THE AMERICAN JOURNAL OF GASTROENTEROLOGY
1, 2000
© 2000 by Am. Coll. of Gastroenterology
Published by Elsevier Science Inc.
Proximal Migration of Esophageal
Acid Perfusions During Waking and Sleep
William C. Orr, Ph.D., Sigrid Elsenbruch, Ph.D., Michael J. Harnish, Ph.D., and Lawrence F. Johnson, M.D.
Thomas N. Lynn Institute for Healthcare Research, Oklahoma City, Oklahoma; and Division of
Gastroenterology, Department of Medicine, University of Alabama, Birmingham, Alabama
OBJECTIVE: Proximal acid migration resulting from gastro-
esophageal reflux has been implicated in aerodigestive com-
plaints and disorders. This study was designed to investigate
the effects of acid volume, posture, and sleep on proximal
esophageal acid migration (drop in pH to ⬍4.0).
METHODS: The study was performed in 15 healthy adults. A
distal esophageal acid perfusion technique to simulate gas-
troesophageal reflux was used. Esophageal acid perfusions
of 1 ml and 3 ml were accomplished at a site 5cm above the
proximal border of the lower esophageal sphincter in the
upright and supine positions during waking, and during
polysomnographically monitored sleep. Esophageal pH was
recorded by two sensors located in the mid- and proximal
esophagus at 10 and 15 cm above the lower esophageal
sphincter.
RESULTS: Acid volume clearly increased the incidence of
migration to the mid and proximal sensors during both
waking and sleep, and also significantly increased acid
clearance time. Posture failed to significantly affect the
incidence of acid migration and acid clearance. Sleep clearly
enhanced migration to the proximal pH sensor of even those
perfusions as small as 1 ml. For instance, 40% of 1 ml
perfusions during sleep migrated to the proximal sensor
compared with ⬍1% during waking. Acid clearance times
were significantly longer during sleep as measured by the
mid- and proximal esophageal pH sensors.
CONCLUSIONS: In healthy individuals, volume enhances the
likelihood of migration to both mid- and proximal esopha-
gus, and significantly prolongs clearance time in the waking
state. Posture appears to be a less significant parameter with
regard to both the incidence of acid migration and acid
clearance. Sleep is a significant risk factor for acid migration
to the proximal esophagus for even minute volumes, and
markedly prolongs acid clearance. (Am J Gastroenterol
2000;95:37– 42. © 2000 by Am. Coll. of Gastroenterology)
INTRODUCTION
The occurrence of aerodigestive complications of gastro-
esophageal reflux (GER) has been amply documented (1–
4). Chronic cough, the exacerbation of bronchial asthma,
and pulmonary complications such as aspiration pneumonia
Vol. 95, No.
ISSN 0002-9270/00/$20.00
PII S0002-9270(99)00728-5
have been linked to complications of GER (1, 2, 5, 6). In
fact, GER has been shown to be among the most common,
though generally unrecognized, causes of chronic cough (5).
Aerodigestive complications of gastric contents refluxed
into the mid- and proximal esophagus would presume some
degree of proximal migration of the refluxant. Although not
always the case, a number of studies have documented
increases in proximal acid contact time in patients with
aerodigestive complications (4, 7, 8 –10). The proximal mi-
gration of refluxed gastric contents certainly increases the
risk of pulmonary aspiration, in addition to enhancing the
probability of other symptoms such as chronic cough. For
example, Patti and colleagues demonstrated that in aspira-
tors, impaired peristalsis caused a higher degree of acid
exposure and delayed acid clearance in the proximal esoph-
agus (6). In addition, Jacob and colleagues demonstrated
that nocturnal proximal acid exposure significantly differ-
entiated patients with reflux laryngitis from a group of
patients with esophagitis and normal controls (9). Recently,
further substantiation of these findings were presented by
Shaker et al., who showed that patients with posterior lar-
yngitis had greater pharyngoesophageal acid contact than
both reflux patients or controls (4).
