An electrophysiological investigation of deglutition in man

This article describes a combined electrophysiological and mechanical method used to measure laryngeal movements and related submental EMG activity during swallowing. The mechanical upward and downward movements of the larynx were detected using a piezoelectric sensor while the submental integrated EMG (SM-EMG) was recorded. Mea- surements were performed in 29 human subjects. The interval between the onsets of the two sensor signal deflections was used as a measure of the time the larynx remained in its superior position during swallowing. In 10 subjects, the cricopharyngeus muscle (CP) of the upper esophageal spinchter showed a continuous tonic EMG activity except during swallowing. All the parameters measured were influenced by the type and volume of the bolus material. The method presented in this study proved its usefulness in the study of the physiology of deglutition as well as in its objective clinical evaluation in patients with dysphagia. 0 1995 John Wiley 8. Sons, Inc.

Key words: crycopharyngeal muscle submental EMG pharyngeal swallowing electrophysiology of swallowing dysphagia

MUSCLE & NERVE 18:1177-1186 1995

AN ELECTROPHYSIOLOGICAL INVESTIGATION OF DEGLUTITION IN MAN

CUMHUR ERTEKIN, Prof, MURAT PEHLIVAN, MD, iBRAHlM AYDOGDU, MD,

ZAFER COLAKOGLU, MD, AYSE SAGDUYU, MD, and NUR YUCEYAR, MD MUSTAFA ERTAS, MD, BURHANETTIN ULUDAG, MD, GURBUZ CELEBI, Prof,

Swallowing is a complex physiological function that has usually been investigated in the past by cinefluoroscopic, videofluoroscopic, and manometric methods. In such studies, swallowing was usually subdivided into three phases: an initial oral phase; a pharyngeal phase; and an esophageal pha~e.~*’*~’’ The oral phase is mainly voluntary and highly variable in duration depending upon taste, environment, hunger, motivation, etc.

The pharyngeal phase encompasses several closely coordinated actions that involve bolus transport from the oropharynx into the esophagus without aspiration. As the bolus is delivered from the oral cavity into the oropharynx, the following events occur su~cessively’~:

From the Departments of Neurology and Biophysics, Ege University Med- ical School Hospital, Bornova, Izmir, Turkey.

Acknowledgments. Supported by a grant from the Turkish Scientific’ and Technological Research Council (TUBITAK), Project No. TBAG-U/17-4.

Address reprint requests to Prof. Cumhur Ertekin, Department of Neurol- ogy and Chnical Neurophysiology , Ege University Medical School Hospi- tal, Bornova, Izmir, Turkey.

Accepted for publication March 24, 1995.

0 1995 John Wiley & Sons, Inc. CCC 0148-639x/95/101177-10

1. Channels to the nasopharynx, oral cavity, and the larynx are closed off.

2. The crycopharyngeal sphincter (or upper esophageal sphincter) is relaxed and the larynx and hyoid bones are lifted upward and anteriorly so that the bolus is engulfed by the pharynx.

3. The sequential peristaltic waves produced by pharyngeal oblique muscles sweep down the entire bolus into the opened upper esophageal sphincter.

After the pharyngeal phase is completed, the pharynx is ready once more to carry on with its respiratory function.

Although the pharyngeal component of swallowing is dependent on central nervous activity it is not always under voluntary control, and many swallows including those that occur spontaneously are subconscious events.

When the pharyngeal phase of swallowing is disturbed by neuromuscular disorders, neurogenic dysphagia may result, leading to the misdirection of the bolus into the nasal, oral, or laryngeal cavi- ties.

The objective evaluation of dysphagia is impor-

Deglutition in Man MUSCLE & NERVE October 1995 1177

tant for the selection of a treatment method.I7 Cineradiographic, videofluorographic, and/or manometric evaluations may be indicated for such patients. But these methods are quite expensive, time consuming, and are often possible to conduct only in clinics of radiology, gastroenterology, and otorhinolaryngology.

Swallowing and its disorders have not been ex- tensively studied by electrophysiological methods. Notable exceptions are perhaps the electromyo- graphical (EMG) studies conducted in man.7*’9-2’ These studies focused on the cricopharyngeal sphincter and/or pharyngeal constrictor muscles.

In the present study, we introduce a convenient electrophysiological method to measure the submental EMG activity associated with the superior and inferior movements of the larynx during a swallow. These measures together reflect the cricopharyngeal sphincter opening and closing during the pharyngeal phase of swallowing.

MATERIALS AND METHODS

Measurements were made in 29 healthy volunteers (14 females, 15 males) most of whom were medical school students and hospital staff who did not have any oropharyngeal or gastrointestinal problems. The subjects ranged in age from 20 to 64 years. The investigation was approved by the ethical com- mittee of our university hospital. Informed con- sent was obtained from each subject.

