Advertisement

Effects of perineural dexmedetomidine combined with ropivacaine on postoperative methadone requirements in dogs after tibial plateau levelling osteotomy: a two-centre study

Published:January 29, 2022DOI:https://doi.org/10.1016/j.vaa.2022.01.004

      Abstract

      Objective

      To evaluate the efficacy of a perineural injection of dexmedetomidine combined with ropivacaine for reducing postoperative methadone requirements in dogs after tibial plateau levelling osteotomy (TPLO).

      Study design

      A prospective, clinical, randomized and blinded trial.

      Animals

      A total of 58 client-owned dogs.

      Methods

      Ultrasound-guided midfemoral sciatic and inguinal femoral nerve blocks with ropivacaine (1 mg kg–1 per nerve block) combined with either dexmedetomidine (0.5 μg kg–1 per nerve block; group DEX) or the same volume of saline (group CON) were performed in dogs undergoing TPLO. Pain was assessed 30 minutes, 2 hours and then every 4 hours for 24 hours after surgery with a validated pain scale (4AVet). Meloxicam (0.15 mg kg–1) was administered intravenously (IV) at recovery. Rescue methadone (0.2 mg kg–1 IV) was administered if a score ≥ 6 (maximal score 18) was recorded and the number of postoperative doses was analysed by Fisher exact tests. The study was performed in parallel at a Veterinary Teaching Hospital (VTH) and a private Veterinary Referral Centre (VRC).

      Results

      Dogs received a total of 22 and 31 postoperative doses of methadone in groups DEX (14 doses at VRC, eight doses at VTH) and CON (28 doses at VRC, three doses at VTH), respectively. Overall, there was no difference in the postoperative rescue analgesia requirements between groups (p = 0.244). At the VRC, dogs received less methadone (p = 0.026) in group DEX compared with group CON, whereas at the VTH, there was no difference between groups (p = 0.216).

      Conclusions and clinical relevance

      Perineural dexmedetomidine combined with ropivacaine did not reduce postoperative methadone requirements in dogs after TPLO, but results may differ from one centre to another. This discrepancy might be linked to variations in clinical practices and questions the validity of results obtained from single-centre randomized controlled trials but applied to different clinical settings.

      Keywords

      Introduction

      Tibial plateau levelling osteotomy (TPLO) is an invasive stifle surgery that includes osteotomy and bone plate application in dogs. This surgical procedure is painful, and effective perioperative analgesia is mandatory (
      • Hoelzler M.G.
      • Harvey R.C.
      • Lidbetter D.A.
      • Millis D.L.
      Comparison of perioperative analgesic protocols for dogs undergoing tibial plateau leveling osteotomy.
      ). Locoregional techniques, such as peripheral nerve blocks, provide better perioperative pain relief and overall veterinary patient comfort than systemic analgesics (
      • Liu S.S.
      • Strodtbeck W.M.
      • Richman J.M.
      • Wu C.L.
      A comparison of regional versus general anesthesia for ambulatory anesthesia: a meta-analysis of randomized controlled trials.
      ;
      • Liu S.S.
      • Wu C.L.
      The effect of analgesic technique on postoperative patient-reported outcomes including analgesia : a systematic review.
      ;
      • Palomba N.
      • Vettorato E.
      • De Gennaro C.
      • Corletto F.
      Peripheral nerve block versus systemic analgesia in dogs undergoing tibial plateau levelling osteotomy: Analgesic efficacy and pharmacoeconomics comparison.
      ); therefore, they have previously been recommended for these surgeries (
      • Hoelzler M.G.
      • Harvey R.C.
      • Lidbetter D.A.
      • Millis D.L.
      Comparison of perioperative analgesic protocols for dogs undergoing tibial plateau leveling osteotomy.
      ;
      • Vettorato E.
      • Bradbrook C.
      • Gurney M.
      • et al.
      Peripheral nerve blocks of the pelvic limb in dogs : A retrospective clinical study.
      ). Locoregional analgesia has the potential to reduce postoperative opioid consumption and typical opioid-related side effects (
      • Campoy L.
      • Martin-flores M.
      • Ludders J.W.
      • et al.
      Comparison of bupivacaine femoral and sciatic nerve block versus bupivacaine and morphine epidural for stifle surgery in dogs.
      ;
      • Bini G.
      • Vettorato E.
      • De Gennaro C.
      • Corletto F.
      A retrospective comparison of two analgesic strategies after uncomplicated tibial plateau levelling osteotomy in dogs.
      ).
      Unfortunately, most of the reported advantages of perineural blocks are short-lived. A single perineural injection of dexmedetomidine, a potent α2-adrenoceptor agonist, combined with a long-acting local anaesthetic agent (LAA), significantly prolongs peripheral nerve block compared with LAA alone (
      • Ping Y.
      • Ye Q.
      • Wang W.
      • et al.
      Dexmedetomidine as an adjuvant to local anesthetics in brachial plexus blocks: A meta-analysis of randomized controlled trials.
      ;
      • Vorobeichik L.
      • Brull R.
      • Abdallah F.W.
      Evidence basis for using perineural dexmedetomidine to enhance the quality of brachial plexus nerve blocks : a systematic review and meta-analysis of randomized controlled trials.
      ). Clinically relevant side effects of dexmedetomidine such as bradycardia or hypotension can be minimized by identification of the minimally effective dose (
      • Vorobeichik L.
      • Brull R.
      • Abdallah F.W.
      Evidence basis for using perineural dexmedetomidine to enhance the quality of brachial plexus nerve blocks : a systematic review and meta-analysis of randomized controlled trials.
      ;
      • Jung H.S.
      • Seo K.H.
      • Kang J.H.
      • et al.
      Optimal dose of perineural dexmedetomidine for interscalene brachial plexus block to control postoperative pain in patients undergoing arthroscopic shoulder surgery: A prospective, double-blind, randomized controlled study.
      ). Dexmedetomidine combined with LAA prolong postoperative pain relief which reduces postoperative opioid consumption and improves the patient’s comfort and well-being after surgery (
      • Abdallah F.W.
      • Brull R.
      Facilitatory effects of perineural dexmedetomidine on neuraxial and peripheral nerve block : a systematic review and meta-analysis.
      ;
      • El-Boghdadly K.
      • Brull R.
      • Sehmbi H.
      • Abdallah F.W.
      Perineural dexmedetomidine is more effective than clonidine when added to local anesthetic for supraclavicular brachial plexus block: a systematic review and meta-analysis.
      ;
      • Vorobeichik L.
      • Brull R.
      • Abdallah F.W.
      Evidence basis for using perineural dexmedetomidine to enhance the quality of brachial plexus nerve blocks : a systematic review and meta-analysis of randomized controlled trials.
      ;
      • Schnabel A.
      • Reichl S.U.
      • Weibel S.
      • et al.
      Efficacy and safety of dexmedetomidine in peripheral nerve blocks A meta-analysis and trial sequential analysis.
      ).
      A limited number of studies have investigated the effect of dexmedetomidine combined with LAA for peripheral nerve blocks in dogs (
      • Bartel A.K.
      • Campoy L.
      • Martin-Flores M.
      • et al.
      Comparison of bupivacaine and dexmedetomidine femoral and sciatic nerve blocks with bupivacaine and buprenorphine epidural injection for stifle arthroplasty in dogs.
      ;
      • Trein T.A.
      • Floriano B.P.
      • Wagatsuma J.T.
      • et al.
      Effects of dexmedetomidine combined with ropivacaine on sciatic and femoral nerve blockade in dogs.
      ). Recently, perineural dexmedetomidine at 1 μg kg–1 combined with ropivacaine 0.5% significantly prolonged the duration of sensory block in dogs (
      • Marolf V.
      • Ida K.K.
      • Siluk D.
      • et al.
      Effects of perineural administration of ropivacaine combined with perineural or intravenous administration of dexmedetomidine for sciatic and saphenous nerve blocks in dogs.
      ). Perineural dexmedetomidine combined with lidocaine effectively prolongs sciatic and femoral nerve blocks in canine patients (
      • Acquafredda C.
      • Stabile M.
      • Lacitignola L.
      • et al.
      Clinical efficacy of dexmedetomidine combined with lidocaine for femoral and sciatic nerve blocks in dogs undergoing stifle surgery.
      ), but the clinical efficacy of dexmedetomidine combined with ropivacaine remains unknown.
      Multicentre studies have the potential to reinforce the reproducibility of data generated and increase the level of scientific evidence. However, two-centre studies may be undertaken as a first step (
      • Petroz G.C.
      • Sikich N.
      • James M.
      • et al.
      A phase I, two-center study of the pharmacokinetics and pharmacodynamics of dexmedetomidine in children.
      ). This type of study has the potential to verify or refute a hypothesis at two different institutions and generate useful data for further multicentre studies (
      • Reid J.
      • Nolan A.M.
      • Hughes J.M.
      • et al.
      Development of the short-form Glasgow Composite Measure Pain Scale (CMPS-SF) and derivation of an analgesic intervention score.
      ).
      The purpose of this study was to investigate the analgesic effects of perineural dexmedetomidine combined with ropivacaine, injected preoperatively at the femoral and sciatic nerves in dogs undergoing TPLO. Primary outcome was the dose of rescue methadone needed to control pain during the first 24 postoperative hours. Secondary outcomes included the intraoperative dose of fentanyl administered and the degree of postoperative sedation and proprioceptive deficits. The working hypothesis was that dogs administered perineural ropivacaine with dexmedetomidine would require less postoperative methadone than dogs administered perineural ropivacaine with saline. Furthermore, it was hypothesized that similar results would be obtained in two different investigation centres.

