Pharyngeal (Zenker's) diverticulum is a disorder of upper esophageal sphincter opening

GASTROENTEROLOGY

1992;103:1229-1235

Pharyngeal (Zenker’s) Diverticulum Is a Disorder of Upper Esophageal Sphincter Opening IAN J. COOK, MARY GABB, VOULA PANAGOPOULOS, GLYN G. JAMIESON, WYLIE J. DODDS, JOHN DENT, and DAVID J. SHEARMAN Departments of Medicine, Gastroenterology, Radiology, and Surgery, University of Adelaide, North Terrace, Adelaide, Australia; and Department of Radiology, Froedtert Memorial Lutheran Hospital, Milwaukee, Wisconsin

Pharyngeal coordination, sphincter opening, and flow pressures during swallowing were investigated in patients with pharyngeal (Zenker’s) diverticula. Fourteen patients with diverticula and 9 healthy age-matched controls were studied using simultaneous videoradiography and manometry. Pharyngeal and upper esophageal sphincter pressures were recorded by a perfused side hole/sleeve assembly. Temporal relationships among swallowing events, extent of sphincter opening during swallowing, and intrabolus pressure during bolus passage across the sphincter were measured. The timing among pharyngeal contraction and sphincter relaxation, opening, and closure did not differ between patients and controls. Sphincter opening was significantly reduced in patients compared with controls in sagittal (P = 0.0003) and transverse (P = 0.005) planes. Manometric sphincter relaxation was normal in patients. Intrabolus pressure was significantly greater in patients than in controls (P = 0.001). It is concluded that Zenker’s diverticulum is a disorder of diminished upper esophageal sphincter opening that is not caused by incoordination or failed pharyngosphincteric sphincter relaxation. Incomplete sphincter opening is likely to cause dysphagia. Increased hypopharyngeal pressures during swallowing are probably important in the pathogenesis of the diverticulum.

L

udlow first described the posterior pharyngoesophageal diverticulum in 1764.’ In 1878, Zenker and Ziemssen proposed that herniation of the pouch proximal to the cricopharyngeus muscle might be caused by high hypopharyngeal pressures, the actual site of herniation being determined by a zone of weakness in the posterior hypopharyngeal wall, Killian’s dehiscence.’ This proposal has remained unconfirmed because deglutitive hypopharyngeal pressures have never been shown to be elevated in these patients. There is a long-held belief

that abnormally high hypopharyngeal pressures could be caused by defective coordination of upper esophageal sphincter (UES) relaxation during pharyngeal propulsion, but consistent demonstration of such incoordination is lacking. The aims of this study were to investigate whether patients with Zenker’s diverticulum and dysphagia have abnormal coordination of pharyngeal and UES motor function and whether increased hypopharyngeal pressures exist during swallowing. Materials and Methods Patients and Controls We studied 14 consecutive patients (9 male, 5 female) with pharyngeal diverticula and dysphagia, none of whom had had previous surgical intervention for their problem. Control data were obtained from 9 healthy volunteer subjects (6 male, 3 female) of similar age who were recruited from a senior citizens’ association. None of the control subjects had dysphagia or any other significant medical illnesses. Ethical approval for the study was granted by the Royal Adelaide Hospital Ethics Review Committee in 1988, and all subjects gave written informed consent.

Videoradiography Patients and control subjects were studied seated. Images of swallows in the lateral and anteroposterior (AP) projections were obtained using 6-in and g-in Phillips image intensifiers (Phillips, Eindhoven, Holland). Fluoroscopic images were recorded on video tape at 25 frames/s by a VHS video recorder (Panasonic, Osaka, Japan) for later analysis. The correction factor for magnification was determined before each study by placing two metallic markers set at a known distance in the field of the image intensifier, above the subjects’s head but in the plane of the UES. Fluoroscopy exposure time was limited to a maximum of 2 minutes per subject, yielding an equivalent to0 1992 by the American

Gastroenterological 0016~5065/92/$3.96

Association

1230 COOK ET AL.

tal-body radiation dose of 4.8 mSv. Subjects swallowed in duplicate 2-, 5-, lo-, and 20-mL boluses of high-density barium suspension [250% (wt/vol), E-Z-HD; E-Z-EM, Inc., Westbury, NY] delivered into the mouth by syringe. Included in the field of view were the incisor teeth anteriorly, hard palate superiorly, cervical spine posteriorly, and proximal cervical esophagus inferiorly. Single swallows were also obtained for each of the four bolus volumes in the AP projection. A water-filled latex glove was held by the patient and positioned loosely against the skin under the chin to prevent flaring of the fluoroscopic image.