The subject sat on an examination coach and was instructed to hold his head in a natural upright position. The EMG activity was recorded on a Medelec Model MS-20 EMG apparatus using bipolar silver-chloride EEG electrodes taped under the chin over the mylohyoid-geniohyoid-anterior digastric muscle complex. The surface muscle activity obtained thus was named submental EMG activity (SM-EMG). The EMG signal was bandpass filtered (between 100 Hz and 10 kHz) amplified, rectified, and then integrated.

Since we also wanted to detect the laryngeal movements, a mechanical sensor that consisted of a simple piezoelectric wafer with a 4 X 2.5 mm rub- ber bulge affixed at its center was placed on the coniotomy region between the cricoid and the thyroid cartilages at the midline. This point could easily be palpated and located on the skin surface. The sensor was fixed onto the neck with medical tape and was belted around it to secure close con- tact with the skin.

The sensor output was connected to another channel of the EMG apparatus. The sensor ampli- fier output was also bandpass filtered (cut-off frequencies 0.01-20 Hz).

The sensor gave two deflections of generally opposing polarity during each swallow, the first of which was often a negative (upward on the screen) deflection. The leading or the trailing edge of the first deflection was used to trigger the delay line circuitry of the recording apparatus so that all signals were time locked to the same instant. Since the SM-EMG activity coincided with the laryngeal upward movement, the rectified-integrated SM-EMG activity was also time locked to the laryngeal sensor signal. The preanalysis time was set at 800 ms and the total sweep time was set at 2 s. In some exper- iments, a 5-s sweep time was also used. At least five successive sensor and EMG signals were recorded for each type of swallow. Swallows were classified as dry swallowing or according to bolus type as liquid swallowing (water), semisolid material (pudding) swallowing, and solid material (hard cookies) swallowing. The signals were examined as single traces and as in the summated form. Their averages were also obtained operating the EMG re- corder in the digital averager mode.

The onset times and peak levels of the averaged sensor signals were measured and the intervals between the first and the second deflections were calculated using the onset times (labeled as 0-2 in Fig. 1) or the peak latencies (labeled as 1-3 in Fig. 1). The duration of the SM-EMG activity during a swallow was determined by measuring the interval between its onset and its peak point (labeled A-B in Fig. l), and by measuring the interval between its peak point and end point (labeled B-C). The total duration and the peak-to-peak amplitude (A- C) of the integrated EMG signals were also measured. Laryngeal sensor deflections and SM-EMG activity were compared with respect to their time relationship.

Swallows were initiated with the bolus (liquid or semisolid) positioned on the tongue and the tongue tip touching the upper incisor^.^ Swallow signals were recorded following delivery from a graduated syringe, of 1, 3, 5, and 10 mL liquid (water) or semisolids (pudding). Some subjects were also asked to swallow solid materials. They were given pieces of hard cookies and all the above swallow correlates were recorded from them as well. However, bolus volumes could not be accu- rately determined in these trials because there always remained some bolus residue in the oral cavity from the previous trial. In many subjects, “dry swallowing” was also investigated.

Cricopharyngeal (CP) muscle EMGs were recorded in 10 subjects. The EMG was recorded using concentric needle electrodes (Medelec dispos- able needle electrode DMC-37; diameter 0.46 mm,

1178 Deglutition in Man MUSCLE & NERVE October 1995

1 mL WATER SWALLOW

1

I

I1 - A C

-ti Cricoid Cartilage

electrode

Thyroid Cartilage

Sensor

FIGURE 1. Laryngeal sensor signals ( I and 111) and integrated submental EMG activity (I1 and IV) during 1-mL water swallowing. Numbers I and II denote the average of five responses; 111 and IV are the same five responses superimposed. The numbers 0-1 define, respectively, the onset and the negative peak of the first deflection of the laryngeal sensor signal. The second deflection of the sensor signal defined between points 2 and 4 represents the downward movement of the larynx. Note the “swallowing jitter” in the second deflection in trace 111. Points A, B, and C (in II) denote the onset, peak and the end of the submental EMG activity. Delay line trigger level was adjusted to the mid-level of the leading edge of the first deflection and the delay was set at 800 ms. Amplitude calibration: 70 pV in I1 and 500 pV in IV; time calibration: 200 ms in all traces (amplitude of sensor signal is unimportant).

recording area 0.07 mm2). The needle electrode was inserted through the skin at the level of the cricoid cartilage, about 1.5 cm lateral to its palpable lateral border in the posteromedial direction. High frequency, tonic EMG activity appeared on the oscilloscope screen as the electrode penetrated the muscle. During dry or liquid material swallowing, this tonic EMG activity disappeared for a short time (400-500 ms). Such a pause in EMG activity during swallowing and the increase in tonic activity prior to and following this pause served as the criteria for correct electrode entry into the CP muscle. CP muscle activity was also rectified, integrated, and averaged during all types of swallowing. The filter settings were the same as those that were used for the SM-EMG activity recording.

