Regional anesthesia (blocking) is a valuable diagnostic aid for localizing lameness in horses when, after thorough physical examination, the origin of pain remains uncertain. Localizing pain to a region enables other diagnostic procedures—including anesthesia of a joint, radiography, ultrasonography, CT, scintigraphy, and MRI—to be used more effectively and economically to identify the cause of lameness.
Before administering regional anesthesia for diagnostic purposes, the clinician should confirm that the horse is consistently and sufficiently lame so that any improvement in gait after blocking can be recognized. At a minimum, the horse should be trotted in a straight line on hard ground and lunged in both directions on soft ground.
Lunging or riding the horse may exacerbate subtle lameness. Lameness in some horses improves or resolves during exercise; these horses can show a false-positive response to regional anesthesia. If a horse is subtly lame, use of a wireless, inertial sensor–based or other video analysis system designed to evaluate lameness can help clarify the results of diagnostic analgesia.
Occasionally, a previously positive flexion test or positive response to hoof testers should be rechecked after a block, or other problem maneuvers for the horse (eg, a lead change, canter departure, sliding stop) can be rechecked under saddle, after blocking, to assess the result.
The use of regional anesthesia for diagnostic purposes requires careful interpretation. It is commonly believed that if a nerve block eliminates lameness, the lesion is located distal to that block. Clinicians should be aware that lesions causing lameness occasionally can be found above the region typically desensitized by a nerve block.
Precise needle placement, the smallest effective volume of anesthetic, and timely evaluation of blocks can help prevent unintended diffusion of local anesthetic and subsequent errors in interpretation.
Mepivacaine HCl (2%) is the agent of choice for local anesthesia in horses. Although it is usually more expensive than lidocaine, mepivacaine generates less tissue reaction and more reliable desensitization. Mepivacaine’s effect lasts 90–120 minutes, longer than lidocaine’s, facilitating the examination of horses that require multiple nerve blocks in one limb or that have lameness in multiple limbs. Importantly, the drug lasts long enough that the second lame limb can be evaluated before the block in the first limb to be examined wears off.
Before the needle is inserted, the site should be cleaned at least with isopropyl alcohol, which is usually sufficient preparation for most nerve blocks of the distal limb. For nerve blocks administered close to a synovial structure, antiseptic preparation using povidone-iodine or chlorhexidine and rinsing with alcohol between scrubs is indicated. More rigorous antiseptic preparation (3–5 minutes) is strongly recommended before performing an intra-articular block.
Although rare, infection can be introduced by the process of blocking. If a skin lesion (eg, dermatitis or a wound) is present on the limb, especially near the injection site, blocking is not advised, because of increased risk for the development of cellulitis.
Most nerves distal to the carpus and hock are anesthetized using a 25‑gauge, 1.6‑cm (5/8‑inch) needle. A larger-gauge (eg, 22‑ or 20‑gauge), 3.8‑cm (1½‑inch) needle is used to anesthetize nerves more proximal on the limb.
If a relatively large-gauge needle is to be used, preliminary subcutaneous deposition of a small amount of local anesthetic with a 25‑gauge needle can prevent the horse’s objection to insertion of the larger-gauge needle later.
To avoid broken or bent needles during perineural administration of local anesthetic, the clinician should always detach the needle from the syringe before insertion. The needle should be directed distally during insertion when anesthetizing nerves in the distal limb. Directing the needle proximally can result in proximal migration of anesthetic and unintended anesthesia of more-proximal branches of the nerve, thus confounding examination results.
Pain is relieved and lameness decreased usually within 10 minutes after a local anesthetic is injected around a sensory nerve in the distal limb, provided that the nerve innervates the painful region of the limb. Anesthesia of large nerves in the proximal limb can take 20–40 minutes.
If the horse’s gait is assessed too early, results of a regional nerve block can be misinterpreted, yielding a false-negative result. Conversely, in the distal limb, anesthetic can migrate proximally to anesthetize more-proximal structures, thus producing a false-positive result and confounding examination results.
