- Email: [email protected]
Abnormal sensory perception in patients with esophageal chest pain
GASTROENTEROLOGY
1986;91:845-52
Abnormal Sensory Perception in Patients With Esophageal Chest Pain JOEL E. RICHTER, CHARLES F. BARISH, and DONALD 0. CASTELL Bowman Gray School of Medicine on Gastroenterology, Winston-Salem,
of Wake Forest University, North Carolina
Many cases of recurrent noncardiac chest pain are considered to be of esophageal origin. The mechanism of this pain is poorly understood and only rarely appears related to abnormal esophageal contractions observed during manometry. We studied the response to intraesophageal balloon distention in 30 patients with noncardiac chest pain (mean age 52 yr) and 30 controls (mean age 41 yr). A polyvinyl balloon (length 30 mm; maximum diameter after IO-ml distention, 25 mm] was positioned 10 cm above the lower esophageal sphincter and inflated with progressively larger volumes of air (each increase = 1 ml). Chest pain was correlated with balloon volumes and pressures, esophageal contractions 5 cm above the balloon, and electrocardiogram changes. Results were as follows: 18 of 30 (60%) patients and 6 of 30 (20%) controls (p < 0.005) experienced chest pain. Electrocardiogram changes ofischemia were not seen. Neither balloon pressures nor esophageal contractions above the balloon differed significantly during pain in the controls or patients. As assessed by pressure-volume curves, esophageal tone was also similar in all subjects regardless of pain production. Patients were more sensitive to smaller volumes of balloon distention; 15 of 18 patients had chest pain at 58 ml volume while controls only noted chest pain at 29 ml Received Jund 7, 1985. Accepted March 27, 1986. Address reqhests for reprints to: Joel E. Richter, M.D., Section on Gastroenterology, Bowman Gray School of Medicine, 300 South Hawthorne Road, Winston-Salem, North Carolina 27103. C. F. Barish’s present address is: Wake Internal Medicine Consultants, 3100 Blue Ridge Road, Suite 300, Raleigh, North Carolina 27612. This work was supported, in part, by grant AM34200-OlAl from the National Institutes of Health. The authors thank John N. Blackwell, M.D., and W. J. Dodds, M.D., for their valuable technical assistance; David Ott, MD., for assistance with the radiographic studies; Chris Dalton, PA-C, for her expertise in the manometry laboratory; and Karen Chatman and Linda Brown for their assistance in the preparation of this manuscript. 0 1986 by the American Gastroenterological Association 0016-5085/86/$3.50
Department
of Medicine,
Section
volume. We therefore arrived at the following conclusions: [a) esophageal balloon distention may provide a nonpharmacologic provocative test for esophageal chest pain; and [b] the mechanism of chest pain in these patients may be related to lower pain threshold to balloon distention, which is independent of esophageal contractions.
The potential for the esophagus to be the site of recurring chest pain has received considerable attention in recent years. The specific mechanisms by which abnormal esophageal function might produce chest pain have not been clearly identified. At least two possible etiologies have been proposed: esophageal mucosal irritation secondary to gastroesophageal reflux, or strong esophageal contractions as suggested in patients with diffuse esophageal spasm. The development of provocative testing techniques has provided some insights and support for the importance of these two mechanisms in the development of esophageal chest pain. Acid perfusion into the distal esophagus (Bernstein test) has been used for many years as a means to reproduce symptoms in patients in whom mucosal acid sensitivity was the proposed origin of their complaints (1).In more recent times, stimulation of forceful esophageal contractions with drugs such as edrophonium (z-4), bethanechol (Z), and ergonovine (4-7) has also been used in an attempt to establish a relationship between abnormal esophageal contractions and chest pain. In the present study we have evaluated the response to graded intraesophageal balloon distention with small volumes of air in an attempt to further elucidate mechanisms of pain production. These studies have indicated the possibility of a third component to the production of pain in these patients: that is, the presence of a lowered visceral pain threshold to esophageal distention.
846 RICHTER ET AL.
GASTROENTEROLOGY Vol. 91, No.
Materials and Methods Preliminary
Balloon
Studies
A variety of experimental balloons were made from a number of materials including different sized Penrose drains, rubber contraceptive devices, portions of rubber gloves, and a standard finger cot. The latter material was found to be most suitable because it easily expanded at low intraesophageal balloon pressures. In vitro studies with rabbit and dog esophagi and in vivo studies in Patas monkeys revealed that this type of balloon would adequately distend the esophagus without producing macroscopic evidence of mucosal or muscular disruption. The polyvinyl balloon finally developed for use in our human studies was constructed from a finger cot (Ansell Inc., Dothan, Ala.) and attached to a standard manometric catheter using silicone glue and 5.0 nylon suture. At IO-ml distention, the balloon measured 30 mm in length and 25 mm in diameter. The catheter orifice 10 cm proximal to the distal orifice opened into the balloon and was perfused only with air. The use of a soft, elastic balloon gave us a large safety margin but raised the possibility that increased esophageal resistance might distort and possibly elongate the balloon. Ideally, all human studies would need to be done under fluoroscopy, but this was neither practical nor ethically possible. Therefore, an in vitro model, using cardboard cylinders of various diameters, was used to test the characteristics of the balloon during graded distention. The proximal tip of the capillary tube opening into the balloon was attached to a three-way stopcock, which was connected to an external transducer with output on a Beckman recorder (Beckman Instruments, Inc., Fullerton, Calif.). Air could then be infused by hand syringe into the balloon at l-ml volume increments and intraballoon pressures recorded by the physiograph. Initially, intraballoon pressures were measured in room air (23°C) up to a maximum volume of 10 ml. Studies were then repeated after placing the balloon in the center of 20-cm-long cardboard cylinders of decreasing diameter (25 mm, 23 mm, 20 mm). All studies were performed twice, and each value was recorded as the average of the two pressure measurements obtained at a specific balloon volume. These values were later compared with intraesophageal pressure-volume measurements during in vivo studies in controls and chest pain patients. Patients
and
Healthy
Volunteers
Thirty patients (mean age 52 yr, range 35-67 yr; 15 men, 15 women) considered by their referring cardiologists to have noncardiac chest pain were studied. Prior cardiac catheterization had revealed normal coronary arteries in 20 patients and nonobstructive coronary artery disease (525% coronary artery diameter narrowing) in 10 patients. Ergonovine was given to 2 patients during cardiac catheterization without inducing chest pain or focal coronary artery spasm. Only those patients who had chest pain predominately at rest received ergonovine, because the possibility of coronary artery spasm was highest in this group (8). Nineteen of the 30 patients (63%) had previous
4
gastrointestinal evaluations (upper gastrointestinal series or endoscopy, or both), which were normal in 10 patients, showed hiatal hernia in 7, and revealed esophagitis in 2. Nine (30%) chest pain patients had abnormal baseline esophageal manometry: 8 patients with the “nutcracker esophagus” (mean distal amplitude 2180 mmHg) and 1 patient with diffuse esophageal spasm. All other patients had normal esophageal motility. These 30 patients were compared with 30 healthy volunteers (mean age 41 yr, range 22-60 yr; 18 men, 12 women) with no history of chest pain. Volunteers were carefully interviewed and excluded if they had any history of frequent heartburn (more than once a month), dysphagia, regurgitation, or antacid use. Additionally, no volunteer had a history of diabetes mellitus, alcoholism, collagen vascular disorders, or neurologic diseases, All volunteers had normal esophageal manometric studies. Written informed consent was obtained from all participants, and the study was approved on February 23, 1984, by the Human Research Review Committee of the Bowman Gray School of Medicine, Winston-Salem, North Carolina. Balloon
Studies
in Humans
Esophageal manometric studies were performed using a round, 8-lumen polyvinyl catheter (diameter 4.5 mm, internal diameter 0.8 mm, Arndorfer Specialties, Inc., Greendale, Wis.). The distal four openings were located at the same level 90” apart, and the four proximal openings were spaced at 5-cm intervals. Each lumen, except for the one associated with the balloon, was continuously perfused with distilled water at a rate of 0.5 ml/min using a low compliance pneumohydraulic capillary infusion system (Arndorfer Specialties, Inc.). The lumen opening into the balloon was filled with air. The catheter was connected to external transducers (Beckman, model 4-327-C, Norcross, Ga.), with output on a Beckman Recorder (model R-612). The manometric system has a pressure rise rate of >250 mmHg/s. Following standard manometry studies, esophageal balloon distention was performed in both chest pain patients and volunteers. The balloon was positioned 10 cm above the lower esophageal sphincter. This permitted recordings of esophageal pressures at the level of the sphincter as well as 5 and 15 cm above the lower esophageal sphincter. The subjects were then given written instructions to hold telling them how to respond to sensations noted during testing. They were given a choice of sensations to pick from (tugging, discomfort, tightness, pressure, or pain) and were advised that they might feel something different or nothing at all. In an attempt to minimize all possibilities of influencing the subjects’ responses, there was no subsequent conversation between the investigator and subjects during the study. During balloon distention, subjects were permitted to dry swallow as necessary. The esophageal balloon was distended in a stepwise fashion using progressively larger volumes of air (each increase = 1 ml) to a total volume of 10 ml. The air was injected rapidly (within 2 s) by syringe and each distention volume was maintained for 10 s. Between each incremental volume increase, the balloon was completely emptied