      Materials and methods

      The study was designed as a prospective, clinical, randomized and blinded trial. The 2010 CONSORT guidelines for transparent reporting of randomized controlled trials (RCTs) have been applied. The University Teaching Hospital (UTH) of Liège in Belgium and the Veterinary Referral Centre (VRC) Medi-Vet SA in Switzerland participated in the study. Ethical approvals were obtained from the competent Belgian authorities (Nr. 2138) for UTH and from the Swiss authority (Nr. 29685, VD3493) for VRC. A signed owner informed consent was collected prior to study participation in both locations.

      Animals

      A total of 30 cases planned for TPLO surgery per centre were recruited. Dogs were considered eligible for study participation if: they were 1–15 years old; weighted 10–50 kg; were classified as American Society of Anesthesiologist physical status score I or II and could be hospitalized for 24 hours after surgery. Exclusion criteria included: invasive reconstructive stifle surgery involving other techniques than TPLO; body condition score ≥ 8 based on the Nestlé Purina Score [https://www.morrisanimalfoundation.org/sites/default/files/filesync/Purina-Body-Condition-System.pdf, (last accessed 17 January 2022)]; any contraindication for locoregional anaesthesia; any contraindication for the use of ropivacaine, dexmedetomidine or meloxicam; aggressive dogs; the presence of skin infection at the injection site for the sciatic or femoral nerve block; owner refusal; or neuromuscular disorders. Dogs were classified as small dogs (≤20 kg) or large dogs (>20 kg).
      Dogs were randomly assigned to two different groups (30 cases at each centre) by drawing a lot from an envelope containing an equal number of papers assigning the dogs to each group. The dexmedetomidine group (DEX) received an ultrasound (US)-guided perineural injection of ropivacaine combined with dexmedetomidine, while the control group (CON) received a US-guided perineural injection of ropivacaine combined with saline 0.9% at the sciatic and femoral nerves.

      Anaesthesia

      The left or right cephalic vein was catheterized, and dogs were premedicated with acepromazine 0.01 mg kg–1 (Prequillan; Arovet AG, Switzerland) and methadone 0.2 mg kg–1 (Methadone; Streuli Pharma AG, Switzerland) administered intravenously (IV). Approximately 10 minutes later, anaesthesia was induced with propofol (Propofol MCT; Fresenius Kabi AG, Switzerland) IV titrated to effect until the anaesthetic plane was sufficient to allow orotracheal intubation with an appropriately sized cuffed endotracheal tube (PVC Endotracheal Tube with Cuff; Kruuse, Denmark). The tube was connected to a circle breathing system (22 mm Compact Extendable breathing system; Intersurgical, Switzerland). The fresh gas flow of 90–100% oxygen was set at 1 L minute–1 and anaesthesia was maintained with an end-tidal concentration of isoflurane (Attane Isoflurane; Provet AG, Switzerland) targeted at 1.3%. Lactated Ringer’s solution (Ringer-Lactat; Fresenius Kabi AG) was administered IV at a rate of 5 mL kg–1 hour–1. A multiparameter monitor including electrocardiogram, pulse oximetry, noninvasive blood pressure, capnography, anaesthetic gas analysis and temperature measurement was used to assess cardiopulmonary variables of the dogs during anaesthesia. Blood pressure was measured by a cuff size of 40% of limb circumference placed around the thoracic limb and measurements were cycled every 3 minutes. Hypotension, defined as mean arterial pressure (MAP) < 60 mmHg, was treated by reduction of end-tidal isoflurane by 0.1% every 5 minutes according to clinical anaesthetic plane. If this was insufficient to restore normotension, a bolus of lactated Ringer’s solution of 10 mL kg–1 was administered over 20 minutes IV. Finally, dobutamine (Dobutrex; Teva Pharma AG, Switzerland) at 5 μg kg–1 minute–1 IV was administered, if necessary, to restore normotension. Bradycardia was defined as heart rate (HR) < 40 beats minute–1. Bradycardia was treated by administration of IV atropine (Atropine Sulfate; Amino AG, Switzerland) at 20 μg kg–1. Bradycardic and hypotensive events during surgery were recorded. HR, respiratory rate (fR) and blood pressure were recorded as baseline value before the start of surgery. Any increase of > 25% in HR, MAP or fR of the recorded baseline value was considered indicative of nociception and led to the intraoperative administration of fentanyl at 2 μg kg–1 IV. Any further increase of HR, MAP or fR 10 minutes after the first dose of fentanyl would lead to a second dose of fentanyl being administered, and so on. The total dose of fentanyl administered during surgery was recorded. Meloxicam (Inflacam; Virbac AG, Switzerland) 0.15 mg kg–1 was administered IV shortly after removal of the endotracheal tube.

