Mechanisms of LOS incompetence in patients with symptomatic gastro-oesophageal reflux

Abstract

Patterns of lower oesophageal sphincter (LOS) function associated with the onset of 644 reflux episodes were recorded and analysed in 67 patients referred for evaluation of gastro-oesophageal reflux (GOR). Patients were studied recumbent, for one hour before and four hours after a standard meal. Transient LOS relaxation was the most prevalent mechanism and overall accounted for 82% of reflux episodes. With increasing severity of oesophagitis, absent basal LOS pressure became a progressively more common mechanism, accounting for 23% of episodes in the patients with severe oesophagitis. Patients commonly exhibited more than one mechanism. The timing of most (69%) LOS relaxations associated with reflux was not compatible with triggering by swallowing. Prolonged transient LOS relaxations were associated with inhibition of oesophageal peristalsis suggesting that this response is produced by neural inhibition. This study suggests the primary importance of transient LOS relaxations as the cause of GOR across the spectrum of severity of reflux disease.

Full text

Gut,
1988,
29,
1020-1028
Alimentary
tract
and
pancreas
Mechanisms
of
lower
oesophageal
sphincter
incompetence
in
patients
with
symptomatic
gastrooesophageal
reflux
J
DENT,
R
H
HOLLOWAY,
J
TOOULI,
AND
W
J
DODDS
From
the
Departments
of
Medicine
and
Surgery,
Flinders
Medical
Centre,
Bedford
Park,
South
Australia
SUMMARY
Patterns
of
lower
oesophageal
sphincter
(LOS)
function
associated
with
the
onset
of
644
reflux
episodes
were
recorded
and
analysed
in
67
patients
referred
for
evalution
of
gastro-
oesophageal
reflux
(GOR).
Patients
were
studied
recumbent,
for
one
hour
before
and
four
hours
after
a
standard
meal.
Transient
LOS
relaxation
was
the
most
prevalent
mechanism
and
overall
accounted
for
82%
of
reflux
episodes.
With
increasing
severity
of
oesophagitis,
absent
basal
LOS
pressure
became
a
progressively
more
common
mechanism,
accounting
for
23%
of
episodes
in
the
patients
with
severe
oesophagitis.
Patients
commonly
exhibited
more
than
one
mechanism.
The
timing
of
most
(69%)
LOS
relaxations
associated
with
reflux
was
not
compatible
with
triggering
by
swallowing.
Prolonged
transient
LOS
relaxations
were
associated
with
inhibition
of
oesophageal
peristalsis
suggesting
that
this
response
is
produced
by
neural
inhibition.
This
study
suggests
the
primary
importance
of
transient
LOS
relaxations
as
the
cause
of
GOR
across
the
spectrum
of
severity
of
reflux
disease.
For
many
years
gastrooesophageal
reflux
(GOR)
was
generally
believed
to
result
from
lower
oesophageal
sphincter
(LOS)
incompetence
caused
by
defective
basal
LOS
tone.'`
This
concept,
however,
fails
to
account
for
the
substantial
proportion
of
patients
with
reflux
disease
in
whom
resting
LOS
pressure
is
normal.7
A
recent
study
has
shown
that
GOR
in
normal
subjects
occurs
almost
exclusively
as
a
result
of
transient
LOS
relaxation,
rather
than
from
defective
basal
LOS
pressure.8
In
a
subsequent
study
of
10
selected
patients
with
erosive
peptic
oeso-
phagitis,
transient
LOS
relaxation
accounted
for
65%
of
reflux
episodes,
the
remainder
of
reflux
episodes
occurring
during
prolonged
periods
of
absent
or
low
basal
LOS
pressure.9
The
aims
of
the
present
study
were
to:
(a)
investigate
possible
varia-
tion
in
the
mechanisms
of
GOR
within
a
large
group
of
patients
exhibiting
a
spectrum
of
severity
of
reflux
disease:
and
(b)
analyse
in
detail,
patterns
of
oeso-
Address
for
correspondence:
Dr
John
Dent,
Director,
Gastroenterology
Unit,
Royal
Adelaide
Hospital,
North
Terrace,
Adelaide,
South
Australia,
5000.
Received
for
publication
29
February
1988.
phageal
motility
associated
with
reflux
events
in
an
effort
to
gain
insight
into
the
mechanism
of
transient
LOSRs.
Methods
STUDY
GROUP
The
study
protocol
outlined
below
was
approved
by
the
Ethical
Review
Committee
of
Flinders
Medical
Centre
in
March
1977.
The
patients
included
in
the
study
were
referred
because
they
represented
a
problem
in
clinical
management.
The
reasons
for
referral
included
troublesome
or
atypical
symptoms,
atypical
oesophageal
ulceration,
and
preoperative
assessment
for
antireflux
surgery.
Patients
with
a
previous
vagotomy,
gastric
resection,
gastro-
enterostomy,
or
antireflux
surgery
were
excluded.