October
MECHANISM
1986
for 5 s. This procedure was done to prevent the esophagus from accommodating to progressive, gradual balloon distention. Unknown to the subjects, a placebo-sham distention was also performed before the actual inflation. A positive response to balloon distention was recorded only when the patients experienced reproduction of their typical chest pain or the volunteers complained of chest pain. Any other sensation was not considered a positive test. If pain was produced by balloon distention, it was localized and correlated with balloon volumes, balloon pressures, and esophageal contraction waves (amplitude and duration) above the balloon. Patients and volunteers not experiencing chest pain had similar parameters measured at maximal balloon distention (lo-ml volume). The reproducibility of pain threshold was examined in the last 15 chest pain patients by repeating the above sequence. Electrocardiographic monitoring was continuously recorded during the balloon distention studies. Two control subjects and 2 patients also had limited balloon distention studies performed during fluoroscopy to further address the possibility that increased esophageal resistance might distort and elongate the balloon. As previously described, the balloon was positioned 10 cm above the lower esophageal sphincter. In the upright position, subjects drank a small amount of high-density barium sulfate suspension to coat the esophageal mucosal surface. Balloon distention studies at volume increments of 6, 8, and10 ml and simultaneous left posterior oblique radiographs were done. The diameter of the balloon with each incremental inflation was determined by correcting for magnification using the 4.5-mm-diameter catheter as a known reference. These values were compared with similar measurements obtained from radiographs of the balloon alone taken after placing it in room temperature (23°C) water and distending with 6-, 8-, and lo-ml volumes of air. Statistical
Analysis
Statistical analysis was performed with the x2 test and Student’s t-test for paired and unpaired samples as appropriate. Results are expressed as mean ? SEM. Unless otherwise specified, all levels of significance have a probability value
Results Human
Studies
Eighteen of 30 (60%) patients experienced chest pain with balloon distention. In contrast, only 6 of the 30 (20%) normal volunteers had any chest pain. No pain was noted in either patients or volunteers with sham distention. This relationship between a history of noncardiac chest pain and pain with balloon distention was highly significant (p < 0.005; x2 analysis). When pain was reproduced, it was localized to the mid or lower sternal region in all patients. Once started, the pain intensified with further increments of volume and was immediately relieved by balloon decompression. In no patients
OF ESOPHAGEAL
CHEST
PAIN
847
loo-
$0. 60.
CUMULATIVE % POSITIVE
ii. 50. 40. 30. F...
20. 10. Pallsnls Normals -.
..
7
. 1
2
r 3
. 4
BALLOON
Figure
I
1 5
1 6
VOLUME
,..a
K 7
8
9
10
(CC)
Esophageal pain with balloon distention. Sixty percent of chest pain patients (solid line] experienced pain, compared to only 20% of normals (broken line). The difference is highly significant (p -C 0.005). Not only do the patients more frequently develop pain with balloon distention, but their pain occurs at smaller volumes, suggesting a lower pain threshold to esophageal distention.
having pain were electrocardiogram changes of ischemia noted. An abnormal baseline manometric study was not associated with a higher frequency of balloon-induced chest pain. Only 4 of 9 (44%) patients, all with the “nutcracker esophagus,” had ballooninduced pain. Intraesophageal balloon volume at the onset of pain also distinguished the patient and control groups, as noted in Figure 1. Fifteen of the 18 chest pain patients experiencing pain noted it at a volume of r8 ml of air. All control subjects with pain did not note their symptoms until at least 9 ml of air distention. The association between a history of noncardiac chest pain and provoked pain at a volume of 18 ml was also highly significant (p < 0.001; x2 analysis]. Reproducibility of balloon distention was tested in the last 15 patients by repeating the study 5 min later. The initial balloon distention replicated the patient’s chest pain in 9 subjects, and all experienced pain during the second study at the same volume or within 1 ml. Figure 2 shows the curves illustrating the relationship between intraesophageal balloon pressure and volume in patients and normal subjects. These curves are compared with the mean pressure-volume curve generated by inflating the balloon on three separate occasions in room air at 23°C. There was no significant difference between the curves obtained in patients and that found in the healthy controls. Furthermore, no difference in pressure-volume curves was found between patients in whom pain occurred during balloon distention and those without pain. All four curves do not begin to diverge until 5-ml volume, at which time it is assumed that the balloon begins to occlude the esophagus (diameter at 5-ml volume during distention in air = 20 mm). The difference observed between the three curves obtained in patients and controls and the
848
RICHTER ET AL.
GASTROENTEROLOGY Vol. 91, No. 4
BALLOON PRESSURE ImmHg)
--
Pls with
pak (16)
-
Balloon In slf
BALLOON
Figure
VOLUME (CC)
2. Balloon pressure-volume curves in the three study groups compared to balloon distention in air. Each point represents the mean balloon pressures at each l-ml increment in balloon volume. There is no significant difference between the curves obtained in patients with or without pain and the normal controls. The significant difference (* = p < 0.05) seen between the patient curves and the balloon-in-air curve represents the compliance of the esophagus.
balloon-in-air curve beyond 5-ml volume represents the effective elastic modulus or compliance of the esophagus. As this measure of esophageal resistance is similar among patient and control groups, it cannot account for the patient’s esophageal chest pain. Intraesophageal balloon pressures were measured at the onset of pain or at maximal distention (10 ml) if no pain developed. The chest pain patients who experienced pain during testing generally showed lower intraesophageal balloon pressures at the onset of pain than did the healthy volunteers who experienced pain (mean pressures: 86 k 4 mmHg vs. 103 + 5 mmHg; p < 0.05, Figure 3), but there was some overlap between groups. This result was somewhat unexpected but could have been predicted by the similar pressure-volume relationship among the three groups, as pain generally occurred at smaller distention volumes in the patients as compared with controls. Chest pain patients and controls who did not experience pain with balloon distention had similar mean intraesophageal balloon pressures at lo-ml maximal distention (89 + 5 mmHg vs. 90 k 4 mmHg) (Figure 3). These data suggest that the esophagi are similar in their elasticity in all study groups, and they do not support the hypothesis of increased esophageal wall tone or “spasm” as the cause of our patients’ chest pains. Figure 4 illustrates a typical esophageal motility tracing obtained during our studies. Both patients and normal subjects experienced increased esophageal contraction activity above the balloon at higher distention volumes. These high-amplitude waves of prolonged duration and with frequent repetitive contractions have been called the “esophageal propulsive force” (9). Because it has been suggested that
“spastic” esophageal contractions proximal to a functional distal obstruction might be the source of esophageal chest pain (lo), we analyzed the characteristics of the esophageal propulsive force associated with the onset of pain, or at lo-ml volume if no pain developed. The maximal individual amplitudes and durations of the esophageal propulsive forces in patients with and without pain and in healthy controls are shown in Figure 5. Mean amplitude and duration was 64 k 8 mmHg and 6.2 t 0.7 s in the patients with pain compared to 55 + 11 mmHg and 5.7 + 0.8 s in the patients without pain, and 59 t 6 mmHg and 5.5 k 0.5 s in the control subjects. There was no significant difference between mean amplitudes or durations of the esophageal propulsive forces among the three groups. Effects of Volume Increments on Balloon Diameter-In Vitro and In Vivo Results Although esophageal compliance was similar in all three subgroups, the soft, elastic characteristics of our balloon still raised the possibility that individual variation in esophageal distention resulting from balloon distortion might account for the patients’ chest pain. Practical and ethical reasons prevented us from investigating all subjects during fluoroscopy, but the following in vivo and in vitro data suggest this was not the cause of our patients’ chest pain. The balloon pressure-volume curve in the 25mmdiameter cylinder was similar to the balloon-in-air curve (Figure 6). This was anticipated because the diameter of the balloon-in-air measured 25 mm when inflated maximally (10 ml). The characteristics
PnEBsLmE WnNB) (lOed
YTRAB*uooN
WllWOUT PAIN
...
130 120 110
b
;
..... .
70 60 60 10 30 20 10
P51lWl~S
EFigure
:
...
B =*
B. ..
P5tI*nts
Normals
70
60 50 40 30 20 10
:.
