Comparison of rostral spread of lumbosacral epidural volume calculated by body weight or length of the vertebral column in small-sized anesthetized dogs

Abstract

The first objective of this prospective, randomized, crossover experimental trial was to compare the rostral spread of lumbosacral epidural volume calculated by body weight (BW) or vertebral column length (LE) in 6 small, isoflurane-anesthetized female beagle dogs (BW: 7.5 to 10.2 kg; LE measured from the occipital crest to the sacrococcygeal space: 46 to 56 cm). The second objective was to assess the response to a noxious stimulus once the dogs recovered from anesthesia and to determine the effects of the injection on cardiopulmonary variables. While in sternal position, dogs were injected through an epidural catheter with a volume mixture of bupivacaine 0.25% and iopamidol 15% based on BW: 0.2 mL/kg or LE: 0.05 mL/cm (< 50 cm) or 0.07 mL/cm (50 to < 70 cm). Rostral spread was determined by counting the number of vertebrae reached by iopamidol using computed tomography. After anesthesia, cardiopulmonary variables, motor function, and responses to nociceptive stimuli were evaluated. Comparisons were completed with mixed linear models and 2-way analysis of variance (ANOVA) (P < 0.05). The volume of injectate (3.29 ± 0.74 versus 1.81 ± 0.21 mL; mean ± SD) and the number of vertebrae (22 ± 2 versus 19 ± 2 vertebrae) reached by iopamidol were significantly greater for LE than for BW. Response to nociception, time to return of pain sensation, motor function, and cardiopulmonary variables were similar between groups. In conclusion, dosing based on LE resulted in larger rostral spread than when based on BW in dogs of small size.

Introduction

Epidural lumbosacral (LS) injection of a local anesthetic with or without an opioid is one of the most common regional techniques used in dogs in clinical practice to provide anesthesia and/or analgesia to the pelvic limb, pelvic area, and abdomen (1–3). The rostral spread of a local anesthetic drug injected into the LS space allows the drug to interact with the spinal cord and emerging spinal nerves in order to exert its analgesic/anesthetic effects (4–7).

For a local anesthetic, the relationship between volume of injectate and rostral spread is linear (8–11) and determines the extent of blockade. A small volume of injectate at the LS space acts close to the site of injection and may only involve sensory blockade of the pelvic area, whereas a larger volume of injectate can be adjusted to reach higher lumbar and lower thoracic segments of the spinal cord to include the abdomen (8,11,12).

In addition to volume of injectate, the degree of epidural rostral spread is influenced by two physical factors: gravity and friction (13). Gravity and friction forces in the epidural space are influenced by inherent factors of the animal, such as body condition score (BCS) and size, position at the time of injection and thereafter, injection at the site of entry of the needle or in a more rostral position with an epidural catheter, and whether the animal is conscious, sedated, or under general anesthesia (8,10–12,14–16).

The epidural volume of injectate can be calculated by two methods, either using body weight (BW; mL/kg) or length of the vertebral column (LE; mL/cm) from the occipital crest to the sacrococcygeal space (8–11,13–16). It has not yet been determined in live dogs whether these methods are interchangeable.

In a recent study, the comparison of both methods of calculation in a hypothetical situation determined that they are not interchangeable in small- and medium-sized dogs (< 25 kg) since larger calculated volumes were obtained based on LE when compared to BW. This difference tended to disappear or revert as size increased (≥ 25 kg) and with higher BCS (obesity) (13). It has not yet been determined whether the methods of calculation result in different volumes of injectate, which could in turn result in different analgesic and cardiopulmonary effects from epidural spread.

Therefore, the objectives of this study were i) to compare in small, anesthetized dogs (≤ 10 kg) the rostral spread of an LS epidural injection volume of bupivacaine–contrast medium calculated by LE or BW using computerized tomography (CT); and ii) to assess the response to a noxious stimulus once the dogs recovered from anesthesia as well as to determine the effects of the injection on cardiopulmonary variables.

We hypothesized that epidural volume calculated using recommended doses based on LE would be greater than that based on BW in small-sized dogs and result in higher rostral spread and desensitization of a greater number of dermatomes, but that cardiopulmonary variables would remain similar in both groups.