This
report
describes
the
findings
in
67
patients
in
whom
technically
satisfactory
oesophageal
mano-
metric
and
pH
recordings
were
obtained.
Studies
in
23
other
patients
were
unsatisfactory
because
of
technical
failure
of
pH
electrodes,
difficulties
with
intubation,
or
failure
of
the
subject
to
complete
the
1020

Mechanisms
of
LOS
incompetence
full
monitoring
period
because
of
intolerance
to
the
catheter
assembly.
EVALUATION
OF
PATIENTS
Patients
were
assessed
initially
by
history
and
physical
examination.
An
upper
gastrointestinal
endoscopy
was
carried
out
if
not
done
previously
and
the
results
clearly
documented
within
six
months
of
referral.
Oesophageal
motility
and
pH
studies
were
done
at
least
one
week
after
endoscopy.
When
a
clear
association
between
symptoms
and
spontaneous
oesophageal
acidification
was
not
established
during
the
motility
and
pH
study,
a
Bernstein
test
was
done
immediately
after
the
completion
of
the
manometric
study.
Based
on
the
clinical
history,
endoscopic
findings,
and
the
relationship
of
symptoms
to
oesophageal
acidification,
the
patients
were
classified
into
four
groups:
those
judged
not
to
have
symptomatic
GOR
(group
A),
and
those
diagnosed
as
having
sympto-
matic
GOR
of
three
levels
of
severity
(groups
B,
C,
D).
Group
A
patients
had
atypical
symptoms,
a
normal
mucosal
appearance
at
endoscopy,
with
symptoms
that
could
not
be
reproduced
during
oesophageal
acidification.
Group
B
patients
had
no
endoscopic
evidence
of
oesophageal
erosions
or
ulceration
but
their
symptoms
were
clearly
provoked
by
oesophageal
acidification
during
either
spontane-
ous
acid
reflux
(six
patients)
or
subsequent
acid
infusion
(Bernstein)
testing
(14
patients).
Group
C
patients
exhibited
patchy
erosions
or
non-confluent
ulceration.
Group
D
patients
had
evidence
of
confluent
ulceration,
columnar
epithelium
(five
patients),
or
a
chronic
oesophageal
ulcer.
MANOMETRIC
AND
OESOPHAGEAL
pH
RECORDINGS
Oesophageal
motility
was
monitored
using
an
8-
lumen
manometric
assembly
that
incorporated
a
sleeve
sensor.`
The
6
cm
long
sleeve
sensor
was
positioned
so
that
it
straddled
the
LOS.
Two
side
holes
1
and
3
cm
below
the
distal
margin
of the
sleeve
recorded
intragastric
pressure.
Side
holes
at
the
upper
margin
of
the
sleeve
and
7,
14,
and
21
cm
more
proximally
monitored
motor
activity
of
the
oeso-
phageal
body.
The
most
proximal
side
hole,
located
in
the
pharynx,
monitored
swallows.
Each
lumen
was
connected
in
series
with
a
Bell
and
Howell
(no
4-3271)
pressure
transducer
and
perfused
with
distil-
led
water
by
a
low
compliance
pneumo-hydraulic
pump.
The
LOS
and
gastric
manometric
channels
were
perfused
at
0-5
mi/min.
The
oesophageal
body
and
pharyngeal
channels
were
perfused
at
0.
13
ml/min,
a
rate
selected
to
yield
reliable
recognition
of
the
occurrence
of
oesophageal
body
and
pharyngeal
contractions
yet
minimise
alterations
to
the
patterns
of
oesophageal
acid
clearance
or
the
rate
of
primary
or
secondary
peristalsis.
Oesophageal
pH
was
monitored
using
either
a
Beckman
(no
39042)
or
a
Microelectrodes
(no
M1-506)
unipolar
intraluminal
pH
electrode
positioned
5
cm
above
the
proximal
margin
of
the
LOS.
The
electrode
was
calibrated
in
buffers
of
pH
4
and
7
before
and
after
each
study.
If
electrode
drift
in
excess
of
0-6
pH
units
occurred,
the
study
was
rejected.
Correction
for
electrode
drifts
of
less
than
0.6
pH
units
was
made
on
the
assumption
that
the
drift
was
linear
during
the
study.
Signals
from
the
pressure
transducers
and
pH
meter
(Electronics
Instuments
Ltd,
no
2320)
were
processed
and
recorded
on
a
Devices
M19
chart
recorder
at
a
paper
speed
of
80
mm/minute.
STUDY
PROTOCOL
Patients
were
studied
after
a
four
hour
fast.
This
fasting
period
was
selected
with
the
aim
of
achieving
a
late
fed
state
of
motor
activity
for
the
first
hour
of
the
study
and
thereby
avoid
the
cyclical
changes
of
LOS
pressure
associated
with
the
phases
of
the
migrating
motor
complex,
that
occur
during
fasting.
"
Anticholinergic
or
antisecretory
drugs
were
stopped
36
hours
before
the
study,
antacids
were
not
allowed
for
12
hours
before
the
start
of
recordings,
and
smoking
was
not
permitted
on
the
day
of
the
study.