..
Norm5lr
ii fSE
I 3
Intraesophageal balloon pressure (mmHg) with and without pain. Left: at the onset of pain, patients with pain have a significantly lower [p < 0.05) mean intraesophageal balloon pressure than controls with pain. Right: patients and normals without pain have similar intraesophageal balloon pressures at lo-ml inflation.
October
MECHANISM
1986
Figure
IOCm
(balloon)
LES
of the balloon pressure-volume curves in the cylinders did not markedly deviate from the mean curve obtained in the patient groups, with and without pain, until cylinder diameters well below 25 mm were tested, i.e., 23 and 20 mm (Figure 6). Analysis of individual pressures for all study subjects at each balloon volume revealed that no pressures fell above the curve described for the 20-mm-diameter cylinder. Nine patients (30%), 5 with pain and 4 without pain, and 10 control subjects (33%) had at least one intraballoon pressure exceeding the curve for the 2%mm cylinder. Therefore, this in vitro model would predict that the maximal decrease in balloon diameter during in vivo studies should have been <2 mm (or 8% of 25 mm) in 41 of 60 (68%) study subjects. Five patients would PROXIYAL DURATIONS (SECONDS)
PROXIMAL AMPLITUDES 0IMl2) 120
-.*
110 100
... .
*
80
.. .
:p,
@J
50
.. ..
40
:.
30
‘:’
20
::.
.
e ..
10 NOFM*lS
?? =~~sE
16
s
.
. .
11
..
90
Figure
*. . .
.. El .. .
pAsYs ‘wATZ’ ? PAY PAP4
,O
.. ..
2 6
...
7
.
:
B
4
.. . . . . . . . . . . .
3
*
2
.. . wol))rlALI
. ...
Q .. .. .. . . zY= PIN
. .
El . .... ‘%z PAW
5. Individual maximal amplitudes (mmHg) and durations [seconds) of the esophageal propulsive force measured 5 cm above the balloon. There is no significant difference between mean amplitudes and durations in patients with or without pain and in normal volunteers.
OF ESOPHAGEAL
4
CHEST
PAIN
849
Example of one of the motility tracings obtained during our studies. The esophageal balloon is positioned 10 cm above the lower esophageal sphincter (LES). The lowest orifice is at the upper level of the LES. The other orifices are located 5 cm above the LES, and 5 cm proximal to the balloon (15 cm above the LES). The high-amplitude and prolonged duration waves above the balloon indicate the esophageal propulsive force (EPF). In the balloon, changes in pressure associated with l-ml volume increases can be seen. Between incremental inflations the balloon is emptied of air. The pressures do not appear to drop back to baseline on the tracing, however, because a stopcock closes the catheter to the transducer during deflation.
be predicted to have had a decrease in balloon diameter between 2 mm and 5 mm (8%-20%) at the onset of their pain. However, a similar diameter change was observed in 4 patients without pain and in 10 control subjects (8 without pain), confirming that increased esophageal resistance was not the cause of these patients’ chest pain. Our limited in vivo studies supported these in vitro results. One control subject, experiencing no chest pain at IO-ml balloon volume, had radiographic evidence of only a 1.6% decrease in balloon diameter. The second control subject noted chest pain at IO-ml balloon volume with an associated 3.4% decrease in balloon diameter. Both patients experienced chest pain at 8-ml balloon volume, at which time balloon diameter had decreased 3.0%~ and 8.4%. These x-ray studies revealed that, rather than narrowing and elongating with increasing volumes, the balloon tended to square off.
Discussion The precise explanation of the origin of pain in esophageal diseases has been an elusive problem. Commonly proposed mechanisms have included stimulation of acid-sensitive chemoreceptors in the esophageal mucosa or changes in esophageal wall tension stimulating mechanoreceptors. The symptoms of heartburn and chest pain in gastroesophageal reflux disease primarily appear to be the result of acid stimulation of chemoreceptors (I). Although acid infusion may occasionally provoke esophageal motility abnormalities as shown by Siegel and Hendrix (ll),more recent studies suggest motility changes are an uncommon cause of acid-induced symptoms (12,131.
850
RICHTER ET AL.
GASTROENTEROLOGY Vol. 91, No. 4
Cylinder
( 20
mm
)
, Cylinder
( 23
mm
)
180 #’ 160
_
/
140
I”
Figure
120 E
? :, 0 t =: m”
100
80
60 40
1
2
3
4 Balloon
5
6 Volume
7
6
6
6. Pressure-volume characteristics of the study balloon under different test conditions. For each curve the balloon was inflated stepwise from 1 to 10 ml. Pressure-volume curves are shown for the balloon inflated in air and inflated within a cardboard cylinder of diameters 25 mm, 23 mm, and 20 mm. Also shown is the mean curve obtained from the balloon distention inside the esophagus of study patients.
lo
( cc)
Chest pain in patients with esophageal motility disorders, either occurring spontaneously or in response to pharmacologic provocation, may be associated with esophageal contractions of increased amplitude, prolonged duration, and/or simultaneous activity along several recording sites (14-16). Diffuse esophageal spasm has been the traditional prototype of these painful esophageal motility disorders. More recently, however, it has been recognized that hypertensive peristaltic contractions (“nutcracker esophaare the most common motilgus, ” “supersqueezer”) ity abnormality found in patients with a history of esophageal chest pain (17, 18). Chest pain presumably could be produced by these excessive esophageal contractions or “spasms” producing either myoischemia or proximal esophageal distention. Myoischemia could result from increased intramural tension inhibiting esophageal blood flow for a critical period (19). A more common etiology for pain of hollow organ origin, however, is distention. Early studies with intraesophageal balloon dilatation produced chest pain referred to the subxiphoid, substernal, or cervical regions (20,21). Although the data are meager, esophageal pain could be the result of dilatation proximal to a fixed functional obstruction, either at the level of the lower esophageal sphincter or from a spastic, nonperistaltic region in the esophageal body (l&22). Recent observations question the importance of abnormal esophageal contractions as the primary etiology of esophageal chest pain. In 9 patients undergoing prolonged esophageal pressure monitoring, Clouse et al. (23) found no relationship between spontaneous chest pain and esophageal contractions. We have recently observed that edrophonium (80 pgikg intravenous bolus) is a useful provocative
test for esophageal chest pain (3). This cholinesterase inhibitor consistently augmented esophageal contractions by increasing wave amplitude and prolonging wave duration, yet we found no correlation between the magnitude of esophageal contractions and the replication of chest pain. The present study indicates that intraesophageal balloon distention is an excellent provocative test for esophageal chest pain. Eighteen of 30 (60%) patients had their pain replicated by balloon distention, compared with only 6 of 30 (20%) asymptomatic volunteers (Figure 1). Test specificity improved to lOOoh with the observation that only chest pain patients (15 of 18) had pain replication with balloon volumes 58 ml of air. Additionally, balloon distention is a simple test, and unlike systemic drugs such as bethanechol, edrophonium, or ergonovine, it directly stimulates only the esophagus. Our studies also provide some interesting insights into the etiology of esophageal chest pain. The balloon pressure-volume curves generated during larger distending volumes in our in vitro and in vivo studies have a much higher slope than the same curves obtained during balloon distention in air (Figures 2 and 6). The studies in rigid cylinders suggest this pressure change is secondary to external compression, i.e., in vivo esophageal tone. This is supported by previous studies in monkeys in which atropine partially inhibited this response, confirming that the additional intraesophageal balloon pressure is a reflection of resistance provided by the esophageal smooth muscle (24). Presumably, if spastic or hypercontractile waves were to produce this increased esophageal tone, symptomatic patients would be expected to have higher intraluminal pressures than asymptomatic subjects. As shown in
October
1986