Materials and methods

Animals

A total of 6 purpose-bred research, ovariohysterectomized female beagle dogs, aged 5.3 to 8.3 y and weighing 7.5 to 10.2 kg, with the same BCS (3 of 5 = moderate) (17) were studied and considered healthy based on physical examination, complete blood (cell) count, and biochemistry profile. The study was carried out in accordance with the guidelines of the Canadian Council on Animal Care and was approved by the Institutional Animal Care Committee at the University of Guelph (AUP #4347).

Study

Dogs were used in a prospective, randomized, crossover study design, in which each received two treatments, with a washout period of at least 5 d between treatments. On the first occasion, 3 dogs received Treatment 1 (epidural volume based on LE) and the other 3 dogs received Treatment 2 (epidural volume based on BW), followed by the reversed order on the second occasion (http://www.random.org).

Volumes of injectate for the epidural injection consisted of 0.2 mL/kg calculated according to BW and 0.05 mL/cm (< 50 cm) or 0.07 mL/cm (50 to < 70 cm) according to LE (13). On each occasion, the dogs were fasted for 12 h for food, but water was available. Body weight was recorded each day before the experiment and LE was recorded under anesthesia while dogs were in sternal recumbency and with pelvic limbs extended caudally, using a metric tape, during the first occasion.

Epidural injection and evaluation

General anesthesia was induced on the CT table by delivering 5% isoflurane (Aerrane; Baxter Corporation, Mississauga, Ontario) in oxygen at 4 to 5 L/min via facemask with a circle system. The trachea was intubated and dogs were connected to an electronically controlled, volume-cycled ventilator (S/5 Aespire 7900 Ventilator; GE Healthcare, Madison, Wisconsin, USA) at a rate of 10 breaths/min, a tidal volume of 15 mL/kg, and an inspiration-to-expiration ratio of 1:3.

The dogs were positioned in sternal recumbency with the pelvic limbs extended caudally and anesthesia maintained with isoflurane in > 95% oxygen with a fresh gas flow of 1 to 2 L/min at an initial end-tidal isoflurane concentration (Fe′Iso) of 1.5%; adjusted accordingly for depth during placement of the epidural catheter. The Fe′Iso was measured from the connection of the endotracheal tube with the circle system using a multiparameter monitor (Datex-Ohmeda S/5 Anesthesia Monitor; GE Healthcare Finland, Helsinki, Finland), calibrated before each experiment with a standardized calibration gas mixture designed for the analyzer (755571-HEL, Calibration gas mixture; GE Healthcare Finland).

In addition, dogs were instrumented for non-invasive oscillometric arterial blood pressure using a cuff width size of approximately 40% of the limb circumference above the tarsal joint on the right pelvic limb (Size 2), rectal thermometer, capnography, pulse oximetry, and side-stream end-tidal CO₂ (Pe′CO₂). These variables were recorded at 5 to 10 min intervals throughout anesthesia during the CT.

A 5- × 5-cm area over the LS space was clipped and aseptically prepared. A stab incision was made over the LS space and a 24-gauge, 25-cm epidural catheter was advanced through a 20-gauge, 4.5-cm Tuohy needle (Epidural Pain Management Kit; MILA, Florence, Kentucky, USA), 1 to 2 cm into the epidural space from the tip of the needle.

The Tuohy needle was left in place and the catheter flushed with 0.1 mL of sterile saline, followed by injection of a test dose of a 1:1 mixture of preservative-free bupivacaine 0.5% (Bupivacaine Injection BP; SteriMax, Oakville, Ontario) and iopamidol contrast medium (IsoVue–300; Bracco Imaging, Anjou, Quebec), 0.2 to 0.3 mL, resulting in a solution of bupivacaine 0.25% and iopamidol 15%.

An initial CT scan (16-slice detector, BrightSpeed CT Scanner; GE Healthcare, USA) was conducted to confirm the correct placement of the epidural catheter and presence of the iopamidol. The remaining final volume was then injected by hand over 20 s and a CT scan of the entire spine was completed within 5 min of the injection.

Images from the CT were acquired and evaluated from the cranial aspect of the cervical spine to just caudal to the LS space, in both soft tissue (standard) and bone windows, by a Board-certified radiologist (MJ) who was unaware of the treatment. Assessment included the most cranial spread of positive contrast medium to determine the number of vertebrae reached by the iopamidol from the LS space. A maximum of 27 vertebrae was possible (7 lumbar + 13 thoracic + 7 cervical) and was always inclusive of the vertebra the iopamidol reached, regardless of being whole or a fraction of it (Figure 1).

Figure 1.

