A MANOMETRIC INVESTIGATION OF THE ALARYNGEAL SOUND SOURCE IN LARYNGECTONIZED SPEAKERS by

The location of the alaryngeal sound source sites were identified infour esophageal speakers by means of radiography and direct intraluminal manometry using miniature pressure transducers. These sound source sites were consistently found to be above the cricopharyngeal sphincter regions and not coincidental with them. The findings from this investigation do /lot support the generally held opinion that the cricopharyngeal sphincter subserves the function of an alaryngea/ sound source. It has been estimated that approximately two-thirds of all laryngectomized persons develop the ability to use alaryngeal speech, albeit with varying degrees of proficiency. However, despite the amount of information obtained over the past 60 to 70 years, relatively little is known about the basic physiology underlying this phonatory system. It is generally accepted that in esophageal voice production the laryngectomee insufflates his esophagus with air by either a sucking, pumping or injection technique. When eructated, this air generates a sound by causing some portion of the alimentary tract above the esophageal air reservoir to vibrate. The exact source of this sound generator remains uncertain. Various locations within the upper alimentary tract have been identified with the alaryngeal sound source at one time or another, among which are the following: (I) between the dorsum of the tongue and the vel urn; (2) between the posterior wall of the oropharynx and the base of the tongue; and (3) between portions of the inferior constrictor muscle (Kallen, 1934). It has even been suggested (Pellegrini, 1957) that no real pseudoglottis or vibrating structure exists. Rather, that the air column itself produces a sound when it passes through a constricted section of the alimentary tract (a phenomenon analogous to that which occurs in organ pipes). The most commonly held opinion, however. is that the location of the alaryngeal sound source in the majority of esophageal speakers lies within the region of the cricopharyngeal sphincter (Moolenaar-Bijl, 1953; Levin, 1962; Snidecor, 1971; Simpson, Smith and Gordon, 1972; Greene, 1974). Support for the inferred phonatory role of the cricopharyngeal sphincter has come from radiologic studies and observations of the structural characteristics of this sphincter. Although it is recognized that the cocopharyngeal sphincter opens reflexively during swallowing and vomiting, voluntary control of the sphincter is believed possible because of its striated muscle composition and its innervation (Putney. 1958; Seeman. 1958; van den Berg and Moolenaar-Bijl. 1959; Levin, 1962; Shipp. 1970; Greene. 1972. 1974). Levin (1962), for example. points out that in addition to its strategic location at the mouth of the esophagus. the cricopharyngeal sphincter can be' 'trained to act as a substitute for the vocal cords" because it contains striated muscle fibers which are innervated by the recurrent laryngeal nerve. Similarly, Greene (1972) accounts for the "extraordinary control and skill achieved by esophageal speakers in control of the pseudo-cords" on the basis that the upper portion of the esophagus is fomled by striated muscle which receives its innervation from the recurrent laryngeal nerve. ElIis (1971) and Sandmark (1973) point out, however. that there is no evidence that the recurrent laryngeal nerve supplies the cricopharyngeal sphincter since its function remains unaffected in persons with bilateral vocal fold paralysis. There is increasing evidence, on the other hand, that this sphincter may be under the control of the autonomic nervous system. 29 nU1Vlf\.1"j LV1VllVlU1"J.'