Perioperative analgesic effects of the erector spinae plane block with bupivacaine or bupivacaine-dexmedetomidine in dogs undergoing hemilaminectomy: A randomized controlled trial

Abstract

Objective

To compare the perioperative opioid requirements among dogs receiving an erector spinae plane (ESP) block with bupivacaine, with or without dexmedetomidine, and a control group.

Animals and procedure

Thirty client-owned, healthy adult dogs undergoing hemilaminectomy were included in this randomized, prospective, blinded clinical study. Dogs were randomly assigned to 1 of 3 treatment groups: Group B, ESP block with bupivacaine; Group BD, ESP block with bupivacaine and dexmedetomidine; and Group C, control. Rescue intra- and postoperative analgesia consisted of fentanyl and methadone, respectively. Postoperative pain was evaluated using the short form of the Glasgow Composite Measure Pain Scale (CMPS-SF).

Results

In Group BD, 0/10 dogs required intraoperative fentanyl, compared to 9/10 in Group C (P < 0.001), whereas 1/10 required postoperative methadone, compared to 9/10 in Group B (P = 0.003) and 10/10 in Group C (P < 0.001). The total amount of intraoperative fentanyl (μg/kg) was 0 (0 to 4) in Group B and 0 (0 to 0) in BD, compared to 6 (0 to 8) in C (P = 0.004 and P < 0.001, respectively). Postoperative methadone (mg/kg) required during the first 12 h was 0.5 (0 to 1.4) in Group B (P = 0.003) and 0 (0 to 0) in BD (P < 0.001), compared to C (P = 0.003 and P < 0.001, respectively).

Conclusion

An ESP block with bupivacaine, with or without dexmedetomidine, was associated with a reduction in perioperative opioid consumption and provided effective acute pain control.

Introduction

The erector spinae plane (ESP) block is an injection of local anesthetic (LA) between the erector spinae muscle group and the transverse processes of the thoracic vertebrae (1). In veterinary medicine, the ESP block was recently described in 4 cadaveric studies in dogs, at the thoracic and lumbar levels, using various approaches (parasagittal and transverse) (2–5). Although controversies regarding the mechanism of action have been discussed, these studies reported consistent staining of the dorsal branches of the spinal nerves (DBSN). In particular, medial and lateral DBSN innervate the epaxial muscles, the vertebral laminae, the facet joints, and the skin of the dorsolateral aspect of the trunk (6). Therefore, an ultrasound-guided ESP block has been used to provide analgesia in dogs undergoing hemilaminectomy (7–9). A preoperative ESP block in dogs undergoing hemilaminectomy was associated with a reduced incidence of intraoperative bradycardia and hypotension, a lower perioperative opioid requirement, and a shorter interval to voluntary food intake (7–10). However, these findings should be interpreted with caution due to their retrospective nature.

The combination of LA with α2-adrenoceptor agonists, such as dexmedetomidine, has recently gained attention as an approach to prolong the sensory nerve blockade. In particular, perineural injections of dexmedetomidine (0.5 or 1 μg/kg) combined with 0.5% ropivacaine or 0.5% bupivacaine, respectively, improved the sensory duration of femoral and sciatic nerve blocks in dogs, with no reported adverse effects (11,12).

After an online literature search using PubMed (http://www.pubmed.gov/), ScienceDirect (https://www.sciencedirect.com), Wiley Interscience (https://onlinelibrary.wiley.com), and Google Scholar (https://scholar.google.com) databases, the authors did not identify any prospective study evaluating the analgesic efficacy of the ESP block in dogs undergoing hemilaminectomy. The primary aim of this study was to evaluate the postoperative methadone requirements in dogs receiving an ESP block versus those not receiving any block, and the secondary aim was to investigate the effects of combining dexmedetomidine with bupivacaine for the ESP block. We hypothesised lower postoperative methadone requirements in dogs receiving the block, and a longer analgesic effect when dexmedetomidine was added to the LA.

Materials and methods

Animals

The study was approved by the Institutional Animal Care and the Clinical Research and Ethical Review Board at The Royal Veterinary College; and signed, written, informed consent was obtained from all owners before the dogs entered the study. This study is reported following the CONSORT guidelines (13) and a flow diagram is provided (Figure 1).

Figure 1.

Consolidated Standards of Reporting Trials (CONSORT) flow diagram of the progress through the phases of a parallel randomized trial of 3 groups.

