Addition of magnesium sulphate to ropivacaine for spinal analgesia in dogs undergoing tibial plateau levelling osteotomy

Accepted Manuscript Title: Addition of magnesium sulphate to ropivacaine for spinal analgesia in dogs undergoing tibial plateau levelling osteotomy Author: C. Adami, D. Casoni, F. Noussitou, U. Rytz, C. Spadavecchia PII: DOI: Reference:

S1090-0233(15)00501-8 http://dx.doi.org/doi: 10.1016/j.tvjl.2015.11.017 YTVJL 4715

To appear in:

The Veterinary Journal

Accepted date:

27-11-2015

Please cite this article as: C. Adami, D. Casoni, F. Noussitou, U. Rytz, C. Spadavecchia, Addition of magnesium sulphate to ropivacaine for spinal analgesia in dogs undergoing tibial plateau levelling osteotomy, The Veterinary Journal (2015), http://dx.doi.org/doi: 10.1016/j.tvjl.2015.11.017. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Addition of magnesium sulphate to ropivacaine for spinal analgesia in dogs undergoing tibial plateau levelling osteotomy C. Adami a, *, D. Casoni b, F. Noussitou c, U. Rytz c, C. Spadavecchia b a

Department of Clinical Sciences and Services, Royal Veterinary College, University of London, Hawkshead Campus, North Mymms, AL97TA Hatfield, Herts, UK b Department of Veterinary Clinical Science, Anaesthesiology and Pain Therapy Division, Vetsuisse Faculty, University of Berne, Länggassstrasse 124, CH-3012 Berne, Switzerland c Department of Veterinary Clinical Science, Surgery Division, Vetsuisse Faculty, University of Berne, Länggassstrasse 124, CH-3012 Berne, Switzerland

* Corresponding author. Tel.: +44 765 4183186. E-mail address: [email protected] (C. Adami).

Page 1 of 20

22

Highlights

23 24 25 26 27 28 29 30 31



32

Abstract

33

  

Spinal magnesium sulfate added to ropivacaine enhances duration and intensity of the analgesia provided by ropivacaine in dogs undergoing TPLO. Spinal magnesium sulfate added to ropivacaine carries the potential for prolonged, though transient, impairment of the motor function. This can result in patient discomfort for several hours after surgery, and also make the post-operative care more demanding and expensive. Further studies may be necessary to identify still effective, though safer, magnesium doses to be administered spinally to dogs.

The aim of this blinded, randomised, prospective clinical trial was to determine

34

whether the addition of magnesium sulphate to spinally-administered ropivacaine would

35

improve peri-operative analgesia without impairing motor function in dogs undergoing

36

orthopaedic surgery. Twenty client-owned dogs undergoing tibial plateau levelling

37

osteotomy were randomly assigned to one of two treatment groups: group C (control,

38

receiving hyperbaric ropivacaine by the spinal route) or group M (magnesium, receiving

39

a hyperbaric combination of magnesium sulphate and ropivacaine by the spinal route).

40

During surgery, changes in physiological variables above baseline were used to evaluate

41

nociception. Arterial blood was collected before and after spinal injection, at four time

42

points, to monitor plasma magnesium concentrations. Post-operatively, pain was

43

assessed with a modified Sammarco pain score, a Glasgow pain scale and a visual

44

analogue scale, while motor function was evaluated with a modified Tarlov scale.

45

Assessments were performed at recovery and 1, 2 and 3 h thereafter. Fentanyl and

46

buprenorphine were administered as rescue analgesics in the intra- and post-operative

47

periods, respectively.

48 49 50

Plasma magnesium concentrations did not increase after spinal injection compared to baseline. Group M required less intra-operative fentanyl, had lower

Page 2 of 20

51

Glasgow pain scores and experienced analgesia of longer duration than group C (527.0

52

± 341.0 min vs. 176.0 ± 109.0 min). However, in group M the motor block was

53

significantly longer, which limits the usefulness of magnesium for spinal analgesia at

54

the investigated dose. Further research is needed to determine a clinically effective dose

55

with shorter duration of motor block for magnesium used as an additive to spinal

56

analgesic agents.

