Abstract
Objectives
The aim of this study was to assess retrospectively the efficacy and complication rate of hindlimb peripheral nerve blocks (PNBs) in cats.
Methods
Clinical records of cats that received PNBs and underwent hindlimb orthopaedic surgery from February 2010 to October 2014 were examined. Type of PNB, type and dose of local anaesthetic used, end-expiratory fraction of isoflurane (FE′Iso) administered, additional intraoperative analgesia, incidence of hypotension, postoperative opioid requirement, postoperative contralateral limb paralysis and neurological complications at the 6 week re-examination were investigated.
Results
Eighty-nine records were retrieved but only 69 were analysed. Four combinations of PNBs were used: 34 lateral preiliac (LPI) approach to lumbar plexus (LP) associated with lumbar paravertebral approach to sciatic nerve (SN); 20 LPI–LP associated with the lateral approach to SN; three LPI–LP associated with gluteal approach to SN; 12 dorsal-paravertebral (DPV) approach to LP associated with lateral SN. Levobupivacaine was used for the majority of PNBs. The mean intraoperative FE′Iso was 1.15%; hypotension was documented in 55.1% of anaesthetics, while 31.8% of cats received fentanyl and/or ketamine intraoperatively. Postoperatively, 72.7% of cats received at least one dose of opioid, while five cats required further postoperative analgesia (ketamine constant rate infusion and/or gabapentin). No cats showed contralateral limb paralysis and neurological complications at the 6 week re-examination. No differences were found when comparing the different PNBs used.
Conclusions and relevance
PNBs contributed to perioperative anaesthesia/analgesia in cats undergoing hindlimb orthopaedic surgery. However, the clinical relevance of intraoperative hypotension needs further investigation.
Introduction
Peripheral nerve blocks (PNBs) are routinely used in humans undergoing orthopaedic surgery and are preferred to neuraxial techniques because of their selective action, shorter hospitalisation time, lower incidence of intraoperative hypotension, postoperative urinary retention and incidence of neurological complications. 1 PNBs are popular also in small animal veterinary anaesthesia, especially in dogs undergoing hindlimb orthopaedic surgery.2–9
Similarly to dogs, nerves arising from the lumbosacral plexus, femoral nerve (FN) and sciatic nerve (SN), in particular, innervate the hindlimb in cats. 10 The FN originates from the ventral branches of the fifth and sixth lumbar (L) nerves, which emerge from the intervertebral foraminae of the vertebral bodies of L5 and L6.11,12 The major component of the FN can be identified at the level of L7 vertebral body within the iliopsoas muscle. Running caudally, the FN enters the abdominal cavity and leaves the latter through the muscular lacuna; then, it divides into three branches, of which the saphenous nerve represents the main sensory one. The SN represents the natural caudal continuation of the lumbosacral trunk, arising from L6–L7–S1 and S2. After passing between the ischiatic tuberosity and the greater trochanter, the SN runs distally, innervating the hindlimb as the peroneal and tibial nerves; peroneal nerve fibres originate from L6–L7, while tibial nerve fibres originate from L6–L7–S1, and in some cases S2. 10
While several approaches have been evaluated,2–4,6–9 and a retrospective multicentre study reported minimal perioperative and no postoperative neurological complications in dogs, 5 only two cadaveric studies using ultrasound-guided techniques have been published in cats.11,12 The paucity of published clinical studies evaluating PNBs in the feline species was also evident in a recent review. 13 For this reason this study will describe techniques and retrospectively assess the use of PNBs implemented as part of perioperative analgesia in cats undergoing hindlimb orthopaedic surgery in a referral centre.
Material and methods
The clinical records of cats that underwent hindlimb orthopaedic surgery at Dick White Referrals between February 2010 and October 2014 were reviewed. Of these, records of cats in which a PNB was used were included in the study. The information presented in Table 1 was recorded in a Microsoft Excel spreadsheet. Incomplete files were excluded from the study.