Figure 3, the opposite result was found; patients experienced pain at lower intraesophageal pressures than the control subjects. Furthermore, esophageal wall resistance did not distinguish patient groups (Figure 2), suggesting that the general responsiveness of the esophagus to stretch could not account for the patients’ pain. Esophageal muscular activity proximal to the balloon was also similar among chest pain patients and controls. Although esophageal distention was not directly measured, an esophageal propulsive force (Figure 4) developed in all subjects. These highamplitude, repetitive contractions of prolonged duration are the result of localized muscular contractions along a short segment of the esophagus immediately proximal to an obstructing bolus (9). The stimulus for this contraction is presumably the stretch of smooth muscle fibers invoked by the balloon. As shown in Figure 5, mean amplitude and duration of the esophageal propulsive force did not separate patients with balloon-induced pain from patients without pain, or from control subjects. Here again, our data do not support either abnormal contractile activity or intraluminal distention above a relative obstruction as the cause of esophageal pain in these patients. Thus, the results of the current study question the prevailing concept that abnormal esophageal contractions or distention are the basis for esophageal chest pain. Rather, our experimental findings suggest that a mechanism of pain in these patients may be related to enhanced sensitivity to distention as a result of lower visceral pain thresholds. In hypothesizing that esophageal chest pain is related to abnormal perception of distention, one is reminded of similar observations in patients with the irritable bowel syndrome. Ritchie (25) and Whitehead et al. (26) observed that patients with irritable bowel experienced lower abdominal pains with rectal balloon volumes not generally perceived by normal subjects. Kullmann and Fielding (27) noted that abnormal perception of colonic distention also distinguished irritable bowel patients from a group of anxious patients without bowel symptoms as well as from normal subjects. As we have observed in the esophagus, these groups found no evidence of abnormal muscular activity or increased gut wall tension. Balloon diameter assessed by radiograph, intraballoon pressures, and colonic motility did not distinguish the healthy subjects from the irritable bowel patients with balloon-induced pain. Discrepant results, however, were reported by Latimer et al. (28), who could find no difference in pain perception with rectal distention among irritable bowel patients compared with psychiatric and healthy control groups without bowel symptoms. It
MECHANISM
OF ESOPHAGEAL
CHEST
PAIN
851
is unclear why Latimer’s group failed to replicate the observations of the other three research teams, although it may be significant that they embedded this perception task in a study of the effects of stressful interview and meal stimulation on colonic activity. Further support for our observations can be found in a preliminary report by Edwards, entitled “The Tender Esophagus” (29). He describes a group of patients with dysphagia in whom symptoms could be produced by inflating a balloon at different points in the esophagus. Not only was their pain duplicated, but these patients were consistently sensitive to a balloon diameter unnoticed by normal subjects or patients with esophagitis. The pathogenesis of the lower visceral pain threshold in patients with esophageal chest pain and the irritable bowel syndrome is poorly understood. The perception and verbal report of pain is a complex process mediated by factors such as anxiety, depression, individual cultural values, and secondary gain (30). Relevant to this point is the recent observation by two laboratories that patients with noncardiac chest pain and distal esophageal motility disorders frequently have psychiatric disorders (31,32). As has been previously reported in the irritable bowel syndrome (33-35), depression, anxiety, and somatization disorders are diagnosed in as many as 84% of patients with esophageal chest pain. This relationship is intriguing and more work remains to be done to understand the complex interaction between gastrointestinal pain, altered visceral pain thresholds secondary to distention, and psychiatric illnesses. Our studies indicate that balloon distention will frequently replicate esophageal chest pain. The mechanism of pain in these patients appears not to be related to abnormal esophageal contractile events, but may represent a lower pain threshold to esophageal distention. In this respect, esophageal pain may share a common mechanism with the lower abdominal pain reported by patients with the irritable bowel syndrome. References Bernstein L, Baker L. A clinical test for esophagitis. Gastroenterology 1958;34:760-81. Benjamin SB, Richter JE, Cordova CM, Knuff TE, Caste11 DO. Prospective manometric evaluation with pharmacologic provocation of patients with suspected esophageal motility dysfunction. Gastroenterology 1983;84:893-901, Richter JE, Hackshaw BT, Wu WC, Caste11 DO. Edrophonium: a useful provocative test for esophageal chest pain. Ann Intern Med 1985;103:14-21. Koch KL, Curry RC, Feldman RL, Pepine CL, Long A, Mathias JR. Ergonovine-induced esophageal spasm in patients with chest pain resembling angina pectoris. Dig Dis Sci 1982; 27:1073-80.
852
RICHTER ET AL.
5. London RL, Ouyang A, Snape W, Goldberg S, Hirschfield SW, Cohen S Provocation of esophageal pain by ergonovine or edrophonium. Gastroenterology 1981;81:10-14. 6. Alban Davis H, Daye MD, Rhoades J, Dart AM, Henderson AH. Diagnosis, of esophageal spasm by ergometrine provocation. Gut 1982;23:89-97. 7. Eastwood GL, Weiner BH, Dickerson WJ, et al. Use of ergonovine to identify esophageal spasm in patients with chest pain. Ann Intern Med 1981;94:768-71. 8. Bertrand ME, LaBlanche JM, Tilmant PY, et al. Frequency of provoked coronary arterial spasm in 1089 consecutive patients undergoing coronary arteriography. Circulation 1982: 65:1299-305. 9. Winship DH, Zboralske FF. The esophageal propulsive force: esophageal response to acute distention. J Clin Invest 1967; 46:1391-401. 10. Benjamin SB, Caste11 DO. Esophageal causes of chest pain. In: Caste11 DO, Johnson LF, eds. Esophageal function in health and disease. New York: Elsevier Science Publishing Co., 1983:85-98. 11. Siegel CI, Hendrix TR. Esophageal motor abnormalities induced by acid perfusion in patients with heartburn. J Clin Invest 1963;42:686-95. 12. Richter JE, Johns DN, Wu WC, Caste11 DO. Are esophageal motility abnormalities produced during intraesophageal acid perfusion test? JAMA 1985;253:1914-7. 13. Verituratos SG, Burns TW, Neighbors BT. Gastroesophageal refluxi are motility changes related to symptoms (abstr)? Gastroentarology 1983;84:1342. 14. Creamer B, Donoghue FE, Code CF. Pattern of esophageal motility in diffuse spasm, Gastroenterology 1958;34:782-96. 15. Roth HP, Fleshier B. Diffuse esophageal spasm. Ann intern Med lB64;61:914-23. 16. Mellow M. Symptomatic diffuse esophageal spasm: manometric follow-up and response to cholinergic stimulation and cholinesterase inhibition. Gastroenterology 1977;73:237-40. 17. Benjamin SB, Gerhart D, Caste11 DO. High amplitude, peristaltic contractions associated with chest pain and/or dysphagia. Gastroenterology 1979;77:478-83. 18. Brand DL, Martin D, Pope CE II. Esophageal manometrics in patients with angina-like chest pain. Am J Dig Dis 1977; 23:300-I. 19. Mellow MH. Effect of isosorbide and hydralazine in painful
GASTROENTEROLOGY Vol. 91, No. 4
20.
21. 22. 23.
24.
25.
26.
27. 28.
29. 30. 31.
32.
33.
34. 35.
primary esophageal motility disorders. Gastroenterology 1982;83:364-70. Kramer P, Hollander W. Comparison of experimental esophageal pain with clinical pain of angina pectoris and esophageal disease. Gastroenterology 1955;29:719-43. Baylis JH, Kauntze R, Trounce JR. Observations on distention of the lower end of the oesophagus. Q J Med 1955;94:143-54. Kaye MD. Anomalies of peristalsis in idiopathic diffuse oesophageal spasm. Gut 1981;22:217-22. Clouse RE, Staiano A, Landau DW, Schlacter JL. Manometric findings during spontaneous chest pain with presumed esophageal “spasms.” Gastroenterology 1983;85:395-402. Barish CF, Richter JE, Blackwell JN, Caste11 DO. Pharmacologic testing of esophageal muscle response to balloon dilatation in primates (abstr). Gastroenterology 1984;86:1020. Ritchie J. Pain from distention of the pelvic colon by inflating a balloon in the irritable colon syndrome. Gut 1973; 14:125-32. Whitehead WE, Engel BT, Schuster MM. Irritable bowel syndrome. Physiological and psychological differences between diarrhea-predominant and constipation-predominant patients. Dig Dis Sci 1982;25:404-13. Kullmann G, Fielding JF. Rectal distensibility in the irritable bowel syndrome. Ir Med J 1981;74:140-2. Latimer P, Campbell D, Latimer M, Sarna S, Waterfall W, Daniel EE. Irritable bowel syndrome: a test of the colonic hyperalgesia hypothesis. f Behav Med 1979;2:285-95. a new syndrome Edwards DAW. “Tender oesophagus”: (abstr). Gut 1982;23:919. Melzack R, Wall PD; The challenge of pain. New York: Basic Books, 1983. Clouse RE, Lustman PJ. Psbchiatric illnesses and contraction abnormalities of the esophagus. N Engl J Med 1982; 3091337-42. Richter JE, Obrecht WF, Bradley LA, Young LA, Anderson KO. Psychological similarities between patients with the nutcracker esophagus and irritable bowel syndrome. Dig Dis Sci 1986;31:131-8. Chaudhary NA, Truelove SC. The irritable bowel syndrome: a study of the clinical features, predisposing causes and prognosis in 130 cases. Q J Med 162;31:813-7. Liss JL, Alpers D, Woodruff RA Jr. The irritable colon syndrome and psychiatric illness. Dis Nerv Syst 173;34:151-7. Drossman DA, Powell DW, Session JT Jr. The irritable bowel syndrome. Gastroenterology 1977;73:811-22.