B — Bupivacaine; BD — Bupivacaine + dexmedetomidine; C — Control; CMPS-SF — Glasgow Composite Measure Pain Scale; ESP — Erector spinae plane; IM — Intramuscular; IPPV — Intermittent positive-pressure ventilation; IV — Intravenous.

The study included 30 dogs with acute progressive painful/ non-painful T3-L3 myelopathy. Physical examination, venous blood gas, creatinine, PCV, and total solids were used to assess the health statuses of the dogs. Inclusion criteria were client-owned dogs > 1 y of age; American Society of Anesthesiologists physical status classification II to III; and intervertebral disc extrusion Hansen Type I or II undergoing thoracic, lumbar, or thoracolumbar hemilaminectomy. Dogs were excluded if they were receiving any behavior-modifying drugs, had the presence of dermatological lesions in the region of the dorsum, required intramuscular premedication, were deep pain negative, or required hemilaminectomy on > 2 consecutive spaces.

Anesthesia

All dogs included in the study underwent magnetic resonance imaging (MRI) for diagnosis and neurolocalization of the intervertebral disc extrusion, followed by surgery under the same general anesthetic.

After aseptic preparation, a catheter was placed in the cephalic vein. The dog was then premedicated intravenously (IV) with 0.2 mg/kg methadone and 1 μg/kg dexmedetomidine. After 10 min, preoxygenation was performed using a flow-by technique and administering oxygen at 6 L/min for 3 min. Anesthesia was induced with propofol titrated to effect, and then dogs were intubated and general anesthesia was maintained with 2.5% sevoflurane in 100% oxygen (initial flow rate of 4 L/min). Volume-control ventilation was set at a tidal volume of 10 mL/kg I:E (inspiratory:expiratory) 1:2, and the respiratory rate was adjusted to maintain normocapnia (Pe’CO₂ 35 to 45 mmHg).

Electrocardiography, heart rate (HR), noninvasive blood pressure, esophageal temperature (T), respiratory rate, capnography, Pe’CO₂, pulse oximetry, and end tidal sevoflurane (Fe’Sevo) were measured and recorded every 5 min from induction to the end of the procedure. Two multiparametric monitors (Expression IP5 multiparametric monitor; MRI-Devices, Philips, Farnborough, UK and Datex-Ohmeda S/5; Datex-Ohmeda, Buckinghamshire, UK) were used during MRI and surgery, respectively. The only vital parameter not monitored during the MRI scan was the ECG. The size of the blood pressure cuff was based on published guidelines and placed in the antebrachium area (14). Hartmann’s solution was administered at 5 mL/kg per hour. After MRI, the dog was transferred to an anesthesia preparation room where the hair was clipped and the skin aseptically prepared. At this stage, the anesthetist (anesthesia nurse or veterinarian) in charge of the case was asked to leave the room to remain blinded with regards to the experimental intervention. The dog was then randomly allocated to a treatment group and the treatment completed. Thereafter, the dog was transferred to the surgical theatre and connected to a rebreathing system and ventilator. Active warming was provided during surgery in case of hypothermia (T < 37°C).

During general anesthesia, hypotension was defined as mean arterial blood pressure (MAP) < 60 mmHg (15) and treated after 2 consecutive readings. If bradycardia (HR < 60 bpm) was present at the time of hypotension, glycopyrrolate was administered IV at 10 μg/kg.

Experimental protocol

Prior to hemilaminectomy, an online tool (www.randomizer.org) was used to randomly assign dogs to 1 of the 3 study groups: bupivacaine (B), bupivacaine + dexmedetomidine (BD), and control (C). In Group B, an ultrasound-guided ESP injection with bupivacaine (Marcain Polyamp Steripack 0.5%; Pfizer), 0.4 mL/kg, was administered; in Group BD, the same block was administered using dexmedetomidine, 1 μg/kg, diluted in bupivacaine 0.5%, with a total volume of 0.4 mL/kg. No locoregional technique was used in dogs in Group C. All blocks were performed unilaterally by 3 anesthesia residents trained for this technique under the direct supervision of a European College of Veterinary Anaesthesia and Analgesia diplomate, using parasagittal approaches described by Portela et al (2020) (3) for the thoracic area, and by Medina-Serra et al (2021) (4) for the lumbar region.