Mean 2 SEM values were calculated for all quantities measured and statistical analyses were performed to assess the effects of different swallow conditions using variance and correlation analyses as appropriate.

RESULTS

Typical records of laryngeal movements obtained from the piezoelectric sensor are shown in traces I and I11 of Figure 1. Submental EMG activity during 1-mL water swallowing is shown in the same figure (traces I1 and IV). The laryngeal movement

signal showed a pattern of two deflections. The first deflection, which coincided with the upward laryngeal movement, was usually but not always a negative (upward) deflection (labeled 0-1 in Fig. 1) while the second deflection, which was associated with the downward movement of the larynx, was a positive (downward) deflection (defined by points 2-34 in Fig. 1).

All the signals were time locked and stabilized on the screen by triggering the oscilloscope with the leading edge of the first deflection. It is apparent from the above description that the interval (0-2 in Fig. 1) between the onsets of two sensor deflections represents the time necessary for the elevation, closure, and upward relocation of the larynx. During this time, the bolus passes through the upper esophageal sphincter which opens only during deglutition. Therefore, the 0-2 interval is probably a measure of the pharyngeal transit time of the bolus between the oropharynx through the cricopharyngeal juncture and the esophagus. l 7

The end of the downward laryngeal movement represented by point 4 in Figure 1 varied consid- erably among subjects as well as among different trials with the same subject swallowing boluses of the same volume. We consider this variability to be an inherent property of the swallowing process and call it, in this article, the “swallowing jitter.” It is measured from the variability of point 3 on the


second deflection (Fig. 1, trace 111). In normal subjects, the average swallowing jitter was found be approximately 82 ms during 3 mL of water swallowing and 134 ms during 10 mL of water swallowing (Table 1).

SM-EMG activity begins approximately 300- 400 ms before the onset of the upward laryngeal movement (with a range of 100-700 ms). Its peak latency (point B) is closely associated with the peak latency of the sensor signal (point 1). Its amplitude reaches its peak during the A-B interval and declines progressively during the two laryngeal sensor deflections. It usually ceases just before or at the onset of the second deflection. However, this activity can be longer or shorter and in some trials can rise again during its declining phase depending on the material swallowed. Such variations in the submental muscle activity change and interfere with the sensor signals. But the onset times of the sensor deflections (point 0 and 2) remain station- ary and can always be distinguished in single records as well as in superimposed and averaged records.

The Relationship between Sensor Signal Deflections and Cricopharyngeal Sphincter Muscle. T h e cricopharyngeus muscle (CP) is the main anatomic correlate of the upper esophageal sphincter and its continuous EMG activity is transiently switched off during s ~ a l l o w i n g . ' ~ ~ ~ ' ~ It is therefore reasonable to expect a temporal relationship between the CP- EMG pause and the sensor signal deflections recorded in this study, since the beginning of the upward movement of the larynx is closely associated with the manometric upper esophageal sphincter re la~a t ion '~ and this movement always

precedes the upper esophageal sphincter opening with a mean interval of 100 ms.

In 10 subjects, we investigated the CP-EMG activity using a concentric needle electrode simultaneously with sensor signal recording. Figure 2 shows such a record obtained from one subject. At rest there is a continuous high frequency EMG activity in the CP muscle, but during a swallow it ceases immediately (second traces in Fig. 2A and B). There was almost always a high (both in frequency and amplitude) EMG activity before and following the pause (about 400 ms in this case), the first burst being always shorter than the second one.

The upward laryngeal movement represented by the first deflection of the sensor signal begins just before the EMG pause of the CP muscle. The CP-EMG pause ends just before or at the onset of the downward laryngeal movement as reflected by the second deflection of the sensor signal (Table 2). Thus the onset times of the first and second deflections (0-2) may be related with the CP muscle EMG pause-that is to say with the upper esophageal sphincter opening. This relationship was demonstrated by a regression equation with a correlation coefficient of r2 = 0.4852. However, this correlation depended on the type of material being swallowed, and the bolus volume effected both the 0-2 interval length and the duration of the CP-EMG pause.

An earlier phase of EMG suppression was observed in some cases. This early suppression was usually not a loss of all the motor unit activity but rather a reduction in its frequency (first pause in Fig. 3). In such cases the first burst of the EMG activity appeared just before the first laryngeal sig-

Table 1. Summary of the electrophysiological findings obtained by swallowing (mean f SEM).'