1986;91:845-52
Abnormal Sensory Perception in Patients With Esophageal Chest Pain JOEL E. RICHTER, CHARLES F. BARISH, and DONALD 0. CASTELL Bowman Gray School of Medicine on Gastroenterology, Winston-Salem,
of Wake Forest University, North Carolina
Many cases of recurrent noncardiac chest pain are considered to be of esophageal origin. The mechanism of this pain is poorly understood and only rarely appears related to abnormal esophageal contractions observed during manometry. We studied the response to intraesophageal balloon distention in 30 patients with noncardiac chest pain (mean age 52 yr) and 30 controls (mean age 41 yr). A polyvinyl balloon (length 30 mm; maximum diameter after IO-ml distention, 25 mm] was positioned 10 cm above the lower esophageal sphincter and inflated with progressively larger volumes of air (each increase = 1 ml). Chest pain was correlated with balloon volumes and pressures, esophageal contractions 5 cm above the balloon, and electrocardiogram changes. Results were as follows: 18 of 30 (60%) patients and 6 of 30 (20%) controls (p < 0.005) experienced chest pain. Electrocardiogram changes ofischemia were not seen. Neither balloon pressures nor esophageal contractions above the balloon differed significantly during pain in the controls or patients. As assessed by pressure-volume curves, esophageal tone was also similar in all subjects regardless of pain production. Patients were more sensitive to smaller volumes of balloon distention; 15 of 18 patients had chest pain at 58 ml volume while controls only noted chest pain at 29 ml Received Jund 7, 1985. Accepted March 27, 1986. Address reqhests for reprints to: Joel E. Richter, M.D., Section on Gastroenterology, Bowman Gray School of Medicine, 300 South Hawthorne Road, Winston-Salem, North Carolina 27103. C. F. Barish’s present address is: Wake Internal Medicine Consultants, 3100 Blue Ridge Road, Suite 300, Raleigh, North Carolina 27612. This work was supported, in part, by grant AM34200-OlAl from the National Institutes of Health. The authors thank John N. Blackwell, M.D., and W. J. Dodds, M.D., for their valuable technical assistance; David Ott, MD., for assistance with the radiographic studies; Chris Dalton, PA-C, for her expertise in the manometry laboratory; and Karen Chatman and Linda Brown for their assistance in the preparation of this manuscript. 0 1986 by the American Gastroenterological Association 0016-5085/86/$3.50
Department
of Medicine,
Section
volume. We therefore arrived at the following conclusions: [a) esophageal balloon distention may provide a nonpharmacologic provocative test for esophageal chest pain; and [b] the mechanism of chest pain in these patients may be related to lower pain threshold to balloon distention, which is independent of esophageal contractions.
The potential for the esophagus to be the site of recurring chest pain has received considerable attention in recent years. The specific mechanisms by which abnormal esophageal function might produce chest pain have not been clearly identified. At least two possible etiologies have been proposed: esophageal mucosal irritation secondary to gastroesophageal reflux, or strong esophageal contractions as suggested in patients with diffuse esophageal spasm. The development of provocative testing techniques has provided some insights and support for the importance of these two mechanisms in the development of esophageal chest pain. Acid perfusion into the distal esophagus (Bernstein test) has been used for many years as a means to reproduce symptoms in patients in whom mucosal acid sensitivity was the proposed origin of their complaints (1).In more recent times, stimulation of forceful esophageal contractions with drugs such as edrophonium (z-4), bethanechol (Z), and ergonovine (4-7) has also been used in an attempt to establish a relationship between abnormal esophageal contractions and chest pain. In the present study we have evaluated the response to graded intraesophageal balloon distention with small volumes of air in an attempt to further elucidate mechanisms of pain production. These studies have indicated the possibility of a third component to the production of pain in these patients: that is, the presence of a lowered visceral pain threshold to esophageal distention.
846 RICHTER ET AL.
GASTROENTEROLOGY Vol. 91, No.
Materials and Methods Preliminary
Balloon
Studies
A variety of experimental balloons were made from a number of materials including different sized Penrose drains, rubber contraceptive devices, portions of rubber gloves, and a standard finger cot. The latter material was found to be most suitable because it easily expanded at low intraesophageal balloon pressures. In vitro studies with rabbit and dog esophagi and in vivo studies in Patas monkeys revealed that this type of balloon would adequately distend the esophagus without producing macroscopic evidence of mucosal or muscular disruption. The polyvinyl balloon finally developed for use in our human studies was constructed from a finger cot (Ansell Inc., Dothan, Ala.) and attached to a standard manometric catheter using silicone glue and 5.0 nylon suture. At IO-ml distention, the balloon measured 30 mm in length and 25 mm in diameter. The catheter orifice 10 cm proximal to the distal orifice opened into the balloon and was perfused only with air. The use of a soft, elastic balloon gave us a large safety margin but raised the possibility that increased esophageal resistance might distort and possibly elongate the balloon. Ideally, all human studies would need to be done under fluoroscopy, but this was neither practical nor ethically possible. Therefore, an in vitro model, using cardboard cylinders of various diameters, was used to test the characteristics of the balloon during graded distention. The proximal tip of the capillary tube opening into the balloon was attached to a three-way stopcock, which was connected to an external transducer with output on a Beckman recorder (Beckman Instruments, Inc., Fullerton, Calif.). Air could then be infused by hand syringe into the balloon at l-ml volume increments and intraballoon pressures recorded by the physiograph. Initially, intraballoon pressures were measured in room air (23°C) up to a maximum volume of 10 ml. Studies were then repeated after placing the balloon in the center of 20-cm-long cardboard cylinders of decreasing diameter (25 mm, 23 mm, 20 mm). All studies were performed twice, and each value was recorded as the average of the two pressure measurements obtained at a specific balloon volume. These values were later compared with intraesophageal pressure-volume measurements during in vivo studies in controls and chest pain patients. Patients
and
Healthy
Volunteers
Thirty patients (mean age 52 yr, range 35-67 yr; 15 men, 15 women) considered by their referring cardiologists to have noncardiac chest pain were studied. Prior cardiac catheterization had revealed normal coronary arteries in 20 patients and nonobstructive coronary artery disease (525% coronary artery diameter narrowing) in 10 patients. Ergonovine was given to 2 patients during cardiac catheterization without inducing chest pain or focal coronary artery spasm. Only those patients who had chest pain predominately at rest received ergonovine, because the possibility of coronary artery spasm was highest in this group (8). Nineteen of the 30 patients (63%) had previous
4
gastrointestinal evaluations (upper gastrointestinal series or endoscopy, or both), which were normal in 10 patients, showed hiatal hernia in 7, and revealed esophagitis in 2. Nine (30%) chest pain patients had abnormal baseline esophageal manometry: 8 patients with the “nutcracker esophagus” (mean distal amplitude 2180 mmHg) and 1 patient with diffuse esophageal spasm. All other patients had normal esophageal motility. These 30 patients were compared with 30 healthy volunteers (mean age 41 yr, range 22-60 yr; 18 men, 12 women) with no history of chest pain. Volunteers were carefully interviewed and excluded if they had any history of frequent heartburn (more than once a month), dysphagia, regurgitation, or antacid use. Additionally, no volunteer had a history of diabetes mellitus, alcoholism, collagen vascular disorders, or neurologic diseases, All volunteers had normal esophageal manometric studies. Written informed consent was obtained from all participants, and the study was approved on February 23, 1984, by the Human Research Review Committee of the Bowman Gray School of Medicine, Winston-Salem, North Carolina. Balloon
Studies
in Humans
Esophageal manometric studies were performed using a round, 8-lumen polyvinyl catheter (diameter 4.5 mm, internal diameter 0.8 mm, Arndorfer Specialties, Inc., Greendale, Wis.). The distal four openings were located at the same level 90” apart, and the four proximal openings were spaced at 5-cm intervals. Each lumen, except for the one associated with the balloon, was continuously perfused with distilled water at a rate of 0.5 ml/min using a low compliance pneumohydraulic capillary infusion system (Arndorfer Specialties, Inc.). The lumen opening into the balloon was filled with air. The catheter was connected to external transducers (Beckman, model 4-327-C, Norcross, Ga.), with output on a Beckman Recorder (model R-612). The manometric system has a pressure rise rate of >250 mmHg/s. Following standard manometry studies, esophageal balloon distention was performed in both chest pain patients and volunteers. The balloon was positioned 10 cm above the lower esophageal sphincter. This permitted recordings of esophageal pressures at the level of the sphincter as well as 5 and 15 cm above the lower esophageal sphincter. The subjects were then given written instructions to hold telling them how to respond to sensations noted during testing. They were given a choice of sensations to pick from (tugging, discomfort, tightness, pressure, or pain) and were advised that they might feel something different or nothing at all. In an attempt to minimize all possibilities of influencing the subjects’ responses, there was no subsequent conversation between the investigator and subjects during the study. During balloon distention, subjects were permitted to dry swallow as necessary. The esophageal balloon was distended in a stepwise fashion using progressively larger volumes of air (each increase = 1 ml) to a total volume of 10 ml. The air was injected rapidly (within 2 s) by syringe and each distention volume was maintained for 10 s. Between each incremental volume increase, the balloon was completely emptied