Despite the numerous studies documenting the impor-
tance of proximal acid migration in producing aerodigestive
complications, the risk factors that facilitate proximal acid
migration have not been adequately described. Previous
studies from our laboratory documented that acid clearance
during sleep was markedly prolonged in normal controls
and in patients with esophagitis (11–13). Prolonged acid
clearance in the supine position during sleep would intu-
itively enhance the risk of proximal esophageal migration.
Thus, sleep itself would be considered a putative risk factor
for proximal migration of refluxed gastric contents. Addi-
tionally, our previous studies documented that other risk
factors for pulmonary complications, including low pH
value and increased acid volume, also produced differential
responses during sleep (11, 12).
In previous human studies, we have used a distal esoph-
ageal acid perfusion technique to simulate gastroesophageal
reflux episodes in healthy volunteers (11, 12). The unique
advantages of this methodology relate to the experimental
control of time of acid perfusion, and acidity and volume of
38 Orr et al.
the acidic solution placed into the distal esophagus. The aim
of the present study was to investigate the effects of state of
consciousness (sleep vs waking), acid volume, and body
position on the incidence of mid- and proximal esophageal
migration, as well as clearance of acid infused into the distal
esophagus, in healthy individuals.
MATERIALS AND METHODS
Subjects
Fifteen healthy volunteers (seven men, eight women) with a
mean age of 33.6 yr (range, 21– 45 yr) participated in the
study. Before enrolling in the study, all potential subjects
participated in a structured interview and filled out several
screening questionnaires relating to history of gastrointes-
tinal (GI) symptoms, current medication use, and sleep
problems. Individuals with a history of GI disease, clinically
significant complaints of heartburn, medication use, or sleep
abnormalities were excluded from participation.
Experimental Protocol
Distal esophageal perfusions of acid were performed in the
upright and supine positions during waking, and in the
supine position during polysomnographically monitored
sleep. Sleep monitoring consisted of the simultaneous re-
cording of the electroencephalogram (EEG), electrooculo-
gram (EOG), chin electromyogram (EMG), and electrocar-
diogram (ECG). Each subject received a total of six
perfusions of 0.1 hydrochloric acid (HCl): a single perfusion
of each 1 ml and 3 ml in the upright and supine positions
during waking, and a single perfusion of each 1 ml and 3 ml
during nonrapid eye movement (NREM) sleep (stages 2, 3,
or 4). Perfusions were not performed during REM sleep
because REM sleep occupies only a minor portion of the
night, and because there is evidence that arousal thresholds
to external or internal stimuli are markedly decreased during
REM sleep (14).
The esophageal perfusions were accomplished through an
opening in the distal tip of the nasoesophageal probe (i.e.,
end opening), which was located 5 cm above the manomet-
rically determined proximal border of the lower esophageal
sphincter (LES). Esophageal pH was continuously moni-
tored from two pH sensors placed 5 cm apart. One sensor
was located 10 cm orad to the LES in the midesophagus, and
the other was located 15 cm orad to the LES in the proximal
esophagus.
The use of the 1-ml and 3-ml volumes was based on
evidence from pilot data. In the pilot study, we compared
infusions of 1, 3, and 5 ml during waking and sleep in 10
healthy volunteers. The results revealed that the 1-ml and
3-ml volumes constituted the smallest volumes for which
proximal migrations could consistently be demonstrated. By
using these small volumes, we hoped to be able to differ-
entiate migration effects at the two pH sensors in the dif-
ferent experimental conditions with maximal sensitivity be-
cause we found that larger volumes (i.e., 5 ml) produced
AJG – Vol. 95, No. 1, 2000
nearly simultaneous drops in pH at both pH sensors. An-
other reason for choosing these small volumes was that the
repeated infusions of larger volumes led to complaints of
heartburn and frequent awakenings in some individuals.