GASTROENTEROLOGY Vol. 103, No. 4

80 PHARYNX 0f 80

Manometry A g-lumen (ID of each lumen, 0.51 mm) silicon rubber/polyvinyl chloride (PVC) manometric catheter (OD, 5 mm) that incorporated a 6-cm sleeve assembly was positioned transnasally with the sleeve straddling the UES. An additional larger center lumen (ID, 1.2 mm) within the manometric catheter was used for passage of the assembly over a guide wire if required. If the catheter tip entered the diverticulum, it was withdrawn and a Teflon catheter with a mouldable tip (Wm. Cook, Sydney, Australia) was passed through the UES under fluoroscopic guidance. A flexible-tipped guide wire was then passed into the esophagus through the Teflon catheter, which was then withdrawn. The manometry catheter could then be passed into the esophagus over the guidewire. The sleeve sensor had a 5- X 3-mm oval configuration to maintain its anteroposterior orientation within the UES and incorporated a total of eight sideholes spaced at 1.5-cm intervals. Tantalum metallic markers (diameter, 0.6 mm; length, 5 mm) inserted into the catheter at each side hole permitted radiographic localization of each recording site. The four proximal side holes recorded pharyngeal pressures. The most distal pharyngeal side hole was situated at the proximal sleeve margin and was orientated posteriorly. Four more side holes were arranged along the sleeve length and aided in positioning the sleeve such that its midpoint was in the center of the resting UES high-pressure zone. With the sleeve positioned in this fashion, during the UES ascent on swallowing the proximal end of the sleeve was seen radiologically to be situated just above the proximal margin of the open sphincter. This could be confirmed manometrically because the appearance of the pressure recording from the side hole at the proximal sleeve margin was hypopharyngeal in character (Figure 1). All catheter lumina were perfused with degassed water by a low-compliance pneumohydraulic perfusion system at 0.6 mL/min. Pressures were measured by transducers (Deseret Medical Inc., Park Davis, UT) and recorded on a 12-channel polygraph (Grass Instrument Co., Quincy, MA), at a paper speed of 100 mm/ s. To avoid irritation from pharyngeal perfusion, the catheter was perfused only when swallows were being evaluated and the pump was turned off between swallows. A purpose-built video digital timer unit (Practel Sales International, Holden Hill, South Australia) imprinted simultaneously elapsed time on the video images in hundredths of seconds and a signal on the pressure tracing each whole second. This method allowed precise temporal correlation of video images with pressure.

0( 8Or SLEEVE WES) 0 t -------------------mm-’0.5 set ’ Figure 1. Manometric recording from a healthy subject during a 5-mL swallow. Al1 side holes are spaced at 1.5cm intervals. Pharyngeal closure (top arrow) was the radiologically determined point at which the tail of the bolus arrives at the side hole 4.5 cm above the mid-sphincter zone during UES opening and served as a temporal reference. Stippled segments represent the time intervals at which the bolus is in contact with the respective side hole. Intrabolus pressure (bottom arrow) is the low-amplitude pressure wave preceding the pharyngeal closure wave. The hatched bar represents the interval of UES opening.

Data Analysis Many of the radiographic and manometric measurements made have been described previously.3 Basal UES pressure and nadir UES pressure during deglutition were referenced to preswallow hypopharyngeal pressures. Basal UES pressure was calculated over a l-minute interval 10 minutes after catheter positioning to permit subject adaptation. Nadir UES pressure during swallowing was determined from “dry” swallows, which were swallows of residual saliva and barium after the initial swallows of 2mL barium boluses. Hypopharyngeal, midintrabolus pressure was defined as the pressure recorded by the side hole immediately proximal to the UES during transphincteric flow at the time point midway between arrival of the bolus head and departure of the bolus tail at that sidehole (Figures 1 and 2). Radiological time reference points also used for analysis were onset of the lingual stripping wave, which was the initial movement of the tongue tip against the maxillary incisors, and midpharyngeal closure, which was defined as the time of pharyngeal wall contraction on the tail of the bolus at the side hole 3 cm proximal to the most distal hypopharyngeal side hole. Fluoroscopy was used to measure UES opening and closure and duration of UES flow. UES opening was defined as the time point of first entry of the bolus head into the UES. Onset of UES relaxation was defined as the time point when the basal UES pressure began to decrease abruptly. Maximum UES relaxation was defined as the time point at which the UES relaxation profile ceased to decrease rapidly and leveled