57 58

Keywords: Analgesia; Dog; Magnesium sulphate; Ropivacaine; Spinal anaesthesia

Page 3 of 20

59 60

Introduction Prevention and control of pain is one of the most important ethical obligations of

61

veterinarians. As a result, various aspects of this fascinating branch of anaesthesia have

62

been explored, and a number of novel techniques have been developed over the past

63

decades to improve peri-operative pain management. It is likely that a multimodal

64

approach has increased efficacy and, consequently, there has been particular interest in

65

agents that, although not classified as analgesics, do exert antinociceptive effects

66

(KuKanich et al., 2013; Madden et al., 2014; Crociolli et al., 2015; Norkus et al., 2015).

67 68

The use of magnesium has generated widespread interest as it could prevent

69

central sensitisation by acting as a non-competitive antagonist at N-methyl-D-aspartate

70

receptors in the dorsal horn, in a voltage-dependent fashion. Magnesium sulphate is

71

commercially available in Europe, and the formulation developed for parenteral use is

72

inexpensive, stable at room temperature and approved for use in dogs. Several studies in

73

both human patients and dogs suggest that magnesium sulphate exerts antinociceptive

74

effects (Bahrenberg et al., 2015), and consistently prolongs the duration of analgesia of

75

various local anaesthetics and opioid combinations when administered via either the

76

epidural or spinal route (Buvanendran et al., 2002; Oezalevli et al., 2005; Arcioni et al.,

77

2007). Additionally, a study in dogs investigating the neurotoxicity of intrathecal

78

magnesium sulphate found that a dose rate of 3 mg/kg did not cause neurological

79

deficits or histopathological changes in the spinal cord (Simpson et al., 1994).

80 81

Overall, these findings supported our hypothesis that a clinical trial investigating

82

the effects of spinally administered magnesium in client-owned dogs would be feasible

83

and ethically acceptable. The aim of this study was to compare the intensity and

Page 4 of 20

84

duration of peri-operative analgesia and motor block in client-owned dogs undergoing

85

elective orthopaedic surgery, after spinal administration of either ropivacaine, or a

86

combination of ropivacaine-magnesium sulphate. Our hypothesis was that the inclusion

87

of magnesium sulphate would provide longer lasting, better quality analgesia than

88

ropivacaine alone, without impairing neurological function of the pelvic limbs and/or

89

prolonging the duration of the motor block.

90 91

Materials and methods

92

Animals

93

Twenty client-owned dogs undergoing elective tibial plateau levelling osteotomy

94

(TPLO) between May 2014 and March 2015 were enrolled in the study. On arrival, a

95

pre-anaesthetic physical examination was performed, as well as venous blood sampling

96

for haematology and chemistry. Exclusion criteria were an American Society of

97

Anaesthesiologists risk category higher than 2, infectious skin diseases affecting the

98

lumbosacral area, and bleeding disorders. The clinical trial was approved by the

99

Committee for Animal Experimentation, Canton of Berne, Switzerland (Approval No.

100

BE11/14, 28 April 2014), and performed with informed owner consent.

101 102 103

Study design The study was designed as an investigator-blinded, block-randomised,

104

prospective clinical trial. Dogs were randomly allocated to one of two treatment groups

105

using a block randomisation method, based on shuffle and drawing of treatment

106

assignments inside an opaque, sealed envelope. One operator not involved in the study

107

was in charge of the allocations list, which was disclosed only at the end of the trial.

108

Page 5 of 20

109

A sample size calculation determined that ten dogs were needed in each treatment

110

group, to achieve a power of 0.9 with an α of 0.05, to detect a minimum difference of 60

111

min in the mean duration of analgesia (defined as the time elapsed from the spinal

112

injection to the first administration of rescue analgesics, either intra-operative fentanyl

113

or post-operative buprenorphine), between groups.