Ultrasound-guided blocks were completed using a 15-6 MHz linear transducer (SonoSite, Bothell, Washington, USA) and the last rib as the reference anatomical landmark to identify the correct space. The target vertebral chosen was the cranial in the surgical site (i.e., T13 in case of T13-L1 hemilaminectomy; L1 in case of L1–L3 hemilaminectomy). The transducer was positioned longitudinal to the spine on the site to be operated and was moved medially until the transverse process was visualised. Using an in-plane technique, an 8-centimeter echogenic insulated needle (22-gauge, 80-millimeter Ultraplex 360 needle; B. Braun Medical, Chapeltown, UK) was advanced through the erector spinae musculature in a craniodorsal-to-caudoventral direction until the tip was visualized in contact with the dorsal aspect of the transverse process. Hydrodissection of the thoracolumbar fascia after a test volume injection of 0.5 mL of LA subtracted from the total volume was used to confirm correct positioning of the needle. Only 1 test injection was done in each case.

Once in the operating theatre, for all dogs, Fe’Sevo was initially targeted at a value of 2.3% before skin incision, to achieve an adequate plane of anesthesia, and then maintained during surgical procedure based of clinical endpoints (blood pressure, palpebral reflex, jaw tone). Five minutes before skin incision, baseline values of HR and MAP were recorded. Rescue analgesia was fentanyl, 2 μg/kg, IV, when a 20% increase above baseline was observed in HR and/or MAP (16). If 3 fentanyl doses were required within 20 min, a constant rate infusion (CRI) of fentanyl, 12 μg/kg per hour, was started. At tracheal extubation, a recovery score was assigned by the main anesthetist, using a published scale (17): 1 — calm, 2 — transient whining or limb movement, 3 — uncoordinated behavior, 4 — agitation. Dogs with a score > 2 received dexmedetomidine, 1 μg/kg, IV.

Postoperative pain assessments were completed by 4 qualified veterinary nurses unaware of the study group allocation, using the short form of the Glasgow Composite Measure Pain Scale (18). Pain scores were assessed immediately after extubation, every hour during the first 4 h post-extubation, and every 4 h during the following 20 h. In case of a score above 5/20, methadone, 0.2 mg/kg, IV, was administered as rescue analgesia. In case of a pain score remaining above the threshold value 20 min after methadone administration, a fentanyl CRI at 3 to 6 μg/kg per hour was started after a loading dose of 2 μg/kg fentanyl, IV. The total amount of opioids required during the first 24 h post-hemilaminectomy was recorded. Non-opioid analgesic drugs such as gabapentin, paracetamol, and NSAIDs, were included in the postoperative analgesic protocol, based on preference of the neurologist, who was blinded to group allocation.

Statistical analyses

Sample size calculation using a priori analysis based on data retrospectively collected from the same institution, assuming probability (power) of 0.8 and α of 0.05, concluded that 10 dogs per group were necessary. The total dose of postoperative methadone was chosen as the outcome of interest for power calculation, assuming a standard deviation of 0.6, based on a retrospective analysis comparing dogs receiving systemic analgesia or an ESP block with bupivacaine for hemilaminectomy during the first 12 h postoperatively. A post-hoc analysis was done to confirm the adequacy of the sample size. Dedicated statistical software was used for sample size calculation and post-hoc power analysis (G*Power version 3.1.9.6; http://www.gpower.hhu.de/en.html).

Data distribution was assessed with Shapiro-Wilk test, with data presented as median (range). Chi-squared and Fisher exact tests for pairwise comparisons, with Bonferroni correction, were used to compare breed distribution; target vertebra for ultrasound-guided needle placement (thoracic versus lumbar); number of intervertebral discs affected (1 versus 2); incidence of hypotension and hypothermia; number of dogs requiring fentanyl, methadone, any rescue opioid; and postoperative non-opioid analgesic drug distribution among groups. Differences in age, weight, duration of general anesthesia, surgical time, baseline HR and MAP, recovery score, total intraoperative dose of fentanyl (excluding CRIs), and postoperative total dose of methadone (total and within 12 h after extubation) were assessed using Kruskal-Wallis and Dunn tests, with Bonferroni correction for pairwise comparisons.

Kaplan-Meier curves were used to assess differences in time to rescue analgesia, both in the intra- and postoperative periods. Log-rank test and Bonferroni correction were used for pairwise comparison between curves. Statistical significance was considered with P < 0.05.

Data analyses were conducted using SPSS software (Version 28; IBM, Armonk, New York, USA). Graphs were generated with Prism software (Version 9.4.1; GraphPad, San Diego, California, USA).