Diagram used to determine the number of vertebrae reached by iopamidol after lumbosacral epidural injection. A maximum of 27 vertebrae was possible (7 lumbar + 13 thoracic + 7 cervical).

Following the CT, the epidural catheter and needle were removed, isoflurane administration was discontinued, and dogs were transported to a different room and maintained in the same sternal position during recovery until full display of normal conscious behavior. Dogs were extubated when spontaneous breathing returned.

Total anesthesia time, from intubation to extubation, was recorded. During recovery, dogs were assessed for 30 min for post-anesthesia sedation by a blinded investigator (FF) using a recovery score of 0 to 3, with 0 indicating no sedation and 3 indicating profound sedation (see Appendix A) (18). Heart rate, assessed by auscultation, oscillometric blood pressure (Cardell Veterinary Vital Signs Monitor, Model 9403; Midmark Corporation, Versailles, Ohio, USA), and rectal temperature were measured at 10 and 30 min during this period.

Appendix A.

Sedation recovery score used during the first 30 min (recovery period) after epidural injection in dogs anesthetized with isoflurane.

Sedation score
0	Bright and alert, no apparent sedation
1	Calm, minimal sedation (quiet but still alert and aware of surroundings, mild resistance to restraint for lateral recumbency, moderate response to voices and touch)
2	Very calm, with moderate sedation (quiet, relaxed, minimal restraint required for lateral recumbency, mild response to voices or touch)
3	Profound sedation (quiet, very relaxed, no restraint necessary for lateral recumbency, does not respond to voices or touch)

Modified from Skelding et al (18).

At 30, 45, and 60 min post-extubation and every 30 min thereafter, dogs were assessed for ambulation by the same blinded investigator using a score from 0 to 3, with 0 indicating normal ambulation and 3 indicating unable to stand and/or ambulate (see Appendix B), until a score of 0 was achieved. Time to full recovery was recorded as time from extubation to a score of 0 for both the sedation and ambulation scores.

Appendix B.

Ambulation score used at the end of recovery period (30 min) after epidural injection in dogs anesthetized with isoflurane.

Ambulation score
0	Normal ambulation
1	Able to walk but shows ataxia and mild proprioceptive deficit of hind limbs
2	Drags one or both hind limbs when walking; marked proprioceptive deficit
3	Unable to stand and/or ambulate

Analgesia assessments were aimed at detecting the return of pain and were completed by the same blinded investigator and started when the post-anesthesia sedation score was 0, but not before 30 min post-extubation. The assessment consisted of applying a noxious stimulus to 11 dermatomes, starting at the tail and moving cranially to the thoracic limb paw. These included the base of the tail (AN1), the pelvic limb paw (AN2), the area surrounding the iliac crest (AN3), the inguinal area near the femoral nerve (AN4), the area close to the transverse process of L₃ and flank (AN5), the paramedian area next to the umbilicus (AN6), along the thirteenth rib (AN7), the paramedian area to the xyphoid (AN8), along the ninth rib (AN9), along the second rib (AN10), and the thoracic limb paw (AN11).

Noxious stimulation consisted of clamping the tail or paws with 24-cm sponge forceps with protective plastic tubing on each jaw (Robbins Instruments, Deptford, New Jersey, USA) or the use of a 12.5-cm hemostat for the remaining dermatomes. The anatomical region was clamped with the instrument, without reaching the first notch, for a period of < 3 s. A positive response to pain consisted of gross purposeful movement or a vocal response elicited during the period of stimulation.

If the evaluator was uncertain of the response, electrical current was applied using alligator clips 5 cm apart in the anatomical area tested with the surgical instrument and consisted of 1 to 3 single stimuli of 30 to 40 volts at 50 cycles/s for 10 ms (S48 Stimulator; Astro-Med, West Warwick, Rhode Island, USA). If gross purposeful movement or a vocal response was elicited before the cycle was completed, the electrical stimulus was discontinued immediately and the response assessed as positive.

A second positive response obtained in the following assessment time indicated complete sensory recovery and the time to a positive response to noxious stimulation was recorded as the time when the first assessment was completed. All dermatomes were assessed until 2 consecutive positive responses were obtained and were no longer stimulated subsequently.

Dogs were administered meloxicam (Metacam; Boehringer Ingelheim Canada, Burlington, Ontario), 0.1 mg/kg subcutaneously (SC), at the end of each experiment and were monitored for normal behavior. Dogs were returned to the supplier when the last dog had completed the study.