Laryngeal sensor SM-EMG activity

No. of 6 2 sensor Swallowing A-0 intervalt A-B duration A-C duration Amplitude Type of swallowing cases interval (ms) jitter (ms) (ms) (ms) (ms) ($4

Dry swallowing 1 mL water 3 mL water 5 mL water 10 mL water 3 mL semisolid 10 mL semisolid 1 mL solid* 3 mL solid 5 mL solid

14 9

26 10 13 15 9 8 8 5

487.8 2 24.2 504.3 2 29.1 562.8 f 23.2 566.6 f 32.0 568.1 2 34.8 516.2 f 31.7 529.0 f 23.1 524.5 f 28.2 504.7 2 27.5 520.8 f 32.2

129.1 k 22.6 60.5 f 11 .O 81.9 t 7.9 93.9 2 28.2

134.5 f 23.9 95.6 t 15.8 70.2 f 7.1

107.8 2 28.1 55.8 f 9.1 81.4 2 20.4

380.0 2 53.5 355.7 2 30.8 294.1 ? 26.7 373.8 ? 32.2 334.4 t 33.4 361.4 ? 34.3 355.3 2 58.3 374.6 f 42.0 370.2 2 41.6 358.6 f 32.8

382.0 ? 58.6 301.5 2 37.3 294.0 -t 24.4 374.1 f 54.6 367.2 2 39.8 310.7 2 28.2 328.8 2 55.0 361.7 2 44.3 305.5 ? 27.2 268.6 2 43.7

901.8 f 60.9 850.0 2 60.8 880.1 f 38.2 912.5 f 43.9 91 2.6 2 46.9 888.3 f 36.0 953.3 f 58.7 865.7 f 60.7 865.0 f 57.6 980.2 ? 66.4

129.9 f 22.9 86.6 t 14.1 75.7 2 7.5 91.1 f 8.6 81.7 2 7.9

111.3 f 13.0 95.7 f 10.1

140.5 f 13.7 132.1 f 10.4 127.4 t 12.0

~ ~~

*A// results were obtained from 29 normal subjects tThe interval between SM-EMG onset and the beginning of the laryngeal sensor signal #Pieces of hard cookies were weighted and their sizes were transferred as "mL "


3 mL SEMISOLID SWALLOW

1.awneeal

CP-EMF

Laryngeal Sensor Signals -

FIGURE 2. Laryngeal sensor signals (top traces in A and B) and CP-EMG activity during 3-mL semisolid swallowing. In (A) the integrated EMG traces are superimposed (second trace). Conventional EMG recordings are shown in (6) (second trace). Amplitude calibration: 200 pV in the second trace in (A); 100 pV in the second trace in (B). Time calibration: 200 ms in all traces.

nal deflection and was followed by the absolute EMG pause (second pause in Fig. 3). Finally, a rebound high frequency and high amplitude EMG activity coincident with or just preceding the onset of the downward laryngeal movement was recorded from the CP muscle.

The following is a chronological account of the electrophysiological events that take place during pharyngeal swallowing. First, the SM-EMG activity begins and attains its peak within 100-300 ms (Fig. 4C). The SM-EMG activity and partial activities of other muscles pull the larynx and the hyoid upward as signaled by the first deflection of the sensor (Fig. 4A). Upward movement of the larynx reaches its limit in about 100-120 ms and the larynx is held in that position for about 400-500 ms. During this period, tonic activity of the CP sphincter muscle is switched off (Fig. 4B). However, before the CP-EMG pause and at the beginning of laryngeal elevation there is a short burst of CP- EMG activity lasting about 200 ms. CP-EMG pause begins just after or at the beginning of the upward movement of the larynx and ends just before it starts its downward movement as demonstrated by the second deflection of the sensor signal. Some- times the CP-EMG activity is increased explosively as the larynx begins to move down. This second burst or rebound activity lasts as long as or even longer than 600-800 ms which exceeds the first burst. During the CP muscle EMG pause, SM-

EMG activity declines progressively until the laryngeal downward movement is signaled. SM-EMG activity lasts about 700-900 ms. But, the sequences of all these events are prone to variability and therefore they give only an estimate of the temporal course of the events. The 0-2 interval of the sensor signal, the duration of the CP-EMG pause, and SM-EMG activity vary with the type and volume of the material swallowed. Therefore, all the statistical results of this study should be viewed with the due caution to the type of material swallowed (Tables 1 and 2).

Effects of Bolus Type and Volume on the Pharyngeal Phase of Swallowing. Both bolus type and bolus volume affected the laryngeal movements as well as the submental and CP muscle EMG parameters; but the essential features of these activities remained the same. Liquid (water), semisolid (pudding), and solid (cookies) boluses affected different parameters of the pharyngeal swallowing responses in a given case, but statistical analyses did not reveal any significant differences among most of these responses, probably due to the large in- terindividual variability. Nevertheless, data from one group of subjects reveal that there is some relationship between bolus properties and measured parameters.