Information extracted and results obtained in cats undergoing hindlimb orthopaedic surgery and in which a peripheral nerve block (PNB) was used as part of perioperative analgesia
Data are reported as mean ± SD or median (range) unless otherwise indicated. Where two values appear within parentheses, the first value represents the median, the second value represents the range (min-max)
FE′IAA = end-expiratory fraction of inhalational anaesthetic agent; ASA = American Society of Anesthesiologists; DSH = domestic shorthair; DLH = domestic longhair; F = female; FS = female spayed; M = male; MN = male neutered; LPI = lateral preiliac; LP = lumbar plexus; DPV = dorsal-paraventral; LPV = lumbar paravertebral; SN = sciatic nerve; NSAIDs = non-steroidal anti-inflammatory drugs; MAP = mean arterial blood pressure
From the incision to the final suturing of the skin, the mean end-expiratory fraction of inhalational anaesthetic agent (FE′IAA) was calculated averaging the readings recorded every 5 mins in each animal. Intraoperative fentanyl (Fentadon [Dechra Veterinary Products] and Sublimaze [Janssen-Cilag]) and/or ketamine (Narketan; Vetoquinol) used to control a nociceptive response – identified as heart rate (HR) or mean arterial blood pressure (MAP) >20% of prestimulus value – were calculated dividing the total amount of each drug administered, measured in µg and mg, respectively, by the duration of surgery (in hours) and then by the body weight of the cat (in kg). Intraoperative hypotension was defined as an MAP – measured using an oscillometric technique – <60 mmHg for more than two consecutive readings obtained at 5 min intervals. Treatments necessary to correct hypotension were grouped as showed in Table 2. Intraoperative success rate was defined as fentanyl consumption <2.1 µg/kg/h.5,14
Incidence and treatment of hypotension, defined as mean arterial blood pressure <60 mmHg for more than two consecutive readings obtained at 5 min intervals, in 38/69 cats undergoing hindlimb orthopaedic surgery and in which a peripheral nerve block (PNB) was used
LPI = lateral preiliac; LP = lumbar plexus; LPV = lumbar paravertebral; SN = sciatic nerve; DPV = dorsal-paravertebral; FE′IAA = end-expiratory fraction of inhalational anaesthetic agent
The amount of methadone (Comfortan [Eurovet Animal Health] and methadone hydrochloride [Martindale Pharma]) and/or buprenorphine (Vetergesic; Alstoe) administered during the first 24 h postoperatively was calculated by dividing the total amount administered, measured in mg, by the cat’s body weight. According to the standard perioperative care in our institution, anaesthetists were responsible for the postoperative analgesic plan: opioids were administered if pain scores were ⩾4/12 using a modified 4AVet scale, 15 or systematically at regular intervals (4–5 h for methadone, 6–8 h for buprenorphine), depending on the animal’s demeanour, severity of pain and type of surgery. Pain assessment was performed at 2 h intervals.
FN and SN were identified using a peripheral nerve stimulator (EZstim II ES400; Life-Tech) applying techniques previously described in dogs (supplementary material). FN was identified within the lumbar plexus (LP) using either a dorsal-paravertebral or a lateral preiliac (LPI) approach.2,8 The SN was identified using one of the following approaches: lateral, gluteal or lumbar paravertebral (LPV).2,3,9 The choice of the PNB used was at anaesthetist’s discretion, owing to the retrospective nature of this study.
Statistical analysis
After normality was assessed using the Kolmogorov–Smirnov test, categorical data were analysed using Fisher’s exact test or the χ2 test, while continuous data were analysed using one-way ANOVA or the Kruskal–Wallis test and the Student’s t-test or the Mann–Whitney U-test, depending on normality and number of groups to be compared (SigmaStat 3.5). A value of P <0.05 was considered to be statistically significant. Results are reported as mean ± SD or median (range).
Results
Eighty-nine records of cats that underwent hindlimb orthopaedic surgery, and in which a PNB was performed preoperatively, were retrieved; however, only 69 files were sufficiently complete to be included in this study. The results of the descriptive statistical analysis are reported in Table 1.