Statistical analysis

All data were tested for normality using Shapiro-Wilk, Kolmogorov-Smirnov, Cramer-von Mises, and Anderson-Darlin procedures. UNIVARIATE and PLOT procedures were used to detect unequal variances, outliers, and other non-random patterns within the data for each group (SAS Version 9.4; SAS Institute, Cary, North Carolina, USA).

A mixed model of residual maximum likelihood was used to compare the effect of the method (LE versus BW dosing) and the period of administration (first and second time) on the number of vertebrae reached by iopamidol. Presence of pain was analyzed with a mixed model of residual maximum likelihood that considered the method and dermatome, with time as the independent variable.

The error structure was selected based on the Akaike Information Criteria among structures offered by SAS using a random effect for each treatment. Cardiopulmonary and temperature variables and sedation and ambulation scores for both methods were analyzed with a mixed model if there were missing values or a 2-way analysis of variance (ANOVA) with repeated measures when all values were recorded and multiple comparisons within treatments were carried out with a Šidák test (GraphPad, Version 9.3.1 for MacOS; GraphPad Software, San Diego, California, USA). Significance was set at P < 0.05.

The sample size was calculated based on the mathematical model for the different methods of calculating the epidural volume (13), using the small-sized (< 10 kg) dog category data. Therefore, to compare the mean volume of injection of 2.2 mL (LE dose) and 1.1 mL (BW dose) with a standard deviation of 0.5, a power of 0.8 (Type-II error) and alpha of 0.05 (Type I-error) required at least 5 dogs.

Results

All 6 dogs completed the study without complication (Table I). One of the dogs weighed 10.2 kg but was kept in the study because the data from this dog was consistent with dogs < 10 kg. Total anesthesia time from intubation until extubation was 24.2 ± 2.1 min in the LE group and 23.7 ± 6.5 min in the BW group.

Table I.

Demographic data of female beagle dogs after a lumbosacral epidural injection of 0.2 mL/kg body weight (BW) and 0.05 mL/cm (< 50 cm) or 0.07 mL/cm (50 to < 70 cm) based on length of the vertebral column (LE) in a randomized crossover fashion of a volume mixture of bupivacaine 0.25% and iopamidol 15%.

Dog number	FT	Age (years)	LE (cm)	BW1 (kg)	BW2 (kg)	Volume for BW (mL)	Volume for LE (mL)
1	LE	5.3	52	10.2	10.2	2.04	3.64
2	BW	7.1	52	9.6	9.5	1.92	3.64
3	LE	5.3	46	7.5	7.5	1.50	2.30
4	LE	5.3	48	7.6	8	1.60	2.40
5	BW	8.3	55	9.6	9.6	1.92	3.85
6	BW	5.3	56	9.4	9.4	1.88	3.92
Mean ± SD		6.1 ± 1.30	51.5 ± 3.9	9.0 ± 1.1	9.0 ± 1.0	1.81 ± 0.21	3.29* ± 0.74

FT — First assigned treatment; BW1 — First occasion; BW2 — Second occasion.

^*Significantly different than volume for BW (P = 0.0012).

During the epidural injection and CT under anesthesia, as well as during the assessment of the presence of pain after anesthesia, cardiopulmonary values and rectal temperature were not significantly different between groups (Figure 2).

Figure 2.

a — Heart rate (HR); b — Respiratory rate (f_R); c — Systolic arterial pressure (SAP); d — Diastolic arterial pressure (DAP); e — Mean arterial pressure (MAP); f — Rectal temperature (TEMP); g — End-tidal CO₂ (Pe′CO₂); and h — End-tidal isoflurane concentration (Fe′Iso) in 6 dogs initially under general anesthesia (GA) for placement of an epidural lumbosacral catheter and injection of a volume mixture of bupivacaine 0.25% and iopamidol 15% and after anesthesia (RE). The injection consisted of 0.2 mL/kg based on body weight (BW; solid black square) on one occasion or 0.05 mL/cm (< 50 cm) or 0.07 mL/cm (50 to < 70 cm) based on length of the vertebral column (LE; open circle) on a different occasion in a randomized crossover fashion. Data are presented as mean ± standard deviation.