In swallowing semisolid boluses the effect of increasing the bolus volume is somewhat different. Larger semisolid boluses prolong the 0-2 interval as in the case of liquid boluses, but it is generally shorter than that observed with liquid boluses (mean 5 SEM: 462.5 + 32.7 ms for 3 mL semisolid versus 540.7 + 22.3 ms for 3 mL water, P < 0.05; calculated from the data of 8 subjects). Another important difference is that the swallowing jitter significantly decreases with larger semisolid boluses, implying that semisolid swallowing is more stable than liquid swallowing with respect to the time course of laryngeal movements.

Solid material swallowing also tends to affect the upward laryngeal movement. Larger solid boluses prolong the time course of laryngeal movements and the swallowing jitter tends to diminish as is the case with semisolid swallowing. In general, it was found that semisolid material swallowing is associated with the least jitter, shortest upward laryngeal movement time (0-2), and longest SM- EMG activity duration. All the sensor signal intervals appeared to be shorter for dry material swallowing than those for liquid and semisolid bolus swallowing, while the SM-EMG activity amplitude was sometimes higher for dry material swal-


Table 2. EMG findings of cricopharyngeus muscle during swallowing (mean 2 SEM).

EMG pause Second burst First burst

a-c' Rising phase

Type of No. of Duration Amplitude a-o interval c-1 interval ( ~ 4 ) s Duration Amplitude d-2 interval swallowing cases (a-c) (rns) (IN (ms)t (ms)S (ms) (ms) ( IN (ms)"

Dry swallowing 6 216.0 t 21.8 47.0 f 8.8 169.5 f 11.2 51.1 2 18.7 406.3 2 244 2 9 z 2 42.0 97.0 f 19.7 107.6 2 21.7 3 rnL water 9 231.5 t 26.3 48.5 f 7.1 167.4 f 26.5 63.82 12.8 481 1 f 30.8 254 6 2 27.3 90.2 t 12.2 38.6 ? 8.0 3 rnL semisolid 10 266.5 t 18.0 41.6 f 8.0 194.3 f 24 3 66.7 2 14.8 425.0 f 24.9 229.5 2 20.6 80.3 r 12.4 23.0 c 6.5 10 rnL water 5 239.5 f 85.7 24.2 2 7.8 194.8 t 65.7 70.4 r 19.7 505.6 -c 48.4 247.2 2 17 7 65.6 2 13.1 152.4 ? 57.8

"(ax) Denotes the onset and the end of the first burst of CP muscle. flnterval from the onset of the first burst of CP muscle to the onset of the first sensor deflexion indicating upward movement of larynx. Slnterval from the peak of the first sensor deflexion to the end of the first CP-EMG burst or the beginning of the EMG pause. §(c-d) Denotes the onset and the end points of the EMG pause of CP muscle. "lnterval from the onset of the second deflexion showing the downward movement of the larynx to the end of CP-EMG pause.

lowing than it was for other types of bolus swallowing. These observations are summarized in Table 3.

Case Reports. A 53-year-old man presented

with complaints of swallowing difficulties for the last 5 months. He had experienced nasal regurgi- tation and coughing upon drinking water 1 month before. He also had dysphonia and partial bleph- aroptosis. Neurological examination on admission revealed weakness in face and neck muscles, dysphonia, and dysphagia. But his palatal and gag reflexes were normal. EMG decremence test and single fiber EMG demonstrated a motor endplate transmission defect indicating myasthenia gravis.

Case 1 (N.A.).

3 mL WATER SWALLOW

2 Laryngeal Sensor Signals

- - I Pause 2 Pause

FIGURE 3. Lsyngeal sensor and the CP-EMG signal during swallowing 3 rnL of water. Top: The laryngeal sensor signal: middle: the CP-EMG pause: bottom: integrated EMG signals superposed. Note the earlier suppression of CP-EMG just before the first burst in the middle trace. Calibrations: 200 rns and 100 pV in all EMG traces.

Electrophysiological evaluation of the patient's dysphagia showed that the duration of upward laryngeal movement (0-2 interval) had increased significantly (i.e., 740 ms while swallowing 3 mL of water) and the swallowing jitter was beyond normal limits (384 ms) while the CP-muscle activity remained within normal levels (Fig. 5). It was con- cluded that the laryngeal elevator muscles were in- volved in the myasthenic process and this produced incoordination between SM and CP muscle activities resulting in dysphagia.