PATHOPHYSIOLOGY

October 1992

mmHg 80 5 ml

Figure 2. Manometric tracing from the distal pharyngeal side hole during holus flow showing the correlation with videoradiographic images to illustrate the holus position at various points along the intrabolus pressure wave. The mid-point of the intrabolus pressure wave (2) corresponds to the pressure within the bolus as it is traversing the opened UES.

off, usually at or just before the nadir UES pressure. Because the proximal sleeve margin projects into the hypopharynx, the sleeve indicates the offset of UES relaxation slightly prematurely.4 Accordingly, termination of UES relaxation was measured from the side hole 1.5 cm distal to the proximal sleeve margin, which was seen fluoroscopitally to lie within the UES at the time of UES closure. Maximal UES diameters were determined in the transverse and sagittal planes. Hyoid movement was expressed as maximal anterosuperior hyoid excursion referenced to its preswallow resting position. The radiological reference point for laryngeal excursion with swallowing was the posterosuperior corner of the tracheal air column, which corresponds to the posterior-inferior margin of the true vocal cords. Maximal UES dimensions during sphincter opening were measured in the mid-sphincter zone, which was found to be 1.6 cm distal to the inferior margin of the true vocal cords. The axial pressure profile and its relationship to the inferior margin of the true vocal cords were determined by performing a slow pull-through with simultaneous recording of manometry and videoradiography as described previously.3 Subject mean data were first calculated from the duplicate values for each volume swallowed. In all cases, statistical inferences were made regarding the disease effect, swallowed bolus volume effect, and disease/volume interaction using a two-way analysis of variance (ANOVA) for repeated measures with mixed design. All values are represented as mean k SEM unless otherwise stated.

Results The mean age of patients 46-83 years) did not from that of the control subjects years). All patients complained years;

range,

(67 years; SD, 13 differ significantly (75 years;

of dysphagia.

SD, 10

The

OF ZENKER’S

DIVERTICULUM

1231

mean sagittal width of the pouch during 5-mL barium swallows was 14 mm (SD, 13 mm; range, 2-46 mm). The mean length of the diverticulum (measured from the inferior aspect of the neck of the diverticulum to its most inferior margin) was 20 mm (SD, 16 mm; range, 4-59 mm). Basal UES pressure did not differ significantly between patients (56.6 + 3.6 mm Hg) and controls (42.6 t 7.5 mm Hg). Neither the location of the UES high-pressure zone relative to the vocal cords nor sphincter length differed significantly between groups. In response to dry swallows, the UES in patients showed complete relaxation in that the nadir UES pressure of patients (1.9 + 0.2 mm Hg) did not differ from that of controls (0.8 + 0.5 mm Hg), The timing of UES relaxation, UES opening, and UES closure and the duration of UES opening did not differ between patients and controls (Table 1). Similarly, timing of the onset of the hyoid bone motion was comparable between groups. Hence, pharyngosphincteric coordination was normal in patients with Zenker’s diverticulum. Inspection of the pharyngeal pressure profiles of patients showed a prominent pressure ramp that preceded the pharyngeal closure wave. This pressure ramp, which corresponded to the intrabolus pressure described in controls, became much larger in response to larger swallowed bolus volumes (Figure 3). In some patients, when large (20 mL) boluses were swallowed the pharynx was incapable of propelling the entire bolus across the UES. In such circumstances, the intrabolus pressure exceeded the pharyngeal closure pressure, and the anterior and posterior pharyngeal walls separated prematurely. This resulted in failure of the aborally migrating stripping wave and retrograde flow of barium into the proximal pharynx (Figure 4). Group mean hypopharyngeal intrabolus pressures (Figure 5) were significantly greater in patients than in controls (P = 0.001). The major determinant of resistance to flow of a liquid through a tube is the diameter of the tube. For all volumes swallowed, maximal sphincter dimensions were reduced significantly in both the sagittal (P = 0.0003) and transverse (P = 0.005) planes (Table 2). The hyoid bone is believed to impart a distracting force on the UES and thereby facilitate UES opening.3,5 To determine whether incomplete UES opening was an intrinsic or extrinsic problem, we measured the extent of hyoid excursion in the anterosuperior direction. No differences in the timing or extent of hyoid excursion were observed between patients and controls. Anterior excursion of the larynx did not differ between patients and controls, being 1.2 k 0.1 cm and 1.1 + 0.1 cm, respec-

1232

GASTROENTEROLOGY Vol. 103. No. 4

COOK ET AL.