Epidural volume and rostral spread

The volume of injectate was significantly greater for LE than for BW (3.29 ± 0.74 mL versus 1.81 ± 0.21 mL; P = 0.0012), which resulted in a significantly larger (P = 0.0488) rostral spread for LE [22 ± 2 vertebrae; 20, 24 (95% confidence intervals)] than for BW (19 ± 2; 17, 21) (Figure 3). The range of vertebrae reached with the LE dose was C4 to T1 and with the BW dose was C5 to T5.

Figure 3.

Comparison of number of vertebrae reached by a volume mixture of bupivacaine 0.25% and iopamidol 15% in 6 dogs administered a lumbosacral epidural injection of 0.2 mL/kg based on body weight (BW) and 0.05 mL/cm (< 50 cm) or 0.07 mL/cm (50 to < 70 cm) based on length of the vertebral column (LE) in a randomized crossover fashion. Each violin plot depicts the frequency distribution of the data, with lines at the median (dashed line) and quartiles (smaller dotted lines) for each group (LE, BW). The larger dotted lines across both violin plots indicate the mean of each group and the difference between the 2 means, and standard deviation is depicted with solid lines on the right axis (3 ± 3.6 vertebrae).

Response to noxious stimulation

Sedation scores of 0 (no sedation) were present in all dogs at 30 min post-extubation and were not significantly different between groups (P = 0.8833). Ambulation scores were not significantly different between groups at any of the periods of assessment (P = 0.4582) and times for a score of 0 for both sedation and ambulation scales were 48.2 ± 18.8 min in the LE group and 35.5 ± 28.6 min in the BW group (P = 0.2904).

The time from recovery until the first assessment for the presence of pain was 36.3 ± 0.8 min in the LE group and 37.7 ± 1.9 min in the BW group and not significantly different (P = 0.2065). Time to positive response to the presence of pain was similar between individual dermatomes and groups (P = 0.1440) and overall time to the presence of pain (37.7 ± 20.4 min for LE and 49.0 ± 20.4 min for BW) was also similar (P = 0.3591). The times for detection to positive response to nociception for each dermatome are shown in Table II.

Table II.

Time to presence of a positive response to noxious stimulation of different dermatomes starting from caudal to cranial in 6 female beagle dogs that received a lumbosacral epidural injection of 0.2 mL/kg based on body weight (BW) and 0.05 mL/cm (< 50 cm) or 0.07 mL/cm (50 to < 70 cm) based on the length of the vertebral column (LE) in a randomized crossover fashion of a volume mixture of bupivacaine 0.25% and iopamidol 15%.

Dermatome	Time to presence of pain (min)
	LE	BW
AN1	46.3 ± 23.4	55.2 ± 23.4
AN2	38.8 ± 23.4	52.7 ± 23.4
AN3	38.8 ± 23.4	52.7 ± 23.4
AN4	36.3 ± 23.4	52.7 ± 23.4
AN5	36.3 ± 23.4	52.7 ± 23.4
AN6	36.3 ± 23.4	52.7 ± 23.4
AN7	36.3 ± 23.4	52.7 ± 23.4
AN8	36.3 ± 23.4	52.7 ± 23.4
AN9	36.3 ± 23.4	40.2 ± 23.4
AN10	36.3 ± 23.4	37.7 ± 23.4
AN11	36.3 ± 23.4	37.7 ± 23.4
Mean ± SD	37.7 ± 20.4	49.0 ± 20.4

AN1 — Base of the tail; AN2 — Pelvic limb paw; AN3 — Area surrounding the iliac crest; AN4 — Inguinal area near the femoral nerve; AN5 — Area surrounding the transverse process of L₃ and flank; AN6 — Paramedian area next to the umbilicus; AN7 — Along the 13th rib; AN8 — Paramedian area to the xyphoid; AN9 — Along the 9th rib; AN10 — Along the 2nd rib; AN11 — Thoracic limb paw.

Discussion

The rostral spread of an LS injectate volume based on LE or BW was successfully determined in small-sized dogs (≤ 10.2 kg). The doses used for LE and BW resulted in a volume of epidural injectate 1.82-fold greater for LE than for BW, which resulted in a larger rostral spread for LE (22 ± 2 vertebrae versus 19 ± 2 vertebrae, respectively).

A greater volume for LE than BW was also determined in another study in small- (< 10 kg) and medium-sized (10 to < 25 kg) dogs with a body condition score (BCS) of 2 to 5 in a mathematical model and the difference for the small-dog category in that study was 2.18-fold (13). The dose for LE and BW should therefore not be considered interchangeable in small-sized dogs.