Four combinations of PNBs were used to block the FN and SN: (1) LPI approach to LP associated with LPV approach to SN (LPI–LP + LPV–SN); (2) LPI–LP associated with the lateral approach to SN (LPI–LP + lateral SN); (3) LPI–LP associated with gluteal approach to SN (LPI–LP + gluteal SN); (4) DPV approach to LP associated with lateral SN (DPV–LP + lateral SN).
Methadone (0.2 mg/kg IV or IM), buprenorphine (0.02 mg/kg IV or IM), dexmedetomidine (2 µg/kg IV or 5 µg/kg IM [Dexdomitor; Orion Pharma]), midazolam (0.3 mg/kg IM [Hypnovel; Roche]) and ketamine (3 mg/kg IM) were used in different combination as preanaesthetic medication (Tables 1–3). The most performed surgeries were: tibial fracture repair (n = 20), tarsal-calcaneal fracture/dislocation repair (n = 13), medial patella luxation repair (n = 11), femoral fracture (n = 10) and cranial cruciate ligament (n = 8) repair. Surgeries were grouped according to the anatomical location (proximal to stifle, stifle and distal to the stifle) for the purpose of statistical analysis (Table 1).
Information extracted and results obtained in cats undergoing hindlimb orthopaedic surgery and in which a peripheral nerve block (PNB) was used as part of perioperative analgesia
Data are reported as mean ± SD or median (range) unless otherwise indicated. Three cats in which the LPI–LP was associated with the gluteal approach to the SN were not considered. Three records of postoperative analgesia administration were missing. Where two values appear within parentheses, the first value represents the median, the second value represents the range (min–max)
P <0.001
P <0.05
0.01 > P >0.001
LPI = lateral preiliac; LP = lumbar plexus; LPV = lumbar paravertebral; SN = sciatic nerve; DPV = dorsal paravertebral; LA = local anaesthetic; FE′IAA = end-expiratory fraction of inhalational anaesthetic agent; ASA = American Society of Anesthesiologists; DSH = domestic shorthair; DLH = domestic longhair; F = female; FS = female spayed; M = male; MN = male neutered; NSAIDs = non-steroidal anti-inflammatory drugs; MAP = mean arterial blood pressure
Comparison between different PNBs was performed; however, LPI–LP + gluteal SN was not considered because the low number of cats (n = 3). Groups were homogeneous for breed, sex, age, body weight and American Society of Anesthesiologists’ status, but not for body condition score (statistically higher in DPV–LP + lateral SN; P = 0.001) and area of surgery (P <0.001; Table 3). While levobupivacaine was used in all cats belonging to the LPI–LP + LPV–SN and LPI–LP + lateral SN groups, bupivacaine and ropivacaine were mainly used in the DPV-LP + lateral SN group. No difference in levobupivacaine dose was found between the LPI–LP + LPV–SN and LPI–LP + lateral SN groups. Isoflurane was used in all cats but two belonging to the LPI–LP + lateral SN and DPV–LP + lateral SN groups. No difference in FE′IAA was found between groups. Similarly, the number of cats in which fentanyl or ketamine were used to control nociception was not different (Table 3). The intraoperative success rate was 81.2%.
Intraoperative hypotension was recorded in 55.1% of cats and was similar between groups (Table 3); it was treated decreasing FE′IAA, administering an antimuscarinic when HR <120 beats per min (glycopyrrolate 10 µg/kg IV [Glycopyrronium Bromide; Martindale Pharmaceuticals]) and/or vasopressors (ephedrine 0.1 mg/kg IV [Ephedrine Hydrochloride; Martindale Pharmaceuticals] or dopamine 5–10 µg/kg/min IV [Dopamine; Hospira UK]) and/or fluid boluses (crystalloid 10 mg/kg or colloid 2 mg/kg), as shown in Table 2. No difference in MAP treatment was found between groups; treatments were effective in all cases.
Postoperatively, 95.6% of cats received non-steroidal anti-inflammatory drugs (NSAIDs) and 72.7% of them at least one dose of opioid. The postoperative requirement of methadone and buprenorphine, and the number of animals in which opioids were administered systematically, was similar (Table 3). Five cats required further postoperative analgesia: two cats (LPI–LP + LPV–SN and LPI–LP + lateral SN) had ketamine constant rate infusion (CRI); two cats had gabapentin (LPI–LP + LPV-SN); one cat required both drugs.