To minimize these errors, the clinician should evaluate the horse’s gait within 15 minutes of administering a regional nerve block in the distal limb, and within 15–30 minutes of administering a regional nerve block in the proximal limb.
Skin sensation should be checked as an index of desensitization before lameness is reevaluated. Individual variation in loss of skin sensation is common, however, and some horses retain skin sensation even when deeper structures have been desensitized. Conversely, skin sensation can be lost even if deeper structures have not been desensitized. Poor correlation between loss of skin sensation and desensitization of deeper structures is more likely when lidocaine, rather than mepivacaine, is administered.
When a regional nerve block is administered in the proximal limb, the horse may stumble because of altered proprioception. If nerves proximal to the hock or carpus are anesthetized, the horse’s gait should be assessed by trotting the horse on a forgiving surface to protect the skin over the dorsum of the fetlock in case the horse stumbles. Alternatively, the skin of the fetlock can be protected with a bandage applied to the distal limb before the horse is trotted on a hard surface.
If the horse is examined while being ridden, the rider should be reminded that the horse could trip or stumble after any nerve block; for blocks administered in the distal limb, however, stumbling is rare.
When regional anesthesia is performed, especially in the distal limb, local anesthetic can be administered inadvertently into a blood vessel, joint, tendon sheath, or bursa. The appearance of blood in the needle’s hub after aspiration before injection indicates that the needle has entered a blood vessel, and redirection is warranted.
Nerves, with accompanying blood vessels, are surrounded by a sheath (the circumneural sheath). Local anesthetic deposited outside of this sheath is likely to be ineffective. Administering anesthetic into a site outside of the neurovascular bundle results in a false-negative conclusion because anesthesia of the nerve will be ineffective or at least delayed.
Regional anesthesia of the distal limb can be accomplished in most horses with minimal restraint, but a good handler is still essential for safety. For fractious horses or for horses previously subjected to regional anesthesia, using a lip twitch or lip chain is prudent. If the horse remains uncooperative with these restraints, a mild anxiolytic drug or sedative can be helpful.
Acepromazine has no analgesic effect in horses and therefore is less likely to influence lameness than is xylazine or detomidine, each of which provides some analgesia. Some clinicians claim that acepromazine accentuates subtle lameness by relaxing the horse.
The extent to which sedation or tranquilization interferes with gait assessment depends on the severity of lameness and the skill of the clinician performing the examination. Even though numerous studies indicate that sedation has little effect on lameness, the effect of sedation on the horse’s gait should be assessed before a block is administered.
A horse should not be restrained in stocks to administer a nerve block to the distal limb, because restraint increases the likelihood of injury to both the clinician and the horse. Nerve blocks of the distal limb are generally safest when performed with the limb held. If a nerve block is performed with the horse’s limb on the ground, the contralateral limb can be held to enhance the safety of the procedure.
A horse should not be restrained in stocks to administer a nerve block to the distal limb, because restraint increases the likelihood of injury to both the clinician and the horse. Nerve blocks of the distal limb are generally safest when performed with the limb held. If a nerve block is performed with the horse’s limb on the ground, the contralateral limb can be held to enhance the safety of the procedure.
A comprehensive lameness examination may require reblocking on sequential days. Repeated blocking can be necessary in the following circumstances:
Pain that is causing lameness is present in both the proximal and the distal limb or in multiple limbs.
Equivocal results were obtained on the first day of examination and require clarification.
Initial blocking identified the general area of pain causing lameness, and the clinician wishes to localize it more precisely with alternative diagnostic anesthesia before proceeding to imaging.
Comprehensive lameness examination in a horse can be time-consuming and costly, and it cannot be achieved in every field setting. For these reasons, the approach to regional anesthesia to localize the site of pain that is causing lameness should be practical and strategic. The client should be consulted at the beginning of the examination to discuss the plan, potential timeline, and possible necessity for referral and/or advanced imaging.
Anesthesia of the Forelimb in Horses
Because perineural anesthesia of horses should generally start distally and progress proximally, the palmar digital nerve (PDN) block (the first block in the sequence) is the most commonly performed regional nerve block of the forelimb.