114 115 116

Anaesthetic protocol and procedures After IM premedication with acepromazine (0.03 mg/kg, Prequillan, Aprovet),

117

an appropriately sized IV catheter was placed in a cephalic vein. General anaesthesia

118

was induced with IV propofol (Propofol, Fresenius Kabi) titrated to effect to enable

119

orotracheal intubation, and maintained with isoflurane (IsoFlo, Abbott) vaporised in an

120

oxygen-air mixture and delivered via a circle system. All dogs received IV lactated

121

Ringer’s solution (Ringer-Lactat, Fresenius Kabi) at a rate of 10 mL/kg/h during

122

anaesthesia. The dorsal pedal artery of the non-surgical pelvic limb was catheterised to

123

allow blood sampling and continuous measurement of the systolic (SAP), mean (MAP)

124

and diastolic (DAP) arterial blood pressures. A multiparametric monitor was used to

125

assess cardiovascular (SAP, MAP, DAP, heart rate [HR]), and respiratory (end-tidal

126

carbon dioxide, PEʹCO2; peak inspiratory pressure, PIP; respiratory rate, RR; tidal

127

volume, TV; inspired fraction of oxygen, FIO2; end-tidal isoflurane tension, PEʹISO)

128

variables, as well as oesophageal temperature (T, °C). Data were manually recorded

129

every 5 min until the end of anaesthesia. The dogs were allowed to breathe

130

spontaneously unless PEʹCO2 was > 45 mmHg, in which case pressure-controlled

131

ventilation with PIP set at 10 cm H2O was used to maintain PEʹCO2 within the normal

132

range. A constant PEʹISO of 1.3%, equivalent to the minimum alveolar concentration

133

(MAC) for the species (Steffey et al., 2007), was targeted during anaesthesia.

Page 6 of 20

134 135

Hypotension, defined as MAP lower than 60 mmHg, was treated with a

136

crystalloid bolus (10 mL/kg lactated Ringer’s delivered IV over 10 min). Non-

137

responsive hypotension was treated initially with a colloid bolus (2 mL/kg tetrastarch

138

delivered IV over 10 min), and then with dopamine infusion, starting at a rate of 5

139

µg/kg/min. The dose was increased by 2.5 µg/kg/min every 10 min until MAP was

140

above 60 mmHg.

141 142

Bradycardia, defined as HR lower than 45 beats per min (bpm), was treated with

143

IV glycopyrronium, 10 µg/kg, administered as a bolus. Any clinical signs compatible

144

with hypermagnesaemia, including cardiac bradyarrhythmias and persistent

145

hypotension, were recorded.

146 147 148

After tracheal extubation, carprofen (4 mg/kg) was administered IV to all dogs. Dogs were discharged from the hospital 24 h after surgery.

149 150 151

Spinal injection Once the plane of anaesthesia was judged as adequate on the basis of clinical

152

assessments (jaw relaxation, absence of active blinking, slight or absent palpebral

153

reflex, immobility and physiological variables within normal ranges for the species),

154

spinal injection was performed by one of two anaesthetists (C.A. or D.C.), who were

155

blinded to the treatment. The dogs were positioned in lateral recumbency with the limb

156

to be operated on in a dependent position, with both pelvic limbs pulled symmetrically

157

cranially to maximise the length of the dorsal lumbar intervertebral spaces. The ilial

158

wings and the dorsal spinous processes of L5, L6 and L7 were used as anatomical

Page 7 of 20

159

landmarks. After surgical preparation of the area, a 75 mm  19 G spinal needle was

160

inserted towards the epidural space, with the bevel facing cranially, through the

161

interspinosus ligament between L6 and L5. The stylet was then withdrawn and the

162

needle slowly advanced until cerebrospinal fluid was observed at the hub of the needle.

163

A ‘dry tap’ after the third attempt of needle insertion was considered an exclusion

164

criterion.

165 166

Treatment groups

167

Group C (control) received ropivacaine (Naropin 1%, AstraZeneca), at a dose of

168

1 mg/kg (0.1 mL/kg). Group M (magnesium) received a mixture of magnesium sulphate

169

(2 g/10 mL, Magnesio Solfato, Galenica Senese), at a dose of 2 mg/kg (equivalent to a

170

volume of 0.01 mL/kg), and ropivacaine at a dose of 1 mg/kg. All treatments were

171

administered spinally.

172 173

For both treatments, the solution for injection was made hypertonic immediately

174

before the injection by adding 50% glucose (Glucose 50% BBraun, 0.002 mL/kg) to the

175

solution. The specific gravity of the solutions, measured with a refractometer, were

176

1.032 and 1.035 at 25 ºC for groups C and M, respectively. The solution was injected

177

over 1 min. Doses and volumes were based on previous reports in both human and

178

veterinary medicine (Oezalevli et al., 2005; Arcioni et al., 2007; Bilir et al., 2007;

179

Sarotti et al., 2011; Sanasilp et al., 2012).