      Surgery

      At the VRC, dogs were operated on by three different residency trained surgeons [two of whom were Diplomates of the European College of Veterinary Surgeons (Dipl. ECVS)] with at least 3 years’ experience as independent surgeons. The canine stifle joint was instrumented with a three-portal method. The arthroscope was inserted laterally to the patellar ligament, the cannula was inserted proximally in the medial joint compartment, and the instrument was inserted medially. The surgical procedure for TPLO followed a standard approach (
      • Slocum B.
      • Slocum T.D.
      Tibial plateau leveling osteotomy for repair of cranial cruciate ligament rupture in the canine.
      ). The type of procedure (arthrotomy versus arthroscopy) performed before TPLO was documented. The presence of meniscal tear lesions was recorded as present or absent during arthroscopy or arthrotomy by the operating surgeon. At the UTH, dogs were operated on by two different board-certified (Dipl. ECVS) experienced surgeons or by a final year surgery resident (PP) under direct supervision of a board-certified surgeon. At the UTH, the joint capsule was not opened prior to TPLO and arthroscopy/arthrotomy were not performed. TPLO surgery followed the same standard procedure as at the VRC.

      Locoregional anaesthesia/analgesia

      A sciatic and femoral US-guided nerve block was performed preoperatively using the approach described by
      • Campoy L.
      • Bezuidenhout A.J.
      • Gleed R.D.
      • et al.
      Ultrasound-guided approach for axillary brachial plexus, femoral nerve, and sciatic nerve blocks in dogs.
      . The nerve stimulator was used to confirm correct needle placement if necessary. Nerve blocks were performed by two different anaesthesiologists familiar with US guidance (UTH: DP-50Vet; Mindray, Germany; VRC: Wireless US-probe B024; Konted, Beijing, China) at each centre (AT UTH, VM at VRC). A diluted preparation of dexmedetomidine was reconstituted for dogs of that group. A volume of 0.1 mL of dexmedetomidine 0.05% (Dexdomitor; Provet AG) was diluted with 0.9 mL of NaCl 0.9% in a 1 mL syringe. Dogs in group DEX received a perineural sciatic injection of 0.2 mL kg–1 of 0.5% ropivacaine (Ropivacain; Fresenius Kabi AG) combined with 0.01 mL kg–1 of diluted dexmedetomidine (equals 0.5 μg kg–1 of dexmedetomidine). The same volume was injected at the femoral nerve. Dogs in group CON received a perineural injection of 0.2 mL kg–1 0.5% ropivacaine combined with 0.01 mL kg–1 of NaCl 0.9% at the sciatic nerve. The same dose and volume were injected at the femoral nerve. After perineural injection, the volume of the extension line and needle was flushed with NaCl 0.9%.

      Postoperative data collection

      At 30 minutes (T0.5), 2 hours (T2) and then every 4 hours (T4, T8, T12, T16, T20, T24) for 24 hours after recovery, the 4Avet pain score (
      • Holopherne-Doran D.
      • Laboissière B.
      • Cogny M.
      Validation of the 4AVet postoperative pain scale in dogs and cats.
      ) was used to assess pain. When the pain score was ≥ 6 out of 18, 0.2 mg kg–1 of methadone was administered IV. The number of methadone doses administered were recorded and compared between groups. At the same evaluation time points, sedation and proprioception were also scored. The sedation score ranged from 0 to 3 (score 0: fully alert and able to stand and walk; score 1: alert and able to maintain sternal recumbency but unable to walk; score 2: drowsy and able to maintain sternal recumbency but unable to stand; and score 3: fast asleep;
      • Campoy L.
      • Martin-flores M.
      • Ludders J.W.
      • et al.
      Comparison of bupivacaine femoral and sciatic nerve block versus bupivacaine and morphine epidural for stifle surgery in dogs.
      ). Proprioception was assessed by positioning the dorsal part of the digits of the anaesthetized limb on the floor to verify the immediate reposition (knuckling reflex). The dogs were not systematically supported during the evaluation. The following scores were used: score 1: absent reposition of the leg; score 2: retarded, weak or diminished reposition; and score 3: immediate reposition of the leg. A score = 1 indicated the presence of motor nerve block, a score = 2 indicated a partial motor nerve block and a score = 3 indicated the absence of motor nerve block. Pain scores, sedation scores and proprioception scores were collected by VM, JS or AT during daytime. Scores collected overnight were completed by final year veterinary students or interns at the UTH, or by experienced veterinary nurses and veterinarians at the VRC. These assessors were previously trained by VM, JS or AT in the use of the 4Avet pain score system. All assessors were blinded to the allocated group.

      Statistical analysis

      An online program [sealedenvelope.com, (last accessed on 17 January 2022)] was used to calculate the number of animals required per experiment. An alpha error of 5% and power of 90% were used based on the duration of sensory sciatic nerve block after perineural injection of ropivacaine combined with dexmedetomidine (321 ± 123 minutes) and ropivacaine combined with saline (171 minutes) (
      • Marolf V.
      • Ida K.K.
      • Siluk D.
      • et al.
      Effects of perineural administration of ropivacaine combined with perineural or intravenous administration of dexmedetomidine for sciatic and saphenous nerve blocks in dogs.
      ). A set up for continuous outcome superiority trial indicated that 15 animals per group were necessary. The statistical analysis was performed with GraphPad Prism Version 5.00 for Windows (GraphPad Software, CA, USA). A Shapiro-Wilk test verified distribution of data. Parametric data are presented as mean ± standard deviation and nonparametric data as median and interquartile range. Demographic data, duration of surgery, anaesthesia, nerve block to extubation, and nerve block to start of arthroscopy/arthrotomy were analysed by unpaired Student t test. Time to first methadone and time to full recovery of proprioception were evaluated by Mann–Whitney U test.
      Fisher exact tests were applied to analyse postoperative methadone doses administered. Confounding factors such as the surgeon, body weight of the dog, presence or absence of meniscal tear lesions and the type of surgery (arthrotomy versus arthroscopy) were analysed with Fisher exact and Chi-square tests. Pain scores, sedation scores and proprioception scores were compared with Friedman’s test with Dunn’s post hoc test for multiple comparison. Data were analysed separately for each centre when indicated. At the UTH, two dogs completed the study twice because they needed TPLO surgery on both pelvic limbs on different occasions. These data were treated as unpaired data. A p value of ≤ 0.05 was significant.