Table 1. Pharyngosphincteric

Coordination During Swallowing Statistical

inferences

(P)

Bolus volume effect

Bolus volume effect

Disease 2mL

Onset of anterior hyoid motion Controls Zenker’s Onset of UES relaxation Controls Zenker’s Complete UES relaxation Controls Zenker’s UES opening Controls Zenker’s Termination of UES relaxation Controls Zenker’s UES closure Controls Zenker’s

5 mL

10 mL

Disease/volume interaction

0.29 -+ 0.03 0.32 + 0.03

0.31 f 0.03 0.44 + 0.04

0.35 + 0.03 0.47 AZ0.05

0.36 k 0.05 0.43 f 0.03

NS

0.0006

NS

0.17 f 0.05 0.25 zlz0.05

0.22 f 0.06 0.33 + 0.05

0.29 k 0.04 0.33 + 0.06

0.39 I!I0.06 0.40 f 0.05

NS

0.0001

NS

0.02 + 0.04 0.05 + 0.03

0.06 ztz0.02 0.16 f 0.04

0.13 + 0.04 0.16 + 0.05

0.25 + 0.05 0.23 f 0.03

NS

0.0001

NS

0.01 * 0.03 0.01 & 0.02

0.09 f 0.04

0.16 ?I 0.03 0.17 + 0.03

0.22 + 0.03 0.17 * 0.02

NS

0.0001

NS

0.14 + 0.05

-0.44 -0.36

+ 0.03 + 0.03

-0.45 -0.29

f 0.04 ?z 0.05

-0.41 + 0.05 -0.38 + 0.05

-0.40 f 0.04 -0.45 f 0.09

NS

NS

0.03

-0.55 -0.52

If: 0.05 + 0.04

-0.57 -0.51

* 0.04 f 0.06

-0.56 k 0.03 -0.52 + 0.05

-0.54 k 0.03 -0.66 + 0.06

NS

NS

0.03

NOTE. Data represent means case, events are referenced to temporal relationships among and negative values to events

+ SEM. The timing of events culminating in sphincter opening and closure are shown in seconds. In each the onset of midpharyngeal closure at the recording site positioned 4.5 cm proximal to the open UES. The these events did not differ between patients and controls. Positive values relate to events occurring before occurring after midpharyngeal closure.

tively, for a lo-mL bolus. However, for 20-mL boluses, the mean anterior laryngeal movement (1.92 f 0.17 cm) was significantly greater than in controls (1.2 + 0.22 cm; P = 0.02). Discussion This study shows that the primary abnormality in patients with Zenker’s diverticula is one of incomplete UES opening. This UES dysfunction causes

NORMAL

20 mL

ZENKERS

0..

Figure 3. Manometric tracing during a lo-mL swallow in a patient showing much higher intrabolus pressure waves (stippled) than those seen in the control subject (lefi). Note that when the patient swallowed a second time to clear residual bolus from the pharynx, this lower-volume bolus was associated with a lower (but still abnormal) intrabolus pressure and a shorter duration of UES opening (bar).

a marked increase in hypopharyngeal intrabolus pressure during the phase of transsphincteric bolus flow. It is likely that occurrence of this relative high pressure before hypopharyngeal contraction causes pouch herniation. Contrary to widespread belief, our findings show no evidence of abnormal coordination between pharyngeal contraction and UES relaxation or opening. This is the first study to show that hypopharyngeal pressure is abnormally increased during bolus transit and supports Zenker’s original hypothesis that the pouch is in fact a pulsion diverticulum.’ Our findings implicate disordered UES function as the prime cause for the increase in hypopharyngeal pressure during swallowing. While the sphincter relaxes manometrically, the UES opens incompletely. The finding that the duration and coordination of UES opening are both normal and that the timing and extent of hyoid motion are normal suggests that the disorder is caused by diminished sphincter compliance rather than pharyngeal incoordination or diminished distraction on the UES. These findings favor a localized mechanical disturbance of the UES rather than neuromuscular incoordination in the pathogenesis of pharyngeal diverticula. The inability of the sphincter to open fully is likely to explain the dysphagia experienced by these individuals. The conclusive nature of the present study lies in