The
patients
remained
recumbent,
either
supine
or
on
their
side,
except
while
eating
the
meal
and
when
voiding.
After
one
hour
of
baseline
recordings,
the
patients
then
ate
a
standard
meal
consisting
of
soft
meat
or
fish,
mashed
potatoes,
cooked
vegetables,
ice
cream
and
180
ml
of
milk.
After
completion
of
the
meal,
oesophageal
motility
and
pH
recordings
were
continued
for
an
additional
4
hours.
Patients
were
not
permitted
to
doze
during
the
recording
because
of the
effect
of
sleep
on
patterns
of
reflux.8
12
Patients
indicated
to
the
observer
when
they
had
symptoms
such
as
heartburn,
regurgitation
or
chest
pain,
and
this
was
noted
on
the
chart.
DATA
ANALYSIS
The
oesophageal
pH
tracing
was
analysed
manually
to
determine
the
time
that
pH
was
less
than
4.
Reflux
episodes
were
also
identified
and
counted,
being
defined
as
either
a
drop
of
oesophageal
pH
below
4
for
at
least
four
seconds,
or
if
oesophageal
pH
was
already
below
4,
as
a
decrease
of
at
least
1
pH
unit
sustained
for
at
least
four
seconds.
The
duration
of
acidification
after
reflux
episodes
was
measured
as
the
time
taken
for
oesophageal
pH
to
return
to
4.
Excluded
from
rigid
application
of
these
criteria
was
the
brief
upward
spike
of
pH
associated
with
peristalsis.
This
spike
could
be
differentiated
from
a
reflux
event
by
its
shape,
the
pattern
of
return
of
1021

Dent,
Holloway,
Toouli,
and
Dodds
oesophageal
pH
to
near
preswallow
levels,
and
its
association
with
peristalsis.
Infrequently,
oseo-
phageal
pH
drifted
downwards
during
a
period
of
several
minutes
and
dropped
below
pH
5
and
occa-
sionally
pH
4.
These
pH
drifts
were
included
in
analysis
of
the
duration
of
oesophageal
acid
exposure
but
were
not
scored
as
reflux
episodes.
The
onset
of
the
usually
abrupt
reduction
of
oesophageal
pH
associated
with
GOR
was
used
as
the
reference
for
analysis
of
the
motility
events
associated
with
reflux.
For
all
reflux
episodes
the
following
variables
of
motility
were
measured:
(1)
LOS
pressure
at
the
onset
of
acid
reflux;
(2)
the
pattern
of
LOS
pressure
for
30
seconds
before
reflux;
and
(3)
the
pattern
of
pharyngeal
and
oesophageal
motility
for
30
seconds
before
the
onset
of
reflux.
Basal
end
expiratory
LOS
pressure,
referenced
to
intragastric
pressure,
was
determined
at
15
minute
intervals
by
taking
a
one
minute
visual
mean
of
the
tracing.
Overall
mean
values
of
LOS
pressure
were
derived
for
each
patient
for
the
one
hour
before
and
four
hours
after
the
meal.
Statistical
analysis
of
the
rate
of
reflux
episodes
and
the
duration
of
oesophageal
acid
exposure
used
the
Wilcoxon's
signed-rank
test
and
the
Mann-Whitney
U-test.
Differences
in
the
proportions
of
the
reflux
episodes
associated
with
the
major
mechanism
of
LOS
incompetence
were
assessed
by
X2
analysis.
Values
for
basal
LOS
pressure
were
assessed
by
analysis
of
variance
and
the
Student's
t
test
for
unpaired
values.
Results
ACID
REFLUX
PATTERN
Overall,
the
results
of
oesophageal
pH
monitoring
supported
the
grouping
of
patients
based
on
the
clinical
and
endoscopic
assessment
of
the
severity
of
GOR
disease.
Patients
in
Groups
B,
C,
and
D
judged
to
have
symptomatic
GOR,
had
significantly
more
postprandial
oesophageal
acid
exposure
than
did
patients
in
Group
A
judged
not
to
have
symptomatic
reflux
disease
(p<001)
(Table).
The
patients
with
reflux
disease
exhibited
a
progressive
increase
in
postprandial
reflux
rate
and
oesophageal
acid
exposure
with
increasing
severity
of
oesophagitis
but
differences
among
these
latter
three
groups
were
not
statistically
significant.
The
reflux
rate
and
duration
of
acid
exposure
increased
significantly
after
the
meal
in
all
groups.
RELATIONSHIP
OF
LOS
PRESSURE
TO
REFLUX
Lower
oesophageal
sphincter
pressure
during
the
period
surrounding
each
reflux
episode
could
be
ana-
lysed
in
644
of
691
episodes
(93%).