A 45-year-old woman presented with complaints of uncontrolled laughing and crying since 9 months prior to admission. Her speech and deglutition were disturbed for the last 6 months and she lost 5 kg of weight. Upon neurological examination she was found to have a na- salldysarthric speech with hypophonia. T h e tongue was wasted and fasciculated while the palatal and gag reflexes were augmented. Her oropharyngeal secretion was increased. Coughing was weak

Case 2 (M.K.).

3 mL WATER SWALLOW

A Laryngeal Sensor Signals

FIGURE 4. Laryngeal sensor signal (A), cricopharyngeal (B), and Subrnental (C) muscle EMG activities during swallowing of 3 rnL of water. EMG traces are the averages of five responses. Amplitude calibration: 100 pV and 20 pV in (B) and (C), respectively: time calibration: 200 rns in all traces.


Table 3. Effect of bolus type and volume on measured swallow oarameters.

SM-EMG activity Increasing bolus duration volume 0-2 Interval Swallow jitter (A-B interval)

Liquid t t .1 Semisolid t .1 t Solid t 4 t

and dysphagia was apparent. Mandibular reflex was brisk. Distal muscles of the upper limbs were found to be somewhat weak and fasciculations were observed throughout all the extremity muscles. Tendon reflexes were brisk. EMG showed dif- fuse anterior horn cell involvement indicating motor neuron disease.

Electrophysiological evaluation of patient’s dysphagia revealed that the duration of upward laryngeal movement (interval 0-2) was significantly prolonged (854 ms while swallowing 3 mL of water; Fig. 6). On the other hand, the duration of the CP-EMG pause was within normal limits demon- strating that the muscle was intact. Yet, such incoordination between the laryngeal elevators and the CP-sphincter muscle was one cause of the patient’s dysphagia.

A 53-year-old man presented with a history of acute stroke for several years. He eventually developed a gait disturbance of small steps. He was also experiencing involuntary laughing and crying from time to time. Dysphagia also appeared especially during drinking liquids. His oropharyngeal secretion was abnormally increased. Neurological examination revealed dysphonia, increased oropharyngeal secretions, slurred speech, and mild hemiparesis on the left side. Labial and tongue movements were slow and limited. Deep tendon reflexes were brisk in all four limbs. Mandibular tap and gag reflexes were also brisk. He was obliged to cough in almost every at- tempt to swallow water and he had wet voice following a swallow. CT-scan showed multiple lacu- nar infarctions.

Electrophysiological evaluation of the patient’s dysphagia revealed that both the duration of upward laryngeal movement and the duration of the CP-EMG pause were within normal limits. But some other abnormal findings were quite evident. For example, the SM-EMG rise time was very short and its amplitude exceeded normal limits. The CP- EMG bursts preceding and following the pause were extremely high in amplitude and some motor unit activity could be observed during the EMG

Case 3 (K.P.).

MYASTHENIA GRAVlS WITH DYSPHAGIA

DRY A

SEMISOLID

FIGURE 5. Laryngeal sensor signals (top traces in each pair) and CP-EMGs (lower traces in each pair) recorded from a myasthenia gravis patient with dysphagia. Sensor signals show that the larynx remains significantly longer in its upper position (Le., the 0-2 interval increases) when swallowing semisolids. Swallowing jitter is also quite enhanced, whereas the CP-muscle behavior remains normal. Amplitude calibration: 100 FV in all EMG traces; time calibration: 200 ms in all traces.

pause (Fig. 7). Interestingly, he had a very low swallow jitter. These findings were quite compati- ble with CP-sphincter muscle “hypertonia” accom- panied by a mild spasticity of the laryngeal elevator muscle.

DISCUSSION

During oropharyngeal swallowing, or the so-called “swallowing reflex” or “swallowing response,”* the following related events occur in the following or-

hyoid bone and the larynx move anteriorly; (3) CP- EMG activity is inhibited; (4) the upper esophageal sphincter begins to relax; and ( 5 ) the sphincter opens.

In the present study, we were able to evaluate some of the above components of the swallowing response. The upward movement of the larynx could be easily monitored by a piezoelectric sensor placed between the thyroid and the cricoid cartilages. The sensor yielded two deflections, the first

der2,3,5,15. . ( 1 ) the larynx moves upward; (2) The


3 rnL WATER SWALLOW

ALS WITH DYSPHAGIA

300 m,

FIGURE 6. Laryngeal sensor signals and CP-EMGs recorded from a normal subject (top two traces) and a patient with amyotrophic lateral sclerosis (ALS) having dysphagia (lower two traces) during swallowing of 3 mL of water. Laryngeal movement signals (upper trace in each pair) and the CP-EMG had no coordination in ALS. The 0-2 interval of the laryngeal sensor signal was significantly prolonged in spite of the very early resumption of CP-muscle activity (i.e., short pause in CP-EMG). Amplitude calibration: 100 pV and 50 pV for the EMG traces (second and fourth traces from the top); time calibration: 200 ms in all traces.