October

19%

mmng

60

I

0

-

Figure 4. Correlation of manometry and radiographic tracings from the same patient shown in Figure 3 during a larger (20-mL) bolus swallow. Note the very high intraholus pressure (lefty in continuity with the open sphincter and with the diverticulum. The pharyngeal contraction “fails” (center) as a consequence of extremely high intraholus pressure, resulting in pharyngeal contraction failure and retrograde holus escape. The patient swallowed again directly afterward (right) to clear most of the residual holus.

the use of a combination of manometry and dynamic radiography that permitted accurate measurement of the relevant pressures and the temporal relationships among oral, pharyngeal, and sphincteric events during swallowing and enabled us to define precisely the pressure acting in the region of the mouth of the diverticulum at the critical time interval when the swallowed bolus is traversing the opened UES. The manometric sleeve/side hole assembly we used can accommodate the axial mobility of the UES during swallowing and gives reliable measurements of basal UES pressure, the timing of UES relaxation onset, and the adequacy of UES relaxation during swallowing.4r6 The perfused pharyngeal side holes used in our study, unlike solid-state intraluminal transducers, underestimate the peak amplitude of pharyngeal contractions,7 but it was not the aim of this study to quantify pharyngeal contraction pressures. It is clear from inspection of the intrabolus pressure waveform (Figure 1) that it has a very low rate of increase so that accurate measurement is well within the capability of the manometric system used in this study. It has been widely believed that pharyngosphincteric incoordination is the cause of the proposed increased hypopharyngeal pressure during deglutition. Although various abnormalities of sphincter coordination have been described, this phenomenon has been poorly reproducible and the precise nature of the incoordination reported has been inconsistent among researchers. Premature closure of the sphincter,“-” in some cases associated with early re-

PATHOPHYSIOLOGY

OF ZENKER’S

DIVERTICULUM

1233

laxation,‘O~*l has been reported. However, the validity of early manometric observations must be questioned because UES pressures were recorded at a single focal recording site positioned within the sphincter. We now know that such methods can yield an apparent early onset of relaxation during orad sphincter movement, which causes the recording site to “drop out” of the UES before opening.4*‘2 Studies reporting early UES relaxation found in some cases that only 50% of swallows within each subject showed incoordination, and early relaxation was not reproduced by other workers.‘3-‘6 The variability of these earlier observations may well be attributable to methodological problems. It has been shown that increases in swallowed bolus volume cause earlier relaxation and opening of the UES, larger sphincter areas, and higher intrabolus pressures, although the timing of UES closure is unaltered.3*‘7 Swallowed bolus volume must therefore be controlled for; the lack of such control in earlier studies may account in part for some of the inconsistencies in the data on the timing of sphincter relaxation. Basal UES pressure has been variably reported to be 10w,‘~,‘~norma1,g~‘o~‘3and high” in these patients. However, basal UES pressure is not likely to be relevant to pathogenesis because it does not correlate with intrasphincteric pressure during swallow-induced relaxation. In the present study, basal UES pressure was comparable between patients and controls. Hence there is no evidence of so-called cricopharyngeal spasm as a factor in the pathogenesis of the pharyngeal pouch. Incomplete manometric relaxation of the UES has been reported by one author in two patients with Zenker’s diverticula,” but UES relaxation has been found to be complete by most investigatorsgs’3,‘6 and in the present study. The present study shows that the UES in patients with diverticula relaxes manometrically but opens incom-

40-l

)+

Controls

-

Zenkere )

30INTRABOLUS PRESSURE (mnW8

20-

+5

_ 0

10

20

BOLUS VOLUME (ml)

Figure 5. Group mean data showing intrabolus pressure for a range of swallowed bolus volumes (mean + SEM). Overall, pressures were significantly greater in patients than in controls (P = 0.001).