      Results

      A total of 58 dogs completed the study, representing 60 cases of TPLO. Of the two dogs underdoing TPLO twice (the two pelvic limbs underwent surgery on different occasions), one was randomized twice to group CON and the other was randomized twice to group DEX. No perioperative complications were observed.

      Demographic data

      Age and weight were normally distributed. There was no significant difference between groups in age (p = 0.404; CON: 6.3 ± 3.0; DEX: 5.7 ± 2.7 years) or weight (p = 0.277; CON: 28.1 ± 9.4; DEX: 30.8 ± 9.5 kg). Breeds of dogs are presented in Table 1.
      Table 1The table shows confounding factors which may have influenced the number of postoperative doses of methadone administered. These factors were analysed dependently (p value in column group CON and group DEX) or independently of group allocation (p value in last column). A total of 58 dogs undergoing tibial plateau levelling osteotomy received a preoperative ultrasound-guided sciatic and femoral nerve block with ropivacaine 0.5% combined with perineural dexmedetomidine at 1 μg kg–1 (group DEX) or combined with the equivalent volume of perineural saline solution (group CON)
      Confounding factorsGroup CON (n = 15)Group DEX (n = 15)p value (CON versus DEX)
      WeightLarge dogs (>20 kg)n = 23

      x = 19
      n = 26

      x = 14
      0.007
      Significant difference between groups (p < 0.05).
      Small dogs (≤20 kg)n = 7

      x = 12
      n = 4

      x = 8
      p value (small versus large)0.040
      Significant difference between groups (p < 0.05).
      0.003
      Significant difference between groups (p < 0.05).
      Meniscal tear (only applicable for VRC)Presentn = 8

      x = 18
      n = 5

      x = 5
      0.123
      Absentn = 10

      x = 10
      n = 7

      x = 9
      p value0.2021
      Surgery (only applicable to VRC)Arthroscopy + TPLOn = 13

      x = 23
      n = 15

      x = 14
      0.167
      Arthrotomy + TPLOn = 2

      x = 5
      n = 0

      x = 0
      p value0.5251
      SurgeonSurgeon 1 (VRC)n = 6

      x = 5
      n = 7

      x = 5
      <0.001
      Significant difference between groups (p < 0.05).
      Surgeon 2 (VRC)n = 6

      x = 10
      n = 3

      x = 0
      Surgeon 3 (VRC)n = 3

      x = 13
      n = 5

      x = 9
      Surgeon 4 (UTH)n = 6

      x = 3
      n = 6

      x = 4
      Surgeon 5 (UTH)n = 1

      x = 0
      n = 1

      x = 0
      Last year resident (UTH)n = 8

      x = 0
      n = 8

      x = 4
      p value<0.001
      Significant difference between groups (p < 0.05).
      0.051
      Intraoperative fentanylRequired at least 1 dosen = 10

      x = 10
      n = 6

      x = 0
      0.1914
      No fentanyl requiredn = 20

      x = 21
      n = 24

      x = 22
      p value10.010
      Significant difference between groups (p < 0.05).
      BreedMixed breed (n = 7)

      Labrador (n = 4)

      Bernese Mountain Dog (n = 5)

      Golden Retriever (n = 3)

      Australian Shepherd (n = 2)

      Other breeds (n = 9)
      Labrador (n = 5)

      German Shepherd (n = 3)

      Golden Retriever (n = 5)

      Australian Shepherd (n = 2)

      American Staffordshire Terrier (n = 3)

      Other breeds (n = 12)
      N/A
      n, number of dogs; N/A, not applicable; TPLO, tibial plateau levelling osteotomy; UTH, University Teaching Hospital; VRC, Veterinary Referral Centre; x, number of postoperative doses of methadone administered.
      Significant difference between groups (p < 0.05).

      Surgery

      At the VRC, all dogs underwent arthroscopy prior to TPLO surgery except two dogs, which underwent arthrotomy and in which the entire joint capsule was opened prior to TPLO. At the UTH, no dogs underwent arthroscopy or arthrotomy prior to TPLO and the joint capsule was not opened. Durations of surgical and anaesthetic procedures are presented in Table 2.
      Table 2Duration of surgical and anaesthetic procedures. A total of 58 dogs scheduled for TPLO surgery at two different centres (30 TPLO cases at VRC); 30 TPLO cases at UTH) were randomized to one of two groups. Dogs were given a preoperative ultrasound-guided sciatic and femoral nerve block with ropivacaine 0.5% combined with perineural dexmedetomidine at 1 μg kg–1 (group DEX) or combined with the equivalent volume of perineural saline solution (group CON). Results are presented as mean ± standard deviation Duration of arthroscopy/tomy was analysed with one sample t test, other durations with unpaired Student t tests
      Duration of procedures (minutes)CentreGroup
      VMCUTHCONDEX
      Arthroscopy/tomy

      TPLO surgery

      Anaesthesia (induction to extubation)

      Perineural injection to extubation

      Perineural injection to first surgical incision
      26.2 ± 10.2

      68.0 ± 2.4

      175.8 ± 3.0

      142.0 ± 3.4

      31.0 ± 1.3
      N/A

      68.7 ± 2.4

      198.3 ± 4.5
      Significant difference between centres (p ≤ 0.001).


      145.3 ± 3.5

      40.5 ± 1.5
      Significant difference between centres (p ≤ 0.001).
      27.0 ± 3.2

      67.2 ± 2.3

      190.7 ± 5.2

      146.3 ± 3.8

      38.5 ± 1.8
      25.3 ± 2.0

      69.5 ± 2.4

      183.5 ± 3.2

      141.0 ± 3.1

      33.0 ± 1.3
      Significant difference between groups (p = 0.016).
      N/A, not applicable; TPLO, tibial plateau levelling osteotomy; UTH, university teaching hospital; VRC, veterinary referral centre.
      Significant difference between centres (p ≤ 0.001).
      Significant difference between groups (p = 0.016).

      Intraoperative data

      The percentage of dogs given at least one dose of fentanyl was 20% (n = 6) in group DEX and 33% (n = 10) in group CON. At the VRC, intraoperative fentanyl was administered three times (n = 2), twice (n = 1) or once (n = 1) in group CON and once (n = 4) in group DEX. At the UTH, intraoperative fentanyl was administered twice (n = 3) or once (n = 3) in group CON and once (n = 2) in group DEX. There was no difference in the number of dogs given intraoperative fentanyl between groups (p = 0.382). A single dose of atropine was administered to one dog in group DEX intraoperatively. Hypotension was recorded in 27% of dogs (n = 8) in group CON and 20% (n = 6) in group DEX. In group CON, reduction of isoflurane was effective in restoring normotension in all but one dog, which needed an additional bolus of isotonic crystalloid. In group DEX, reduction of isoflurane was effective in three dogs but three other dogs with hypotension needed a single isotonic crystalloid bolus to restore normotension. Dobutamine was not required in any dog.