Palmar Digital Nerve Block
The PDN block is performed with the horse’s limb held (see PDN block image). A 25‑gauge, 1.6‑cm (5/8‑inch) needle is inserted directly over the palpable neurovascular bundle, approximately 1 cm proximal to the cartilage of the foot. The needle is directed distally, and 1 mL of local anesthetic is deposited near the junction of the nerve and the cartilage of the foot.
Courtesy of Dr. John Schumacher.
The PDN block anesthetizes not only the entire sole but also the horse’s entire foot, including the distal interphalangeal (coffin) joint. This block may also partially desensitize at least part of the pastern joint or even the fetlock joint, especially if a large volume of local anesthetic (> 2.5 mL) is injected over each nerve.
Abaxial Sesamoid Nerve Block
If the horse’s lameness is not ameliorated with a PDN block, most clinicians proceed to an abaxial sesamoid (or basisesamoid) nerve block (see abaxial sesamoid nerve block image). With this block, the clinician can anesthetize the palmar digital nerves by injecting 1.5–2 mL of local anesthetic at the level of the base of the proximal sesamoid bones or slightly more proximally along the abaxial surface, where the neurovascular bundle is easily palpated at this location.
Courtesy of Dr. John Schumacher.
The abaxial sesamoid nerve block desensitizes structures distal to it and is also likely to anesthetize a portion of the fetlock joint. Positive response to this nerve block after a lack of response to the PDN block localizes the lameness to the pastern and fetlock.
Low Palmar Nerve Block
The low palmar nerve block, or low four-point nerve block, is performed after a negative response to the abaxial sesamoid nerve block. This nerve block is usually performed with the horse bearing weight on the limb, but it can be performed with the limb held.
For the low palmar nerve block, the medial and lateral palmar nerves are anesthetized using a 25‑gauge, 1.6‑cm (5/8‑inch) needle to deposit 1.5 mL of local anesthetic over each palmar nerve where it lies subcutaneously at the dorsal border of the deep digital flexor tendon. Care should be taken to direct the needles subcutaneously and far enough proximally to avoid inadvertent penetration of the digital flexor tendon sheath.
To complete the low four-point nerve block (see fetlock region block sites image), 1 mL of local anesthetic is deposited at the distal end of each splint bone, where the palmar metacarpal nerve emerges. For this portion of the block, the needle is directed beneath the button of the splint bone.
Courtesy of Dr. Jamie Textor.
A positive response to a low four-point nerve block, performed after a negative response to an abaxial sesamoid nerve block, generally localizes lameness to the fetlock. However, lameness caused by proximal suspensory desmitis can sometimes be ameliorated after a low four-point nerve block. This observation can be explained by proximal migration of local anesthetic along the lateral palmar nerve to anesthetize the deep branch of this nerve, which innervates the proximal portion of the suspensory ligament.
High Palmar Nerve Block
The high palmar nerve block, or high four-point nerve block, can be performed when the low four-point nerve block has failed to improve lameness. With the limb held or bearing weight, the medial and lateral palmar and palmar metacarpal nerves are anesthetized 2.5 cm (1 inch) distal to the carpometacarpal joint.
To anesthetize each palmar nerve when performing a high four-point nerve block, a 25‑gauge, 1.6‑cm (5/8‑inch) needle is inserted through fascia to where the nerve lies near the dorsal border of the deep digital flexor tendon, and 2 mL of anesthetic is deposited over the nerve. The palmar metacarpal nerves are anesthetized slightly distal to the carpometacarpal joint by inserting a 20‑ to 22‑gauge, 3.8‑cm (1½‑inch) needle into the angle formed by the junction of the third metacarpal bone and the second or fourth metacarpal bone.
The high two-point nerve block (see high two-point nerve block image) is an alternative to the high four-point nerve block. Because the palmar metacarpal nerves arise from the lateral palmar nerve, the lateral palmar nerve can be blocked more proximally on the medial surface of the accessory carpal bone to desensitize three of the four nerves in the four-point block. The medial palmar nerve is then blocked separately, as described above.