For each subject, the order of perfusions (1 ml vs 3 ml)
was randomly determined, and that order was accomplished
within each condition (i.e., supine waking, upright waking,
or supine sleep). All subjects were studied initially during
waking in the supine position, followed by the upright
position, and then in the supine position during subsequent
sleep. For both waking and sleep perfusions, at least 20 min
were allowed in between perfusions if no migration oc-
curred. When migration occurred, the next perfusion was
not performed until acid clearance (pH ⬎4) was evident for
a minimum of 15 min.
PROCEDURES. The study lasted from 4 PM to 6 AM. On
the day of the study, subjects arrived at the laboratory at 4
PM, at which time the LES was manometrically determined,
and the nasoesophageal tube was placed. At 6 PM, partic-
ipants were fed a standardized meal. At 7:30 PM, subjects
were prepared for polysomnography, and at approximately
8 PM, the waking perfusions were started. Polysomno-
graphic monitoring during the waking perfusions was ac-
complished to ensure that subjects remained awake through-
out the waking perfusions. Subjects retired to bed at
approximately 10:30 PM, after the waking perfusions were
completed. Both waking and sleep perfusion were remotely
accomplished from an adjacent monitoring room. This was
done to avoid anticipatory reactions to perfusions by the
subjects, such as swallowing during the waking perfusions,
or arousals from sleep due to external events during the
conduct of sleep perfusions. All perfusions were performed
manually at a rate of 1 ml per 10 s. Subjects were allowed
to sleep spontaneously throughout the night. The sleep study
ended at 6 AM, after which time subjects were free to leave
the laboratory.
DATA ANALYSIS. Polysomnographic recordings were
scored according to internationally accepted sleep scoring
criteria (15). Two primary measures were used for analysis
of pH data, each separately analyzed for the two pH sensors
(i.e., mid and proximal): the percentage of perfusions re-
sulting in a migration episode (in all subjects combined),
and mean acid clearance time in minutes. A migration
episode was considered present if esophageal pH decreased
below a pH value of 4 at either of the two pH sensors within
10 min of the perfusion. The percent of perfusions resulting
in a migration episode for each sensor in each condition for
all subjects combined was determined as follows: number of
migration episodes occurring at a sensor in a particular
condition in all subjects combined/total number of perfu-
sions in the same condition in all subjects combined ⫻ 100.
Although the study was originally designed to measure
clearance time in all conditions, the 1-ml perfusions did not
consistently result in migration from the distal perfusion site
AJG – January, 2000
Figure 1. Percentage of acid perfusions (1 ml vs 3 ml) resulting in
migration to the midesophageal pH sensor in the upright and
supine positions during waking. The Cochran’s Test revealed a
significant difference in the distribution of migration episodes (p ⬍
0.05).
to the mid pH sensor. Therefore, the percentage of perfu-
sions that resulted in a migration episode was used as a
primary measure. These data are dependent and dichoto-
mous, which makes them unsuitable for standard statistical
comparisons. Therefore, the Cochran’s test was used for
statistical analysis. This test assesses the equality of corre-
lated (repeated) observations where the dependent variable
is dichotomous (16).
The acid clearance time was considered to be the time (in
seconds) from a drop of esophageal pH ⬍ 4 until its rise to
⬎4 for ⱖ30 s. For statistical analysis of clearance times
data, t tests for dependent samples were used. Clearance
time results are expressed as mean ⫾ SEM.
Additional measures included migration latency, which
was the time (in seconds) from the onset of acid perfusion
to the time a drop in pH to ⬍4 was detected at the mid or
proximal sensor, respectively. To assess migration velocity,
the time (in seconds) was measured from the pH drop at the
midesophageal sensor to detection of a pH drop at the
proximal sensor. To statistically compare migration laten-
cies and migration velocities between the upright and supine
positions, between volumes, and between waking and sleep,
paired t tests were carried out.
RESULTS
Waking
Of all perfusions performed in the waking state, 63% (38 of
60) resulted in migration to the mid sensor, whereas only
23% (14 of 60) resulted in migration to the proximal sensor.