180 181

Assessment of nociception

182

Intra-operatively, any increase in HR, MAP and/or RR of 20% above baseline

183

values (defined as the values recorded before skin incision, after PEʹISO values of 1.3%

Page 8 of 20

184

had been recorded consecutively for 15 min) were considered indicative of nociception.

185

When such increases were seen for at least two of these parameters, fentanyl (3 µg/kg,

186

IV) was administered as rescue analgesia.

187 188

Post-operatively, pain was assessed with a modified multifactorial pain score

189

(Sammarco et al., 1997; Adami et al., 2012) and the Glasgow pain scale (Holton et al.,

190

2001). Additionally, a 10 cm visual analogue scale (VAS) with end points labelled

191

‘worst pain imaginable’ (0) and ‘no pain’ (10) was used. Cut-off values to administer

192

rescue buprenorphine (Temgesic, 10 µg/kg IV) were one or more pain scores exceeding

193

40% of the maximum value possible (>4 for the VAS, >6 for the Sammarco pain score,

194

and > 10 for the Glasgow pain scale). Neurological function of the pelvic limbs and the

195

degree of motor block were assessed with a modified Tarlov scale (Table 1)

196

(Buvanendran et al., 2002).

197 198

The assessments were performed as soon as the dogs were conscious enough to

199

respond to stimulation (vocal call and incitement to sit or stand up) and then at 60, 120,

200

180, 240 and 300 min thereafter. The evaluations were performed by one of two

201

observers (C.A. and D.C.), who were blinded to the treatment. During preliminary tests,

202

comparable pain and motor block scores were determined by the two observers when

203

independently evaluating the same dogs.

204 205 206

Blood sampling Arterial blood was sampled from the indwelling arterial catheter before the

207

spinal injection (baseline), and then 15, 60, 120 and 240 min thereafter. The samples

208

were placed in 1.3 mL heparinised tubes, which were centrifuged for 5 min at 3000 g

Page 9 of 20

209

and the obtained plasma frozen at -20 °C. Total plasma magnesium concentrations

210

(mmol/L) were determined from dogs in group M within 10 months of sample

211

collection, using a commercial colorimetric assay (Magnesium Gen 2, Roche

212

Diagnostics).

213 214

Statistical analysis

215

Normality of data was tested with the Kolmogorov-Smirnov test and the

216

Shapiro-Wilk test. Repeated measures ANOVA, followed by Tukey Kramer’s multiple

217

comparison test, were used for the plasma magnesium concentrations, for physiological

218

variables and for post-operative pain and Tarlov scores, with treatment (group) and time

219

of data collection as factors. Physiological variables used for statistical analysis were

220

recorded at three predefined time points: (1) before the beginning of surgical stimulation

221

(used as baseline); (2) immediately after skin incision; and (3) immediately after the

222

beginning of tibial osteotomy.

223 224

The duration of anaesthesia and analgesia, as well as the number of intra-

225

operative fentanyl and post-operative buprenorphine boluses received by each group,

226

were tested using either one way ANOVA followed by a Bonferroni multiple

227

comparison test, or Kruskal Wallis ANOVA on ranks followed by Dunn’s test. The

228

proportion of dogs within each group showing hypotension and/or bradyarrhythmias

229

were analysed with Fisher’s exact test.

230 231 232

Commercially available software (NCSS-2007, SigmaStat; SigmaPlot 12, Systat Software) were used. P values < 0.05 were considered statistically significant.

233

Page 10 of 20

234 235

Results Eleven female dogs (seven of which were neutered) and nine males (six of

236

which were neutered) were enrolled in the study. The dogs weighed 35.5 ± 22.0 kg,

237

with a mean age of 12.5 ± 5.8 years. Data for age, bodyweight, intra-operative fentanyl

238

requirement, duration of anaesthesia and analgesia and plasma magnesium

239

concentrations were normally distributed. The number of cases per treatment group was

240

equally distributed between the two observers.