The
remaining
7%
could
not
be
analysed
because
of
movement
arte-
Table
1
Rate
of
reflux
episodes
and
duration
of
oesophageal
acid
exposure
Rate
of
reflux
episodes
Duration
of
acid
exposure
Patient
group
Preprandial
Postprandial
Preprandial
Postprandial
(no)
(nlh)
(nlh)
(%)
(%)
GroupA
0.2(0.1)
0.5(0.
1)*
0.6(0.4)
0.9(0.4)*
(n=
15)
GroupB
1.1(0.8)
2.2(0.2)t
1.3(1
0)
9-0(3.8)tt
(n=20)
GroupC
1.1(0.4)
2-9(05)tt
2.3(1.2)
10-7(2-8)tl:
(n=
18)
GroupD
1.1
(04)t
4-9(1.0)tt
17
(09)t
18
6(5
1)t:
Data
expressed
as
x
(SE).
Differed
significantly
from
preprandial
value:
*p<0.05,
tp<0
01.
Differed
significantly
from
Group
A
value:
tp<001.
fact
or
several
motor
events
occurring
within
a
few
seconds
surrounding
the
onset
of
the
reflux
event.
A
consistent
relationship
existed
between
LOS
pressure
and
the
occurrence
of
reflux.
In
95%
of
reflux
episodes
LOS
pressure
was
undetectable
(<2
mmHg)
at
the
onset
of
the
pH
drop.
The
undetect-
able
LOS
pressure
occurred
through
two
main
mechanisms.
In
82%
of
episodes
the
LOS
underwent
complete
relaxation
against
a
background
of
easily
recordable
basal
LOS
pressure
(type
1)
(Fig.
1).
In
13%
of
episodes,
undetectable
basal
LOS
pressure
existed
for
at
least
30
seconds
before
the
reflux
event
(type
2).
Mean
resting
LOS
pressure
before
type
1
reflux
episodes
was
10.1
(1.0)
mmHg,
whereas
that
before
type
2
reflux
episodes
was
<2
mmHg.
In
the
5%
of
instances
when
reflux
occurred
in
the
presence
of
detectable
LOS
pressure
(type
3)
the
reflux
occurred
simultaneous
with
pressure
transients
generally
caused
by
deep
inspiration
but
occasionally
by
coughing
or
straining
(Fig.
2).
Resting
LOS
pressure
at
the
time
of
these
pressure
transients
was
almost
invariably
low
(<5
mmHg),
and
there
was
usually
manometric
evidence
of
an
hiatal
hernia.
Reflux
was
not
observed
when
strains
occurred
while
LOS
pressure
was
>10
mmHg.
The
proportion
of
reflux
episodes
associated
with
the
three
types
of
reflux
mechanism
differed
signifi-
cantly
among
the
study
groups
(X2=55.21
p<001)
(Fig.
3).
The
Group
A
patients
refluxed
exclusively
via
the
type
1
mechanism.
This
mechanism
was
also
the
most
common
pattern
associated
with
reflux
episodes
occurring
in
patients
with
reflux
disease.
With
increasing
severity
of
reflux
disease,
however,
absent
basal
LOS
pressure
(type
2)
became
increas-
ingly
more
important
and
accounted
for
23%
of
reflux
episodes
in
group
D.
The
three patterns
were
not
mutually
exclusive
in
any
given
patient.
Even
though
many
patients
exhibited
periods
of
absent
1022

Mechanisms
of
LOS
incompetence
Oesoph
pH
6]
mm
Hg
Pharynx
40]
O0]
801
80-
Oesoph
body
O0
0-
LOS
80-
0
Qesoph
pH
23
mmHg
Pharynx
20
Oesoph
body
Stomach
40]
Time
(min)
Fig.
1
Gastro-oesophageal
reflux
occurring
during
a
spontaneous
transient
lower
oesophageal
sphincter
(LOS)
relaxation.
Abrupt
relaxation
of
the
LOS
occurred
without
any
apparent
antecedent
motor
event
in
the
pharynx,
oseophageal
body
orstomach.
Shortly
after
LOS
relaxation
was
complete,
the
pH
electrode
indicated
the
occurrence
of
acid
reflux.
LOS
pressure,
particularly
in
the
first
post
prandial
hour,
these
patients
also
had
periods
of
moderate
LOS
pressure,
often
in
excess
of
10
to
15
mmHg,
most
commonly
during
the
third
and
fourth
post-
prandial
hours.
In
the
presence
of
measurable
LOS
pressure,
however,
reflux
continued
to
occur
because
of
LOSRs.
Only
one
patient
refluxed
exclusively
because
of
undetectable
LOS
pressure.
PATTERNS
OF
LOS
RELAXATION
ASSOCIATED
WITH
REFLUX
Detailed
analysis
of
the
timing
of
motor
events
in
the
pharynx,
oesophageal
body
and
LOS,
in
relation
to
the
onset
of
type
1
reflux
episodes,
revealed
6
distinct
patterns
of
LOSR.
The
most
common
pattern
was
an
abrupt
LOSR
that
occurred
without
swallowing
or
any
detectable
motor
activity
in
the
oesophageal
body
(spontaneous,
transient
LOSR)
(Fig.