Results

A total of 33 dogs were recruited for this study and allocated to groups with a ratio of 1:1:1. A total of 3 dogs were excluded from the study: 1 due to behavioral reasons that did not allow a reliable postoperative pain assessment, and 2 due to missing post-operative data. Descriptive statistics are in Tables 1 and 2. There were no significant differences among groups for age, weight, duration of general anesthesia, surgical time, pre-incisional HR and MAP, median Fe’Sevo, recovery score, breed, number of intervertebral discs affected, target vertebra for ultrasound-guided needle placement, incidence of hypotension and hypothermia, and postoperative non-opioid analgesic drugs. All hypotensive dogs were also bradycardic and received glycopyrrolate shortly after the beginning of the MRI scan. None of the dogs included in the study required additional dexmedetomidine administered during the recovery period. The number of non-opioid analgesic drugs (single treatment, 2 or 3 drugs concurrently) did not differ among groups (P = 0.775).

Table 1.

Descriptive statistics of the studied population of dogs. Results expressed as median (range).

Variable	Group C (n = 10)	Group B (n = 10)	Group BD (n = 10)	P-value
Age (y)	5.5 (4 to 8)	5 (3 to 7)	4.5 (3 to 12)	0.449
Weight (kg)	6.45 (3.5 to 11.3)	9.65 (4.88 to 40)	10.2 (4.8 to 16.2)	0.071
GA duration (min)	215 (180 to 240)	195 (100 to 270)	180 (135 to 320)	0.365
Surgical time (min)	165 (120 to 210)	150 (60 to 240)	120 (70 to 300)	0.412
Baseline HR (bpm)	55 (35 to 90)	53 (40 to 83)	58 (35 to 80)	0.810
Baseline MAP (mmHg)	75 (60 to 90)	75 (60 to 85)	80 (65 to 90)	0.592
Median Fe’Sevo (%)	2.3 (2.2 to 2.4)	2.3 (2.2 to 2.3)	2.3 (2.2 to 2.3)	0.909
Recovery score	1 (1 to 2)	1 (1 to 1)	1 (1 to 2)	0.328

Group C — Control; Group B — Erector spinae plane block with bupivacaine; Group BD — Erector spinae plane block with bupivacaine and dexmedetomidine.

Fe’Sevo — End tidal sevoflurane; GA — General anesthesia; HR — Heart rate; MAP — Mean arterial blood pressure.

Table 2.

Descriptive statistics of the studied population of dogs. Results expressed as number of dogs.

Variable	Group C (n = 10)	Group B (n = 10)	Group BD (n = 10)	P-value
Dachshunds (n)	8	4	5	0.259
Hemilaminectomy 2 spaces (n)	3	3	3	1.000
Target vertebra for the block (nT; nL)	5T; 5L	5T; 5L	7T; 3L	0.723
Hypothermia (n)	7	6	4	0.562
Hypotension (n)	3	5	2	0.498
Postoperative gabapentin (n)	4	3	4	1.000
Postoperative paracetamol (n)	7	6	7	1.000
Postoperative meloxicam (n)	7	8	7	1.000

Group C — Control; Group B — Erector spinae plane block with bupivacaine; Group BD — Erector spinae plane block with bupivacaine and dexmedetomidine.

L — Lumbar. T — Thoracic.

The number of dogs requiring rescue opioids (intraoperatively, postoperatively and overall), and the amounts of intraoperative fentanyl and postoperative methadone administered as rescue analgesia are in Table 3. The number of dogs requiring intraoperative fentanyl differed for Groups BD and C (P < 0.001). The number of dogs requiring methadone postoperatively differed for Groups B and BD (P = 0.003) and Groups BD and C (P < 0.001). When the overall perioperative opioid requirement was considered, the number of dogs receiving a rescue dose of either fentanyl or methadone differed between Groups B and BD (P = 0.003) and Groups BD and C (P < 0.001).

Table 3.

Dogs requiring perioperative rescue opioids, intraoperative fentanyl, and postoperative methadone boluses administered as rescue analgesia. Results expressed as n or median (range).

Variable	Group C (n = 10)	Group B (n = 10)	Group BD (n = 10)
Dogs requiring intraoperative fentanyl (n)	9	3	0a
Dogs requiring postoperative methadone (n)	10	9	1a,b
Dogs requiring perioperative opioids (n)	10	9	1a,b
Intraoperative fentanyl (μg/kg)	6 (0 to 8)	0 (0 to 4)a	0 (0 to 0)a
First 12 h postoperative methadone (mg/kg)	0.6 (0.2 to 0.8)	0.5 (0 to 1.4)a	0 (0 to 1.2)a
12 to 24 h postoperative methadone (mg/kg)	0.7 (0 to 0.08)	0.4 (0 to 0.8)	0 (0 to 0.8)a
Total 24 h postoperative methadone (mg/kg)	1.3 (0.2 to 1.4)	0.5 (0 to 1.4)	0 (0 to 1.2)a

Group C — Control; Group B — Erector spinae plane block with bupivacaine; Group BD — Erector spinae plane block with bupivacaine and dexmedetomidine.