1234 COOK ET AL.

Table 2. Maximum

GASTROENTEROLOGY

UES Dimensions

During

Vol. 103,No. 4

Opening Statistical

inferences

(P)

Bolus volume 2 mL Sagittal [mm) Controls Zenker’s Transverse (mm) Controls Zenker’s

10 mL

5mL

Disease effect

20 mL

Bolus volume effect

Disease/volume interaction

7.8I!0.6 4.5f 0.5

9.5f 0.7 6.2+ 0.4

10.6f 0.8 6.4f 0.8

11.7f 0.8 7.1+ 1.1

0.0003

0.0001

NS

16 IL0.8 11.1f 0.8

16.5zk0.8 12.4f 1.3

17.5f 0.7 14 f 1.2

17.6+ 0.6 15 + 2.1

0.005

0.0004

NS

NOTE. Data represent means + SEM. Maximal sphincter patients compared with controls.

dimensions

pletely. Other pathogenetic theories such as multiple swallows against a closed sphincter,‘“~‘g delayed UES relaxation,” and increased sagittal mobility of the cricopharyngeus away from the prevertebral fascia” were not substantiated by the present study. The final extent of UES opening is determined by the opposing force of intrabolus pressure acting against the intrinsic stiffness or compliance of the UES. Our finding that UES opening is diminished while intrabolus pressure is increased indicates that the UES offers increased resistance to stretching by the distending forces within the bolus. Because with a standardized bolus volume the duration of UES opening in patients and controls is comparable, the calculated UES flow rates for patients did not differ from those of controls. Hence, to maintain a constant flow rate in the context of a reduced sphincter area, the intrabolus pressure must increase. The tonic neural stimulation of the cricopharyngeus muscle is normally inhibited transiently during UES relaxation.” Partial or complete failure of this inhibition of active cricopharyngeal muscle tone during transsphincteric flow might result in a functional loss of sphincter compliance. Alternatively, morphological abnormalities such as cricopharyngeal hypertrophy or fibrosis might render the muscle less able to distend during bolus passage independently of any abnormality of innervation. Diminished distraction forces by the suprahyoid muscles on the cricoid cartilage has been excluded as a cause of poor UES opening because hyoid bone excursion in these patients was normal. While the appropriate studies to address the question of failed inhibition of the cricopharyngeus are awaited, morphological studies on the cricopharyngeus muscle in these patients have shown structural changes including necrosis and fibrosis,23 indicating the likelihood of a structural rather than a functional sphincter disorder. Abnormal neural innervation or a disturbance in the patterned response generator in the medullary swallow centerz4 is unlikely to be relevant in the

in both sagittal and transverse

planes were significantly

reduced in

pathogenesis of this condition because coordination and the bolus volume-dependent temporal shifts in swallow events are both preserved. Although there is no doubt that UES opening is diminished, the possibility remains that extrinsic compression from the pouch itself may at least in part contribute to this. If pressure exerted by the pouch alone were responsible for incomplete sphincter opening, one would expect to see only a reduction in UES sagittal dimensions. Patients showed significantly reduced UES dimensions in both sagittal and transverse planes, consistent with a circumferential defect rather than a unidirectional extrinsic compressive effect. We have encountered a patient in whom pouch resection alone without fullthickness cricopharyngeal myotomy did not normalize sphincter opening or intrabolus pressure. Conversely, cricopharyngeal myotomy alone without resection of the diverticulum can normalize both UES opening and intrabolus pressure without influencing pouch dimensions (I. J. Cook, unpublished observations). A similar reduction in UES opening, also accompanied by elevated intrabolus pressure, is seen in dysphagic patients with cricopharyngeal bars without accompanying diverticula.25 Although it is tempting to speculate that the cricopharyngeal bar might in some instances be a precursor to diverticulum formation, there is no current evidence to confirm this. This study of a homogeneous group of patients with an isolated disorder of UES function emphasizes that loss of sphincter compliance is an important cause of dysphagia. These findings also show the potential usefulness of intrabolus pressure measurement as an indirect measure of sphincter compliance in studies of UES pathophysiology. References I. Ludlow A. A case of obstructed deglutition from a preternatural dilatation of and bag formed in the pharynx. Inquiries 1764;3:85-101.