Figure 1 shows the percent of perfusions (i.e.,1 ml vs 3 ml)
resulting in migration to the mid sensor in the upright and
supine positions during waking. Cochran’s test revealed a
significant difference in the distribution of these observa-
tions (p ⬍ 0.05). The results clearly support that volume is
the major determinant of migration to the midesophageal
Proximal Esophageal Acid Perfusions 39
Figure 2. Percentage of acid perfusions (1 ml vs 3 ml) resulting in
migration to the proximal pH sensor in the upright and supine
positions during waking. The Cochran’s Test revealed a significant
difference in the distribution of migration episodes (p ⬍ 0.05).
sensor, as evidenced by the greater percent of migration
with the 3-ml volume, and very small differences between
the upright and supine positions. Figure 2 shows the percent
perfusions migrating to the proximal sensor under the same
conditions depicted in Figure 1. For the proximal sensor,
there was also a significant difference in the distribution of
migration episodes (p ⬍ 0.05). Similar to findings for the
midesophageal sensor, the proximal sensor results support
an effect of volume, but not posture, on proximal migration.
That is, the percent of migrations is relatively similar be-
tween the upright and supine positions for both volumes.
Compared with the smaller volume, the larger volume sub-
stantially enhanced the incidence of migration.
Because the majority of 1-ml perfusions performed in the
waking state did not demonstrate migration, acid clearance
times could be analyzed only for 3-ml perfusions. Statistical
analysis of the clearance times for the midesophageal sensor
(3-ml perfusions) revealed no significant difference between
the upright and supine positions (3.5 ⫾ 0.08 min vs 6.1 ⫾
2.1 min). Also, there was no significant difference between
the clearance times from the proximal and midesophageal
sensors (5.0 ⫾ 0.8 min vs 2.6 ⫾ 0.8 min).
Sleep
Complete sleep perfusion data (i.e., for 1 and 3 ml) were
available for 11 of the 15 participants. Incomplete data
resulted from technical and other data acquisition problems,
such as instances when infusions were accidentally per-
formed during sleep stages other than stages 2, 3, or 4 of
NREM sleep or during an arousal from sleep, or pH record-
ing failures. Of all completed perfusions for 1 ml and 3 ml
performed during sleep, 77% (17 of 22) resulted in migra-
tion to the mid sensor, and 73% (16 of 22 perfusions)
resulted in migration to the proximal sensor. Figure 3 shows
the percent of perfusions (i.e., 1 ml vs 3 ml) resulting in
migration to the midesophageal sensor during supine wak-
ing and sleep. Whether during waking or sleep, volume
clearly enhanced the incidence of migration to the
midesophageal sensor. Cochran’s test showed a significant
40 Orr et al.
Figure 3. Percentages of acid perfusion (1 ml vs 3 ml) resulting in
migration to the midesophageal pH sensor during waking and
sleep. The Cochran’s Test revealed a significant difference in the
distribution of migration episodes (p ⬍ 0.05).
difference in the distribution of these observations (p ⬍
0.05). In addition, the data suggest that migration to the mid
pH sensor was not severely affected by sleep with either the
1- or 3-ml perfusions. Figure 4 shows the percentage of
migrations to the proximal sensor after acid perfusion of 1
ml versus 3 ml. Sleep clearly increased the incidence of
migration to the proximal sensor for both volumes. Coch-
ran’s test supports this interpretation by showing a signifi-
cant difference in the distribution of these observations (p ⬍
0.05). In addition, volume also appeared to increase the
percentage of migration to the proximal sensor independent
of sleep.
Compared with waking, acid clearance times for the 3-ml
perfusions were significantly longer during sleep for both
midesophageal (45.1 ⫾ 6.5 min vs 3.4 ⫾ 1.1 min) and
proximal sensors (36.6 ⫾ 7.3 min vs 2.6 ⫾ 0.6 min) (p ⬍
0.05). A comparison of clearance times from the two sensors
during sleep (3-ml perfusions) revealed that acid clearance
was significantly faster from the proximal esophageal sensor
(36.6 ⫾ 7.3 min) when compared with the midesophageal
sensor (45.1 ⫾ 6.5 min) (p ⬍ 0.05).
