Deep Tendon Reflex: The Tools and Techniques. What Surgical Neurology Residents Should Know

The deep tendon reflex (DTR) is a key component of the neurological examination. However, interpretation of the results is a challenge since there is a lack of knowledge on the important features of reflex responses such as the amount of hammer force, the strength of contraction, duration of the contraction and relaxation. The tools used to elicit the reflexes also play a role in the quality of the reflex contraction. Furthermore, improper execution techniques during the DTR assessment may alter the findings and cloud the true assessment of the nervous system. Therefore, understanding the basic principles and the key features of DTR allows for better interpretation of the reflex responses. This paper discusses the brief history of reflexes, the development of the reflex hammer, and also the key features of a reflex response encompassing the amplitude of force needed to elicit a reflex response, the velocity of contraction, the strength of contraction, and the duration of contraction and relaxation phases. The final section encloses the techniques of eliciting DTR in the upper extremities, trunk, and lower extremities, and the interpretation of these reflexes.


A Brief History
Reflexes have been studied for centuries and it begins with Aristotle and Galen. French physiologist, François Magendie in Paris and surgeon-anatomist, Charles Bell in London were among the first to identify the sensory and motor function of spinal nerve roots (2). In the late 18th century, Robert Whytt, Johann Unzer and Prochaska set up the reflex concept (1,3) while the concept of the 'reflex arc' was formulated in the 1830s by Marshall Hall (1790-1857). However, the concept underwent multiple developments and refinements conducted by various physicians in the late 1800s to reach the neurological examination concept that is applied today. Charles Sherrington played a central role in understanding the reflex activity from his vast physiologic investigations in the late 1800s to early 1900s (1,2).

Muscle Stretch Reflexes or Deep Tendon Reflexes
In 1859, Silas Weir Mitchell was the first person to appreciate the response elicited by percussed muscles with a hammer. The term 'reflex' was not used at that time and the percussing action was performed on muscles instead of the tendons. Jean-Martin Charcot recognised the diagnostic significance of hyperactive and absent muscle stretch reflex; however, this was only studied systematically by Erb and Westphal 16 years later and the findings were published in the German Archives of Psychiatry and Nervous Diseases.

The Tools
Percussion Hammers (3,4) In 1761, Viennese physician, Leopold Auenbrugger (1722-1809) described the use of percussion on the chest, back and abdomen as an aid to medical diagnosis. This technique is inspired by the thumping of wine casks to determine the level of the fluid.
In 1826, French Pierre Adolphe Piorry (1794-1879) invented a pleximeter, a resonator in the form of small ivory, metal, cedar or river disk, that was placed on the chest and struck with a finger. Two years later, a percussion hammer was devised by a Scottish physician, Sir David Barry (1781-1836) to strike the pleximeter. However, the percussion hammer was not widely adopted.

Introduction
A reflex is an involuntary, unlearned, repeatable response to a specific stimulus that does not require any input from the brain (1). The deep tendon reflex (DTR), also known as a myotatic reflex, is a sequence of lengthening, contraction, and relaxation of a group of muscles. A DTR comprises of a reflex arc, which is a neural pathway that controls a reflex. The reflex arc is made up of five components: The muscle spindle is a receptor within the muscle that detects changes in the length of the muscle. The muscle spindle consists of a noncontractile centre portion and intrafusal muscle fibres which make up the contractile portion. Tapping the tendon will cause stretching of the muscle spindle, activating it, leading to the propagation of an action potential to the spinal cord via afferent Ia fibres through the dorsal horn. In the spinal cord, the afferent nerve fibre synapses with a-motoneurons that supply the agonist muscles and also synapses with an inhibitory neuron that inhibits the antagonistic muscle group. This causes a concomitant relaxation of the antagonistic muscle as the agonist muscle groups contract.
The firing of the afferent fibres (reflecting the sensitivity of the central portion) depends on the length of the intrafusal fibres. The intrafusal fibres are controlled by gamma motoneurons, which are influenced by the cortex, cerebellum and various brainstem nuclei. This forms the suprasegmental control that modulates reflex activity. The higher centres receive information from the muscle spindles. For example, the dorsal spinocerebellar tract conveys the information about the proprioceptive organs from the muscle spindle to the cerebellum. In return, the higher centres modulate the segmental activity through gamma motoneurons. This, in turn, regulates the quality and amount of information received as the 'sensitivity' of the central portion depends on the length of the intrafusal fibres.