October
MECHANISM
1986
for 5 s. This procedure was done to prevent the esophagus from accommodating to progressive, gradual balloon distention. Unknown to the subjects, a placebo-sham distention was also performed before the actual inflation. A positive response to balloon distention was recorded only when the patients experienced reproduction of their typical chest pain or the volunteers complained of chest pain. Any other sensation was not considered a positive test. If pain was produced by balloon distention, it was localized and correlated with balloon volumes, balloon pressures, and esophageal contraction waves (amplitude and duration) above the balloon. Patients and volunteers not experiencing chest pain had similar parameters measured at maximal balloon distention (lo-ml volume). The reproducibility of pain threshold was examined in the last 15 chest pain patients by repeating the above sequence. Electrocardiographic monitoring was continuously recorded during the balloon distention studies. Two control subjects and 2 patients also had limited balloon distention studies performed during fluoroscopy to further address the possibility that increased esophageal resistance might distort and elongate the balloon. As previously described, the balloon was positioned 10 cm above the lower esophageal sphincter. In the upright position, subjects drank a small amount of high-density barium sulfate suspension to coat the esophageal mucosal surface. Balloon distention studies at volume increments of 6, 8, and10 ml and simultaneous left posterior oblique radiographs were done. The diameter of the balloon with each incremental inflation was determined by correcting for magnification using the 4.5-mm-diameter catheter as a known reference. These values were compared with similar measurements obtained from radiographs of the balloon alone taken after placing it in room temperature (23°C) water and distending with 6-, 8-, and lo-ml volumes of air. Statistical
Analysis
Statistical analysis was performed with the x2 test and Student’s t-test for paired and unpaired samples as appropriate. Results are expressed as mean ? SEM. Unless otherwise specified, all levels of significance have a probability value
Results Human
Studies
Eighteen of 30 (60%) patients experienced chest pain with balloon distention. In contrast, only 6 of the 30 (20%) normal volunteers had any chest pain. No pain was noted in either patients or volunteers with sham distention. This relationship between a history of noncardiac chest pain and pain with balloon distention was highly significant (p < 0.005; x2 analysis). When pain was reproduced, it was localized to the mid or lower sternal region in all patients. Once started, the pain intensified with further increments of volume and was immediately relieved by balloon decompression. In no patients
OF ESOPHAGEAL
CHEST
PAIN
847
loo-
$0. 60.
CUMULATIVE % POSITIVE
ii. 50. 40. 30. F...
20. 10. Pallsnls Normals -.
..
7
. 1
2
r 3
. 4
BALLOON
Figure
I
1 5
1 6
VOLUME
,..a
K 7
8
9
10
(CC)
Esophageal pain with balloon distention. Sixty percent of chest pain patients (solid line] experienced pain, compared to only 20% of normals (broken line). The difference is highly significant (p -C 0.005). Not only do the patients more frequently develop pain with balloon distention, but their pain occurs at smaller volumes, suggesting a lower pain threshold to esophageal distention.
having pain were electrocardiogram changes of ischemia noted. An abnormal baseline manometric study was not associated with a higher frequency of balloon-induced chest pain. Only 4 of 9 (44%) patients, all with the “nutcracker esophagus,” had ballooninduced pain. Intraesophageal balloon volume at the onset of pain also distinguished the patient and control groups, as noted in Figure 1. Fifteen of the 18 chest pain patients experiencing pain noted it at a volume of r8 ml of air. All control subjects with pain did not note their symptoms until at least 9 ml of air distention. The association between a history of noncardiac chest pain and provoked pain at a volume of 18 ml was also highly significant (p < 0.001; x2 analysis]. Reproducibility of balloon distention was tested in the last 15 patients by repeating the study 5 min later. The initial balloon distention replicated the patient’s chest pain in 9 subjects, and all experienced pain during the second study at the same volume or within 1 ml. Figure 2 shows the curves illustrating the relationship between intraesophageal balloon pressure and volume in patients and normal subjects. These curves are compared with the mean pressure-volume curve generated by inflating the balloon on three separate occasions in room air at 23°C. There was no significant difference between the curves obtained in patients and that found in the healthy controls. Furthermore, no difference in pressure-volume curves was found between patients in whom pain occurred during balloon distention and those without pain. All four curves do not begin to diverge until 5-ml volume, at which time it is assumed that the balloon begins to occlude the esophagus (diameter at 5-ml volume during distention in air = 20 mm). The difference observed between the three curves obtained in patients and controls and the
848
RICHTER ET AL.
GASTROENTEROLOGY Vol. 91, No. 4
BALLOON PRESSURE ImmHg)
--
Pls with
pak (16)
-
Balloon In slf
BALLOON
Figure
VOLUME (CC)
2. Balloon pressure-volume curves in the three study groups compared to balloon distention in air. Each point represents the mean balloon pressures at each l-ml increment in balloon volume. There is no significant difference between the curves obtained in patients with or without pain and the normal controls. The significant difference (* = p < 0.05) seen between the patient curves and the balloon-in-air curve represents the compliance of the esophagus.
balloon-in-air curve beyond 5-ml volume represents the effective elastic modulus or compliance of the esophagus. As this measure of esophageal resistance is similar among patient and control groups, it cannot account for the patient’s esophageal chest pain. Intraesophageal balloon pressures were measured at the onset of pain or at maximal distention (10 ml) if no pain developed. The chest pain patients who experienced pain during testing generally showed lower intraesophageal balloon pressures at the onset of pain than did the healthy volunteers who experienced pain (mean pressures: 86 k 4 mmHg vs. 103 + 5 mmHg; p < 0.05, Figure 3), but there was some overlap between groups. This result was somewhat unexpected but could have been predicted by the similar pressure-volume relationship among the three groups, as pain generally occurred at smaller distention volumes in the patients as compared with controls. Chest pain patients and controls who did not experience pain with balloon distention had similar mean intraesophageal balloon pressures at lo-ml maximal distention (89 + 5 mmHg vs. 90 k 4 mmHg) (Figure 3). These data suggest that the esophagi are similar in their elasticity in all study groups, and they do not support the hypothesis of increased esophageal wall tone or “spasm” as the cause of our patients’ chest pains. Figure 4 illustrates a typical esophageal motility tracing obtained during our studies. Both patients and normal subjects experienced increased esophageal contraction activity above the balloon at higher distention volumes. These high-amplitude waves of prolonged duration and with frequent repetitive contractions have been called the “esophageal propulsive force” (9). Because it has been suggested that
“spastic” esophageal contractions proximal to a functional distal obstruction might be the source of esophageal chest pain (lo), we analyzed the characteristics of the esophageal propulsive force associated with the onset of pain, or at lo-ml volume if no pain developed. The maximal individual amplitudes and durations of the esophageal propulsive forces in patients with and without pain and in healthy controls are shown in Figure 5. Mean amplitude and duration was 64 k 8 mmHg and 6.2 t 0.7 s in the patients with pain compared to 55 + 11 mmHg and 5.7 + 0.8 s in the patients without pain, and 59 t 6 mmHg and 5.5 k 0.5 s in the control subjects. There was no significant difference between mean amplitudes or durations of the esophageal propulsive forces among the three groups. Effects of Volume Increments on Balloon Diameter-In Vitro and In Vivo Results Although esophageal compliance was similar in all three subgroups, the soft, elastic characteristics of our balloon still raised the possibility that individual variation in esophageal distention resulting from balloon distortion might account for the patients’ chest pain. Practical and ethical reasons prevented us from investigating all subjects during fluoroscopy, but the following in vivo and in vitro data suggest this was not the cause of our patients’ chest pain. The balloon pressure-volume curve in the 25mmdiameter cylinder was similar to the balloon-in-air curve (Figure 6). This was anticipated because the diameter of the balloon-in-air measured 25 mm when inflated maximally (10 ml). The characteristics
PnEBsLmE WnNB) (lOed
YTRAB*uooN
WllWOUT PAIN
...
130 120 110
b
;
..... .
70 60 60 10 30 20 10
P51lWl~S
EFigure
:
...
B =*
B. ..
P5tI*nts
Normals
70
60 50 40 30 20 10
:.
..