Figure 4. Percentages of acid perfusion (1 ml vs 3 ml) resulting in
migration to the proximal pH sensor during waking and sleep. The
Cochran’s Test revealed a significant difference in the distribution
of migration episodes (p ⬍ 0.05).
AJG – Vol. 95, No. 1, 2000
Migration latencies from the perfusion port to the mid-
esophageal and proximal sensors were not calculated for the
upright position (both volumes) or for the 1-ml volume in
the supine position because the number of detected migra-
tions to the proximal sensor were too small (N ⫽ 1, N ⫽ 1,
and N ⫽ 4, respectively). Migration latency and migration
velocity for the 3-ml perfusions from the distal perfusion
port to the midesophageal and proximal sensors were not
affected by state of consciousness. Moreover, during sleep,
acid volume did not significantly affect the migration la-
tency or migration velocity.
DISCUSSION
In this study, we investigated the role of body position,
volume of infused acid, and state of consciousness (sleep vs
waking) on the proximal migration of acid infused into the
distal esophagus. Our results offer important information
about parameters of acid migration in the esophagus. The
data support that in the waking state, volume clearly en-
hances the likelihood of migration from the distal perfusion
site to both midesophageal and proximal pH sensors, and
significantly prolongs clearance time (see Figs. 1 and 2).
The results concerning the role of posture are less clear.
Although the supine position may pose a greater risk than
the upright position, the analysis did not reveal a statistically
significant effect of posture on the incidence of migration
and acid clearance, suggesting a minor role for posture.
However, the small number of observations, which may
have resulted in low statistical power, must be taken into
account when interpreting this finding. Nevertheless, even if
present, the effect of posture appears to be small, and
volume is clearly the most important parameter for migra-
tion in the waking state.
Sleep profoundly affects acid migration. For instance,
during waking, 1-ml perfusions resulted in migration to the
proximal sensor in fewer than 1% of individuals, whereas
this very small volume migrated 15 cm into the proximal
esophagus during sleep in more than 40% of individuals.
These results lend strong support to the role of sleep as a risk
factor facilitating acid migration to the proximal esophagus.
Although migration to the midesophagus does not seem to
be differentially affected by sleep per se, sleep clearly
enhances migration to the proximal esophagus for both 1-ml
and 3-ml perfusions (see Fig. 4). In addition, there is a
marked and significant prolongation of acid clearance from
both sensors during sleep, whereas acid clearance is rela-
tively rapid in the waking state. Volume is also a critical
determinant of proximal acid migration during sleep, with a
marked increase in the incidence of migration after perfu-
sion of the larger volume (see Figs. 3 and 4). Fortunately, as
a homeostatic protective mechanism, proximal acid clear-
ance is faster than noted in the midesophagus.
There are numerous studies supporting the notion that
sleep is a risk factor for pulmonary aspiration, and therefore
by definition for the proximal migration of refluxed gastric
AJG – January, 2000
contents. Huxley and colleagues showed that depressed
consciousness clearly enhanced pharyngeal aspiration, and
Pelligrini et al. demonstrated that individuals with symp-
toms suggestive of proximal acid migration (e.g., a sour
taste in the mouth) had a high incidence of prolonged acid
clearance during the sleeping period (17, 18). Previous
studies from our laboratory utilizing distal esophageal per-
fusion of larger acid volumes documented that sleep delayed
acid clearance, and that arousals from sleep facilitated acid
clearance (11–13). Both the volume of intraesophageal acid
as well as low pH values of the perfusate (a pH of 1 ⬎ 3 ⬎
5) facilitates arousal from sleep (11–13). Collectively, these
data support the notion that depressed state of conscious-
ness, i.e., sleep, alters responsiveness to acid in the esoph-
agus, and that homeostatic protective mechanisms related to
intraesophageal acid clearance facilitate arousal from sleep.