241 242

Duration of anaesthesia was 267.3 ± 36.0 min (mean ± standard deviation) in

243

group M and 282 ± 36.0 min in group C; this difference was not statistically significant

244

(P = 0.37). Group M experienced significantly longer analgesia (527.0 ± 341.0 min;

245

Fig. 1) and required fewer intra-operative fentanyl boluses (median 0, range 0-1; Fig. 2)

246

than group C (176.0 ± 109.0 min, P = 0.015; median 1.5, range 0-4 boluses, P = 0.0018;

247

respectively).

248 249

Intra-operative physiological variables remained within normal ranges for the

250

species and no differences were detected between treatments or time points (Fig. 3).

251

However, one dog in group M and one dog in group C showed moderate sinus

252

bradycardia (40 beats per min) and arterial hypotension (MAP, 50 and 55 mmHg,

253

respectively) shortly after the spinal injection, which responded to glycopyrronium and

254

colloid administration, respectively. None of the dogs required rescue buprenorphine

255

before the final pain assessment.

256 257 258

Post-operatively, there were no significant differences between groups or time points in VAS (P = 0.36 and P = 0.57) and Sammarco (P = 0.17 and P = 0.16) pain

Page 11 of 20

259

scales (Fig. 4). However, group M had significantly lower scores for the Glasgow pain

260

scale (P = 0.012) and the Tarlov scale (P = 0.049) compared to group C (Fig. 4). The

261

Glasgow (P = 0.08) and the Tarlov (P < 0.001) scores significantly increased over time

262

in both groups. Two dogs in group M showed a persistent motor block, accompanied by

263

loss of deep pain sensation, which lasted 24 and 18 h; neurological function of the

264

pelvic limbs normalised progressively and no long-term complications were observed,

265

although these two dogs required longer hospitalisation and were discharged 72 h after

266

surgery.

267 268

Total plasma magnesium concentrations remained within physiological ranges

269

for the species (Fig. 5) and no significant differences were observed between subjects (P

270

= 0.015) or between time points (P = 0.61).

271 272 273

Discussion The main finding of this study is that spinal administration of magnesium

274

potentiates the analgesia provided by ropivacaine, in dogs undergoing elective

275

orthopaedic surgery. However, magnesium also prolongs the duration of motor block,

276

which makes it a less attractive adjunctive analgesic for peri-operative pain in client-

277

owned dogs, especially in those undergoing TPLO, as they are frequently large breed

278

dogs. Persistent motor block is likely to cause discomfort and to increase the costs of

279

hospitalisation. This is in contrast with the findings of most reports focusing on the

280

neuroaxial use of magnesium in both humans and canine patients, which indicated a

281

lack of effect of magnesium on motor function when administered by either epidural or

282

spinal routes (Buvanendran et al., 2002; Yousef and Amr, 2010; Shahi et al., 2014;

283

Bahrenberg et al., 2015). Nonetheless, magnesium was found to prolong and enhance

Page 12 of 20

284

brachial plexus motor block when used with lidocaine in human patients (Haghighi et

285

al., 2015).

286 287

The mechanisms by which magnesium causes motor block are unknown.

288

Magnesium sulphate is an inorganic salt which readily dissolves in water and becomes

289

almost completely dissociated across a wide pH range, from the low pH of the stomach

290

to the neutral pH of extracellular and cerebrospinal fluids1. An in vitro study on isolated

291

mammalian dorsal root ganglion neurons showed that bivalent and trivalent metal

292

cations transiently block voltage-activated calcium channel currents (Busselberg et al.,

293

1994). The ionised magnesium released by its salt could have acted so, blocking the

294

calcium currents by altering the resting potential of the neuronal membrane within the

295

spinal cord.

296 297

Another possible explanation is that, because the magnesium sulphate solution

298

used was hyperosmolar, it might have altered the osmotic homeostasis of cerebrospinal

299

fluid and spinal cord, leading to axonal shrinking and transient neurological

300

dysfunction. In vitro studies have shown that osmotically perturbed neurons are capable

301

of regulating their membrane capacitance, structural organisation and topology, and that

302

these changes are reversible (Wan et al., 1995; Mills et al., 1998). Furthermore,

303

dynamic changes in neuronal volume and surface area caused by osmotic manipulation

304

of isolated ganglia resulted in blockade of transmembrane sodium channels (Mills et al.,

305

1998), which is also a well-recognised mechanism of action by which local anaesthetics

306

interrupt sensory and motor transmission. For most solutions, however, osmolarity and 1

See: http://george-eby-research.com/html/stability_constants.html. (accessed 8 June 2015).