^aDifferent from Group C (P < 0.05).

^bDifferent from Group B (P < 0.05).

Dogs in Groups B and BD received a lower dose of fentanyl (μg/kg) than dogs in Group C (P = 0.004 for Group B versus C, and P < 0.001 for Group BD versus C). A fentanyl CRI was required in 3/30 (3.3%) dogs, all allocated to Group C. When the dose of postoperative methadone required during the first 12 h was compared among groups, dogs in Groups BD and B received a lower dose compared to those in Group C (P = 0.004 for Group B versus C, and P < 0.001 for Group BD versus C). Regarding the dose of methadone received between 12 and 24 h and the total dose of postoperative methadone, the only significant difference was between Groups BD and C (P = 0.009).

Postoperative Glasgow Composite Measure Pain Scale scores are in Figure 2, including data for dogs receiving postoperative rescue analgesia. Differences in pre- and postoperative time to first rescue analgesic treatment in the 3 groups are in Figures 3 and 4. Based on anesthetists’ notes, 8/10 dogs in Group C received the first rescue fentanyl bolus at the time of skin incision and 1/10 received it during muscle incision. In Group B, 3/10 dogs received fentanyl during removal of disc material from the spinal canal. No dog reacted intraoperatively in Group BD.

Figure 2.

Box-and-whisker plot of Glasgow Composite Measure Pain Scale (CMPS-SF) scores in the 3 groups at the first 4 h post-extubation (t1, t2, t3, t4) and every 4 h (t8, t12, t16, t20, t24) during the first 24 h postoperatively.

C — Control; B — Bupivacaine; BD — Bupivacaine + dexmedetomidine.

The CMPS-SF reported also included dogs that received opioid rescue analgesia as well as other postoperative analgesics.

Figure 3.

Kaplan-Meier curve of the time to first intraoperative fentanyl rescue bolus in the 3 groups over time, defined as minutes intraoperatively.

C — Control; B — Bupivacaine; BD — Bupivacaine + dexmedetomidine.

Figure 4.

Kaplan-Meier curve of the time to first postoperative methadone rescue bolus in the 3 groups over time, defined as hours postoperatively.

C — Control; B — Bupivacaine; BD — Bupivacaine + dexmedetomidine.

For time-to-event analysis, there were differences among groups in the time of first fentanyl (P < 0.001) and methadone (P < 0.001) doses. There were differences between Groups B and C (P = 0.001) and Groups BD and C (P < 0.001) for fentanyl, and between Groups B and C (P < 0.001), BD and C (P < 0.001), and B and BD (P = 0.001) for methadone.

No complications other than those commonly related to general anesthesia were recorded for any of the dogs at any time.

Discussion

The findings of our study agreed with those of previous studies (8,9) in which an ESP block with bupivacaine reduced the intraoperative rescue fentanyl administration and methadone requirements during the first postoperative 12 h in dogs. Intraoperatively, none of the dogs in Groups B or BD received fentanyl during skin incision and muscle dissection, consistent with cadaveric studies describing staining of the DBSN (2–4).

Methadone use from the 12th to the 24th postoperative hour was significantly lower in dogs from Group BD compared to Group C, but was not different when comparing Groups BD versus B, or Groups B versus C. Furthermore, the time to first rescue analgesia postoperatively was significantly longer in Group BD compared to Group B and the control group. Although dexmedetomidine could have contributed to prolonging the sensory blockade in Group BD, the limited number of dogs included in the study could have masked differences between groups.

Notwithstanding the paramount role of full μ-agonist opioids in the management of postoperative pain (19), their use can result in several adverse effects. In fact, the use of these drugs in dogs has been associated with higher incidence of bradycardia, hypoventilation, ileus, postoperative nausea and vomiting, dysphoria, prolonged hospitalization, and hyperalgesia (8,20). Although clinical effects of some of these side effects in dogs are still to be clarified, a reduction of the total dose of opioids administered perioperatively could be an important factor in the overall well-being of hospitalized dogs.

Based on our findings, we inferred that the ESP block has an important role in a multimodal analgesic protocol for dogs undergoing hemilaminectomy, and that the adjunction of dexmedetomidine at 1 μg/kg could prolong postoperative analgesia, in agreement with studies in humans (21,22).