The rostral spread of any substance in the epidural space is the result of dynamic forces that affect cerebral spinal fluid (CSF) motion and pressures in that space. These forces include compliance of the spinal canal; physical activity and position of the individual; the cardiovascular system (blood pressure, cardiac output, and heart rate); the respiratory system (rate and effort of breathing); and systemic absorption of the substance (16,19,20).

Several studies have determined a time-dependent behavior of rostral epidural spread of contrast medium for up to 45 min using radiographic imaging (16) or for up to 20 to 30 min using CT (15,21). The number of vertebrae reached gradually increases in the first 20 to 30 min post-injection (15,16,21) and then decreases 30 min after injection due to systemic absorption of the contrast medium (16).

The LE doses used in this study were based on studies in pigs and dogs in which a linear regression equation was generated by injecting the contrast diatrizoate mixed with lidocaine 2% in 2-mL increments at the LS to determine the rostral spread radiographically immediately after the injection to help predict the volume of injectate needed to reach T10 (vertebra 11) (8), which represents dermatomes that include abdominal analgesia/anesthesia in addition to the pelvis and pelvic limb.

In contrast, for pelvis and pelvic limb, the inclusion of up to L3 (vertebra 5) is all that is required (22). In the present study, in both the LE and BW groups, the iopamidol spread exceeded the predicted site (T10) by reaching C4 to T1 with the LE dose and C5 to T5 with the BW dose. This extent of rostral spread did not correspond to the sensory block elicited by bupivacaine 0.25% in this study.

This study demonstrated similar rostral spread to that obtained in medium-sized dogs injected with 0.2 mL/kg of contrast medium in other studies, 12.8 kg (10 to 15 kg) (15) and 12.0 kg (10.9 to 14.0 kg) (21), but less rostral spread than that obtained in small-sized dogs of 7.6 ± 1.1 kg in another study (16). It is difficult to draw comparisons between studies as different methodologies are used to determine the speed of injection, time to imaging, use or not of epidural catheter, and the method for counting the number of vertebrae.

In the present study, dogs were injected faster (over 20 s) using an epidural catheter and assessed sooner (within 5 min) than in other studies (0.6 to 1.0 mL/min for injection and imaging at 20 to 45 min). In addition, only one other study used an epidural catheter (21) and different methods for counting the number of vertebrae reached were used in all studies (15,16,21).

The epidural distribution of the indicator (contrast medium or dye) is not always uniform around the spinal cord. In some studies, including this one, the number of vertebrae reached was based on the presence of the indicator without regard for uniform spread (14,15). In other studies, this was calculated as the mean of the dorsal and ventral spread (10,16) or by approximating the number to the nearest half vertebral body (21,23), the number corresponding to the ventral column (8), or the number according to a set percentage of coverage around the spinal cord, e.g., 90% (11).

In the present study, 11 dermatomes were used to assess the analgesic effects of epidural bupivacaine. The methods for assessing analgesia used herein have also been used in other studies (10,24). In certain dermatomes, it was easier to use electrical current to interpret when responses to the mechanical stimulation were not definitive. This occurred in very few instances and in anatomical areas that were more diffusely innervated, such as the flank and the inguinal area.

The main difference in assessment of the response to nociception in this study and in other studies was that the observer was blinded to the treatments, and it was an all-or-none response, which is more representative of detecting the presence of pain perception than the degree of analgesia.

Depending on the type of indicator used, it has been shown that the number of analgesic dermatomes did not correspond to the rostral spread of that indicator (10,16). A 1:1 mixture of bupivacaine 0.5% with methylene blue 1% dosed at 0.2, 0.4, 0.6, and 0.8 mL/kg resulted in a larger rostral spread of dye of 18 ± 7 (T3), 22 ± 2 (C6), 26 ± 2 (C2), and 26 ± 1 (C2) vertebrae, respectively, than for analgesia, which was assessed for 30 min post-injection and resulted in a maximum of 5.0 ± 3.3 vertebrae (L3), 14.2 ± 4.2 (T7), 20.2 ± 1.3 (T1), and 21.0 ± 0 (C7), respectively (10).

The need to dilute the bupivacaine 0.5% by adding an equal volume of the contrast medium and the resulting reduction in concentration to 0.25% was a limitation of this study and of other studies. It is most likely associated with the incomplete and short-lasting sensory block, despite the extensive rostral spread of iopamidol with doses calculated based on both LE and BW. In humans, the contrast medium used was highly correlated (r = 0.97) with analgesic spread when effective concentrations of local anesthetic were used (25).