Neither postoperative contralateral paralysis at 1 h after tracheal extubation nor neurological complications at 6 week re-examination were recorded in any cat.
Discussion
This is the first clinical study evaluating the use of PNBs in cats. The recorded FE′IAA and the low requirement of antinociceptive drugs together with the absence of postoperative neurological complications support the use of PNB as part of perioperative anaesthesia/analgesia in cats undergoing hindlimb orthopaedic surgery. Further, we demonstrated that PNB techniques previously described in dogs are effective also in cats.
The FE′IAA sparing effect can reflect the efficacy of PNBs used to provide antinociception. However, other important factors, such as preanaesthetic medications used, presence of hypothermia, hypotension and its treatment, use of local anaesthetic and their systemic absorption, administration of further antinociceptive drugs, etc, need to be considered, complicating this evaluation. While the enrolment of a control group would facilitate the interpretation of the results, this was not possible in this study as regional anaesthesia (PNB or neuraxial) was routinely used in our institution before 2010; further, it was deemed questionable to enrol a control group prospectively just for the benefit of this study. For this reason, and according to the 3Rs concept, 16 it could be useful to compare our results with a study in which the effect of a brachial plexus block on FE′Iso in cats undergoing orthopaedic surgery was prospectively clinically evaluated. 14 In that study, a combination ketamine (3 mg/kg IV) and midazolam (0.3 mg/kg IV) was used as preanaesthetic medication, and fentanyl CRI (2 µg/kg/h) was administered to provide further antinociception. The FE′Iso necessary to maintain anaesthesia and administered according to changes of clinical signs – HR, respiratory rate and arterial blood pressure – ranged between 1.5% and 1.7% in the control group and between 1.0% and 1.1% in the PNB group. Two cats in the control group received additional fentanyl boluses. Our results agree with those obtained by Mosing et al: 14 the block of the FN and SN can be used to provide antinociception and produce an FE′IAA sparing effect measurable in a clinical setting in cats undergoing orthopaedic hindlimb surgery.
In dogs undergoing hindlimb orthopaedic surgery and in which LPI–LP + LPV–SN was performed, the mean FE′IAA of isoflurane used to maintain anaesthesia was 1.07%. 9 Using the same PNB technique, the mean FE′IAA of isoflurane was 1.12% in cats. Despite no attempt being made to find the minimum FE′IAA to prevent movement or sympathetic response to surgery, but considering isoflurane’s different potency between the two species, 17 it appears that PNBs allow the maintenance of anaesthesia with relative lower FE′IAA in cats, and this sparing effect is confirmed by comparing our results with those of Mosing et al. 14 Ketamine, used as part of preanaesthetic medication in 8/69 cats, but not in dogs, might have contributed to further reducing FE′IAA. However, even excluding those cats, the mean FE′IAA was 1.1%, reflecting a significant sparing effect. However, as mentioned above, the effect of other variables affecting the FE′IAA sparing effect should be also considered before drawing any firm conclusion. However, as no differences were found comparing the wide range of preanaesthetic medications used, cats’ body temperature at tracheal extubation, incidence of hypotension and its treatment between the three groups reported herein, it is possible to assert that all three combination of PNB techniques described here were similarly effective in control the nociceptive response during surgery.
Efficacy of a PNB depends on the volume of local anaesthetic administered and the precision – distance between tip of the needle and nerve – of the injection. 18 The mean dose of levobupivacaine used in this study was greater than the 1.63 mg/kg used in dogs to perform LPI–LP, 9 and might partially explain the higher intraoperative success rate. According to Coulomb’s law, the smaller the distance between tip of the needle and nerve is, the lower is the intensity of the current, measured in mA, required to stimulate that nerve, if the tip does not contact the epineurium or has not been positioned intraneurally; 19 in order to increase the success rate of a PNB, it has been suggested to inject local anaesthetic only when muscle contraction is maintained using a current threshold 0.2 <mA <0.6. 18 In our study, local anaesthetic was injected when the muscle contraction was maintained using a median mA of 0.4; unfortunately, Vettorato et al did not specify the exact value of mA at injection, but it was stated that injections were performed when 0.2 <mA <0.6. 9 Therefore, it is possible that nerves were also more precisely identified in cats. Injecting a larger volume of local anaesthetic closer to the nerve may have resulted in a greater intraoperative success rate.
