Materials and Methods

Horses—Twelve adult horses owned by the university were included in the study. In the main portion of the study, 9 horses (age range, 5 to 29 years old; weight range, 510 to 660 kg [1,122 to 1,452 lb]) were used. Three additional horses were euthanized for reasons unrelated to the study, and limbs were used to determine evidence of and the potential route responsible for anesthetic agent diffusion to the region of the proximal suspensory ligament.

For the main portion of the study, the hair on both forelimbs of each of the 9 horses was clipped, and injection sites were aseptically prepared with chlorhexidine solution^a and alcohol prior to injection. Five horses were sedated with xylazine hydrochloride^b (0.3 to 0.5 mg/kg [0.14 to 0.23 mg/lb], IV) prior to performance of the LPBs. All horses received 1 dose of phenylbutazone^c (4 to 5 mg/kg [1.8 to 2.3 mg/lb], PO) on completion of the study. The horses were returned to their regular housing and were monitored daily for any signs of lameness for 7 days. The Institutional Animal Care and Use Committee at Colorado State University approved the experimental protocols used in the study.

LPB administration—The LPBs were performed by injection of a combination of mepivacaine hydrochloride^d and iohexol^e (50:50 dilution) with the third metacarpal bone supported and the MCP joint in an extended position (n = 4 limbs) or in a weight-bearing position (14 limbs), depending on the compliance of the horse. Prior to the study, 1 limb of each horse was randomly assigned via coin toss to receive 2 mL of the solution/injection site, and the other limb received 4 mL of the solution/injection site. A 5/8-inch (1.59-cm), 25-gauge needle (n = 12 limbs/6 horses) or 1-inch (2.54-cm), 22-gauge needle (6 limbs/3 horses) was used for each injection. Needle size was not randomly assigned because the 1-inch, 22-gauge needles were used initially and then changed to the 5/8-inch, 25-gauge needles because the smaller needles provided better handling and more consistent injections.

Placement of the needles to achieve the LPB was performed on the basis of a published description.⁶ To anesthetize the medial and lateral palmar metacarpal nerves, the needle was placed just distal to the distalmost aspect of the small metacarpal bones between the third metacarpus and the suspensory ligament. The needle was directed at a 45° angle proximad in an attempt to deposit the anesthetic agent–contrast medium solution under the distal end of the small metacarpal bones. To anesthetize the medial and lateral palmar nerves, the needle was placed in the groove between the suspensory ligament and the deep digital flexor tendon at the level of the distalmost aspect of the small metacarpal bones. The needle was directed perpendicular to the skin. The 1-inch needles were inserted approximately halfway. The 5/8-inch needles were often inserted all the way to the hub. All injections were performed by 1 author (KAS).

On completion of the injections (designated as 0 minutes), the limbs were cleaned with alcohol to remove any contrast medium from the skin. A horizontal, linear, radiopaque marker was placed on the dorsolateral aspect of the third metacarpal bone at the level of the injections. The marker was held in place with nonelastic tape for the duration of the study. Each horse was walked or jogged a distance of approximately 46 m (50 yards) in the intervals between time points at which radiographic images were obtained.

Radiography—Prior to obtaining the first radiographic image, the proximolateral aspect of each limb was labeled with a radiopaque number for identification. Lateral preinjection radiographic views of the metacarpal region were obtained in 14 of the 18 limbs. Preinjection radiographic views were not obtained for the first 2 horses evaluated because initially we did not expect that such views were necessary. The purpose of the preinjection radiographic views was to ensure that there were no radiographic artifacts that could be confused with subtle lines of contrast medium. Lateral radiographic views of the metacarpal region of each limb were obtained at 5, 15, 30, 60, 90, and 120 minutes after administration of the LPBs. All radiographic images were obtained with a digital radiography unit^f by the same person (KTS).