      Postoperative methadone administration

      No differences in postoperative methadone requirement were observed between groups (p = 0.244). When data were analysed separately for both centres, the following results were observed. The total need for postoperative methadone was higher in group CON than in group DEX (p = 0.026) at the VRC but not at the UTH (p = 0.216). The number of methadone doses administered during the first 24 postoperative hours were 31 and 22 in groups CON and DEX, respectively. Dogs received more postoperative doses of methadone at the VRC (28 doses in group DEX; 14 doses in group CON) than at the UTH (eight doses in group DEX, three doses in group CON). At the VRC, in group CON two dogs received six postoperative doses and one dog received five postoperative doses. The total cumulative dose of methadone (mg kg–1) administered is illustrated in Fig. 1. A total of 64% (19/30) and 57% (17/30) of dogs did not receive methadone during the postoperative evaluation period of 24 hours in groups DEX and CON, respectively. The percentage of dogs that required at least one dose of methadone is presented in Fig. 2. The time to administration of the first postoperative dose of methadone did not differ between groups (p = 0.575).
      Figure 1
      Figure 1Cumulative postoperative methadone consumption (mg kg–1) of 58 dogs after tibial plateau levelling osteotomy. Dogs received either a preoperative ultrasound-guided sciatic and femoral nerve block with ropivacaine 0.5% combined with perineural dexmedetomidine at 1 μg kg–1 (group DEX) or saline solution (group CON). Postoperative pain was assessed by 4Avet pain scores at predetermined time points (T0.5-T24 hours) starting from removal of the endotracheal tube.
      Figure 2
      Figure 2The proportion (%) of dogs (from the total of 60 tibial plateau levelling cases) which required at least one dose of postoperative rescue analgesia (methadone 0.2 mg kg–1 intravenously) after surgery. The numbers in parentheses are the number of dogs given two or more doses of methadone at each evaluation time points. Postoperative pain was assessed by 4Avet pain scores at predetermined time points (T0.5–T24 hours) starting from removal of the endotracheal tube. A preoperative ultrasound-guided sciatic and femoral nerve block with ropivacaine 0.5% combined with (group DEX; n = 30) or without (group CON; n = 30) dexmedetomidine at 1 μg kg–1 injected perineurally. Note that the maximal value on the y-axis was set at 50% for illustration purposes.

      Confounding factors

      Confounding factors which might have influenced the postoperative consumption of methadone are presented in Table 1. Dogs weighing ≤ 20 kg needed significantly more methadone than dogs weighing > 20 kg. The operating surgeon and the administration of intraoperative fentanyl in group DEX also significantly influenced the number of postoperative methadone doses administered.

      Postoperative pain scores

      There was no significant difference in pain scores at the different evaluation time points within groups (CON: p = 0.220 and DEX: p = 0.059) or between groups (p = 0.071). At T16, pain scores were lower (p < 0.05) at the UTH [2 (1-3)] than at the VRC [4 (2-5.25)], but there was no significant difference at other evaluation time points between both centres.

      Sedation

      Sedation scores were higher in group CON (p < 0.05) at T0.5 (1 [0–2]) than at T2 (0 [0–1]) and in group DEX (p < 0.05) at T0.5 [1 (1-2)] than at T2 [0 (0-1)]. The sedation scores were not statistically different between groups at any evaluated time points. The sedation score was the same as baseline in all dogs latest at T12 in group CON and at T20 in group DEX.

      Proprioception

      Proprioceptive deficits (score = 2) were observed in one dog in group CON at T24. At hospital discharge (36 hours after nerve blocks), proprioception had returned to normal (score = 3) in this dog. All other dogs had normal proprioception (score = 3) at T24. There was no significant difference in time to full recovery of proprioception (p = 0.152) or in the duration of proprioceptive deficits at the different evaluation time points between groups (p > 0.05). When data were analysed separately for both centres, at the VRC, the time to full recovery of proprioception was longer in group DEX than in group CON (p = 0.004) but not at the UTH (p = 0.882). At the VRC, a significant difference between proprioception scores was present at T8 between groups (p < 0.05). The time to recovery of proprioception at the VRC is illustrated in Fig. 3.
      Figure 3
      Figure 3Recovery of proprioception after ultrasound-guided sciatic and femoral nerve assessed at different time points after endotracheal tube removal in dogs at the veterinary referral centre (n = 30) recovering from tibial plateau levelling osteotomy. For detailed legend, see . A score of 1 indicated absent proprioception, a score of 2 indicated partial or retarded proprioception and a score of 3 indicated recovered proprioception. The figure illustrates recovery of proprioception of dogs at a single veterinary referral centre. ∗Significant differences between groups (p < 0.05).

