Blinking is one of the most frequent motor actions that humans perform every day. It involves the rapid activation of the orbicularis oculi muscle and relaxation of the levator palpebrae muscle. Simply counting the rate of spontaneous blinking and recording the electromyographic (EMG) activity and eyelid movement with EMG electrodes can furnish very relevant information that is of value in the diagnosis of neurologic disorders. However, for neurophysiologic testing, it is customary to induce blinking by applying a sensory stimulus. The most typical method is to apply an electric shock to the supraorbital nerve. Unilateral stimuli give rise to bilateral blink reflexes in the orbicularis oculi muscles. This permits separate assessment of the afferent and the efferent arms of the reflex circuit. The blink reflex, together with other cranial nerve reflexes, such as the jaw jerk and the masseteric silent period, permits the neurophysiologic evaluation of many disorders affecting the cranial nerves and brainstem. Additionally, the blink reflex can be used to examine various functions that are either integrated in, or mediated by, the brainstem. The physiology and clinical utility of the blink reflex are reviewed in this chapter. Other brainstem reflexes, such as the jaw jerk and the masseter inhibitory reflex, are also considered briefly.
Blinking is a frequent human action that occurs spontaneously or in response to various sensory inputs. Blink and corneal reflexes are elicited by finger tap on the forehead or light touch of the cornea, and the resulting response can be observed clinically. Recording the EMG activity from the orbicularis oculi provides for a more complete evaluation of the reflex circuit. The most commonly used sensory stimulus for eliciting the blink reflex in the laboratory is a brief electrical shock applied to the supraorbital nerve. The reflex response to that stimulus consists of two separate components: an early ipsilateral R1 and a later bilateral R2 ( Fig. 19-1, A ). The two responses have a different afferent circuit in the brainstem ( Fig. 19-1, B ), based on observation of the abnormalities generated with various focal disorders. The afferent input goes to facial motor neurons in the facial nucleus in the pons.
The R1 response relates to an oligosynaptic pontine reflex, whereas R2 is relayed through a more complex route including many interneurons in the pons and lateral medulla. R1 serves as a more reliable measure of nerve conduction along the cranial nerves and reflex pathways, with only one or two interneurons between the trigeminal input and the responding facial motor neurons. Analysis of the ipsilateral and contralateral R2 components helps to localize any lesion to the afferent or efferent reflex arc. Involvement of the trigeminal nerve causes an afferent pattern of abnormality, which consists in a delay or diminution of R2 bilaterally after stimulation on the affected side. Diseases of the facial nerve give rise to an efferent pattern with alteration of R2 only on the affected side, regardless of the side of stimulation. A pattern of abnormality not reflecting either an afferent or efferent lesion often indicates the presence of a lesion in the brainstem.
Blink reflexes can be obtained by electrical stimulation of cranial nerves other than the supraorbital nerve. Stimulation of the infraorbital nerve evokes R1 in some cases and R2 in most subjects, with similar latencies to those elicited by supraorbital nerve stimulation. Shocks applied to the mental nerve elicit R1 or R2 inconsistently and with variable latency. When electrically induced, the blink reflex is believed to result from activation of cutaneous type II afferents. However, the blink reflex can also be induced by activation of other nerve fibers using specific stimuli. This is the case for nociceptive trigeminal afferents. Ellrich and co-workers first pointed out the possibility that some components of the blink reflex were elicited by nociceptive inputs. Katsarava and colleagues, using a specially designed electrical stimulator to activate selectively or preferentially small nociceptive fibers of the trigeminal nerve, induced a nociceptive blink reflex response consisting of a bilateral R2 response at a latency of about 50 msec. Other authors subsequently confirmed that the blink reflex can be induced by activation of nociceptive trigeminal afferents by using laser or contact heat stimulators. A mechanical tap over the supraorbital notch or to the glabella also elicits a blink reflex of similar characteristics as the one evoked by electrical stimulation. However, although the stimulus is a gentle tap, the response is considered an exteroceptive reflex rather than a stretch reflex; it probably is relayed via the same polysynaptic reflex pathways as the electrically elicited blink reflex. Magnetic stimulation over the supraorbital nerve or at various scalp sites where it can activate trigeminal skin afferents can be used as an alternative in patients who cannot tolerate electrical stimulation.