Recently, the intraoperative success rate of PNBs in dogs was reported to be 77%. 5 Using the same criteria, the overall success rate of LPI–LP + LPV–SN and DPV–LP + lateral SN in dogs was 78.5%. 9 In our study, fentanyl was not the only antinociceptive drug used; thus comparison with the results obtained in dogs is more complicated; therefore, we decided to consider cases in which ketamine was administered as failure of PNB even if this may not have been the case. According to the definition proposed by Vettorato et al, 5 PNBs could be considered successful in 56/69 cats (81.2%), suggesting a similar intraoperative success rate between dogs and cats. Further, 47/69 cats (68.1%) did not require the administration of antinociceptive drugs. This result is similar to that obtained in dogs: 67.1% and 52.1% of animals in which LPI–LP + LPV–SN and DPV–LP + lateral SN were used did not receive antinociceptive medications, respectively. 9 Nevertheless, the retrospective nature of this study, the use of the several combinations of preanaesthetic medications, differences in body temperature, incidence, severity and treatment of hypotension may have affected the results obtained and their interpretation.
The incidence of hypotension under anaesthesia has been reported as 7.0% and 8.5% for dogs and cats, respectively. 20 Similarly, the incidence of hypotension reported in dogs undergoing orthopaedic surgery, and in which a PNB was used, was 7.8%. 5 In particular, it was 10.4% and 5.7% when LPI–LP + LPV–SN and DPV–LP + lateral SN were used, respectively. 9 In our study, hypotension was significantly more frequent despite a higher reduction of FE′IAA: >50% of cats were considered hypotensive but no difference was found comparing the three groups. Similarly, the incidence of hypotension – defined as value <80 mmHg and measured using a Doppler technique – in cats receiving a brachial plexus block was 30%. 14 This result might reflect a physiological difference of anaesthetised cats compared with dogs, the accuracy of the blood pressure monitoring equipment, the incidence of relative bradycardia associated with the use of an α2 agonist and could reflect the overall higher mortality rate of cats. 21 Alternatively, it is possible that the local anaesthetic absorbed systemically caused some degree of cardiovascular depression, as well as potentially reducing anaesthetic requirements. The depressant effect of intravenous lidocaine has been documented; 22 it would be expected that levobupivacaine might cause a greater depressant effect, although the extent of intravascular absorption can only be speculated. Finally, the difference may simply reflect the fact that the population of cats considered in this study was different from the one of dogs and cats previously reported.5,9,21 While the majority of dogs were healthy and underwent elective tibial plateau level osteotomy, the majority of cats required surgery to reduce fractures following a trauma. As cats are generally more difficult to assess clinically, it is possible that dehydration and blood loss concurrent to the trauma had not been adequately estimated and replaced prior to surgery, exacerbating the development of intraoperative hypotension. Further, the presence of a subclinical cardiomyopathy and/or myocardial contusion following the trauma may have gone unnoticed and resulted in hypotension once the cat was anaesthetised. Nevertheless, despite several approaches having been used to correct hypotension, no differences were found between groups.