Courtesy of Dr. John Schumacher.
The high two-point nerve block is performed, with the limb held or bearing weight, with a 25‑gauge, 1.6‑cm (5/8‑inch) needle inserted approximately 2 cm distal to the distal border of the accessory carpal bone (Wheat technique). The needle can be inserted in either a medial-to-lateral or a palmar-to-dorsal direction, and 2 mL of local anesthetic is deposited within the fascia. Alternatively, the lateral palmar nerve can be blocked more proximally, along the medial surface of the accessory carpal bone, by directing a 25-gauge needle to a subcutaneous position (Castro technique).
Compared with the high four-point nerve block, the high two-point nerve block can be performed more rapidly and requires fewer needle sticks. The disadvantage of the Castro technique is the possibility of desensitizing the entire carpus. The disadvantage of the Wheat technique is inadvertent injection and desensitization of the carpal sheath.
Regardless of which proximal metacarpal regional block is performed, the distal aspect of the carpus may also be desensitized. This potential effect must be considered when responses to blocks in this region are interpreted. Conversely, intra-articular anesthesia of the intercarpal joint can also eliminate pain in the region of the proximal portion of the suspensory ligament.
Proximal Forelimb Nerve Blocks
If the site of pain that is causing lameness cannot be localized by the nerve blocks described above, most clinicians perform intra-articular blocks of the carpus, elbow, or shoulder. The order in which these synovial structures are desensitized is not important, because the response to each joint block can be considered independent of the others.
A potential drawback to performing intra-articular anesthesia is that local anesthetics are toxic to cartilage and synoviocytes. The clinical importance of this toxicity has not yet been determined for horses.
As an alternative to intra-articular anesthesia of the carpus, the median and ulnar nerves can be anesthetized simultaneously to exclude pain distal to the elbow as the cause of lameness (see median and ulnar nerve block sites image).
Courtesy of Dr. Jamie Textor.
Anesthesia of the Pelvic Limb in Horses
Distal Pelvic Limb Nerve Blocks
For lameness in the pelvic limb, the plantar digital nerve blocks and the low four-point nerve block are performed as described above for the forelimb. However, an anatomical difference has implications for the low four-point block in the hindlimb: the medial and lateral dorsal metatarsal nerves (branches of the deep peroneal nerve) course adjacent to the long digital extensor tendon and innervate the skin over the dorsal aspect of the distal portion of the metatarsus.
After local anesthetic is deposited for a low four-point nerve block at the distal aspect of the splint bones, the needle can be redirected dorsolaterally or dorsomedially toward the margin of the extensor tendon, and an additional 2 mL of local anesthetic can be deposited subcutaneously at each site to anesthetize the medial and lateral dorsal metatarsal nerves. When performed in this manner, this block is often referred to as a “low six-point nerve block.”
However, many clinicians perform only a four-point nerve block when evaluating a horse with pelvic limb lameness, because the dorsal metatarsal nerves are thought to innervate only skin and the dorsal cortex of the third metatarsal bone.
The proximal aspect of the metatarsus can be desensitized in a manner similar to that used to desensitize the proximal aspect of the metacarpus—ie, with a high four-point nerve block (or high six-point nerve block, if the dorsal metatarsal nerves are included). In this region, however, most clinicians instead perform a single nerve block, the deep branch of the lateral plantar nerve block (see deep branch lateral plantar nerve block image).
Courtesy of Dr. Jamie Textor.
A block of the deep branch of the lateral plantar nerve may inadvertently desensitize the tarsometatarsal joint. Conversely, anesthesia of the tarsometatarsal joint may result in inadvertent anesthesia of the deep branch of the lateral plantar nerve.
Imaging of both the hock and the proximal portion of the metatarsus is indicated for horses that respond positively to a block of the deep branch of the lateral plantar nerve or to direct anesthesia of the tarsometatarsal joint.