Weight: 80 g-140 g
This is the first reflex hammer demonstrated to the Philadelphia Neurological Society on 27 February 1888. The original model was shaped as a flattened cone with a base similar to the striking surface of the ulnar side of the palm, often used to elicit tendon jerk and a rounded apex to elicit biceps jerk (3,4). It was made of moderately soft rubber and encircled by a metal band, connecting to a rigid straight handle which finished as an open loop. Dr Taylor believed that a rigid handle allowed better control as the elastic handle produced inconsistent force.
In 1920, the open-loop handle was replaced with a solid and sharpened tip ( Figure 2) to allow for testing of the cutaneous reflexes (4), eliciting chest sounds and percussing the abdomen. The American Academy of Neurology incorporated the Taylor hammer into its logo ( Figure 3).  In 1841, a German clinician, Max A Wintrich (1812-1882) introduced a percussion hammer, known as the Wintrich hammer (Figure 1), which was then widely used throughout Europe. The hammer underwent many modifications using different materials, weight and shape. In 1912, the hammer was refurbished by Ebstein into a new model called the 'reflex and sensibility tester' for striking the tendons, with a pin on one end. Reflex Hammers (2,3,4) As the practice of testing the DTR grew, so did the development of a variety of percussion hammers. The early percussion hammers were too light and produced inconsistent results. Berliner and Troemner then suggested that an ideal instrument should be heavier to facilitate a brief but forceful percussion without eliciting pain and striking the tendon over a large surface.
The newly refurbished hammers were heavier, and some incorporated a handle and a pointed tip to test cutaneous reflexes, a ruler along the handle, and a pin, brush, or other tools for testing sensation. Some hammers are combined with a long, flexible handle to increase their effectiveness. Tendons were best struck directly, leading to the development of softer rubber striking surfaces and a more refined technique.