Page 13 of 20

307

specific gravity usually change in parallel, and therefore this explanation is less likely

308

because the solution for injection in both groups had very similar specific gravity.

309 310

Spinal administration of local anaesthetics has been shown to provide adequate

311

analgesia to dogs undergoing orthopaedic procedures (Sarotti et al., 2011), and is a

312

commonly used technique in clinical practice. For this reason, ropivacaine was selected

313

for use in the positive control group in this trial.

314 315

Spinal administration of magnesium did not increase total plasma magnesium

316

concentrations in the dogs enrolled in this trial. However, one limitation of our methods

317

is that plasma magnesium concentrations do not correlate with tissue concentrations,

318

with the exception of interstitial fluid and bone, nor does it reflect total body

319

magnesium (Elin, 2010). Moreover, only total magnesium, rather then the ionised,

320

biologically active form of the ion, could be measured. Another limitation is that blood

321

was collected over a relatively short period of time; more frequent sampling over a

322

longer period, though not feasible in client-owned animals, would have provided a more

323

complete picture of magnesium uptake and distribution. However, because clinical signs

324

compatible with hypermagnesaemia were not observed, it is reasonable to assume that

325

ionised magnesium stayed within acceptable ranges for the species.

326 327

Although cardiovascular variables remained within physiologically acceptable

328

limits, spinal injection in both groups resulted in a transient decrease in heart rate and

329

arterial blood pressure. Additionally, one dog in each group experienced persistent

330

hypotension and bradycardia, which required treatment with colloids and

331

anticholinergics. Administration of ropivacaine by the spinal route might result in

Page 14 of 20

332

decreased sympathetic outflow to the cardiovascular system (Levin et al., 1998).

333

However, because the incidence of cardiovascular side effects did not differ between

334

treatments, it is unlikely that they were caused by magnesium.

335 336

Conclusions

337

The addition of magnesium sulphate to spinal ropivacaine increased the intensity

338

and the duration of peri-operative analgesia in dogs undergoing orthopaedic surgery, but

339

the potential for prolonged motor block could limit its utility in clinical practice. Further

340

research might help to identify a dose with similar analgesic effects but with less

341

potential for prolonged motor block.

342 343 344 345

Conflict of interest statement None of the authors have financial or personal relationships with individuals or organisations that could inappropriately influence or bias the content of the paper.

346 347 348

Acknowledgements The authors thank Dr. Christopher Seymour for kindly revising this manuscript.

349 350 351 352 353 354 355 356 357 358 359 360 361 362

References Adami, C., Veres-Nyéki, K., Spadavecchia, C., Rytz, U., Bergadano, A., 2012. Evaluation of peri-operative epidural analgesia with ropivacaine, ropivacaine and sufentanil, and ropivacaine, sufentanil and epinephrine in isoflurane anesthetized dogs undergoing tibial plateau levelling osteotomy. The Veterinary Journal. 194, 229-234. Arcioni, R, Palmisani, S., Tigano, S., 2007. Combined intrathecal and epidural magesium sulfate supplementation of spinal anesthesia to reduce post-operative analgesic requirements: A prospective, randomized, double-blind, controlled trial in patients undergoing major orthopaedic surgery. Acta Anaesthesiologica Scandinava 51, 482-489.