      Discussion

      Results of this study suggest that perineural US-guided injection of dexmedetomidine combined with ropivacaine 0.5% at the sciatic and femoral nerve does not reduce postoperative methadone consumption in dogs undergoing TPLO. The working hypothesis must therefore be rejected. However, different results were obtained at the two veterinary centres. Methadone requirement appeared to be significantly less when dexmedetomidine was added to ropivacaine in one centre but not in the other. This leads to the rejection of the secondary hypothesis as well.
      Clinical practices regarding TPLO cases at both centres were different. Prior to TPLO surgery, the joint capsule is not routinely opened at the UTH, whereas arthroscopy or, less frequently, arthrotomy is performed at the VRC. The canine stifle joint is innervated by medial, lateral and sometimes posterior articular nerves. Some dogs might have additional innervation coming from femoral and/or obturator nerves (
      • O'Connor B.L.
      • Woodbury P.
      The primary articular nerves to the dog knee.
      ). Opening or partially opening the stifle joint might disrupt the innervation of the capsule. A meniscal tear lesion is a painful condition in dogs (
      • Dillon D.E.
      • Gordon-Evans W.J.
      • Griffon D.J.
      • et al.
      Risk factors and diagnostic accuracy of clinical findings for meniscal disease in dogs with cranial cruciate ligament disease.
      ) and might influence the degree of postoperative pain and postoperative methadone requirements. Differences in practice at both centres might have impacted the degree of postoperative pain in dogs and could explain the greater need for postoperative rescue analgesia at the VRC.
      The analysis of confounding factors revealed that some factors might significantly influence postoperative methadone requirements. A population needs to be standardized to avoid selection bias. Selecting cases is challenging in clinical trials because common veterinary caseloads include dogs of different weight, size and breed. Dogs interact differently with humans depending on their breed. Dogs of certain breeds might be shy, stoic, bouncy or happy and interact differently during pain score evaluation (
      • Ellwood B.
      • Murison P.J.
      Investigating the effect of anxiety on pain scores in dogs.
      ). In this study, small dogs needed more postoperative doses of methadone than large dogs. Dogs of small size have higher metabolism and higher basal HR (
      • Middleton R.P.
      • Sebastien Lacroix S.
      • Scott-Boyer M.P.
      • et al.
      Metabolic differences between dogs of different body sizes.
      ) which might have led to faster elimination of drugs from the body. The statistical analysis revealed that the operating surgeon might have impacted on the number of postoperative methadone doses needed for each dog. In group DEX, intraoperative administration of fentanyl led to a reduction in the number of postoperative methadone injections. A synergistic action between perineural dexmedetomidine and IV fentanyl is possible. Fentanyl has the potential to control intraoperative nociception during anaesthesia (
      • Murphy M.R.
      • Olson W.A.
      • Hug Jr., C.C.
      Pharmacokinetics of 3H-fentanyl in the dog anesthetized with enflurane.
      ). However, fentanyl-induced hyperalgesia is also a described phenomenon and needs to be considered (
      • Rivat C.
      • Laulin J.P.
      • Corcuff J.B.
      • et al.
      Fentanyl enhancement of carrageenan-induced long-lasting hyperalgesia in rats: prevention by the N-methyl-D-aspartate receptor antagonist ketamine.
      ). The amount of intraoperative fentanyl might have influenced the degree of postoperative methadone requirement. The mentioned confounding factors need to be considered as they might have influenced the postoperative number of doses of methadone administered to each dog.
      The differences observed at both centres questions the clinical validity of results from single-centre clinical trial. Prospective single-centre RCTs are carefully designed investigations which can change clinical practice and improve patient care (
      • Saturni S.
      • Bellini F.
      • Braido F.
      • et al.
      Randomized controlled trials and real life studies. approaches and methodologies: a clinical point of view.
      ). Methodological problems, centre-specific bias and errors in data reporting such as blinding, randomization or inclusion/exclusion criteria might affect the validity of the results (
      • Youssef N.C.M.
      • Reinhart K.
      • Sakr Y.
      The pros and cons of multicentre studies.
      ). The efficacy of treatments in single-centre RCTs might be larger than that in multicentre RCTs (
      • Dechartres A.
      • Boutron I.
      • Trinquart L.
      • et al.
      Single-center trials show larger treatment effects than multicenter trials: evidence from a meta-epidemiologic study.
      ;
      • Unverzagt S.
      • Prondzinsky R.
      • Peinemann F.
      Single-center trials tend to provide larger treatment effects than multicenter trials: a systematic review.
      ) and the transposition of results of RCT from one centre to another can be misleading. Multicentre studies with high-quality reporting are therefore encouraged as they have the potential to increase the level of scientific evidence and reduce potential confounder bias (
      • Evans D.
      Hierarchy of evidence: a framework for ranking evidence evaluating healthcare interventions.
      ). Their results are usually reproducible, but these studies can be challenging to conduct (
      • Irving S.Y.
      • Curley M.A.
      Challenges to conducting multicenter clinical research: ten points to consider.
      ). The initial goal was to design this study as a multicentre trial. An explanatory video of the project was presented at the international congress of the Association of Veterinary Anaesthetists in March 2018 in Grenada to recruit different centres to participate. Unfortunately, no veterinary hospital expressed an interest in collaboration. The time-consuming nature of the study, the inability to motivate a responsible person to conduct the study or the restricted number of institutions with experience in US-guided nerve blocks might explain the lack of participation. After peer discussion, it was decided that the study be performed in parallel at two different centres only.