On each radiographic view, the maximum extent of contrast medium diffusion was measured from the points of injection for the palmar nerves and palmar metacarpal nerves in proximal and distal directions. Each point of injection was identified by use of the radiopaque marker that had been placed on the dorsolateral aspect of the third metacarpus immediately after the anesthetic agent–contrast medium solution injections. Proximal and distal extents of contrast medium diffusion from the injection sites were measured on the radiographic images; total extent of diffusion was calculated as the sum of those 2 measurements (proximal plus distal diffusion). The extents of proximal diffusion and total diffusion of the palmar metacarpal nerve block and the extents of proximal diffusion and total diffusion of the palmar nerve block were analyzed further. Four measurements were analyzed for each limb. Measurements were performed separately by 2 individuals (KAS and AVM). The measurements determined by the authors were then averaged for further analysis. Subtle proximal diffusion away from the primary bulk of contrast medium was also noted, and the presence of contrast medium within the DFTS or MCP joint was determined from the radiographic views.

Diffusion of MB solution in forelimbs—Methylene blue solution^g was injected in both forelimbs of 3 horses to determine the potential route responsible for anesthetic agent diffusion to the region of the proximal suspensory ligament. Because of the propensity of MB solution to spread via fascial planes, only 1 mL of a 0.05% solution of MB was used for each nerve in all horses. In 2 of the 3 horses, the MB solution was injected in a manner similar to the method used for horses in the main portion of the study. In the third horse, the MB solution was injected 1 to 1.5 cm proximal to the distalmost aspect of the small metacarpal bones to simulate the block for the medial and lateral palmar metacarpal and palmar nerves. All 3 horses were then euthanized with pentobarbital sodium^h (88 mg/kg [40 mg/lb], IV) for reasons unrelated to this study and were not used in the main portion of the study. The forelimbs were dissected within 2 hours after euthanasia and within 3 hours after injection.

Statistical analysis—The association of diffusion distance following anesthesia of each of the 4 nerves with the volume of anesthetic agent–contrast medium solution injected, time required for diffusion, needle size (gauge), or position of limb was evaluated via linear regression analysis. Prior to performance of linear regression analysis, the data for diffusion distances were evaluated for normality and linearity. A logistic regression analysis was performed to evaluate the probability of placement of contrast medium within the DFTS and MCP joint with the volume of anesthetic agent–contrast medium solution used, time required to diffuse, needle size, and position of the limb. In both analyses, volume of anesthetic agent–contrast medium solution used, time required to diffuse, needle size, position of limb, and evaluator were included as fixed effects. Both analyses were performed using bivariable as well as multivariable approaches. Limb side nested within horse was included as a random effect to control for the repeated measurements on the horses over time. Data analysis was performed by use of commercially available statistical software.ⁱ Diffusion distances were expressed as the adjusted mean and 95% CI. The adjusted mean was the mean (mean outcome in a specific category of the variable) after controlling for the effect of other variables in the model. A value of P < 0.05 was considered significant.

Results

Contrast medium diffusion—The LPBs were placed successfully in all horses. None of the horses in the study developed clinical signs of lameness or synovial inflammation or infection following the nerve blocks. Fourteen preinjection and 108 postinjection radiographic views were evaluated. There were 18 limbs evaluated for each of the 6 postinjection time points. Minimal diffusion of the bulk of the perineural injections was evident over time. There were 4 measurements of interest: the extents of proximal diffusion and total diffusion of the palmar metacarpal nerve block and extents of proximal diffusion and total diffusion of the palmar nerve block. For all measurements of interest, there was a significant difference between the mean diffusion distances at 5 and 30 minutes after LPB injections. However, there was no significant difference between findings at the 15-minute time point and the 5-minute time point for any measurement of interest. The mean proximal and total diffusion distances of the contrast medium injected over the palmar metacarpal nerves were 3.96 cm (95% CI, 3.75 to 4.17 cm) and 6.61 cm (95% CI, 6.40 to 6.82 cm), respectively. The mean proximal and total diffusion distances of the contrast medium injected over the palmar nerves were 4.33 cm (95% CI, 4.12 to 4.54 cm) and 7.06 cm (95% CI, 6.85 to 7.27 cm), respectively.