1),
and
accounted
for
31%
of
type
1
episodes.
These
LOSRs
had
a
mean
duration
of
14-9
(0.65)
s.
Almost
as
common
(22%)
was
a
LOSR
after
a
normal
peri-
80
oi
80.
80]
O-.
LOS
401
01
Stomach
401
O0
Time
(min)
Fig.
2
Gastro-oesophageal
reflux
occurring
in
the
presence
of
detectable
basal
lower
oesophageal
sphincter
(LOS)
pressure
and
a
hiatal
hernia.
In
this
example,
the
oesophagus
is
short
and
the
proximal
side
hole
that
is
normally
in
the
proximal
oesophageal
body
is
in
the
pharynx.
Both
gastric
pressure
tracings
show
an
intrathoracic
pattern
of
respiratory
oscillation
indicating
the
presence
of
an
hiatal
hernia.
Reflux
occurs
during
a
large
drop
in
intrathoracic
pressure
caused
by
deep
inspiration.
This
drop
appears
to
be
transmitted
to
the
LOS
and
to
momentarily
negate
the
pressure
barrier
created
by
LOS
pressure.
staltic
sequence
(postswallow
transient
LOSR),
in
which
the
LOS
showed
a
normal
contraction
after
swallow
induced
relaxation,
but
then
exhibited
immediate
complete
relaxation
(Fig.
4).
Occasion-
ally
(5%)
similar
transient
LOSRs
were
observed
after
secondary
peristalsis
or
spontaneous
synchro-
nous
contractions
in
the
distal
oesophageal
body.
A
minority
(28%)
of
LOSRs
were
preceded
by
a
pharyngeal
swallow
signal.
Reflux
episodes
rarely
occurred
during
LOSRs
associated
with
a
normal
complete
peristaltic
sequence
(swallow
induced
LOSR).
Although
swallow
induced
LOSRs
accounted
for
13%
of
type
1
reflux
episodes
overall,
31
of
the
65
episodes
occurred
in
one
patient.
A
proportion
(15%)
of
reflux
episodes,
however,
did
1023
1'I

Dent,
Holloway,
Toouli,
and
Dodds
100
80
_s
U)
60
[L4,
20
0-
(22)
A
(167)
(11
B
Patient
group
LOSR
Absent
LOSP
S
Strain
a
92)
C
X_
c
(263)
D
Fig.
3
Mechanisms
of
refiux
among
the
four
patient
groups.
Each
bar
represents
the
percentage
of
total
refiux
episodes.
The
numbers
in
parentheses
indicate
the
total
number
of
reflux
episodes
for
each
group.
occur
during
LOSRs
associated
with
defective
oeso-
phageal
peristalsis
(failed
primary
peristalsis),
in
which
swallowing
appeared
to
induce
LOS
relaxation
but
either
failed
to
elicit
any
detectable
peristalsis,
or
the
peristaltic
wave
failed
to
traverse
the
entire
oesophageal
body.
These
LOSRs
were
of
longer
duration
than
those
associated
with
normal
peri-
stalsis.
Rarely
(3%),
reflux
occurred
during
LOSRs
associated
with
a
salvo
of
swallows
in
rapid
succes-
sion
(multiple
swallows).
Lastly,
reflux
episodes
were
associated
occasionally
with
a
gradual
loss
of
basal
LOS
pressure
during
an
interval
of
up
to
30
seconds.
These
episodes,
classified
as
LOS
pressure
drifts
as
opposed
to
LOSRs,
accounted
for
11%
of
type
1
reflux
episodes
and
invariably
occurred
in
a
setting
of
low
basal
LOS
pressure
of
around
5-10
mmHg.
The
distribution
of
mechanisms
underlying
type
1
reflux
episodes
among
the
different
patient
groups
is
depicted
in
Figure
5.
Although
there
were
no
signifi-
cant
differences
amongst
the
groups,
with
increasing
severity
of
reflux
disease
there
was
a
tendency
for
a
greater
proportion
of
reflux
episodes
to
result
from
slow
downward
LOS
pressure
drifts
and
swallow
induced
LOSRs.
Reflux
episodes
that
occurred
when
baseline
intra-
oesophageal
pH
was
less
than
4
were
associated
with
the
same
patterns
of
LOS
pressure
changes
as
those
observed
with
other
reflux
episodes.
Downward
drifts
in
oesophageal
pH,
however,
occurred
without
any
associated
LOS
relaxation
or
straining.
They
usually
occurred
in
a
setting
of
resting
intra-
esoph
pH
2
mmHg
Phorynx
4o3
40
80
Oesoph
]
body
0i
80.
80.
LOS
]
iV~
Stomach
400
Time
(min)
Fig.
4
Two
episodes
of
gastrooesophageal
reflux.
Thefirst
reflux
episode
occurred
during
a
prolonged
spontaneous
lower
oesophageal
sphincter
relaxation
(LOSR).