The present study confirms and extends these inferences by
demonstrating that sleep enhances the probability of acid
migration to the proximal esophagus, and also prolongs
mid- and proximal esophageal acid clearance. These find-
ings are consistent with an increased risk of pulmonary
aspiration. Although previous data from our laboratory have
shown that specific mechanisms (i.e., arousal and swallow-
ing responses) are in place to prevent pulmonary aspiration
during sleep, these do not seem to fully compensate for the
detrimental effects of sleep on acid clearance and the like-
lihood of proximal acid migration of even a very small acid
volume, i.e., 1 ml. In the waking state, on the other hand,
there would seem to be little risk of aspiration because there
is nearly simultaneous acid clearance from both the proxi-
mal and the midesophagus.
Although there appears to be adequate justification for
assuming that proximal esophageal or pharyngoesophageal
acid contact is important in producing aerodigestive symp-
toms, the role of body position, however, appears to be
somewhat controversial. Some studies have identified the
supine position as an important risk factor in proximal
esophageal acid contact. The study by Jacob and colleagues
clearly showed that only proximal esophageal acid contact
in the supine position distinguished a group of patients with
reflux laryngitis from a group of patients with erosive esoph-
agitis (9). Similarly, in a study by McNally et al., patients
with chronic hoarseness had increased supine reflux and
prolonged acid clearance in the midesophagus (19). Con-
versely, Shaker and his group noted that pharyngeal acid
contact time was significantly increased in patients with
posterior laryngitis, and this was noted only in the upright
position (4). Although no significant differences were found
in this study with regard to proximal versus distal GER
episodes and acid contact time, other studies have shown
that proximal esophageal acid exposure is noted primarily in
the upright position (2, 20, 21). Our study results suggest
that position has only a small effect on migration, and that
volume and sleep are much more powerful determinants of
proximal acid migration. Clearly, volume is the most im-
portant issue for migration in the waking state, and acid
Proximal Esophageal Acid Perfusions 41
migration to the more proximal esophagus is facilitated by
both increased volume and sleep. In fact, migration to the
proximal esophagus is enhanced during sleep even with the
smallest volumes (1 ml).
The limitations of this study relate to the question of how
well the perfusion model reflects the circumstances of phys-
iological or pathophysiological sleep-associated reflux. Un-
fortunately, the volume of the ‘natural’ sleep-associated
refluxate in healthy individuals or patients is unknown, and
therefore it is not clear whether the two volumes chosen in
this study are physiological. Similarly, the ‘natural’ reflux-
ate may differ in other characteristics, for example in acid-
ity, from the infused acid in this study, and the chosen
perfusion rate of 10 s for 1 ml may differ from the speed of
physiological reflux events. However, other studies have
also used very small, in fact smaller, amounts of acid, in an
attempt to study the mechanisms of acid clearance in the
human esophagus (22), and one may assume that the esoph-
ageal clearance mechanisms are the same for various vol-
umes and various acidities. Indirect support for the validity
of the perfusion model in the study of nighttime reflux
comes from the fact that, like our healthy study volunteers,
patients undergoing 24-h pH monitoring do not typically
wake up to report heartburn despite long periods of low
esophageal pH. This may be due to a less acidic pH of the
refluxate, and/or a desensitization of the esophagus, and/or
to the possibility that the volume of their refluxate may in
fact be as small as those used in our study.
ACKNOWLEDGMENT
This research was supported by the Thomas N. Lynn Insti-
tute for Healthcare Research.
Reprint requests and correspondence: William C. Orr, Ph.D.,
Thomas N. Lynn Institute for Healthcare Research, 5300 North
Independence, Suite 130, Oklahoma City, OK 73112.
Received May 6, 1999; accepted Sep. 3, 1999.
REFERENCES
1. Pack AI. Acid: A nocturnal bronchoconstrictor? Am Rev
Respir Dis 1990;141:1391–2.
2. Allen CJ, Newhouse MT. Gastroesophageal reflux and chronic
respiratory disease. Am Rev Respir Dis 1984;129:645–7.