The mean (and 95% CI) diffusion distances for the palmar metacarpal and palmar nerve blocks in relation to needle size (gauge), limb position, and volume of anesthetic agent–contrast medium solution were evaluated (Table 1). The diffusion distances for the palmar metacarpal nerve block were not significantly affected by needle size or injection volume. Both the proximal and total diffusion distances for the palmar metacarpal block were significantly influenced by limb position. Injections performed with the third metacarpal bone supported and the MCP joint in an extended position resulted in greater diffusion distances than those associated with injections performed in a weight-bearing position. The diffusion distances for the palmar nerve block were not significantly affected by needle size or limb position. The proximal and total diffusion distances of the palmar nerve block were significantly greater when 4 mL of solution was injected, compared with findings when 2 mL of solution was injected.

Evidence of subtle, linear proximal diffusion of contrast medium was detected radiographically in 17 of the 18 limbs (Figure 1). This diffusion was apparent at the 5-minute time point in 16 limbs and at the 15-minute time point in the remaining limb and was detected for up to 90 minutes after LPB administration. The presence of subtle proximal diffusion of contrast medium was not affected by the volume of anesthetic agent injected, size of the needle, or position of the limb when the needle was placed.

View Full Size

Lateral radiographic view of the metacarpal region of the left forelimb of a horse obtained 5 minutes after injection of 2 mL of mepivacaine hydrochloride–iohexol solution (50:50 dilution) to achieve an LPB. Injections were administered at 4 sites in the forelimb to affect the medial and lateral palmar and palmar metacarpal nerves. Arrows indicate areas of subtle proximal diffusion of contrast medium at this time point. The contrast visible at the lower right is associated with the primary injection sites for the palmar and palmar metacarpal nerves as indicated by the horizontal linear radiopaque object (wire). The radiopaque object present in the midmetacarpal region was placed to identify the limb (ie, horse 3).

Citation: Journal of the American Veterinary Medical Association 239, 10; 10.2460/javma.239.10.1334

• Download Figure
• Download figure as PowerPoint slide

Intrasynovial deposition—Contrast medium was identified in the MCP joint in 8 limbs and in the DFTS in 7 limbs (Figure 2). Only 2 horses received LPBs after which contrast medium was not visible in the DFTS or MCP joint of either limb. In 3 limbs, contrast medium was detected in both the MCP joint and DFTS. Injection into the DFTS was not significantly associated with volume of anesthetic agent injected (P = 0.62), size of needle (P = 0.64), or position of limb (P = 0.54) when the needle was placed. Injection into the MCP joint also did not correlate with volume of anesthetic agent injected (P = 0.33), size of needle (P = 0.60), or position of limb (P = 0.27) when the needle was placed.

View Full Size

Lateral radiographic view of the region of the MCP joint in the right forelimb of a horse obtained 30 minutes after injection of 2 mL of mepivacaine hydrochloride–iohexol solution to achieve an LPB. Injections were administered at 4 sites in the forelimb to affect the medial and lateral palmar and palmar metacarpal nerves. Notice that contrast medium is visible in the DFTS (large arrow) and MCP joint (small arrow). The horizontal linear radiopaque object (wire) was placed to indicate the site of primary injection.

Citation: Journal of the American Veterinary Medical Association 239, 10; 10.2460/javma.239.10.1334

• Download Figure
• Download figure as PowerPoint slide

Diffusion of MB solution in cadaver limbs—In all 3 horses, MB solution was easily identified in the region of the palmar nerves and palmar metacarpal nerves during limb dissection. The MB solution was detected in the DFTS in 4 of 6 limbs (Figure 3) and in the MCP joint in 2 of 6 limbs. No MB solution was evident in the MCP joints or DFTSs of the third horse in which the nerve block in each forelimb was performed 1 to 1.5 cm proximal to the distalmost aspect of the small metacarpal bones to simulate the block for the medial and lateral palmar metacarpal and palmar nerves. Movement of the MB solution up all limbs via small vessels was visible (Figure 4). Histologic examination confirmed that these small vessels were lymphatic vessels.

View Full Size

Photograph of the palmar aspect of the left forelimb from a horse that was injected with 1 mL of a 0.05% solution of MB in a manner similar to the method used to achieve an LPB in horses in the main portion of the study. The horse was subsequently euthanized; the limb was dissected within 2 hours after euthanasia and within 3 hours of injection. Notice that MB solution is present within the DFTS.