      Dexmedetomidine is often used in conjunction with LAA for perineural injection in human patients undergoing orthopaedic surgeries. Doses of approximately 1 μg kg–1 prolong nerve block by approximately 200 minutes (
      • Kirksey M.A.
      • Haskins S.C.
      • Cheng J.
      • Liu S.S.
      Local anesthetic peripheral nerve block adjuvants for prolongation of analgesia: A systematic qualitative review.
      ). Reviews and meta-analysis have shown that dexmedetomidine combined with LAA prolongs postoperative analgesia and reduces 24 hour cumulated consumption of analgesic agents compared with LAA alone (
      • Vorobeichik L.
      • Brull R.
      • Abdallah F.W.
      Evidence basis for using perineural dexmedetomidine to enhance the quality of brachial plexus nerve blocks : a systematic review and meta-analysis of randomized controlled trials.
      ;
      • Schnabel A.
      • Reichl S.U.
      • Weibel S.
      • et al.
      Efficacy and safety of dexmedetomidine in peripheral nerve blocks A meta-analysis and trial sequential analysis.
      ,
      • Wang K.
      • Wang L.J.
      • Yang T.J.
      • et al.
      Dexmedetomidine combined with local anesthetics in thoracic paravertebral block: A systematic review and meta-analysis of randomized controlled trials.
      ). In dogs, a dose of 1 μg kg–1 was proposed for this clinical trial (
      • Marolf V.
      • Ida K.K.
      • Siluk D.
      • et al.
      Effects of perineural administration of ropivacaine combined with perineural or intravenous administration of dexmedetomidine for sciatic and saphenous nerve blocks in dogs.
      ). The 24 hour number of postoperative doses of methadone was used as primary outcome in this study because repeated postoperative administration of methadone might negatively affect the comfort of dogs (
      • Bini G.
      • Vettorato E.
      • De Gennaro C.
      • Corletto F.
      A retrospective comparison of two analgesic strategies after uncomplicated tibial plateau levelling osteotomy in dogs.
      ).
      In experimental dogs, sensory but not motor nerve block was prolonged when dexmedetomidine was added to ropivacaine for peripheral nerve blocks (
      • Marolf V.
      • Ida K.K.
      • Siluk D.
      • et al.
      Effects of perineural administration of ropivacaine combined with perineural or intravenous administration of dexmedetomidine for sciatic and saphenous nerve blocks in dogs.
      ). However, in humans, results are contradictory.
      • Dai W.
      • Tang M.
      • He K.
      The effect and safety of dexmedetomidine added to ropivacaine in brachial plexus block: A meta-analysis of randomized controlled trials.
      reported a longer duration of motor block for supraclavicular brachial plexus block but not for axillary or intermuscular brachial plexus block. The duration of motor block seems to be dependent on the type of nerve block. In the present study, contradictory results were also observed between groups, because time to full recovery of proprioception was longer in group DEX at the VRC but not at the UTH. Additionally, a significant difference in proprioception was observed at T8 between groups at the VRC. Prolonged motor nerve block and potentially prolonged sensory nerve block is possible, but again, the differences between centres do not allow any conclusions to be drawn at present.
      In humans, dexmedetomidine combined with LAA for peripheral nerve blocks can induce sedation, bradycardia or hypotension in a dose-dependent manner (
      • Rancourt M.P.
      • Albert N.T.
      • Côté M.
      • et al.
      Posterior tibial nerve sensory blockade duration prolonged by adding dexmedetomidine to ropivacaine.
      ;
      • Lin Y.N.
      • Li Q.
      • Yang R.M.
      • et al.
      Addition of dexmedetomidine to ropivacaine improves cervical plexus block.
      ;
      • Keplinger M.
      • Marhofer P.
      • Kettner S.C.
      • et al.
      A pharmacodynamic evaluation of dexmedetomidine as an additive drug to ropivacaine for peripheral nerve blockade: A randomised, triple-blind, controlled study in volunteers.
      ). There was no difference between groups in sedation or hypotension, and only one dog was given atropine in group DEX. Possible systemic side effects related to perineural dexmedetomidine administration need to be considered but do not appear clinically relevant at 1 μg kg–1 in dogs.
      Some limitations need to be mentioned. A validated French pain scale (4Avet) was used to evaluate postoperative pain. This pain scale is regularly used in the clinical setting at the VRC, but the UTH usually uses the short form of the Glasgow composite pain scale. Changing clinical practice might have affected postoperative pain evaluation, possibly explaining the lower need for postoperative methadone found at the UTH than at the VRC. A significant difference in pain scores between centres was only present at T16. Validated pain scales are designed to produce reproducible results if applied by different persons, but interobserver and/or centre variability cannot be excluded. Final year veterinary students may not be sufficiently experienced to use pain scores despite initial training. This may have influenced the evaluation of postoperative pain and thus the requirement of rescue methadone. At the VRC, the same out of hours staff were available during the study period, whereas it changed for every dog at the UTH. US-guided nerve blocks were performed by two different anaesthesiologists with different experience, skills and equipment. This might have affected the quality and duration of nerve blocks.
      The analysis of the entire data set of this two-centre study showed that the use of dexmedetomidine as an adjuvant to ropivacaine for femoral and sciatic nerve block did not reduce the postoperative consumption of methadone in dogs after TPLO surgery. The separate analysis of data from both centres revealed that perineural dexmedetomidine reduced postoperative methadone administration in one centre but not in the other. Despite efforts to standardize study design, variations in clinical practice at both centres such as differences in surgical procedures for cranial cruciate ligament repair, assessors collecting postoperative scores, duration of procedures, weight and breeds of dogs or operating surgeon were noted. Those variations probably explain differences observed in the primary outcome and make it difficult to draw conclusions about the clinical benefit of perineural dexmedetomidine.