The blink reflex may also occur as a response to auditory stimuli. The response can be limited to the orbicularis oculi, the so-called auditory blink reflex (ABR). Alternatively, depending on the intensity of the sound and the subject’s reactivity, the response can be part of a more generalized startle reaction. The auditory blink reflex and the startle response follow two different brainstem circuits. Nevertheless, many similarities exist between these two types of response. The same is probably true for the responses of the orbicularis oculi to electrical stimuli applied to peripheral nerves in the limbs. In this case, the orbicularis oculi response is known as the somatosensory blink reflex but it also could be part of a startle reaction to the somatosensory stimuli. Figure 19-2 shows the blink response recorded in the orbicularis oculi to various types of stimuli.
The blink reflex is a powerful tool to measure the excitability of the trigemino-facial reflex arc and, through that, the physiologic correlates of certain functions (and the pathophysiology of dysfunctions) that are either integrated in, or mediated by, the brainstem. There are two main techniques for using the blink reflex to measure excitability of brainstem interneurons or their modulation by rostral inputs: the blink reflex excitability recovery curve (BRER) and blink reflex inhibition by a prepulse (BRIP).
BRER is examined using the double-shock technique. The first stimulus (conditioning) induces a transient change in the excitability of reflex circuits, and the second stimulus (test), delivered at varying interstimulus intervals with respect to the first, is used as a probe stimulus. The BRER curve is obtained by plotting the size of the response elicited by the test stimulus as a percentage of the response to the conditioning stimulus for all intervals tested (usually between 100 and 1000 msec). With paired supraorbital nerve electrical stimuli, R1 is affected little by the conditioning shock, whereas R2 usually is abolished completely from 0 to 200 or 300 msec, and then slowly recovers to reach about 30 to 50 percent at the 500-msec interval and 70 to 90 percent at the 1500-ms interval. Since the same facial motor neurons subserve both R1 and R2, the different behavior of R1 and R2 is attributed commonly to differences in the interneuronal net. Because interneuronal excitability is under the control of rostral structures, including the basal ganglia, patients with disorders such as parkinsonian syndromes and many forms of dystonia show an enhancement of the R2 excitability recovery curve. The circuits by which the basal ganglia influence the excitability of brainstem interneurons are not understood completely. A pathway comprising the nucleus raphe magnus and the superior colliculus has been suggested on the basis of the relationship between saccades and gaze shifts.
BRIP is examined using methods described mainly in regard to the prepulse effect on the startle reaction. A prepulse is any low-intensity stimulus that is unable to cause a recordable response by itself but that induces changes in the response to a subsequent suprathreshold stimulus. This effect is considered to be due to the attentional shift required to process the information brought about by the prepulse. Healthy subjects probably are able to integrate at the subcortical level impulses generated by the environmental conditions of daily life, such as visual, acoustic, or somatosensory impulses. These environmental impulses may adopt the role of prepulse stimuli and cause inhibition of undesired motor reactions, which otherwise would interfere with sensory processing of relevant inputs. A prepulse leads to inhibition of the R2 component in the blink reflex to supraorbital stimuli, and also to an increase in the amplitude of the R1 component at a relatively short interstimulus interval between prepulse and supraorbital nerve stimuli. This effect on R1 is not obvious with other stimuli. The dissociation of the prepulse effects on the R1 and R2 components of the blink reflex suggests that the effects of a prepulse take place in sensory integration centers at a pre-motor neuronal level.
The circuit mediating prepulse inhibition is not well known but it is likely related to the pedunculopontine tegmental nucleus. Using the electrodes inserted in the subthalmic nucleus for repetitive deep brain stimulation in patients with Parkinson disease, Costa and co-workers found that the latency of the prepulse effect on the blink reflex was between 0 and 5 msec, suggesting that the effect was mediated by a nearby structure. Since the electrical stimulus first depolarizes axons that are perpendicular to the electrical field, the authors suggested that a likely site of the prepulse effect is the fiber bundle that links the globus pallidus internus with the pedunculopontine tegmental nucleus. It is also possible that some neuronal groups in the subthalamic nucleus projecting to the pedunculopontine nucleus were influenced by the electrode. The excitability of the prepulse circuit can be modulated by inputs from the basal ganglia through the output connections from the globus pallidus internus to the pedunculopontine tegmental nucleus, and from this to the nucleus reticularis pontis caudalis. Disorders of the basal ganglia can affect both the BRER and BRIP through these specific circuits. Figure 19-3 shows a hypothetical schematic explanation of the circuits engaged in the basal ganglia modulation of BRER and BRIP.