Proximal Pelvic Limb Nerve Blocks
If a horse remains lame after the blocks described above have been performed, the peroneal and tibial nerves can be blocked simultaneously to determine whether lameness is caused by pain in the region of the hock. The tibial and peroneal nerves can also be blocked separately to more accurately identify the site of pain that is causing lameness:
If lameness is ameliorated after a tibial nerve block (see tibial nerve block image) but not after a low four-point or six-point block, the suspensory ligament is the likely source of pain.
Courtesy of Dr. John Schumacher.
If lameness is ameliorated after a peroneal nerve block (or fibular nerve block; see peroneal nerve block image) but not after a low four-point or six-point block, one or more of the tarsal joints is the likely site of pain that is causing lameness.
Courtesy of Dr. John Schumacher.
The tibial nerve in the horse can be anesthetized approximately 10 cm proximal to the point of the hock on the medial aspect of the limb, where it lies on the caudal surface of the deep digital flexor muscle, cranial to the calcaneal tendon. A 20‑gauge, 3.8‑cm (1½‑inch) needle is used to deposit 20 mL of mepivacaine HCl at this site, in several planes in the fascia surrounding the nerve.
The deep peroneal nerve can be blocked on the lateral aspect of the limb, approximately 10 cm proximal to the point of the hock in the groove formed by the lateral and long digital extensor muscles. A 20‑gauge, 3.8‑cm (1½‑inch) needle is directed caudomedially until it contacts the caudal edge of the tibia, and 10–15 mL of mepivacaine HCl is deposited.
The distal limb should be bandaged for protection before tibial or peroneal blocks are evaluated because horses frequently stumble when these nerves are anesthetized.
Intra-Articular Anesthesia in Horses
In general, intra-articular anesthesia (ie, a joint block) is performed in horses to identify a specific joint as the site of pain that is causing lameness. A joint block is performed after strict preparation and with aseptic technique; a new bottle of local anesthetic is recommended, and some clinicians also inject a prophylactic antimicrobial drug.
Some clinicians prefer to rely on regional anesthesia rather than on intra-articular anesthesia to localize pain in horses because of reports documenting toxicity of local anesthetics on cartilage and synovium. Mepivacaine and ropivacaine appear to be less chondrotoxic than other commonly administered local anesthetics.
Volumes of anesthetic used for joint blocks are determined by the size of the joint; in the authors’ opinion, however, overdosing is likely when common recommendations are followed. The commonly administered doses are 2 mL for the tarsometatarsal joint, 5 mL for the distal interphalangeal joint, 8 mL for the fetlock joint, and up to 30 mL for the medial compartment of the femorotibial joint of the stifle.
Needles used for joint blocks are 22 gauge or larger and usually 2.5–3.8 cm (approximately 1–1½ inches) long. Synovial fluid may not be visible even if the needle is in the joint. In general, joint blocks can be evaluated 15 minutes after injection.
For most joints in horses, resistance to intra-articular anesthetic injection is minimal when the needle is correctly placed (see intra-articular anesthesia images showing interphalangeal joint block sites, radiocarpal and intercarpal joint block sites, tarsal joint block sites, and stifle joint block sites). One exception is arthritic joints of the distal aspect of the hock, where joint penetration can be difficult and resistance to injection may be notable. The joints most commonly desensitized by intra-articular anesthesia are the distal interphalangeal joint, carpal joints, hock joints, fetlock, and the compartments of the stifle.
Interphalangeal joint block sites in a horse. 1. Distal interphalangeal joint: A 22‑gauge, 2.5‑cm (1‑inch) needle is quickly inserted in the direction of the arrow, with the limb bearing weight. Synovial fluid is often visible in the needle hub. 2. Proximal interphalangeal joint: A 20‑ to 22‑gauge, 3.8‑cm (1½‑inch) needle is inserted lateral to the extensor tendon and proximal to the joint margin, in the direction of the arrow and along the dorsal surface of the proximal phalanx, with the limb bearing weight. The distal and proximal interphalangeal joints can also be accessed palmarly; however, these approaches are more challenging.
Courtesy of Dr. Jamie Textor.