      Authors’ contributions

      VM: study design, collection of signed owner consent, performance of nerve block, postoperative scores collection, statistical analysis, data preparation, data interpretation, manuscript writing. JS: collection of signed owner consent, TPLO surgery, postoperative scores collection, data preparation, critical revision of the manuscript, final approval of manuscript. PP: collection of signed owner consent, TPLO surgery, data preparation, critical revision of the manuscript, final approval of manuscript. CS: study design, critical revision of the manuscript, final approval of manuscript. AT: performance of nerve block, postoperative scores collection, data preparation, data interpretation, critical revision of the manuscript, final approval of manuscript. CS: study design, data interpretation, critical revision of the manuscript, final approval of manuscript.

      Conflict of interest statement

      The authors declare no conflict of interest.

      References

        • Abdallah F.W.
        • Brull R.
        Facilitatory effects of perineural dexmedetomidine on neuraxial and peripheral nerve block : a systematic review and meta-analysis.
        Br J Anaesth. 2013; 110: 915-925
        • Acquafredda C.
        • Stabile M.
        • Lacitignola L.
        • et al.
        Clinical efficacy of dexmedetomidine combined with lidocaine for femoral and sciatic nerve blocks in dogs undergoing stifle surgery.
        Vet Anaesth Analg. 2021; 48: 962-971
        • Bartel A.K.
        • Campoy L.
        • Martin-Flores M.
        • et al.
        Comparison of bupivacaine and dexmedetomidine femoral and sciatic nerve blocks with bupivacaine and buprenorphine epidural injection for stifle arthroplasty in dogs.
        Vet Anaesth Analg. 2016; 43: 435-443
        • Bini G.
        • Vettorato E.
        • De Gennaro C.
        • Corletto F.
        A retrospective comparison of two analgesic strategies after uncomplicated tibial plateau levelling osteotomy in dogs.
        Vet Anaesth Analg. 2018; 45: 557-565
        • Campoy L.
        • Martin-flores M.
        • Ludders J.W.
        • et al.
        Comparison of bupivacaine femoral and sciatic nerve block versus bupivacaine and morphine epidural for stifle surgery in dogs.
        Vet Anaesth Analg. 2012; 39: 91-98
        • Campoy L.
        • Bezuidenhout A.J.
        • Gleed R.D.
        • et al.
        Ultrasound-guided approach for axillary brachial plexus, femoral nerve, and sciatic nerve blocks in dogs.
        Vet Anaesth Analg. 2010; 37: 144-153
        • Dai W.
        • Tang M.
        • He K.
        The effect and safety of dexmedetomidine added to ropivacaine in brachial plexus block: A meta-analysis of randomized controlled trials.
        Medicine (Baltimore). 2018; 97e12573
        • Dechartres A.
        • Boutron I.
        • Trinquart L.
        • et al.
        Single-center trials show larger treatment effects than multicenter trials: evidence from a meta-epidemiologic study.
        Ann Intern Med. 2011; 155: 39-51
        • Dillon D.E.
        • Gordon-Evans W.J.
        • Griffon D.J.
        • et al.
        Risk factors and diagnostic accuracy of clinical findings for meniscal disease in dogs with cranial cruciate ligament disease.
        Vet Surg. 2014; 43: 446-450
        • El-Boghdadly K.
        • Brull R.
        • Sehmbi H.
        • Abdallah F.W.
        Perineural dexmedetomidine is more effective than clonidine when added to local anesthetic for supraclavicular brachial plexus block: a systematic review and meta-analysis.
        Anesth Analg. 2017; 124: 2008-2020
        • Ellwood B.
        • Murison P.J.
        Investigating the effect of anxiety on pain scores in dogs.
        Vet Anaesth Analg. 2022; 49: 135-142
        • Evans D.
        Hierarchy of evidence: a framework for ranking evidence evaluating healthcare interventions.
        J Clin Nurs. 2013; 12: 77-84
        • Hoelzler M.G.
        • Harvey R.C.
        • Lidbetter D.A.
        • Millis D.L.
        Comparison of perioperative analgesic protocols for dogs undergoing tibial plateau leveling osteotomy.
        Vet Surg. 2005; 34: 337-344
        • Holopherne-Doran D.
        • Laboissière B.
        • Cogny M.
        Validation of the 4AVet postoperative pain scale in dogs and cats.
        Vet Anaesth Analg. 2010; 37: 1-17
        • Irving S.Y.
        • Curley M.A.
        Challenges to conducting multicenter clinical research: ten points to consider.
        AACN Adv Crit Care. 2008; 19: 164-169
        • Jung H.S.
        • Seo K.H.
        • Kang J.H.
        • et al.
        Optimal dose of perineural dexmedetomidine for interscalene brachial plexus block to control postoperative pain in patients undergoing arthroscopic shoulder surgery: A prospective, double-blind, randomized controlled study.
        Medicine (Baltimore). 2018; 97: 1-10
        • Keplinger M.
        • Marhofer P.
        • Kettner S.C.
        • et al.
        A pharmacodynamic evaluation of dexmedetomidine as an additive drug to ropivacaine for peripheral nerve blockade: A randomised, triple-blind, controlled study in volunteers.
        Eur J Anaesthesiol. 2015; 32: 790-796
        • Kirksey M.A.
        • Haskins S.C.
        • Cheng J.
        • Liu S.S.
        Local anesthetic peripheral nerve block adjuvants for prolongation of analgesia: A systematic qualitative review.
        PLoS One. 2015; 10: 1-23
        • Lin Y.N.
        • Li Q.
        • Yang R.M.
        • et al.
        Addition of dexmedetomidine to ropivacaine improves cervical plexus block.
        Acta Anaesthesiol Taiwan. 2013; 51: 63-66
        • Liu S.S.
        • Wu C.L.
        The effect of analgesic technique on postoperative patient-reported outcomes including analgesia : a systematic review.
        Anesth Analg. 2007; 105: 789-808
        • Liu S.S.
        • Strodtbeck W.M.
        • Richman J.M.
        • Wu C.L.
        A comparison of regional versus general anesthesia for ambulatory anesthesia: a meta-analysis of randomized controlled trials.
        Anesth Analg. 2005; 101: 1634-1642
        • Marolf V.
        • Ida K.K.
        • Siluk D.
        • et al.
        Effects of perineural administration of ropivacaine combined with perineural or intravenous administration of dexmedetomidine for sciatic and saphenous nerve blocks in dogs.
        Am J Vet Res. 2021; 82: 449-458
        • Middleton R.P.
        • Sebastien Lacroix S.
        • Scott-Boyer M.P.
        • et al.
        Metabolic differences between dogs of different body sizes.
        J Nutr Metab. 2017; 4535710
        • Murphy M.R.
        • Olson W.A.
        • Hug Jr., C.C.
        Pharmacokinetics of 3H-fentanyl in the dog anesthetized with enflurane.
        Anesthesiology. 1979; 50: 13-19
        • O'Connor B.L.
        • Woodbury P.
        The primary articular nerves to the dog knee.
        J Anat. 1982; 134: 563-572
        • Palomba N.
        • Vettorato E.
        • De Gennaro C.
        • Corletto F.
        Peripheral nerve block versus systemic analgesia in dogs undergoing tibial plateau levelling osteotomy: Analgesic efficacy and pharmacoeconomics comparison.
        Vet Anaesth Analg. 2020; 47: 119-128
        • Petroz G.C.
        • Sikich N.
        • James M.
        • et al.
        A phase I, two-center study of the pharmacokinetics and pharmacodynamics of dexmedetomidine in children.
        Anesthesiology. 2006; 105: 1098-1110
        • Ping Y.
        • Ye Q.
        • Wang W.
        • et al.
        Dexmedetomidine as an adjuvant to local anesthetics in brachial plexus blocks: A meta-analysis of randomized controlled trials.
        Medicine (Baltimore). 2017; 96e5846
        • Rancourt M.P.
        • Albert N.T.
        • Côté M.
        • et al.
        Posterior tibial nerve sensory blockade duration prolonged by adding dexmedetomidine to ropivacaine.
        Anesth Analg. 2012; 115: 958-962
        • Reid J.
        • Nolan A.M.
        • Hughes J.M.
        • et al.
        Development of the short-form Glasgow Composite Measure Pain Scale (CMPS-SF) and derivation of an analgesic intervention score.
        Animal Welfare. 2007; 16: 67-104
        • Rivat C.
        • Laulin J.P.
        • Corcuff J.B.
        • et al.
        Fentanyl enhancement of carrageenan-induced long-lasting hyperalgesia in rats: prevention by the N-methyl-D-aspartate receptor antagonist ketamine.
        Anesthesiology. 2002; 96: 381-391
        • Saturni S.
        • Bellini F.
        • Braido F.
        • et al.
        Randomized controlled trials and real life studies. approaches and methodologies: a clinical point of view.
        Pulm Pharmacol Ther. 2014; 27: 129-138
        • Schnabel A.
        • Reichl S.U.
        • Weibel S.
        • et al.
        Efficacy and safety of dexmedetomidine in peripheral nerve blocks A meta-analysis and trial sequential analysis.
        Eur J Anaesthesiol. 2018; 35: 745-758
        • Slocum B.
        • Slocum T.D.
        Tibial plateau leveling osteotomy for repair of cranial cruciate ligament rupture in the canine.
        Vet Clin North Am Small Anim Pract. 1993; 23: 777-795
        • Trein T.A.
        • Floriano B.P.
        • Wagatsuma J.T.
        • et al.
        Effects of dexmedetomidine combined with ropivacaine on sciatic and femoral nerve blockade in dogs.
        Vet Anaesth Analg. 2017; 44: 144-153
        • Unverzagt S.
        • Prondzinsky R.
        • Peinemann F.
        Single-center trials tend to provide larger treatment effects than multicenter trials: a systematic review.
        J Clin Epidemiol. 2013; 66: 1271-1280
        • Vettorato E.
        • Bradbrook C.
        • Gurney M.
        • et al.
        Peripheral nerve blocks of the pelvic limb in dogs : A retrospective clinical study.
        Vet Comp Orthop Traumatol. 2012; 25: 314-320
        • Vorobeichik L.
        • Brull R.
        • Abdallah F.W.
        Evidence basis for using perineural dexmedetomidine to enhance the quality of brachial plexus nerve blocks : a systematic review and meta-analysis of randomized controlled trials.
        Br J Anaesth. 2017; 118: 167-181
        • Wang K.
        • Wang L.J.
        • Yang T.J.
        • et al.
        Dexmedetomidine combined with local anesthetics in thoracic paravertebral block: A systematic review and meta-analysis of randomized controlled trials.
        Medicine (Baltimore). 2018; 97e13164
        • Youssef N.C.M.
        • Reinhart K.
        • Sakr Y.
        The pros and cons of multicentre studies.
        J Crit Care. 2008; 12: 120-122