There are various theories to explain the analgesic effect of dexmedetomidine, though the actual mechanism remains unclear. First, dexmedetomidine can cause local vasoconstriction, thus delaying absorption and prolonging the effect of the co-administered LA (23). The second theory regards the direct effects of dexmedetomidine on peripheral nerve activity, such as blockade of the hyperpolarization-activated cation current and attenuation of the acute LA-induced perineural inflammation (24). A third hypothesised mechanism is inhibition of nociception by reducing release of neurotransmitters such as substance P and glutamate (25).

Paravertebral and epidural spread of the injectate have been described in humans after ESP block (26). Although never reported in canine cadaveric studies using a parasagittal approach, this spread could be a mechanism of action of this block in our population.

The dose of dexmedetomidine chosen for this study was based on a study by Marolf et al (2021) (11), in which the duration of sensory blockade was prolonged when this drug was co-administered perineurally at this dose, without any reported adverse effects. Further studies testing various doses of dexmedetomidine as an adjunctive drug for this block are needed to investigate any dose-effect relationship in dogs.

The present study had several limitations. The sample size was adequate for assessing the number of methadone doses needed postoperatively, but statistical power calculation could be lower for other comparisons. The main limitation of our study was the inclusion of dogs undergoing hemilaminectomy in various thoracic and lumbar intervertebral spaces, representing a lack of standardization of the surgical site. Anatomical differences among regions may have an important role in the spread of LA in the fascial plane. Portela et al (2020) (3) reported staining of 4 (2–7) medial branches and 4 (2–5) lateral branches after injection of 0.3 mL/kg at the level of the ninth thoracic vertebra, whereas Medina-Serra et al (2021) (4) reported staining of 2 (2–3) DBSN using the transversal approach and 0 (0–3) DBSN using the parasagittal approach with the injection of 0.4 mL/kg at the level of the 4th lumbar vertebra. In our study, there were no significant differences among groups in the distribution of thoracic versus lumbar surgery. However, as this was not considered the main outcome of the study, the sample size may not have been adequately powered for these data.

Although the distribution of breeds was not different among groups from a statistical perspective, most dogs in Group C were dachshunds. One limitation of the study is a possible Type II error when assessing differences in the number of dogs of this breed between groups, since the study was not powered for this outcome. Dachshunds have previously been associated with lower HR under anesthesia (27), which may have affected the intraoperative rescue analgesia threshold.

Another limitation was the use of dexmedetomidine as premedication agent in all dogs, which could have affected the duration of nerve blockade in both Groups B and BD. There are some reports of a prolonged sensory and motor blockade when dexmedetomidine was administered IV in dogs (28). However, the dose used in the present study was lower than that in a previous paper (11) that did not report any effect of systemic dexmedetomidine administered IV at least 1 h before a peripheral block in dogs.

Concurrent administration of other analgesic drugs, such as gabapentin, meloxicam, and paracetamol, is considered another limitation of our study since these have disparate analgesic and sedative properties that could have influenced the study results. Thus, the postoperative pain scores should be interpreted with caution. However, there were no significant differences among groups in concurrent drug administration.

A previous retrospective study in dogs investigating the effects of ESP block reported postoperative benefits up to 48 h after surgery (8). In light of this finding, limiting the postoperative period of this study to the first 24 h post-extubation could be considered another limitation. A longer study interval may have yielded more data about the long-term effects of the dexmedetomidine co-administration. A final limitation was the heterogeneity of neurosurgeons and anesthetists involved with the cases. Even though all used the same surgical technique, differences in level of experience and tissue handling may have affected the extent of tissue trauma, nociception, and pain among dogs. Three anesthesia residents performed the ESP blocks, and this variability could have influenced the effectiveness of the technique. To minimize this potential source of bias, all injections were completed under the supervision of the same European College of Veterinary Anaesthesia and Analgesia diplomate.

Notwithstanding the limitations discussed, the authors believe that the overall potential for bias can be considered inherent to the nature of the study, as the design reflected clinical practice in the referral hospital where the study was done. Further prospective studies are needed to evaluate the efficacy of ESP block with bupivacaine and dexmedetomidine in dogs undergoing hemilaminectomy.

In conclusion, the ESP block with bupivacaine, with or without dexmedetomidine, was associated with a reduction in the perioperative opioid consumption in dogs undergoing hemilaminectomy, and provided effective acute pain control. CVJ