A
swallow
(arrow),
that
occurs
after
the
onset
of
the
LOSR,
initiates
a
peristaltic
sequence
that
fails
to
propagate
along
the
complete
length
of
the
esophagus.
The
second
episode
occurred during
a
transient
LOS
relaxation
(LOSR)
immediately
after
completion
of
a
normal
primary
peristaltic
sequence
(postswallow
LOSR).
oesophageal
pH
<5.0
and
often
during
a
prolonged
interval
without
spontaneous
swallowing.
OESOPHAGEAL
BODY
MOTOR
FUNCTION
DURING
TRANSIENT
LOSRS
Initial
analysis
of the
mechanisms
underlying
spon-
taneous
transient
LOSRs
suggested
that
during pro-
longed
LOSRs,
swallows
did
not
trigger
a
complete
peristaltic
sequence.
This
impression
was
therefore
evaluated
systematically.
To
avoid
effects
on
peri-
staltic
success
by
surrounding
swallows,
and
to
ensure
that
the
LOSR
was
of
sufficient
duration
to
encompass
an
entire
peristaltic
sequence,
only
transient
LOSRs
with
the
following
features
were
analysed:
(i)
duration
greater
than
15
seconds,
(ii)
a
single
swallow
occurring
after
the
onset
of
complete
LOS
relaxation,
and
(iii)
the
swallow
being
evaluated
was
separated
by
at
least
15
seconds
from
adjacent
swallows.
As
a
control,
the
first
swallow
that
occur-
red
more
than
one
minute
after
termination
of
the
LOSR,
and
that
was
separated
by
at
least
15
seconds
from
adjacent
swallows,
was
analysed
for
complete-
1024
t
-A-

Mechanisms
of
LOS
incompetence
(a)
pH
]
Pharynx
]
]
Oescph
]
body
1
-
-
-
LOS
]-
_
A
B
C
D
0
6lo
14°/o
13%.
(e)
pH
]
-<
Pharynx]
h<
]
Oesoph
___A
body
LOS
1
,,
-
(b)
-I-
=.
390/0o
.251
31/
(f)
A4
3`lo
(f)
(c)
A-i
L
LA-+1
LJ-~
L-J_-1
i-
32%.
131.
300/o
23`/.
(g)
~AA
-1
1/1
A
0
0
B
2`%
110%0
C
12`/o
3`/0
D
23`%
3`%0
(d)
--1
271.
24%.
131
6/%
uOS'
LOSA
0
50/0
3`/o
10/0
Fig.
5
Schemashowingthedifferentpatterns
of
oesophageal
and
lower
oesophageal
sphincter
(LOS)
motor
function
associated
with
type
1
reflux
events.
The
numbers
below
each
pattern
indicate
the
percentage
of
total
type
1
episodes
for
each
pattern
for
each
group.
The
broken
horizontal
line
indicates
intragastric
pressure.
The
vertical
broken
line
indicates
the
onset
of
the
reflux
event.
(a)
LOS
pressure
drift,
(b)
spontaneous
transient
LOS
relaxation
(LOSR),
(c)
LOSR
occurring
immediately
aftera
normal
peristaltic
sequence
(postswallow
transient
LOSR),
(d)
LOSR
associated
with
afailedprimary
peristalticsequence,
(e)
reflux
during
a
normalperistalticsequence,
(f)
LOSR
after
spontaneous
synchronous
contractions
in
the
distal
oesophagus,
(g)
LOSR
induced
by
multipleswallows.
ness
of
the
peristaltic
sequence.
Twenty
three
LOSRs
in
15
patients
fitted
the
above
criteria.
Mean
duration
of
these
LOSRs
was
23
(1)
second,
and
20
of
the
23
LOSRs
were
associated
with
a
reflux
episode.
Swallows
occurring
during
these
LOSRs
triggered
complete
peristaltic
sequences
in
only
two
instances
(9%)
compared
with
100%
of
instances
with
control
swallows
(p<0.01)
(Fig.
6).
In
the
other
91%
of
swallows
during
these
prolonged
LOSRs,
the
peri-
staltic
wave
was
usually
observed
at
only
the
most
proximal
oesophageal
lead.
BASAL
LOS
PRESSURE
An
inverse
relationship
existed
between
basal
LOS
pressure
and
the
severity
of
reflux
disease.
However,
During
LOSR
Control
25
50
75
100
Sw(al
lows
with
peristoltic
response
("/,)
Fig.
6
Analysis
of
23
primary
peristalticsequences
during
transient
lower
oesophageal
sphincter
relaxation
(LOSR).
The
horizontal
bars
indicate
the
percentage
ofperistaltic
sequences
in
which
a
contraction
was
recorded
at
that
recordingsite.
The
asterisk
indicates
that
the
most
proximal
recordingsite
was
located
in
the
proximal
oesophagus
in
only
five
instances.
In
the
rest,
the
oesophageal
body
was
too
short
to
permit
allfour
recording
sites
to
be
in
the
oesophageal
body.
only
mean
basal
LOS
pressure
in
Group
D
patients
(9.7
(2.1)
mmHg)
differed
significantly
from
that
in
group
A
patients
(16.0
(2.5)
mmHg,
p<0.01).
