ± 1.26 for CNT, IVA, PNA and GA groups, respectively. Scores of VAS (mean ± SD) were significantly lower with PNA (19 ± 16) versus IVA (69 ± 36) or CNT (81 ± 14). Significantly lower LIFT (mean ± SD) occurred with PNA (20 ± 22) versus CNT (54 ± 24). No horses with PNA treatment required any additional sedations during the procedure, while treatments CNT and IVA required a median (range) of 1.5 (0-3) and 1 (0-4) additional sedations, respectively. Conclusions: This study concluded that PNA was most effective in providing comfort to horses undergoing IVRLP under standing sedation. General or regional anaesthesia with IVRLP did not have any significant effect on synovial amikacin concentrations or rPkp. Relevance: The comfort of horses undergoing standing IVRPL can be increased by performing perineural anaesthesia prior to the treatment. Use of general anaesthesia to improve antimicrobial synovial concentrations during IVRLP is not justified based on this study.