3. Wiener GJ, Koufman JA, Wu WC, et al. Chronic hoarseness
secondary to gastroesophageal reflux disease: Documentation
with 24-h ambulatory pH monitoring. Am J Gastroenterol
1989;84:1503– 8.
4. Shaker R, Milbrath M, Ren J, et al. Esophagopharyngeal
distribution of refluxed gastric acid in patients with reflux
laryngitis. Gastoenterology 1995;109:1575– 82.
5. Irwin RS, Zawacki JK, Curley FJ, et al. Chronic cough as the
sole presenting manifestation of gastroesophageal reflux. Am
Rev Respir Dis 1989;140:1294 –300.
6. Patti MG, Debas HT, Pellegrini CA. Esophageal manometry
and 24-hour pH monitoring in the diagnosis of pulmonary
42 Orr et al.
aspiration secondary to gastroesophageal reflux. Am J Surgery
1992;163:401– 6.
7. Ing AJ, Ngu MC, Breslin AB. Chronic persistent cough and
clearance of esophageal acid. Chest 1992;102:1668 –71.
8. Ing AJ, Ngu MC, Breslin AB. Pathogenesis of chronic per-
sistent cough associated with gastroesophageal reflux. Am J
Respir Crit Care Med 1994;149:160 –7.
9. Jacob P, Kahrilas PJ, Herzon G. Proximal esophageal pH-
metry in patients with ‘reflux laryngitis’. Gastroenterology
1991;100:305–10.
10. Kamel PL, Kahrilas PJ. Outcomes of antireflux therapy for the
treatment of chronic laryngitis. Ann Otol Rhinol Laryngol
1995;104:550 –5.
11. Orr WC, Robinson MG, Johnson LF. The effect of esophageal
acid volume on arousals from sleep and acid clearance. Chest
1991;99:351– 4.
12. Orr WC, Johnson LF. Responses to different levels of esoph-
ageal acidification during waking and sleep. Dig Dis Sci
1998;43:241–5.
13. Orr WC, Johnson LF, Robinson MG. The effect of sleep on
swallowing, esophageal peristalsis, and acid clearance. Gas-
troenterology 1984;86:814 –9.
14. Steriade M. Brain electrical activity and sensory processing
during waking and sleep states. In: Kryger M, Roth T, Dement
W, eds. Principles and practice of sleep medicine.
Philadelphia: W.B. Saunders, 1994:105–24.
AJG – Vol. 95, No. 1, 2000
15. Rechtschaffen A, Kales A. A manual of standardized ter-
minology, techniques, and scoring system for sleep stages
of human subjects. National Institutes of Health Publication
No. 204. Washington, DC: U.S. Government Printing Of-
fice 1968.
16. Cochran WG. The comparison of percentages in matched
samples. Biometrika 1950;37:256 – 66.
17. Huxley EJ, Viroslav J, Gray WR, et al. Pharyngeal aspiration
in normal adults with depressed consciousness. Am J Med
1978;64:564 – 8.
18. Pelligrini CA, DeMeester TR, Johnson LF, et al. Gastroesoph-
ageal reflux and pulmonary aspiration: Incidence, functional
abnormality, and results of surgical therapy. Surgery 1979;86:
110 –9.
19. McNally PR, Maydonovitch CL, Prosek RA, et al. Evaluation
of gastroesophageal reflux as a cause of idiopathic hoarseness.
Dig Dis Sci 1989;34:1900 – 4.
20. Ossakow SJ, Elta G, Bogdasarian R, et al. Esophageal reflux
and dysmotility as the basis for persistent cervical symptoms.
Ann Otol Rhinol Laryngol 1987;96:387–92.
21. Castal OL, Castell JA, Castell DO. Frequency and site of
gastroesophageal reflux in patients with chest symptoms.
Chest 1994;106:1793–96.
22. Shaker R, Kahrilas PJ, Dodds WJ, et al. Oesophageal clear-
ance of small amounts of equal or less than one millilitre of
acid. Gut 1992;33:7–10.