The Taylor Hammer
Inventor: John Madison Taylor (1855-1931), a neurologist from Philadelphia Introduced: 1888 Manufacturer: Snowdon (Brothers' Surgical Instruments Co.) of Philadelphia model. It weighs 100 g and is 22 cm long with a width of 8 cm for the head. The large head is designed for large tendons of extensor surfaces such as patellar, Achilles and triceps reflexes, and also to elicit periosteal and joint reflexes. The smaller head is used for percussion of flexor tendons such as biceps humeri, biceps femoris and semitendinosus. The hammer has a smooth handle that may be used as a tongue blade and a sharpened edge to elicit cutaneous and vascular reflexes. However, Ebstein opposed the usage of the hammer shaft as a tongue blade for hygienic reasons. The metal and nickel-plated hammer ( Figure 6) has a sufficiently heavy head and a large striking surface covered with rubber. The shape is like a hatchet with a tapered handle to test skin reflexes. The Babinski/Rabiner Hammer (3,4) Inventor: Joseph Francois Babinski, French neurologist Introduced: 1896 Babinski described two types of hammers ( Figure 7) with the same 20 cm-25 cm long nickel-plated steel handle. One of the hammers a heavy metallic head fixed to a flattened oval handle with a length of 17 cm (3,4). This is also one of the earliest hammers to allow for testing sensation and reflexes.
The hammer can be used to test temperature by warming the handle that is made of hard rubber and a metal head that remains cold. There is also a 1.5 cm sharp triangular spearhead under a removable cap and a dull rounded rubber point to examine for anaesthesia or hyperaesthesia. It is also incorporated with a two-point discriminator, a metallic cap to provide a hard surface, and camel's hair-brush underneath the cap for a soft surface (3,4). The hammer is all metal with replaceable rubber knobs on both ends of the head ( Figure 5) and is based on the older French has a disk encased by a peripheral rubber ring fixed at the end while for the other hammer, the disk is replaced with a rectangular plate in the same plane as the handle for an easier fit into the pocket.  The hammer ( Figure 9) is designed with a blunt handle, and the heads were formed by a rubber cylinder encircled by a metal. The hammer was modelled after the Henry Vernon chest percussion hammer (introduced in 1858). The hammer consists of a slender 8-inchlong flexible bamboo cane handle with a 1.5-inch rubber ring fitted brass disk on one end ( Figures  10 and 11). The other end is sharpened to elicit cutaneous reflexes. This resulted in a heavy, springy and completely painless hammer.
Marshall and Little (5) conducted a study to examine the peak tap force exerted by the Taylor hammer, the Queen Square hammer and the Babinski hammer. The findings showed that the Taylor hammer exerted the smallest peak tap force due to the lower mass and shorter handle. The peak tap force by the Queen Square hammer elicited a larger reflex response and this could be due to the flexible handle yielding a longer duration tap compared to the rigid handles. Hence, the Queen Square hammer and the Babinski hammer were suggested to be superior for reflex testing. Using an instrumented tendon hammer with a force sensor, Zhang et al. (7) found that the reflex torque (output) depends on the tapping force (input). A larger amplitude tapping force (within limits) produces larger amplitude reflex torques. The tapping force threshold is 20 N in normoreflexic patients. This yields a reflex torque of 4 Nm. In hyperreflexic patients, the tapping force threshold is lower, 13 N-15 N resulting in a larger reflex torque of 10 Nm-11 Nm.
These findings were reproduced in a clinical study conducted by Marshall and Little (5) investigating the typical tap forces used by clinicians to assess the patellar reflex. The peak tap forces: i) 0 N-20 N for hyperreflexia ii) 21 N-50 N for normoreflexia iii) > 50 N for hyporeflexia A proper technique does not require excessive force (8). Electrophysiological studies of the H-reflex recorded at the soleus muscle showed that an increase in the amplitude of stimulation increases the H-wave. Further increase in the stimulation amplitude resulted in the M-response. When the M-response is maximal, the H-wave vanishes.
This occurs because the large amplitude of electrical current stimulates the afferent Ia fibres and also the thinner a-motor neurons. The action potentials in the a-motor neurons propagate to the spinal cord (antidromic), collide with the action potentials of the evoked reflex response (orthodromic) and cause a partial to complete cancellation effect of the evoked reflex response (9,10).
On the electromyography studies, the onset latency of DTRs is about 25 msec-45 msec depending on the distance of the stretched muscle from the spinal cord and whether the tendon was tapped, or the muscle was mechanically stretched (10).
In a study of the ankle jerk reflex, the peak torque was 6.3 Nm ± 2.6 Nm for young adults whereas elderly subjects showed a peak torque of 3.6 Nm-4.3 Nm. The half contraction time in the young adults and the elderly was 27.9 msec-28.9 msec. The half relaxation time for the young adults was 30.0 msec-43.8 msec and 36.4 msec-44.3 msec in the elderly (11). (12,13,14) Biceps Reflex ( Figure 12    Reflexes from the trunk are minimally obtained or absent in normal individuals.  If the reflex is exaggerated, the response may be obtained by tapping the tendon just above the patella (suprapatellar reflex). This is done by placing the examiner's index finger on the upper border of the patella and tapping the finger to push down the patella. In a reflex spread, the knee extension is accompanied by adduction of the hip. An inverted patellar reflex occurs with contraction of the hamstrings and knee flexion (femoral nerve lesion). The Westphal's sign is the absence of patellar reflex. This reflex was first described between 1870 and 1871 by Wilhelm Erb, a German neurologist. The findings were published in the German Archives for Psychiatry and Nervous Disease in 1875 in collaboration with Carl Westphal. Erb correctly explained the reflex to be mediated by a reflex arc, using the term 'Patellarsehnenreflex', which means patellar tendon reflex. Later on, William Gowers from London coined the term 'knee-jerk' in 1881. Sherrington (1891-1892) and Dutch neurologist, Jan van Gijn confirmed that the knee-jerk is a true spinal reflex (1, 2, 3).