Methods and Normative Data
Subjects should be resting on a bed or comfortable chair. Surface electrodes are used for both stimulation of the nerve and recording of the response from the orbicularis oculi muscle. Electrical shocks can have a duration of 0.1 or 0.2 msec and an intensity 2 to 3 times above perception threshold (usually ranging between 50 and 100 V or 5 to 10 mA, assuming a skin resistance of 10 kOhm). The R1 response is usually present with these stimulation parameters but, if not, a good strategy is to request a mild voluntary contraction or apply paired shocks with interstimulus intervals of 3 to 5 msec. The R2 response is habituated easily and, therefore, it is best to give single stimuli with enough separation between them to provide for full recovery of excitability in the circuit. Stimuli applied unexpectedly may be more effective for eliciting the R2 responses. The active recording electrode is placed on the lower aspect of the orbicularis oculi muscle and the reference electrode is placed 2 to 3 cm laterally. Careful placement of the electrodes helps reduce the stimulus artifact that may be bothersome with supraorbital nerve stimulation. An optimal frequency response ranges from 10 Hz to 10 kHz for the recording of either the R1 or R2 component. Increasing the low-frequency cutoff to 100 Hz or even more may be helpful in sharpening the contrast of the responses against baseline, but cutting low frequencies may diminish response amplitude and delay the onset latency. The ground electrode can be placed on the chin or the ears or in another position in the body where it does not interfere with the recordings. Impedance should be checked and electrode placement adjusted if artifacts are problematic.
Normative values for latency of the responses should be obtained in each neurophysiologic laboratory before any patients are studied. The reference values used in the author’s laboratory, obtained using a standard technique, are summarized in Table 19-1 . The R2 response of the side ipsilateral to the stimulus is usually of larger amplitude and duration than that obtained contralaterally (see Fig. 19-1 and data in Table 19-1 ). The ratio between R2 responses generated in the same orbicularis oculi after ipsilateral and contralateral stimulation is thus larger than 1. Interestingly, latency values for R1 obtained in infants are usually longer than in adults, despite a considerably shorter reflex arc. In contrast to adults, an R2 response is present in only two-thirds of neonates, mostly on the side ipsilateral to the stimulus, and rarely in premature babies.
|Mean latency (msec)||11.7 (0.4)||35.1 (2.3)||35.3 (2.4)|
|Amplitude (mV)||0.7 (0.6)||0.9 (0.7)||0.7 (0.4)|
|Duration (msec)||5.4 (0.5)||22.1 (17.2)||18.7 (14.9)|
|Latency upper limit (msec)||13.0||42.2||43.1|
|Latency difference between sides (msec)||0.5 (0.1)||3.2 (2.2)||3.5 (2.8)|
Jaw jerk and masseteric silent period
The mandibular reflex, or jaw jerk, is the only monosynaptic reflex available for electrophysiologic testing in the cranial muscles. It is elicited by a mechanical tap over the mandible with a tendon reflex hammer. The jaw jerk is difficult to evaluate simply by inspection. It is impossible to discern whether there are any abnormalities restricted to one side, or even whether the reflex is present at all in patients who show a response of the lips as well. EMG monitoring of the masseter or temporalis muscle response is of paramount importance in the evaluation of suspected brainstem lesions. The mandibular reflex circuit is also unusual regarding the location of the involved cell bodies. In contrast to those of all other muscles in the body, the proprioceptive neurons of the jaw muscles lie within the neuraxis, protected by the blood–brain barrier from peripheral circulating agents. This is important regarding the diagnosis of certain disorders due to immunologic involvement of the sensory neurons of the Gasserian ganglia.