The effectiveness of sensory and motor block with an LS epidural local anesthetic depends on both volume and concentration of the anesthetic. Although volumes of 0.2, 0.4, and 0.6 mL/kg of bupivacaine 0.25% resulted in more rostral sensory blockade, sensory responses to noxious stimulation were never absent due to the diluted form of the bupivacaine (10).

Increasing concentrations from 0.25% to 0.5% and 0.75% resulted in faster onset and duration, as well as a higher degree of sensory and motor block (24,26), which demonstrates that bupivacaine 0.25% does not cause a complete sensory block. Epidural bupivacaine 0.25% is commonly administered to women to provide analgesia during labor, although frequent top-ups are required to maintain the degree of analgesia (27), whereas epidural bupivacaine 0.5% can be used alone to provide anesthesia for caesarean section (28).

Bupivacaine concentrations of at least 0.5% result in effective motor and sensory block (24,29). Motor block assessed in humans by a quantitative method using isometric measurements of force of extension and flexion of the leg joints to 90% of the control value (before epidural) demonstrated no difference in the return to sensory and motor function for bupivacaine (29).

In the present study, the duration of block with bupivacaine 0.25% was determined by the return of complete motor function (score of 0 for ambulation). During the recovery phase in our study, the effects of isoflurane anesthesia could have influenced the coordination of the dogs and prolonged the duration of motor block. Times of motor block in this study, however, were shorter than those in other studies using the same dose and bupivacaine concentration (10,24). Differences in these times are also due to the endpoints for assessing motor block in those other studies in which ataxia was considered part of the motor block versus complete inability to ambulate as was used in the present study.

Dogs were fully ambulatory at 48.2 ± 18.8 min in the LE group and 35.5 ± 28.6 min in the BW group post-extubation. Although not significant, this difference was most likely related to the higher volume of bupivacaine administered in the LE group. In another study, 0.2 mL/kg of bupivacaine 0.25% caused complete motor block in only 33% of the dogs at 10 min post-injection and the duration of the motor block (from ataxia to complete paralysis) was 72 ± 55 min (24).

In the present study, the use of general anesthesia to facilitate placement of the epidural catheter and carry out CT scanning was a major limitation to our objective of determining cardiopulmonary effects and the degree of motor and sensory block. This meant that the immediate post-injection assessments had to be delayed until the dog had fully recovered. As a result, analgesia was assessed approximately 35 min post-extubation in both the LE and BW group, which meant not earlier than 40 min after the epidural injection of bupivacaine 0.25%. Due to the design of the present study, the onset and immediate effects on sensory and motor function post-extubation were missed since positive responses to nociceptive stimulation were already present at 37.7 ± 20.4 min for LE and 49.0 ± 20.4 min for BW post-extubation.

Despite a high rostral spread with either dosing method, cardiopulmonary variables and rectal temperature values were within acceptable limits (30,31) and were similar in both groups. In one study, rectal temperature was lower in a group of conscious dogs dosed based on LE than in dogs dosed based on BW, although rectal temperature was still ≥ 37°C, which was assumed to be the result of sympathetic block and vasodilation (32). In dogs (18.9 ± 3.3 kg) administered LS bupivacaine 0.25% at 0.2 or 0.4 mL/kg under isoflurane anesthesia, mean arterial pressure and cardiac output values were acceptable with both doses for up to 60 min (33).

In this study, pulmonary function was only assessed by respiratory rate and none of the dogs appeared to have negative effects of the rostral spread of bupivacaine 0.25%. This was most likely because, at this concentration and these doses (BW: 0.2 mL/kg, LE: 0.37 mL/kg), the effect on motor fibers to the respiratory muscles is minimal (10,33).

In conclusion, this study demonstrated that volumes of epidural injectate in small-sized dogs were greater when doses are calculated based on LE than when based on BW. As this results in more rostral spread of the anesthesia, the methods cannot be considered interchangeable. Cardiopulmonary function was similar regardless of the method used to calculate epidural volume. Due to the design of this study, however, it was not possible to demonstrate whether the difference in epidural volume had an impact on the immediate assessment of analgesia because recovery from anesthesia delayed the start of this assessment.

Funding Statement

The authors thank William Sears for statistical analysis and advice and the Ontario Veterinary College Pet Trust for funding this project.