Page 15 of 20

363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409

Bahrenberg, A., Dzikiti, B.T., Fosgate, G.T., Stegmann, F.G., Tacke, S.P., Rioja, E., 2015. Antinociceptive effects of epidural magnesium sulphate alone and in combination with morphine in dogs. Veterinary Anaesthesia and Analgesia 42, 319-328. Bilir, A., Gulec, S., Erkan, A., Ozcelik, A., 2007. Epidural magnesium reduces postoperative analgesic requirement. British Journal of Anaesthesia 98, 519-523. Busselberg, D., Platt, B., Michael, D., Carpenter, D.O., Haas, H.L., 1994. Mammalian voltage-activated calcium channel currents are blocked by Pb2+, Zn2+, and Al3+. Journal of Neurophysiology 71, 1491-1497. Buvanendran, A., McCarthy, R., Kroin, J., Leong, W., Perry, P., Tuman, K.J., 2002. Intrathecal magnesium prolongs fentanyl analgesia: A prospective, randomized, controlled trial. Anesthesia and Analgesia 95, 661-666. Crociolli, G.C., Cassu, R.N., Barbero, R.C., Rocha, T.L., Gomes, D.R., Nicácio, G.M., 2015. Gabapentin as an adjuvant for postoperative pain management in dogs undergoing mastectomy. Journal of Veterinary Medical Science 77, 1011-1015. Elin, R.J., 2010. Assessment of magnesium status for diagnosis and therapy. Magnesium Research 23, 194-198. Haghighi, M., Soleymanha, M., Sedighinejad, A., Mirbolook, A., Nabi, B.N., Rahmati, M., Saheli, N.A., 2015. The effect of magnesium sulfate on motor and sensory axillary plexus blockade. Anesthesiology and Pain Medicine 5, e21943. Holton, L., Reid, J., Scott, E., Pawson, P., Nolan, A., 2001. Development of a behaviour-based scale to measure acute pain in dogs. Veterinary Record 148, 525-531. KuKanich, B., 2013. Outpatient oral analgesics in dogs and cats beyond nonsteroidal antiinflammatory drugs: An evidence-based approach. Veterinary Clinic of North America: Small Animal Practice 43, 1109-1125. Levin, A., Datta, S., Camann, W.R., 1998. Intrathecal ropivacaine for labor analgesia: a comparison with bupivacaine. Anesthesia and Analgesia 87, 624-627. Madden, M., Gurney, M., Bright, S., 2014. Amantadine, an N-Methyl-D-Aspartate antagonist, for treatment of chronic neuropathic pain in a dog. Veterinary Anaesthesia and Analgesia 41, 440-441. Mills, L.R., Morris, C.E., 1998. Neuronal plasma membrane dynamics evoked by osmomechanical perturbations. The Journal of Membrane Biology 166, 223-235.

Page 16 of 20

410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

Norkus, C., Rankin, D., Warner, M., KuKanich, B., 2015. Pharmacokinetics of oral amantadine in greyhound dogs. Journal of Veterinary Pharmacology and Therapeutics 38, 305-308. Oezalevli, M., Cetin, T., Unlugenc, H., Guler, T., Isik, G., 2005. The effect of adding intrathecal magnesium sulphate to bupivacaine-fentanyl spinal anaesthesia. Acta Anaesthesiologica Scandinava 49, 1514-1519. Sammarco, J., Conzemius, M., Perkowski, S., Weinstein, M.J., Gregor, T.P., Smith, G.K., 1996. Postoperative analgesia for stifle surgery: A comparison of intraarticular bupivacaine, morphine, or saline. Veterinary Surgery 25, 59-69. Sarotti, D., Rabozzi, R., Corletto, F., 2011. Efficacy and side effects of intraoperative analgesia with intrathecal bupivacaine and levobupivacaine: A retrospective study in 82 dogs. Veterinary Anaesthesia and Analgesia 38, 240-251. Shahi, V., Verma, A.K., Agarwal, A., Singh, C.S., 2014. A comparative study of magnesium sulfate vs dexmedetomidine as an adjunct to epidural bupivacaine. Journal of Anaesthesiology, Clinical Pharmacology 30, 538-542. Simpson, J., Eide, T., Schiff, G., Clagnaz, J.F., Hossain, I., Tverskoy, A., Koski, G., 1994. Intrathecal magnesium sulfate protects the spinal cord from ischemic injury during thoracic aortic cross-clamping. Anesthesiology 81, 1493-1499. Steffey, E. and Mama, K., 2007. III Pharmacology In: Tranquilli W, Thurmon J,Grimm K, eds. Lumb and Jones' Veterinary Anesthesia and Analgesia. Yousef, A.A., Amr, Y.M., 2010. The effect of adding magnesium sulphate to epidural bupivacaine and fentanyl in elective caesarean section using combined spinal-epidural anaesthesia: A prospective double blind randomised study. International Journal of Obstetric Anesthesia 19, 401-404. Wan, X., Harris, J.A., Morris, C.E., 1995. Responses of neurons to extreme osmomechanical stress. The Journal of Membrane Biology 145, 21-31.