--J\.J. ... ....,1'. UI. .I.,.u~'-". I.J'I.J'J even though these structures are composed of striated muscle (Lund, 1968; Ellis, 1971; Inglefinger, 1973; Sandmark, 1973). It has been logical to assume that great care should be taken during surgery to spare the cricopharyngeal sphincter and its innervation, thereby preserving its sound producing capabilities (van den Berg and Moolenaar-Bijl, 1959; Hunt. 1964; Brodnitz, 1971; Greene, • 1972). However, there is increasing physiologic evidence which suggests that a functionally intact sphincter may, in fact. be a deterrent to the acquisition of esophageal speech. The cricopharyngeal sphincter is in tonic contraction during its normal physiologic resting state and, as such, it generates a zone of elevated pressure at the pharyngoesophageal junction (Winans, 1972; Pope, 1974). This high pressure zone, which is from two to four centimeters in length, acts as a physiologic sphincter against gastric and esophageal regurgitation offood (except when vomiting) during conditions of increased abdominal pressure. In addition, this physiologic barrier also prevents air from getting into the esophagus during inspiration (Dey and Kirchner, 1961; Inglefinger, 1973; Winans et a!., 1974). This last described function of the cricopharyngeal sphincter is clearly incompatible with esophageal phonation, since getting air into the esophagus is precisely what the laryngectomee must do in order to set this sphincter into vibration. How, then, do laryngectomees overcome this elevated sphincter pressure to insufflate their esophagi? Dey and Kirchner (1961) and Winans et al. (1974) suggest that the surgically traumatized cricopharyngeal sphincter has less control over the passage of air into the esophagus. As a result, air can more easily be forced past the sphincter into the esophagus by increased oral and pharyngeal pressure. Physiologic support for this explanation has come from these investigators. Using infusion manometry, Winans et a!. (1974) demonstrated that mean resting cricopharyngeal sphincter pressures in laryngectomees (20.6 ± 2.8 mm.Hg) were significantly lower than in normal subjects (39.4 ± 4.2 mm.Hg). Furthermore, when they compared the mean resting sphincter pressures in good and poor esophageal speakers, they found that the good speakers had significantly lower pressures (13.1 ± l.4mm.Hg) than the poor speakers (29.6 ± 6.2 mm.Hg). In view of the above kinds of information, it would probably be accurate to say that we do not have a clear understanding about the vibratory or alaryngeal sound source in laryngectomized speakers. A major reason for this lack of knowledge has been the relative inaccessibility of the pharyngoesophageal region to direct physiologic investigation. There have been a number of attempts to study the esophageal phonatory mechanism directly by means of radiography (Diedrich and Youngstrom, 1966; Simpson et a!., 1972), electromyography (Shipp, 1970) and intraluminal balloon manometry (Salmon, 1965; Zinner and Fleshier, 1972; Weinhouse, 1973). The overall results of these studies remain equivocal, largely because of the limited physiologic perspective provided by these measu~ement techniques. ,; If we are to understand why some laryngectom~e~ are more proficienr than others as esophageal speakers, while still others are upable to develop this skill at all, it ~~ill be necessary to investigate the fundamental physidlogic factors underlying alaryngeaLphonation, i.e., the location and nature of the alaryngeal sound source. Measurement tec~iques must be used that can rapidly and accurately transduce tissue pressure changes whi~h are associated with phonation. The purpose of this study was to identify the locatio~of the alaryngeal sound source within the upper alimentary tract of four laryngectomized ~p~akers and to describe some of the characteristics of these vibratory sites using radiography and recently developed techniques of intraluminal ma,nometry. .' I." i , , METH6~ Subjects i Four laryngectomized individuals who met the fOllqting criteria were selected as su· this study: \ ~., 30 . ., t. ~1 . cts for \ GASEK: A MA NO METRIC INVESTIGATION OF THE ALARYNGEAL SOUND SOURCE IN LAR YNGECTOMIZED SPEAKERS (l) Subjects must not have had, in addition to their laryngectomies., pharyngectomies or extensive revisions of the cervical esophagus or tongue. (2) Subjects must have been able to phonate, on demand, successive repetitions of the syllable Ipal at a repetition rate of approximately one per second. In addition, they must have been able to sustain the vowel portion of this syllable for at least one second. The four subjects who were included in this study consisted of two females, ages 52 and 60, and two males, ages 48 and 62. The elapsed time between surgery and the present investigation for the two females was approximately five and six years, while for the males it was four and six years. All four subjects used esophageal speech as their primary means of communication.