Stimulation of the stretch receptors induces not only the jaw jerk but also an inhibitory response in the contracting masseter muscles. This is known as the masseter silent period and may correlate with the reaction that occurs when there is an unexpected pinch of our tongue or lips, as while eating. However, in patients with ill-fitting dentures, this silent period may not be reliable. A standardized way of measuring the inhibitory masseteric reflexes is to apply an electrical shock to the mentalis nerve. This gives rise to a long-duration silent period, also called the masseter inhibitory reflex (MIR), which is divided into two phases by a burst of EMG activity. Afferent impulses reach the pons via the sensory mandibular or maxillary root of the trigeminal nerve. The first inhibitory period (MIR1; 10- to 13-msec latency) probably is mediated by inhibitory interneurons located close to the ipsilateral trigeminal motor nucleus and projecting onto jaw-closing motor neurons bilaterally. The whole circuit lies in the mid-pons. The afferents for the second inhibitory period, the MIR2 (40- to 50-msec latency), descend in the spinal trigeminal tract and connect to a polysynaptic chain of excitatory interneurons, probably located in the medullary lateral reticular formation. The last interneuron of the chain is inhibitory and gives rise to ipsilateral and contralateral collaterals that ascend to the spinal trigeminal complexes of both sides, to reach the trigeminal motor neurons. Masseter inhibitory reflexes cannot be tested by clinical procedures alone.
The MIR is also consensual—that is, a unilateral stimulus gives rise to bilateral responses. The MIR can provide information on the functional state of the trigemino-trigeminal inhibitory circuits. The MIR also can be elicited by stimulation of the infraorbital nerve with similar characteristics to that obtained by stimulation of the mentalis nerve. Other stimuli, such as magnetic stimulation over the scalp or laser stimulation to the face, can give rise to a transient inhibition of masseter muscle contraction.
Methods and Normative Data
The EMG evaluation of the jaw jerk requires use of a reflex hammer that electronically triggers the oscilloscopic sweep. The reflex responses are recorded simultaneously from the right and left masseter muscles using pairs of surface electrodes, the active one placed over the muscle belly at the angle of the mandible, and the reference one placed over the mastoid process or ear lobe. Mean latency of the reflex response is variable on both sides, depending on the strength of the tap and whether the masseter muscle is at rest or tonically contracted. If the reflex response is not obtained at rest, a good strategy is to request the subject to close the mouth slowly while the mandible is tapped. Since reflex latencies vary with successive trials, comparison of simultaneously recorded right-sided and left-sided responses is more meaningful than absolute values, which are of the order of 6 to 8 msec in normal subjects.
Mandibular tapping induces a silent period with a latency of between 10 and 14 msec, which duration is rather variable, depending in part on the strength of the tap and the level of activity in the masseter muscles. The MIR usually is obtained by electrical stimulation of the mentalis nerve. If the MIR is to be used for electrodiagnostic purposes, the subject must exert a steady background voluntary contraction, and the amount of inhibition should be quantified (e.g., the area of suppression). To do that, the signals must be either full-wave rectified and averaged, or examined by superimposing several trials. As with the blink reflex, the MIR can provide information on the site of the lesion by analyzing whether the pattern of abnormality is afferent, mixed, or efferent. Some subjects may not be able to activate their masseter muscles in isolation and contract, instead, the perioral muscles. In these cases, no suppression of the EMG activity can be elicited with either a tap to the chin or an electrical stimulus, because there is no silent period in most facial muscles. This should not be mistaken for an abnormal MIR.
MIRs cannot be tested by clinical procedures alone and, in some patients, testing the MIR may be the only way of revealing trigeminal or brainstem dysfunction. For the reflex responses of the masseter muscles to have clinical relevance, standardized methods of stimulation and recording should be used. Again, normative values should be obtained in every neurophysiologic laboratory. Table 19-2 provides the reference values obtained in the author’s laboratory and Figure 19-4 shows an example of the jaw jerk, the masseter silent period induced by a chin tap, and the MIR induced by electrical stimuli to the mentalis nerve.
|Side 1||Side 2||Maximum Interside Difference|
|Jaw jerk mean latency (msec) *||5.6 (1.1)||5.6 (1.1)||0.7|
|Masseteric silent period latency (msec) *||11.8 (1.2)||11.8 (1.2)||0.8|
|Masseteric silent period duration (msec) *||18.7 (7.1)||18.7 (7.1)||6.8|
|MIR1 latency (msec)||14.5 (2.1)||14.5 (2.1)||0.8|
|MIR1 duration (msec)||19.8 (5.5)||20.1 (5.7)||3.8|
|MIR2 latency (msec)||44.6 (2.8)||44.9 (3.2)||4.1|
|MIR2 duration (msec)||22.6 (9.1)||22.2 (9.3)||7.4|
The study of the blink reflex to supraorbital nerve stimulation helps in the assessment of many disorders involving the trigeminal and facial nerves and the brainstem. Latency and amplitude of the responses are the most clinically useful parameters but a study of excitability might be important in some central nervous system (CNS) disorders.