Sites for blocking the radiocarpal and intercarpal joints in a horse. 1. Radiocarpal joint: A 21‑gauge, 3.8‑cm (1½‑inch) needle is inserted in the direction of the arrow, with the limb held in flexion, through the palpable depression between the extensor tendons. 2. Intercarpal joint: A 21‑gauge, 3.8‑cm (1½‑inch) needle is inserted in the direction of the arrow, with the limb held in flexion, through the palpable depression between the extensor tendons or lateral to the lateral digital extensor tendon.
Courtesy of Dr. Jamie Textor.
Tarsal joint block sites in a horse. 1. Tarsometatarsal joint: A 22‑gauge, 2.5‑cm (1‑inch) needle is inserted to its hub in the direction of the arrow, just proximal to the head of the fourth metatarsal bone (MTIV). 2. Distal intertarsal joint: A 25‑gauge, 1.6‑cm (5/8‑inch) needle is inserted to its hub, if possible, perpendicular to the skin surface, either through or near the cranial margin of the cuneal tendon. 3. Tibiotarsal joint: A 21‑gauge, 3.8‑cm (1½‑inch) needle can be inserted perpendicular to the skin, in either the plantarolateral pouch or the dorsomedial pouch. This joint block is aided, and the pouches are more easily identified, when tibiotarsal effusion is present; the horse in this photograph does not have joint effusion.
Courtesy of Dr. Jamie Textor.
Stifle joint block sites in a horse. Although the medial femorotibial and femoropatellar joints generally communicate, all three compartments should be blocked separately to ensure complete desensitization of the stifle. 1. Medial femorotibial joint: A 21‑gauge, 3.8‑cm (1½‑inch) needle is inserted to its hub, if possible, perpendicular to the skin surface, in the space between the medial patellar ligament and the medial collateral ligament, proximal to the tibial plateau. The space is sometimes palpable as a depression, or as a bubble if joint effusion is present. 2. Femoropatellar joint: The joint can be accessed by either of two common approaches, both using a 21‑gauge, 3.8‑cm (1½‑inch) needle. The first approach is to insert the needle in the direction of the arrow, aiming beneath the patella in the trochlear groove. The second approach is to insert the needle perpendicular to the skin, at the point marked with the number 2 inside a circle. The needle contacts the lateral surface of the femur approximately 2 cm caudal to the palpable lateral trochlear ridge. The needle should be retracted slightly from the bone to either inject anesthetic or obtain synovial fluid. 3. Lateral femorotibial joint: A 21‑gauge, 3.8‑cm (1½‑inch) needle is inserted to its hub, if possible, perpendicular to the skin surface, lateral to the lateral patellar ligament, and proximal to the tibial plateau.
Courtesy of Dr. Jamie Textor.
The carpus, tarsus, and stifle joints in horses are each composed of multiple compartments: the carpus has three, the tarsus four, and the stifle three. The communication between multiple joint compartments must be considered when these joints are to be blocked.
Key Points
Regional anesthesia is an essential part of lameness examination in horses; however, results often can be misinterpreted.
Mepivacaine HCl provides better anesthesia and less tissue reaction than does lidocaine HCl in regional or intra-articular anesthesia.
Poor correlation between loss of skin sensation and desensitization of deeper structures is more likely when lidocaine, rather than mepivacaine, is administered for a nerve block.
For More Information
Moyer W, Schumacher J, Schumacher J. A Guide to Equine Joint Injections and Regional Anesthesia. 5th ed. Academic Veterinary Solutions; 2011.
Pilsworth R, Dyson S. Where does it hurt? Problems with interpretation of regional and intra-synovial diagnostic analgesia. Equine Vet Educ. 2015;27(11):595-603.
Bassage LH II, Ross MW. Diagnostic analgesia. In: Ross M, Dyson S, eds. Diagnosis and Management of Lameness in the Horse. 2nd ed. Elsevier Science; 2011:100-135.
Schumacher J, Boone L. Local anaesthetics for regional and intra-articular analgesia in the horse. Equine Vet Educ. 2021;33(3):159-168.
Baxter GM, ed. Adams & Stashak’s Lameness in Horses. 7th ed. John Wiley & Sons; 2020.
Also see pet owner content regarding lameness in horses.