Med Obs

October 1992

2. Zenker FA, Ziemssen H Von. Dilatations of the esophagus. In: Cyclopaedia of the practice of medicine, Volume 3. London: Low, Marston, Searle & Rivington, 1878:46-68. 3. Cook IJ, Dodds WJ, Dantas RO, Massey B, Kern MK, Lang IM, Brasseur JG, Hogan WJ. Opening mechanisms of the human upper esophageal sphincter. Am J Physiol 1989;20:G748G759. 4. Kahrilas PJ, Dodds WJ, Dent J, Logemann JA, Shaker R. Upper esophageal sphincter function during deglutition. Gastroenterology 1988;95:52-62. 5. Jacob P, Kahrilas PJ, Logemann JA, Shah V, Ha T. Upper esophageal sphincter opening and modulation during swallowing. Gastroenterology 1989;97:1469-1478. 6. Kahrilas PJ, Dent J, Dodds WJ, Hogan WJ, Arndorfer RC. A method for continuous monitoring of upper esophageal sphincter pressure. Dig Dis Sci 1987;32:121-128. 7. Dodds WJ, Hogan WJ, Lydon SB, Stewart ET, Stef JJ, Arndorfer RC. Quantitation of pharyngeal motor function in normal human subjects. J Appl Physiol 1975;39:692-696. 8. Ardran GM, Kemp FH, Lund WS. The aetiology of the posterior pharyngeal diverticulum: a cineradiographic study. J Laryngol Otol 1964;78:333-349. 9. Ellis FH, Schlegal JF, Lynch VP, Payne WS. Cricopharyngeal myotomy for pharyngoesophageal diverticulum. Ann Surg 1969;170:340-349. 10. Lichter I. Motor disorder in pharyngoesophageal pouch. J Thorac Cardiovasc Surg 1978;76:272-275. 11.Hurwitz AL, Nelson JA, Haddad JK. Oropharyngeal dysphagia: manometric and tine esophagraphic findings. Dig Dis 1975;20:313-324. 12.Isberg A, Nilsson ME, Schiratzki H. Movement of the upper esophageal sphincter and a manometric device during deglutition: a cineradiographic investigation. Acta Radio1 Diagn 1985;26:381-388. 13.Kodicek JM, Creamer B. A study of pharyngeal pouches. J Laryngol Otol 1961;75:406-411. 14.Duranceau A, Rheault MJ, Jamieson GG. Physiological response to cricopharyngeal myotomy and diverticulum suspension. Surgery 1983;94:655-662. 15. Alstrup P, Pederson SA, Hansen JB. Pharyngoesophageal diverticula. In: Sorensen HR, Japsen 0, Pedersen SA, eds. The

PATHOPHYSIOLOGY

OF ZENKER’S DIVERTICULUM

1235

function of the esophagus. Odense, Denmark: Odense University, 1973:60-63. 16. Knuff TE, Benjamin SB, Caste11 DO. Pharyngeosophageal (Zenker’s) diverticulum: a reappraisal. Gastroenterology 1982;82:734-736. 17. Cook IJ, Dodds WJ, Dantas RO, Kern MK, Massey BT, Shaker R, Hogan WJ. Timing of videofluoroscopic, manometric events, and bolus transit during the oral and pharyngeal phases of swallowing. Dysphagia 1989;4:8-15. 18. Hunt PS, Connell AM, Smiley TB. The cricopharyngeal sphincter in gastric reflux. Gut 1970;11:303-306. 19. Wilson CP. Pharyngeal diverticula, their cause and treatment. J Laryngol Otol 1962;76:151-180. 20. Cross FS, Johnson GF, Gerein AN. Esophageal diverticula: associated neuromuscular changes in the esophagus. Arch Surg 1961;3:525-533. 21. Dohlman G, Mattsson 0. The role of the cricopharyngeal muscle in cases of hypopharyngeal diverticula. AJR 1959;81:561-569. analysis of reflex 22. Doty RW, Bosma JF. An electromyographic deglutition. J Neurophysiol 1956;19:44-60. 23. Cook IJ, Blumbergs P, Cash K, Jamieson GG, Shearman DJ. Structural abnormalities of the cricopharyngeus muscle in patients with pharyngeal (Zenker’s) diverticulum. J Gastroenterol Hepatol 1992 [in press). 24. Miller AJ. Deglutition. Physiol Rev 1982;62:129-184. 25. Dantas RO, Cook IJ, Dodds WJ, Kern MK, Lang IM, Brasseur JG. Biomechanics of cricopharyngeal bars. Gastroenterology 1990;99:1269-1274.

Received January 9, 1992. Accepted April 29,1992. Address requests for reprints to: Ian J. Cook, M.D., Gastroenterology Department, St. George Hospital, Sydney, Australia 2217. Supported by the Gwendolyn Michell Research Fellowship (University of Adelaide), the National Health and Medical Research Council of Australia, and the Royal Adelaide Hospital Research Fund. Also supported in part by National Institutes of Health grants ROl-DK25731 and ROl-DC00669 (to W.J.D.). This study would not have been possible without the commitment of the members of the Mitcham Senior Citizens Association, Adelaide, who volunteered to provide data on healthy agematched controls,