Norm5lr
ii fSE
I 3
Intraesophageal balloon pressure (mmHg) with and without pain. Left: at the onset of pain, patients with pain have a significantly lower [p < 0.05) mean intraesophageal balloon pressure than controls with pain. Right: patients and normals without pain have similar intraesophageal balloon pressures at lo-ml inflation.
October
MECHANISM
1986
Figure
IOCm
(balloon)
LES
of the balloon pressure-volume curves in the cylinders did not markedly deviate from the mean curve obtained in the patient groups, with and without pain, until cylinder diameters well below 25 mm were tested, i.e., 23 and 20 mm (Figure 6). Analysis of individual pressures for all study subjects at each balloon volume revealed that no pressures fell above the curve described for the 20-mm-diameter cylinder. Nine patients (30%), 5 with pain and 4 without pain, and 10 control subjects (33%) had at least one intraballoon pressure exceeding the curve for the 2%mm cylinder. Therefore, this in vitro model would predict that the maximal decrease in balloon diameter during in vivo studies should have been <2 mm (or 8% of 25 mm) in 41 of 60 (68%) study subjects. Five patients would PROXIYAL DURATIONS (SECONDS)
PROXIMAL AMPLITUDES 0IMl2) 120
-.*
110 100
... .
*
80
.. .
:p,
@J
50
.. ..
40
:.
30
‘:’
20
::.
.
e ..
10 NOFM*lS
?? =~~sE
16
s
.
. .
11
..
90
Figure
*. . .
.. El .. .
pAsYs ‘wATZ’ ? PAY PAP4
,O
.. ..
2 6
...
7
.
:
B
4
.. . . . . . . . . . . .
3
*
2
.. . wol))rlALI
. ...
Q .. .. .. . . zY= PIN
. .
El . .... ‘%z PAW
5. Individual maximal amplitudes (mmHg) and durations [seconds) of the esophageal propulsive force measured 5 cm above the balloon. There is no significant difference between mean amplitudes and durations in patients with or without pain and in normal volunteers.
OF ESOPHAGEAL
4
CHEST
PAIN
849
Example of one of the motility tracings obtained during our studies. The esophageal balloon is positioned 10 cm above the lower esophageal sphincter (LES). The lowest orifice is at the upper level of the LES. The other orifices are located 5 cm above the LES, and 5 cm proximal to the balloon (15 cm above the LES). The high-amplitude and prolonged duration waves above the balloon indicate the esophageal propulsive force (EPF). In the balloon, changes in pressure associated with l-ml volume increases can be seen. Between incremental inflations the balloon is emptied of air. The pressures do not appear to drop back to baseline on the tracing, however, because a stopcock closes the catheter to the transducer during deflation.
be predicted to have had a decrease in balloon diameter between 2 mm and 5 mm (8%-20%) at the onset of their pain. However, a similar diameter change was observed in 4 patients without pain and in 10 control subjects (8 without pain), confirming that increased esophageal resistance was not the cause of these patients’ chest pain. Our limited in vivo studies supported these in vitro results. One control subject, experiencing no chest pain at IO-ml balloon volume, had radiographic evidence of only a 1.6% decrease in balloon diameter. The second control subject noted chest pain at IO-ml balloon volume with an associated 3.4% decrease in balloon diameter. Both patients experienced chest pain at 8-ml balloon volume, at which time balloon diameter had decreased 3.0%~ and 8.4%. These x-ray studies revealed that, rather than narrowing and elongating with increasing volumes, the balloon tended to square off.
Discussion The precise explanation of the origin of pain in esophageal diseases has been an elusive problem. Commonly proposed mechanisms have included stimulation of acid-sensitive chemoreceptors in the esophageal mucosa or changes in esophageal wall tension stimulating mechanoreceptors. The symptoms of heartburn and chest pain in gastroesophageal reflux disease primarily appear to be the result of acid stimulation of chemoreceptors (I). Although acid infusion may occasionally provoke esophageal motility abnormalities as shown by Siegel and Hendrix (ll),more recent studies suggest motility changes are an uncommon cause of acid-induced symptoms (12,131.
850
RICHTER ET AL.
GASTROENTEROLOGY Vol. 91, No. 4
Cylinder
( 20
mm
)
, Cylinder
( 23
mm
)
180 #’ 160
_
/
140
I”
Figure
120 E
? :, 0 t =: m”
100
80
60 40
1
2
3
4 Balloon
5
6 Volume
7
6
6
6. Pressure-volume characteristics of the study balloon under different test conditions. For each curve the balloon was inflated stepwise from 1 to 10 ml. Pressure-volume curves are shown for the balloon inflated in air and inflated within a cardboard cylinder of diameters 25 mm, 23 mm, and 20 mm. Also shown is the mean curve obtained from the balloon distention inside the esophagus of study patients.
lo
( cc)
Chest pain in patients with esophageal motility disorders, either occurring spontaneously or in response to pharmacologic provocation, may be associated with esophageal contractions of increased amplitude, prolonged duration, and/or simultaneous activity along several recording sites (14-16). Diffuse esophageal spasm has been the traditional prototype of these painful esophageal motility disorders. More recently, however, it has been recognized that hypertensive peristaltic contractions (“nutcracker esophaare the most common motilgus, ” “supersqueezer”) ity abnormality found in patients with a history of esophageal chest pain (17, 18). Chest pain presumably could be produced by these excessive esophageal contractions or “spasms” producing either myoischemia or proximal esophageal distention. Myoischemia could result from increased intramural tension inhibiting esophageal blood flow for a critical period (19). A more common etiology for pain of hollow organ origin, however, is distention. Early studies with intraesophageal balloon dilatation produced chest pain referred to the subxiphoid, substernal, or cervical regions (20,21). Although the data are meager, esophageal pain could be the result of dilatation proximal to a fixed functional obstruction, either at the level of the lower esophageal sphincter or from a spastic, nonperistaltic region in the esophageal body (l&22). Recent observations question the importance of abnormal esophageal contractions as the primary etiology of esophageal chest pain. In 9 patients undergoing prolonged esophageal pressure monitoring, Clouse et al. (23) found no relationship between spontaneous chest pain and esophageal contractions. We have recently observed that edrophonium (80 pgikg intravenous bolus) is a useful provocative
test for esophageal chest pain (3). This cholinesterase inhibitor consistently augmented esophageal contractions by increasing wave amplitude and prolonging wave duration, yet we found no correlation between the magnitude of esophageal contractions and the replication of chest pain. The present study indicates that intraesophageal balloon distention is an excellent provocative test for esophageal chest pain. Eighteen of 30 (60%) patients had their pain replicated by balloon distention, compared with only 6 of 30 (20%) asymptomatic volunteers (Figure 1). Test specificity improved to lOOoh with the observation that only chest pain patients (15 of 18) had pain replication with balloon volumes 58 ml of air. Additionally, balloon distention is a simple test, and unlike systemic drugs such as bethanechol, edrophonium, or ergonovine, it directly stimulates only the esophagus. Our studies also provide some interesting insights into the etiology of esophageal chest pain. The balloon pressure-volume curves generated during larger distending volumes in our in vitro and in vivo studies have a much higher slope than the same curves obtained during balloon distention in air (Figures 2 and 6). The studies in rigid cylinders suggest this pressure change is secondary to external compression, i.e., in vivo esophageal tone. This is supported by previous studies in monkeys in which atropine partially inhibited this response, confirming that the additional intraesophageal balloon pressure is a reflection of resistance provided by the esophageal smooth muscle (24). Presumably, if spastic or hypercontractile waves were to produce this increased esophageal tone, symptomatic patients would be expected to have higher intraluminal pressures than asymptomatic subjects. As shown in
October
1986