of which signaled the upward movement and upward position of the larynx during the 0-2 interval shown in Figure 1. The downward movement of the larynx marking the end of one swallow was signaled by the second, downward deflection of the sensor. That the above laryngeal movement signals correspond to the physiological steps of a swallow response is clearly demonstrated by the integrated EMG activity simultaneously recorded from the muscles that take part in the response. Submental EMG electrodes usually picked up the mixed activity of the striated muscles located sub- mandibularly (i.e., the mylohyoid-geniohyoid- anterior digastric muscle complex). These muscles fire concurrently to initiate the swallow Since they function as laryngeal elevators and the beginning of laryngeal elevation is considered as the first event defining pharyngeal ~wallowing,~*~*'' the temporal relationship observed between the SM-EMG activity and the first deflection of the laryngeal movement sensor signal clearly indicates the correspondence between the first deflection of the sensor signal and the upward movement of the larynx which is produced by the activity of the submental muscle complex. Additional evidence of this

SUPRABULBAR PALSY WITH DYSPHAGIA 3 mL water swallow

Laryngeal

sed

Laryngeal Sensor Signals

Averaged

Superimposed CP-FMG

1111 I

FIGURE 7. A case of suprabulbar palsy with dysphagia. Two bursts of CP-EMG activity with significantly increased ampli- tudes before and after its pause were recorded (second, fourth, and fifth traces from the top). Laryngeal movements were normal during deglutition and swallowing jitter was quite low. Some motor unit bursts were clearly visible during the of CP-EMG pause (bottom traces). Amplitude calibration: 100 pV, 70 pV, and 50 pV for the second, fourth, and fifth (EMG) traces respectively; time calibration: 200 ms in all traces.

correspondence is our recordings of the cricopharyngeal muscle sphincter activity. The activity of this muscle is sustained by motor units firing at high frequencies during rest, and is ceased during deglutition. Any bolus or dry swallowing produces a pause in the CP muscle EMG. T w o bursts of EMG activity appear in this muscle, one before and one after the pause. These findings have been also described in various previous studies conducted in animals 1.6,lO. 13,15,16 and in man.7*991s21

Our new observation in the present study is the suppression of activity that precedes the first burst in the CP-EMG. It is not observed in all cases or under all conditions but varies in extent from a fall in frequency to a complete cessation. This suppression is probably related to a voluntary effect or expectancy. Further studies are needed to explain this phenomenon.

The CP-EMG pause and the pre- and post- pause bursts can be easily explained. Many inves- tigators agree that CP muscle tonic activity and its pause during deglutition are closely associated with the upper esophageal sphincter relaxation and opening that are observed by manometric


techniques2,".' .' (see also the opposite view of Goyal et aL8).

During deglutition, CP muscle activity and the movement sensor signal also show a temporal correlation. The first burst of CP muscle activity ap- pears synchronously with the beginning of a swallow or just before it, while the first deflection of the movement sensor signal starts just before the EMG pause. The EMG pause, in turn, ends just before or during the second deflection of the movement sensor signal after which the CP activity bursts for a second time.

The above observations support our view that the first deflection of the motion sensor signals the upward movement of the larynx and that the second deflection is related to its downward movement. Furthermore, the period between the onsets of the two deflections (the 0-2 interval) gives in- formation about the period of the upward position of the larynx during deglutition. Indeed, the mean values (about 500 ms) of the 0-2 interval that we have measured closely agree with the results of manometric and barium swallow fluoroscopic s t ~ d i e s ~ . ' ~ conducted to determine this period.

The movement sensor signal and the 0-2 interval together with the SM-EMG and sometimes the CP-EMG allowed us to study the physiological mechanisms of deglutition. These measurements provide a simple and inexpensive method to evaluate dysphagia patients in an EMG laboratory. Movement sensor signals and EMG activities are affected by bolus properties. Increasing bolus size produces a significant increase in the 0-2 interval and CP-EMG pause and influences the rising time and duration of the SM-EMG activity. These findings agree with the results of fluoroscopic and manometric studies which have reported that increased bolus size is associated with progressive increases in the maximal superior hyoid and laryngeal movement^.^,' .'*12 Certain specific variables of swallowing are also affected by dry, liquid, and semisolid boluses. Bolus properties also effect the 0-2 interval length and the SM-EMG activity which are consistent with the results of a videofluoroscopic-manometric study.4 Such effects on deglutition and its physiological mechanisms, which we have determined in this study, will be published elsewhere in more detail.

The methods presented in this article are now being tested in different types of pathologies in- volving dysphagia. The three dysphagia cases reported above are examples of swallowing disorders having different origins, that is, myasthenia gravis, motor neuron disease, and suprabulbar palsy.