Citation: Journal of the American Veterinary Medical Association 239, 10; 10.2460/javma.239.10.1334

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Photograph of the lateral aspect of the left forelimb from another horse that was injected with 1 mL of a 0.05% solution of MB in a manner similar to the method used to achieve an LPB in horses in the main portion of the study. Notice that the MB solution has tracked proximally within the lymphatic vessels.

Citation: Journal of the American Veterinary Medical Association 239, 10; 10.2460/javma.239.10.1334

• Download Figure
• Download figure as PowerPoint slide

Discussion

The purpose of the present study was to investigate proximal or distal tissue diffusion of anesthetic agent following administration of LPBs in horses. To enable measurement of the extent of diffusion, LPBs were performed by use of a combination of anesthetic agent and contrast medium, and forelimbs were monitored radiographically at intervals after injection. Results of the present study indicated that in the forelimbs of horses, the proximal diffusion of local anesthetic agent administered to achieve an LPB is minimal. Following LPB injections, one should expect the most proximal aspect of the anesthetic agent bulk to extend to no more than 5 cm proximal to the injection site, a location that is not near the proximal suspensory ligament region. In addition, diffusion of the anesthetic agent (represented by diffusion of contrast medium in the present study) did not appear to increase over time beyond the first 30 minutes. Inadvertent intrasynovial deposition of contrast medium occurred during this study and should be considered a risk associated with this type of perineural block. Alternative techniques should be used in an effort to minimize the risk of intrasynovial deposition of anesthetic agent.

In equine forelimbs, the proximal extent of the DFTS is a few centimeters proximal to the MCP joint¹⁶ or within the distal fourth of the metacarpus.¹⁷ The DFTS in horses can extend proximal to the distal aspect of the small metacarpal bones; therefore, it should not be surprising that contrast medium entered the DFTS in some horses in the present study. Some publications mention that location of needle placement for the medial and lateral palmar nerve blocks should be more proximal than the distalmost aspect of the small metacarpal bone^1,2,7 to avoid the DFTS, and this technique is supported by the findings of this study. Clinically, this more proximal needle location is well tolerated by patients and would likely avoid deposition of anesthetic agent into the DFTS without the risk of anesthetic agent diffusing too far proximally.

The palmar joint capsule of the MCP joint has been reported to extend proximally to the distal end of the small metacarpal bones.¹⁶ Nearly all publications describe the technique for anesthetizing the medial and lateral palmar metacarpal nerves just distal to the distalmost aspect of the small metacarpal bones,^1,3–7 yet the potential for inadvertent intrasynovial deposition of anesthetic agent is not mentioned. It was therefore surprising that, in 8 of the 18 limbs injected in the present study, contrast medium was detected within the MCP joint. The contrast medium within the MCP joint was quickly identified in some horses, whereas in others, the radiopacity within the MCP joint was subtle at first and increased in intensity during the first 30 minutes after LPB administration. This suggests the possibility of diffusion of contrast medium into the MCP joint following the LPB. To the authors' knowledge, there are no reports of diffusion of contrast medium into a synovial structure as a result of a perineural injection. Therefore, it seems more plausible that there was direct placement of contrast medium into the MCP joint, as likely occurred with some of the DFTSs. Penetration of the MCP joint would most likely occur if the needle was not directed up under the distal end of the small metacarpal bones when the palmar metacarpal nerves were being blocked. It may also occur in horses where the palmar pouch of the MCP joint is more proximally located than typically expected or in horses that have effusion of the joint.