Figure 3, the opposite result was found; patients experienced pain at lower intraesophageal pressures than the control subjects. Furthermore, esophageal wall resistance did not distinguish patient groups (Figure 2), suggesting that the general responsiveness of the esophagus to stretch could not account for the patients’ pain. Esophageal muscular activity proximal to the balloon was also similar among chest pain patients and controls. Although esophageal distention was not directly measured, an esophageal propulsive force (Figure 4) developed in all subjects. These highamplitude, repetitive contractions of prolonged duration are the result of localized muscular contractions along a short segment of the esophagus immediately proximal to an obstructing bolus (9). The stimulus for this contraction is presumably the stretch of smooth muscle fibers invoked by the balloon. As shown in Figure 5, mean amplitude and duration of the esophageal propulsive force did not separate patients with balloon-induced pain from patients without pain, or from control subjects. Here again, our data do not support either abnormal contractile activity or intraluminal distention above a relative obstruction as the cause of esophageal pain in these patients. Thus, the results of the current study question the prevailing concept that abnormal esophageal contractions or distention are the basis for esophageal chest pain. Rather, our experimental findings suggest that a mechanism of pain in these patients may be related to enhanced sensitivity to distention as a result of lower visceral pain thresholds. In hypothesizing that esophageal chest pain is related to abnormal perception of distention, one is reminded of similar observations in patients with the irritable bowel syndrome. Ritchie (25) and Whitehead et al. (26) observed that patients with irritable bowel experienced lower abdominal pains with rectal balloon volumes not generally perceived by normal subjects. Kullmann and Fielding (27) noted that abnormal perception of colonic distention also distinguished irritable bowel patients from a group of anxious patients without bowel symptoms as well as from normal subjects. As we have observed in the esophagus, these groups found no evidence of abnormal muscular activity or increased gut wall tension. Balloon diameter assessed by radiograph, intraballoon pressures, and colonic motility did not distinguish the healthy subjects from the irritable bowel patients with balloon-induced pain. Discrepant results, however, were reported by Latimer et al. (28), who could find no difference in pain perception with rectal distention among irritable bowel patients compared with psychiatric and healthy control groups without bowel symptoms. It
MECHANISM
OF ESOPHAGEAL
CHEST
PAIN
851
is unclear why Latimer’s group failed to replicate the observations of the other three research teams, although it may be significant that they embedded this perception task in a study of the effects of stressful interview and meal stimulation on colonic activity. Further support for our observations can be found in a preliminary report by Edwards, entitled “The Tender Esophagus” (29). He describes a group of patients with dysphagia in whom symptoms could be produced by inflating a balloon at different points in the esophagus. Not only was their pain duplicated, but these patients were consistently sensitive to a balloon diameter unnoticed by normal subjects or patients with esophagitis. The pathogenesis of the lower visceral pain threshold in patients with esophageal chest pain and the irritable bowel syndrome is poorly understood. The perception and verbal report of pain is a complex process mediated by factors such as anxiety, depression, individual cultural values, and secondary gain (30). Relevant to this point is the recent observation by two laboratories that patients with noncardiac chest pain and distal esophageal motility disorders frequently have psychiatric disorders (31,32). As has been previously reported in the irritable bowel syndrome (33-35), depression, anxiety, and somatization disorders are diagnosed in as many as 84% of patients with esophageal chest pain. This relationship is intriguing and more work remains to be done to understand the complex interaction between gastrointestinal pain, altered visceral pain thresholds secondary to distention, and psychiatric illnesses. Our studies indicate that balloon distention will frequently replicate esophageal chest pain. The mechanism of pain in these patients appears not to be related to abnormal esophageal contractile events, but may represent a lower pain threshold to esophageal distention. In this respect, esophageal pain may share a common mechanism with the lower abdominal pain reported by patients with the irritable bowel syndrome. References Bernstein L, Baker L. A clinical test for esophagitis. Gastroenterology 1958;34:760-81. Benjamin SB, Richter JE, Cordova CM, Knuff TE, Caste11 DO. Prospective manometric evaluation with pharmacologic provocation of patients with suspected esophageal motility dysfunction. Gastroenterology 1983;84:893-901, Richter JE, Hackshaw BT, Wu WC, Caste11 DO. Edrophonium: a useful provocative test for esophageal chest pain. Ann Intern Med 1985;103:14-21. Koch KL, Curry RC, Feldman RL, Pepine CL, Long A, Mathias JR. Ergonovine-induced esophageal spasm in patients with chest pain resembling angina pectoris. Dig Dis Sci 1982; 27:1073-80.
852
RICHTER ET AL.
5. London RL, Ouyang A, Snape W, Goldberg S, Hirschfield SW, Cohen S Provocation of esophageal pain by ergonovine or edrophonium. Gastroenterology 1981;81:10-14. 6. Alban Davis H, Daye MD, Rhoades J, Dart AM, Henderson AH. Diagnosis, of esophageal spasm by ergometrine provocation. Gut 1982;23:89-97. 7. Eastwood GL, Weiner BH, Dickerson WJ, et al. Use of ergonovine to identify esophageal spasm in patients with chest pain. Ann Intern Med 1981;94:768-71. 8. Bertrand ME, LaBlanche JM, Tilmant PY, et al. Frequency of provoked coronary arterial spasm in 1089 consecutive patients undergoing coronary arteriography. Circulation 1982: 65:1299-305. 9. Winship DH, Zboralske FF. The esophageal propulsive force: esophageal response to acute distention. J Clin Invest 1967; 46:1391-401. 10. Benjamin SB, Caste11 DO. Esophageal causes of chest pain. In: Caste11 DO, Johnson LF, eds. Esophageal function in health and disease. New York: Elsevier Science Publishing Co., 1983:85-98. 11. Siegel CI, Hendrix TR. Esophageal motor abnormalities induced by acid perfusion in patients with heartburn. J Clin Invest 1963;42:686-95. 12. Richter JE, Johns DN, Wu WC, Caste11 DO. Are esophageal motility abnormalities produced during intraesophageal acid perfusion test? JAMA 1985;253:1914-7. 13. Verituratos SG, Burns TW, Neighbors BT. Gastroesophageal refluxi are motility changes related to symptoms (abstr)? Gastroentarology 1983;84:1342. 14. Creamer B, Donoghue FE, Code CF. Pattern of esophageal motility in diffuse spasm, Gastroenterology 1958;34:782-96. 15. Roth HP, Fleshier B. Diffuse esophageal spasm. Ann intern Med lB64;61:914-23. 16. Mellow M. Symptomatic diffuse esophageal spasm: manometric follow-up and response to cholinergic stimulation and cholinesterase inhibition. Gastroenterology 1977;73:237-40. 17. Benjamin SB, Gerhart D, Caste11 DO. High amplitude, peristaltic contractions associated with chest pain and/or dysphagia. Gastroenterology 1979;77:478-83. 18. Brand DL, Martin D, Pope CE II. Esophageal manometrics in patients with angina-like chest pain. Am J Dig Dis 1977; 23:300-I. 19. Mellow MH. Effect of isosorbide and hydralazine in painful
GASTROENTEROLOGY Vol. 91, No. 4
20.
21. 22. 23.
24.
25.
26.
27. 28.
29. 30. 31.
32.
33.
34. 35.
primary esophageal motility disorders. Gastroenterology 1982;83:364-70. Kramer P, Hollander W. Comparison of experimental esophageal pain with clinical pain of angina pectoris and esophageal disease. Gastroenterology 1955;29:719-43. Baylis JH, Kauntze R, Trounce JR. Observations on distention of the lower end of the oesophagus. Q J Med 1955;94:143-54. Kaye MD. Anomalies of peristalsis in idiopathic diffuse oesophageal spasm. Gut 1981;22:217-22. Clouse RE, Staiano A, Landau DW, Schlacter JL. Manometric findings during spontaneous chest pain with presumed esophageal “spasms.” Gastroenterology 1983;85:395-402. Barish CF, Richter JE, Blackwell JN, Caste11 DO. Pharmacologic testing of esophageal muscle response to balloon dilatation in primates (abstr). Gastroenterology 1984;86:1020. Ritchie J. Pain from distention of the pelvic colon by inflating a balloon in the irritable colon syndrome. Gut 1973; 14:125-32. Whitehead WE, Engel BT, Schuster MM. Irritable bowel syndrome. Physiological and psychological differences between diarrhea-predominant and constipation-predominant patients. Dig Dis Sci 1982;25:404-13. Kullmann G, Fielding JF. Rectal distensibility in the irritable bowel syndrome. Ir Med J 1981;74:140-2. Latimer P, Campbell D, Latimer M, Sarna S, Waterfall W, Daniel EE. Irritable bowel syndrome: a test of the colonic hyperalgesia hypothesis. f Behav Med 1979;2:285-95. a new syndrome Edwards DAW. “Tender oesophagus”: (abstr). Gut 1982;23:919. Melzack R, Wall PD; The challenge of pain. New York: Basic Books, 1983. Clouse RE, Lustman PJ. Psbchiatric illnesses and contraction abnormalities of the esophagus. N Engl J Med 1982; 3091337-42. Richter JE, Obrecht WF, Bradley LA, Young LA, Anderson KO. Psychological similarities between patients with the nutcracker esophagus and irritable bowel syndrome. Dig Dis Sci 1986;31:131-8. Chaudhary NA, Truelove SC. The irritable bowel syndrome: a study of the clinical features, predisposing causes and prognosis in 130 cases. Q J Med 162;31:813-7. Liss JL, Alpers D, Woodruff RA Jr. The irritable colon syndrome and psychiatric illness. Dis Nerv Syst 173;34:151-7. Drossman DA, Powell DW, Session JT Jr. The irritable bowel syndrome. Gastroenterology 1977;73:811-22.