Spinal & withdrawal reflexes
This page describes all of the myotactic (stretch) reflexes that have been described in veterinary neurology. Several of the reflexes listed below can be difficult to elicit, even in normal patients. As such, the results should be interpreted in light of the remainder of the neurological exam.
NOTE: The patellar and withdrawal reflexes are the most reliable, so many neurologists perform only these tests. The other "reflexes" shown below may not truly be reflexes and results can be very inconsistent. They are described here to be complete. If performed, the results should evaluated in light of other neurological exam findings.
Reflexes are generally graded on a scale of 0–4 as described below.
- 0 = absent
- 1 = decreased
- 2 = normal
- 3 = exaggerated
- 4 = clonic (repetitive movement of the limb)
Reflex grading is a subjective assessment. What I call normal, you might think is exaggerated. If it is 100% clear that the reflex is exaggerated, the lesion would be cranial to the intumescence of the limb being tested (cranial to C6 for the thoracic limbs and L4 for the pelvic limbs), except with some caveats. Some older patients have decreased to absent patellar reflexes without obvious femoral nerve signs (see patellar reflex below) and very anxious patients may have exaggerated to clonic reflexes. If the clonic reflexes are real, these patients should have reduced postural reactions in the limb and/or gait abnormalities, while the postural reactions in an anxious dog would be normal.
What’s most important is to determine whether the reflex is decreased or absent because this means that the lesion is at the level of the intumescence of the limb being tested (C6-T2 for thoracic limbs and L4-S1 for pelvic limbs) or more peripherally located (peripheral nerve, neuromuscular junction, muscle).
The patellar and withdrawal reflexes are the most reliable reflexes in the pelvic limbs, while the others listed below are less consistent. However, I can almost always elicit the other reflexes listed below in many normal dogs and cats.
Nerve tested: Femoral nerve, L4-L6
Technique: Place the patient into lateral recumbency and hold the “up” leg gently off the ground with the stifle slightly flexed. Strike the straight patella tendon with a reflex hammer.
Normal response: Rapid extension of the stifle.
- Some normal patients will have a reduced patellar reflex on the “up” leg. You should always test the “down” leg at the same time and then compare the response to when it was the “up” leg. Read this tips & tricks post for more info.
- Many normal older patients will have a reduced to absent patellar reflex.
- Levine JM, Hillman RB, Erb HN, et al. The influence of age on patellar reflex response in the dog. J Vet Intern Med 2002:16(3), 244–246
- Patients with an L6-S1 or peripheral sciatic nerve lesion may have an exaggerated patellar reflex (pseudo-hyperreflexia). All limb reflexes are a balance between agonist (in this case, extensor muscles of the stifle) and antagonistic muscles (flexor muscles of the stifle). Sciatic nerve dysfunction leads to decreased tone of the stifle flexor muscles. As a result, there is decreased counterbalance to the quadriceps muscle when the patellar reflex is performed leading to an exaggerated response. Theoretically, any reflex listed below can be exaggerated if there is decreased antagonistic muscle function. However, in clinical practice, I’ve only seen this with the patellar reflex.
The thoracic limb reflexes can be very unreliable. As a result, many neurologists only perform withdrawal reflex testing in the thoracic limbs. That being said, I can frequently elicit an extensor carpi radialis and biceps reflex in normal patients. If absent, I believe it if there are other clinical signs to support a C6-T2 lesion localization. The triceps reflex is extremely unreliable and I don’t often perform this test.
The withdrawal reflex is very reliable in the pelvic limbs and is the most reliable reflex of the thoracic limbs.
Nerves tested: Primarily sciatic nerve (L6-S1), but also the femoral nerve (L4-L6) and the ventral motor branches innervating the psoas major muscle for hip flexion.
Technique: Pinch the toes of the pelvic limbs and watch for withdrawal of the limb.
Normal response: Withdrawal of the limb with flexion of the hip, stifle, and hock.
Note: Flexion of the hip will still occur with complete sciatic nerve dysfunction when the medial toe is pinched due to cutaneous innervation from the saphenous nerve, a branch of the femoral nerve.
Technique: Pinch the toes of the pelvic & thoracic limbs and watch for withdrawal of the limb.
Normal response: Withdrawal of the limb with flexion of the shoulder, elbow, and carpus.
- Withdrawal of the limb does not equate to intact nociception. Withdrawal involves only a local spinal reflex, while pain perception requires conscious recognition of, and a behavioral response to, a noxious stimulus.
- If the hips and stifle flex, but the hock does not, consider an L6-S1 or peripheral sciatic nerve disorder. I frequently see this with disease of the cauda equina, most commonly due to an intervertebral disk protrusion/herniation.
- If the other limb extends while performing the withdrawal reflex, this is called crossed extension and it is an abnormal response in a recumbent animal. Crossed extension is an upper motor neuron sign, indicating the lesion is cranial to the intumescence of the limb being tested (cranial to L4 in the pelvic limbs and C6 in the thoracic limbs).
- Some anxious patients will have a “positive” crossed extension that is not real/pathological.
- Crossed extension is only normal in standing patients. Consider this. . .you are walking barefoot through your house and step on a tack. Before you even consciously recognize the pain, the limb that stepped on the tack flexes to move away from the source of pain while the opposite limb extends to maintain balance. If you are lying down and I stick your foot with the tack, there’s no need to extend your other leg (except to kick me!).