Electrophysiological evaluations of these patients reflect significantly different features in each case. Therefore, we conclude that electrophysiological evaluations of dysphagia can provide the clinician with the following opportunities:

Mild or modest dysphagia can be objectively evaluated and documented. It is even possible to disclose "silent dysphagia." The development and time course of dysphagia can be objectively and quantitatively followed-up. Electrophysiological quantities obtained from different kinds of dysphagia cases might help in the decisionmaking and man- agement of such cases.

REFERENCES

1 . Asoh R, Goyal RK: Manometry and electromyography of the upper esophageal sphincter in the opposum. Gastroen- terology 1978;74:51&520.

2. Cook IJ: Cricopharyngeal function and dysfunction. Dys- phagta 1993;8:244251.

3. Cook IJ, Dodds WJ, Dantas RO, Massey B, Kern MK, Lang IM, Brasseur JG, Hogan WJ: Opening mechanism of the h u m a n u p p e r esophageal sphincter . A m J Physiol 1989:257(Gastrointest Liver Physzol 20):G7484759.

4. Dantas RO, Kern MK, Massey BT, Dodds WJ, Kahrilas PJ, Brasseur JG, Cook IJ, Lang 1M: Effect of swallowed bolus variables on oral and pharyngeal phases of swallowing. Am J Physrol 1990;258(Gastroinkst Liver Physiol 21):G675-G681.

5.

6.

7.

8.

9.

10.

1 1 .

12.

13.

14.

15.

Dodds WJ, Stewart ET, Logemann JA: Physiology and radiology of the normal oral and pharyngeal phases of swallowing. AJR 1990; 354:953-963. Doty RW, Bosma JF: An electromyographic analysis of reflex deglutition. J Neurophyszol 1956; 19:44-60. Elidan J , Schochina M, Gonen B, Gay I: Manometric and electromyography of the pharyngeal muscles in patients with dysphagia. Arch Otolayngol Head Neck Surg 1990;116: 9 1 0-9 1 3. Goyal RJ, Martin SB, Shapiro J , Spechler SJ: The role of cricopharyngeus muscle in pharyngoesophageal disorders. Dysphagza 1 993 ; 8 : 252-258. Hellemans J, Vantrappen G , Janssen J : Electromyography of the esophagus, in Vantrappen G, Hellemans J (eds): Dzi- emes of the Esophagus. New York, Springer-Verlag, 1974, pp 270-285. Jacob P, Kahrilas PJ, Logemann JA, Shah V, Ha T: Upper esophageal sphincter opening and modulation during swallowing. Gastroenterology 1989;97: 1469-1478. Kahrilas PJ, Dodds WJ, Dent J , Logeman A, Shaker R: Upper esophageal sphincter function during deglutition. Gastroenterology 1 988 : 95 : 5 2-62, Kahrilas PJ, Logemann JA: Volume accommodation during swallowing. Dysphagza 1993:8:259-265. Kawasaki M, Ogura JH, Takenovchi S: Neurophysiologic observations of normal deglutition 11, its relations to allied phenomena. Laqngoscope 1964;74: 1766-l780. Kilman WJ, Goyal RK: Disorders of pharyngeal and upper esophageal sphincter motor function. Arch Intern Med 1976; 136:592-60 1 . Lang IM, Dantas RO, Cook IJ, Dodds WJ: Videoradio- graphic, manometric and electromyographic analysis of canine upper esophageal sphincter. Am J Physiol 1991; 260(Gastrointest Liuer Physiol 23):G9 1 1 4 9 19.


16. Levitt MN, Dedo HH, Ogura JH: The cricopharyngeus muscle, an electromyographic study in the dog. Laryngo- scope 1965;75: 122-136.

17. Logemann J: Evaluation and Treatment of Swallowing Disor- ders. Austin, TX, Pro-ed, 1983.

18. Murakami Y, Fukuda H, Kirchner JA: The cricopharyngeus muscle, an electrophysiological and neuropharmaco- logical study. A d a Otolaryngol 1972;31 l(supp1): 1-19.

19. Shipp T, Deatsch WW, Robertson K: Pharyngoesophageal

1186 Deglutition in Man

muscle activity during swallowing in man. Laryngoscope 1970;SO: 1-16.

20. Tanaka E, Palmer J , Siebens A: Bipolar suction electrodes for pharyngeal electromyography. Dysphagza 1986; 1 : 39-40.

21. Van Overbeek JJ, Wit HP, Paping RH, Segenhout HM: Simultaneous manometry and electromyography in the pharyngoesophageal segment. Laryngoscope 1985;95:582- 584.

MUSCLE & NERVE October 1995

An electrophysiological investigation of deglutition in man

Documents