Achilles Reflex or Ankle Jerk
Striking point: Achilles tendon Myotome: S1/S2 Peripheral nerve: Tibial nerve Effector: Gastrocnemius muscle, soleus muscles Technique: i) In a patient who is seated or lying on a bed, place the thigh in a moderately abducted and externally rotated position with the knee flexed. If the patient is supine, place the patient's legs into a frog-leg position ( Figure 18) with knees apart and ankles close together. Another position is to cross the examined leg over the shin or ankle (figure four position, Figure 19) The deep abdominal reflexes were described by Gerhardt in 1895. In normal individuals, this reflex is minimally present. If the deep abdominal reflexes are exaggerated but the superficial abdominal reflexes are absent, this suggests a corticospinal tract lesion above T6. If the reflex is exaggerated, a 'lower response' or puboadductor reflex occurs. The ipsilateral adductor muscles contract with some hip flexion. (12,13,14) Patellar The Achilles reflex was also described by Erb and Westphal in an article published in the German Archives for Psychiatry and Nervous Disease in 1875 (1, 2, 3).

Reflexes of the Lower Extremity
The following reflexes are less significant as they may be difficult to be elicited in normal patients. These reflexes are significant with unilateral absence or the presence of exaggeration which suggest corticospinal tract disease.
Adductor Reflex (Figure 20   In a hyperactive reflex, tapping other areas of the sole (medioplantar reflex) will elicit the reflex. With reflex spread, striking the Achilles tendon may cause knee flexion. Paradoxical ankle reflex occurs when tapping the anterior aspect of the ankle produces reflex. It is important to note that studies by O'Keefe et al. (16) and Schwartz et al. (17) showed that the plantar strike technique is a more reliable method in elderly patients.  This reflex helps to determine if an absence of ankle jerk is due to peripheral neuropathy (preserved external hamstring reflex) or due to radiculopathy (absent lateral hamstring reflex).

Sustained clonus
Notes: The '+' after the number is to distinguish from muscle testing. The Jendrassik manoeuvre is a reinforcement technique, whereby the patient attempts to pull the hands apart with fingers hooked together. The effects last 1 sec-6 sec and is maximal for only 300 msec

Interpretation
The DTR depends on the integrity of the upper motor neuron and lower motor neuron. An exaggerated DTR suggests an upper motor neuron disease and a reduction or loss of DTR suggests a lower motor neuron disease. Certain patients may have absent or exaggerated reflexes. However, on their own, these findings do not confirm a neurological disease. They are only significant if the reflex amplitude is asymmetrical, unusually brisk, or absent compared to other spinal levels and if they are associated with other neurological signs to indicate either an upper motor neuron or lower motor neuron disease.
Other findings such as reflex spread is defined as the increase in the reflexogenic zone. The reflex tested is present but is accompanied by a response in other muscles. For example, the biceps reflex is obtained by tapping the clavicle and is accompanied by flexion of the fingers. An inverted reflex is seen when the reflex tested is absent but creates a contraction in other muscles. An inverted brachioradialis reflex is seen when the afferent limb of the reflex arc is impaired. There is an absence of supination and flexion of the elbow with twitching or flexion of the hand and fingers.
A video (https://youtu.be/Fz4kAa4Wj_g) has been produced to demonstrate the examination techniques described in this article.

Conclusion
The DTR remains an important aspect of the neurological exam despite the use of modern imaging. The reflex hammers have undergone multiple modifications throughout the years to produce a heavy, high-quality tool with a flexible handle to give a crisp blow to effectively stretch the tendon. More importantly, clinicians should master the correct technique to perform the art of eliciting DTR as this provides valuable information on the condition of the nervous system. The use of correct tools and techniques increases the validity of the findings. A good understanding of the physiology of the DTR allows precise interpretation of the clinical findings.