Postoperative methadone was not required in 72.9% and 66.3% of dogs in which LPI–LP + LPV–SN and DPV–LP + lateral SN were performed, respectively. 9 In those dogs, interns and kennel nurses assessed pain every 2 h for the first 24 h postoperatively using a validated pain scoring system. 23 Unfortunately, we were able to monitor the postoperative analgesic effect of PNBs only in 44/66 cats enrolled as methadone and/or buprenorphine were administered systematically in the remaining 22 owing to the extension of the trauma or poly-trauma, or because of the cat’s demeanour; 18/44 (40.9%) did not require additional opioid analgesia. The overall median dose of methadone and buprenorphine administered was 0.60 and 0.05 mg/kg/24 h, respectively; no difference in opioid consumption was found comparing the three groups. This is higher than the median postoperative methadone consumption (0.21 mg/kg/24 h) after either LPI–LP + LPV–SN or DPV–LP + lateral SN in dogs. 9 The difference could be explained by the poor specificity of the method used to assess postoperative pain and the general difficulty to assess pain in cats, the different species or, as previously hypothesised, by the presence of a more substantial trauma in cats.
Postoperative pain was insufficiently controlled with opioids and NSAIDs in five cats, and therefore further analgesic drugs were administered. While in three cats (two belonging to the LPI–LP + LPV-SN group and one belonging to the LPI–LP + lateral SN group) this was expected owing to the extension and severity of the trauma, in the remaining two cases (belonging to the LPI–LP + LPV–SN and LPI–LP + gluteal SN groups) this was unexpected. Both cats underwent surgery to repair a tibiotarsal fracture; dexmedetomidine combined either with buprenorphine or methadone was used as preanaesthetic medication, and anaesthesia lasted 180 and 210 mins, respectively. Despite PNB being considered successful intraoperatively, as mean FE′IAA was 0.75% and 1.00%, respectively, and no additional antinociceptive drugs were required, pain manifested shortly after tracheal extubation: gentle touching of the cervical-thoracolumbar area and both hindlimbs was consistently eliciting an aggressive response. Further, cats were in alternating states of rest and sleepiness with signs of breakthrough, unbearable pain exhibited as yowling and leaping. It is our opinion that those animals may have been experiencing central sensitisation in the dorsal horn of the spinal cord: continuous nociceptive inputs caused by the trauma and/or nerve damage may induce temporary or permanent changes in the dorsal horn of the spinal cord, leading to hyperalgesia and allodynia. 24 Similar cases have been previously described in cats,25,26 and were successfully treated with gabapentin. Therefore, gabapentin (10 mg/kg PO q8h) was administered and resulted in a dramatic improvement of the animal analgesic status. Neither cat showed any signs of neuropathic pain or neurological deficit at the 6 weeks re-examination.
Extradural spread has been described as a possible complication of FN block using DPV–LP in dog cadavers. 2 Clinically, temporary postoperative paraplegia associated to extradural spread has been reported in 1/95 dog in which DPV–LP was used. 5 Further, 3/96 dogs in which LPI-LP + LPV–SN was used were paraplegic after surgery, 9 even if Portela et al did not identify staining at the level of the intervertebral foramen after LPI–LP in their experimental study. 8 No cats in this study showed postoperative bilateral paraplegia; however, considering the analogy between the neuroanatomy of dogs and cats, 10 and the use of similar PNB techniques, extradural diffusion should also be considered as a potential complication of DPV–LPI or LPI–LP + LPV–SN in the feline species, and therefore monitored postoperatively.
In humans, the reported incidence of temporary postoperative paraesthesia following PNB is 15%, 27 while long-term neurological complications were identified in 2/1065 (0.19%) patients. 28 In a multicentre study performed on 265 dogs no neurological complications were detected postoperatively. 5 Similarly, no cats showed neurological deficits at the 6 weeks re-examination, confirming the low risk of neurological complications following PNB of the hindlimb in small animals. However, the number of animals included, the retrospective nature of the study and the difficulty of recognising mild neurological deficits must be considered when interpreting this finding.
Conclusions
Considering the intraoperative FE′IAA sparing effect observed, the low intraoperative requirement of fentanyl and ketamine and the absence of neurological complications, PNBs should be considered as part of perioperative anaesthesia/analgesia in cats undergoing hindlimb orthopaedic surgery.
Supplemental Material
Click here for Supplementary Material
Description of techniques used to block the femoral nerve and the sciatic nerve in cats
Footnotes
Supplementary material
Description of techniques used to block the femoral nerve and the sciatic nerve in cats.
Conflict of interest
The authors do not have any potential conflicts of interest to declare.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