Trigeminal Nerve Lesions
The R1 response of the blink reflex is relatively stable with repeated trials and therefore is better suited for assessing nerve conduction through the trigeminal and facial nerves. Analysis of R2, however, helps in localizing the lesion in the trigeminal nerve or nuclei. In these cases, R2 is slowed, diminished, or absent bilaterally when the affected side of the face is stimulated (afferent delay), whereas the responses are normal to stimulation of the unimpaired nerve.
Selective lesions of trigeminal sensory branches are relatively infrequent. They may occur as a result of traumatic nerve injuries or ischemia in relation to arteritis or inflammation in connective tissue diseases. The ophthalmic branch of the trigeminal nerve may be involved together with the oculomotor nerve in the wall of the cavernous sinus and in the Tolosa–Hunt syndrome. The infraorbital and mental branches of the trigeminal nerve can be damaged at the infraorbital foramen or the mental foramen, when the nerves cross the skull, in the so-called numb-cheek or numb-chin syndrome. These syndromes typically are associated with metastatic infiltration of bone in patients with cancer of various types, but the most common causes are lesions resulting from dental procedures or bone infection. The appropriate tests should be chosen according to lesion localization. While the blink reflex can be used to assess the ophthalmic and infraorbital branches, the MIR can provide evidence for involvement of the mentalis branch or infraorbital branches. In fact, the MIR is a trigemino-trigeminal reflex. Testing the MIR may be the only way in some patients of revealing trigeminal or brainstem dysfunction. As in blink reflex studies, the pattern of abnormality (afferent, mixed, or efferent) provides information on the site of the lesion. However, the efferent pattern of lesion is extremely rare except in conditions such as a purely motor trigeminal neuropathy and hemimasticatory spasm.
Involvement of the trigeminal nerve is relatively frequent in patients with Sjögren syndrome. These patients may have a focal or multifocal axonal lesion of the trigeminal nerve or an antigen-mediated immunologic neuronopathy involving neurons in the Gasserian ganglion. Even though both types of lesions have certain clinical similarities, the underlying pathophysiology is quite different. Brainstem reflex testing ideally should indicate the site of the damage. In lesions involving the Gasserian ganglion, the jaw jerk is preserved but the cutaneous-mediated reflex responses are abnormal. In focal or multifocal lesions of the trigeminal nerve or its branches, a branch-compatible distribution of the abnormalities is seen when various brainstem reflexes are studied in combination. A variable pattern of lesions may also be found in other disorders—for instance, in systemic sclerosis, where the damage may include both intra- and extra-axial lesions.
No abnormalities usually are detected with the study of brainstem reflexes in patients with idiopathic trigeminal neuralgia. In contrast, R1 was abnormal on the affected side of the face in 10 of 17 patients with tumor, infection, or other demonstrable causes of facial pain. Lesions may be induced in neurons of the Gasserian ganglion following therapeutic thermocoagulation or compression. In accordance with the fact that thermocoagulation predominantly affects small fibers while compression affects large fibers, Cruccu and associates found that patients with thermocoagulation had predominant impairment of corneal reflexes whereas patients treated with compression had predominantly abnormal jaw jerks.
Brainstem vascular lesions may cause injuries in many sites involving the nuclei and tracts of the trigeminal nerve. The most frequent and also the most studied of them is Wallenberg syndrome, which involves the nucleus ambiguus, the spinothalamic tract, and the spinal trigeminal nucleus. Depending on the exact location of the lesion, patients with Wallenberg syndrome may present with a slightly different picture regarding clinical, imaging, and electrophysiologic abnormalities. Again, an afferent pattern of lesion is observed in the R2 responses of the blink reflex but it is important to note that the R1 response, which is integrating in the upper brainstem, is spared ( Fig. 19-5 ).