Page 17 of 20

446

Figure legends

447 448

Fig. 1. Duration of analgesia (min) in dogs receiving a spinal injection of either

449

ropivacaine alone (group C, n = 10) or a combination of ropivacaine and magnesium

450

(group M, n = 10). The asterisk indicates a statistically significant difference between

451

treatments (P < 0.05). The lines indicate median values. The upper and lower boxes

452

indicate the 75% and 25% of the values which fall below the upper and lower quartiles,

453

respectively. The upper and lower whiskers indicate the maximum and minimum

454

values.

455 456

Fig. 2. Number of intra-operative fentanyl boluses (3 µg/kg each bolus) administered to

457

20 dogs receiving a spinal injection of either ropivacaine alone (group C, n = 10) or a

458

combination of ropivacaine and magnesium (group M, n = 10). The lines indicate

459

median values. The upper and lower boxes indicate the 75% and 25% of the values

460

which fall below the upper and lower quartiles, respectively. The upper and lower

461

whiskers indicate the maximum and minimum values. The dots indicate the outliers.

462

The asterisks indicate a statistically significant difference between treatments (P <

463

0.05).

464 465

Fig. 3. Values for heart rate (HR, beats per min [bpm]) and mean arterial pressure

466

(MAP, expressed in mmHg) for 20 dogs receiving a spinal injection of either

467

ropivacaine alone (group C, n = 10) or a combination of ropivacaine and magnesium

468

(group M, n = 10). Data were recorded at three different time points: (1) before the

469

beginning of surgical stimulation (used as baseline); (2) immediately after skin incision;

470

and (3) immediately after the beginning of tibial osteotomy. The lines indicate median

Page 18 of 20

471

values. The upper and lower boxes indicate the 75% and 25% of the values which fall

472

below the upper and lower quartiles, respectively. The upper and lower whiskers

473

indicate the maximum and minimum values.

474 475

Fig. 4. Values for Sammarco pain score, Glasgow pain scale, Visual Analogue Scale

476

(VAS) and Tarlov scale for 20 dogs receiving a spinal injection of either ropivacaine

477

alone (group C, n = 10) or a combination of ropivacaine and magnesium (group M, n =

478

10). Data were recorded at four different time points: at recovery, as soon as the dogs

479

were conscious enough to be examined (time point 1), and then 1, 2 and 3 h after that

480

(time points 2, 3 and 4, respectively). The lines indicate median values. The upper and

481

lower boxes indicate the 75% and 25% of the values which fall below the upper and

482

lower quartiles, respectively. The upper and lower whiskers indicate the maximum and

483

minimum values. The asterisks and the daggers indicate statistically significant

484

differences between treatments and between time points, respectively (P < 0.05).

485 486

Fig. 5. Mean values ( standard deviations) of total plasma magnesium concentrations

487

(mmol/L) for 20 dogs receiving a spinal injection of either ropivacaine alone (group C,

488

n = 10) or a combination of ropivacaine and magnesium (group M, n = 10). Blood was

489

sampled at the following time points: before spinal injection (0, baseline), and then at

490

15, 60, 120 and 240 min thereafter.

491

Page 19 of 20

492

Table 1

493

Tarlov’s scale (modified from Buvanendran et al., 2002) used for neurological

494

assessment

495

Grade

Description

Grade 0

Flaccid paraplegia, no movements of the pelvic limbs, possible loss of bowel/urinary bladder control

Grade 1

Spastic paraplegia with moderate or vigorous purposeless movements of the pelvic limbs. No sitting, unable to walk

Grade 2

Good movements of the pelvic limbs but unable to stand

Grade 3

Able to stand but unable to walk normally, hips and pelvic limbs obviously unstable, moderate to severe ataxia

Grade 4

Able to stand and walk normally, some muscle weakness of the pelvic limbs may be seen

496 497 498

Page 20 of 20