Modification of the technique for anesthetizing the medial and lateral palmar metacarpal nerves is difficult to achieve. In the present study, MB solution was injected in the forelimbs of 3 horses prior to euthanasia. In one of those horses, a more proximal needle placement was evaluated; the MB solution was injected 1 to 1.5 cm proximal to the distalmost aspect of the small metacarpal bones to investigate the potential effect of anesthetic agent administered at that location on the palmar metacarpal nerves and palmar nerves. Dissection of these limbs revealed successful deposition of the MB solution around the medial and lateral palmar metacarpal nerves and medial and lateral palmar nerves. This more proximal needle placement was well tolerated when used to affect the palmar nerves. It was not well tolerated when used to affect the palmar metacarpal nerves as frequent contact with the periosteum on the small metacarpal bones or the palmar surface of the third metacarpal bone resulted in a strong negative response from the horse. In the few clinical cases in which this altered block technique was performed, the horses have reacted with apparent resentment. Because this altered technique for anesthetizing the palmar metacarpal nerves is technically challenging and is not well tolerated by patients, this modification of the technique may not be clinically acceptable. The authors suggest that inadvertent MCP joint placement of anesthetic agent could be avoided by directing the needle upward, keeping the needle placed superficially, and ensuring that a subcutaneous bleb can be seen.

In the present study and 2 studies by Nagy et al,^13,14 subtle proximal diffusion of the contrast medium in a linear fashion from the injection sites was evident radiographically. The presence of MB solution within the lymphatic vessels, which reflected the path of the contrast medium in the forelimb radiographic views in both studies, strongly suggests that the anesthetic agent diffuses proximally in injected limbs via that route. Harkins et al¹⁸ reported that the highest dose of mepivacaine that resulted in no effect was 2 mg/site. This would suggest that > 0.1 mL of mepivacaine would have to be present at the precise site of injury to desensitize the proximal suspensory ligament. On the basis of the small diameter of the lymphatic vessels and the limited amount of anesthetic agent traveling through them at any given time, it would seem unlikely that the amount of anesthetic agent within the lymphatic vessels would be sufficient to diffuse out of the vessel and desensitize the region of the proximal suspensory ligament.

Variables including limb position, needle size, and volume of injected solution were evaluated for their effect on the extent of contrast medium diffusion in the present study. Limb position was not randomly assigned but was based on the compliance of each horse because the authors usually perform this nerve block with the horse in a weight-bearing position. The study findings indicated that when limbs were injected with the third metacarpal bone supported and the MCP joint in an extended position, the diffusion distance of injected contrast medium for the palmar metacarpal nerve block was significantly greater than that observed after administration of the nerve block in weight-bearing limbs. In a non–weight-bearing position, the suspensory ligament may be less taut, thereby allowing for greater diffusion. For either the palmar metacarpal nerve block or palmar nerve block, needle size had no significant effect on the extent of contrast medium diffusion. In the initial stage of the study, the authors used 22-gauge needles, suspecting that the larger-bore needles would be necessary because of the viscosity of the mepivacaine hydrochloride–iohexol solution. However, it soon became apparent that 25-gauge needles were easier to handle and their function was not affected by the viscosity of the solution. Not surprisingly, the greater the volume of solution injected, the greater the diffusion distance from the injection site; however, injections of 4 mL of solution/nerve resulted in significantly greater diffusion for the palmar nerve block only. This should be taken into consideration when repeating an injection of anesthetic agent to overcome a failed nerve block.

On the basis of the findings of the present study, aseptic preparation of the skin at the site of an LPB is recommended prior to needle placement because of the potential of intrasynovial injection of the anesthetic agent. Placement of the needle for local anesthesia of the medial and lateral palmar nerves should be at least 1 cm proximal to the distalmost aspect of the small metacarpal bones, especially in horses that lack obvious DFTS effusion. Clinicians should expect that the extent of anesthetic agent diffusion in a proximal direction from the site of needle placement will be approximately 5 cm, and this should be taken into consideration when a more proximal needle placement is used. For the palmar metacarpal nerve block, needle placement at the distalmost aspect of the small metacarpal bones is appropriate, but the needle should be directed proximad, and a subcutaneous bleb should develop at the injection site. Finally, given that the extent of proximal diffusion of anesthetic agent can be as much as 5 cm, the soft tissue structures proximal to the MCP joint should not be ruled out as the source of lameness when an LPB results in partial improvement. Results of the present study have suggested that localization of lameness to a region distal to the distalmost aspect of the small metacarpal bone following an LPB may not always be accurate. However, it appears unlikely that this particular technique will anesthetize lesions in the proximal suspensory ligament region.