Basal
LOS
pressure
decreased
immediately
after
the
meal
in
all
groups,
however,
a
tendency
existed
for
LOS
pressure
to
return
to
and
exceed
fasting
values
after
the
second
postprandial
hour.
Discussion
In
this
study
of
a
large
heterogeneous
group
of
patients
with
symptomatic
GOR
we
have
examined
in
detail
the
oesophageal
motor
events
surrounding
episodes
of
acid
reflux.
Our
findings
indicate
that
transient
LOSRs
are
the
major
mechanism
of
reflux
in
patients
with
severe
ulcerative
peptic
oesophagitis
as
well
as
symptomatic
GOR
disease.
The
proportion
of
reflux
episodes
occurring
via
this
mechanism
in
patients
with
erosive
of
ulcerative
eosophagitis
is
similar
to
the
55%
we
observed
in
our
previous
study
of 10
patients
with
oesophagitis
of
comparable
severity.9
The
finding
that
patients
judged
not
to
have
symptomatic
GOR
refluxed
exclusively
during
transient
LOSRs
is
in
keeping
with
the
findings
of
our
previous
study
in
normal
subjects.'
Furthermore,
the
mix
of
patterns
of
transient
LOSR,
the
rate
of
reflux
.................................
.EM-
..
...
...............
...................
..............
............
71
....................
.
........
L-
__-
_-_-
1025

Dent,
Holloway,
Toouli,
and
Dodds
episodes
and
the
duration
of
acid
exposure
were
comparable
with
those
in
the
normal
subjects.
Thus
whilst
the
patients
in
Group
A
are
not
strictly
control
subjects,
they
are
nevertheless
a
useful
control
group
with
which
to
compare
to
other
patients
with
reflux
disease.
As
noted
by
others,5
basal
LOS
pressure
was
lower
in
patients
with
severe
degrees
of
oesophagitis.
Additionally,
absent
basal
LOS
pressure
became
a
progressively
more
important
mechanism
of
reflux
the
more
severe
the
oesophagitis.
The
actual
propor-
tion
of
reflux
episodes
that
occurred
by
this
mecha-
nism
was
small,
however,
even
in
patients
with
severe
oesophagitis,
and
only
one
patient
refluxed
exclu-
sively
by
this
means.
In
all
patients
who
had
absence
of
basal
pressure,
this
was
an
intermittent
phenome-
non,
but
they
continued
to
reflux
during
periods
of
detectable
LOS
pressure
because
of
LOSRs.
This
finding
provides
a
possible
explanation
for
the
failure
of
drugs
that
raise
basal
LOS
pressure
to
control
reflux,'3
and
suggests
that
the
pharmacotherapy
of
disordered
motor
function
in
reflux
disease
would
be
better
directed
at
control
of
transient
LOSRs
than
towards
augmentation
of
basal
LOS
pressure.
Detailed
analysis
of
oseophageal
motility
sur-
rounding
the
reflux
events,
revealed
six
principle
patterns
of
LOS
relaxation.
Consistent
with
our
previous
findings89
and
those
of
others'4
in
normal
subjects,
the
majority
of
LOSRs
were
not
directly
preceded
by
a
pharyngeal
swallow
signal.
This
find-
ing,
however,
is
at
variance
with
two
recent
studies
which
have
suggested
that
the
majority
of
reflux
episodes
are
directly
related
to
swallowing.'5`6
The
reason
for
this
disparity
is
not
readily
apparent,
but
may
result
in
part
from
suboptimal
methodology
and
data
analysis
in
the
previous
studies.
One
of
these
studies
was
severely
limited
by
recording
methods
because
the
single
side
hole
sensor
used
to
record
LOS
pressure
was
considered
to
have
monitored
LOS
pressure
adequately
during
only
18%
of
the
reflux
episodes.'5
In
the
other
study,'6
details
of
the
analysis
of
the
timing
of
swallowing
relative
to
the
onset
of
the
reflux
episode
were
not
specified.
The
timing
of
the
pharyngeal
swallow
signal
relative
to
the
onset
of
the
LOSR,
however,
is
crucial.
Our
analysis
reveals
that
most
of
the
swallows
that
preceded
reflux
episodes
actually
occurred
after
the
onset
of
the
transient
LOSR
and
may
have
been
the
consequence
of
reflux.
Such
swallows
could
not
have
been
the
trigger
of
transient
LOSRs
since
the
LOSRs
associated
with
normal
primary
peristalsis
had
their
onset
after
pharyngeal
contraction
was
recorded.
Reflux
episodes
were
recorded
only
rarely
during
swallow
induced
LOS
relaxation
associated
with
a
normal
peristaltic
sequence,
either
because
reflux
did
not
occur
or
because
the
oncoming
wave
of
contrac-
tion
limited
any
refluxate
to
the
distal
oesophagus
beyond
the
pH
electrode.
Although
a
small
propor-
tion
of
LOSRs
appear
to
be
directly
swallow
related,
most
of
these
are
associated
with
failed
peristaltic
sequences.
The
inhibition
of
oesophageal
motor
body
activity
that
we
have
shown
to
accompany
prolonged
transient
LOSRs
raises
the
possibility
that
some
of
these
apparently
failed
peristaltic
sequences
may
have
resulted
from
the
chance
association
of
a
spontaneous
transient
LOSR
and
a
swallow.
On
occasions,
swallows
occurred
after
the
occur-
rence
of
transient
LOSRs
but
just
before
the
onset
of
oesophageal
acidification.
The
role
of
such
swallows
in
the
triggering
of
reflux
is
not
clear.
The
factors
that
maintain
gastrooesophageal
competence
during
apparent
complete
LOS
relaxation
are
not
adequately
understood.
Possibly,
shortening
of
the
oesophagus
induced
by
swallowing'`
may
exert
traction
on
the
LOS
and
help
overcome
any
residual
forces
which
maintain
sphincter
competence
during
LOS
relaxation.
The
abruptness
of
transient
LOSRs
suggests
that
they
are
mediated
by
a
neural
mechanism.
This
notion
is
supported
by
our
recent
demonstration
in
the
dog
that
blockade
of
both.
cervical
vago-
sympathetic
bundles'8
and
light
general
anaesthesia'9
abolishes
transient
LOSRs
associated
with
belching,
and
that
gaseous
gastric
distension
does
not
provoke
these
events
in
patients
with
achalasia.21
This
view
is
further
supported
by
our
analysis
of
oesophageal
body
motor
function
during
prolonged
transient
LOSRs.
The
demonstration
of
refractoriness
of
the
smooth
muscle
oesophagus
is
only
readily
explain-
able
by
neural
inhibition.
The
duration
of
transient
LOSRs
suggests
that
this
inhibition
may
be
stronger
and
of
longer
duration
than
that
which
occurs
during
normal
swallow
induced
LOS
relaxation.
Relatively
long
lasting
LOS
and
oesophageal
body
inhibition
would
be
expected
to
facilitate
reverse
transit
in
the
oesophagus,
consistent
with
the
proposed
normal
role
of
transient
LOSRs,
that
of
controlled
orad
venting
of
gastric
contents.2'
Little
is
known
about
triggers
for
transient
LOSRs.
Gastric
distension
is
the
most
potent
stimulus
yet
identified.2'22
Our
data
indicate
that
most
transient
LOSRs
are
not
triggered
by
swallowing
as
they
are
not
associated
with
a
manometrically recordable
pharyngeal
signal.
It
has
been
suggested
recently,
that
transient
LOSRs
may
result
from
partial
activa-
tion
of
the
excitatory
component
of
primary
peri-
stalsis
with
more
complete
stimulation
of
LOS
inhibitory
mechanisms
mediated
by
a
long
train
vagal
stimulus.
14
23
The
inhibition
of
oesophageal
peristalsis
that
we
observed
during
prolonged
LOSRs
could
be
explained
by
a
long
train
vagal
discharge
but
in
the
opossum,
transient
LOSRs
associated
with
partial
1026

Mechanisms
of
LOS
incompetence
1027
activation
of
the
swallow
reflex
were
largely
incomplete
and
of
shorter
duration
than
those
associ-
ated
with
complete
swallows.'
This
finding
is
in
marked
contrast
to
spontaneous
LOSRs
in
the
present
study
which
were
complete
and
of
longer
duration
than
swallow
induced
LOSRs,
and
argues
against
the
hypothesis
that
all
LOSRs
result
from
partial
or
complete
expression
of
the
swallow
reflex.
The
basal
LOS
hypotonia
present
in
some
patients
with
peptic
oesophagitis,
has
been
ascribed
to
either
dysfunction
of
the
circular
smooth
muscle24
or
defective
tonic
neural
control.'
Absent
basal
LOS
pressure
was
never
present
consistently
throughout
the
study
period
in
any
of
our
patients,
all
of
whom
manifested
at
some
time
detectable
LOS
pressure
sufficient
to
prevent
reflux.
Indeed
many
patients
with
undetectable
fasting
or
immediately
postpran-
dial
basal
LOS
pressures
subsequently
developed
considerable
LOS
pressure.
This
pressure
was
often
in
the
region
of
15
to
20
mmHg
in
the
second or
third
postprandial
hour
and
was
maintained
until
the
end
of
the
study.
Thus,
the
sphincter
muscle
in
these
patients
was
capable
of
normal
contraction
for
sustained
periods
of
time,
indicating
adequate
con-
tractility
of
the
LOS
smooth
muscle,
thereby
imply-
ing
a
defect
in
neural
control.
Defective
neural
control
of
basal
LOS
pressure
also
provides
a
plaus-
ible
explanation
for
the
increased
prevalence
of
LOS
pressure
drifts
that
we
observed
in
patients
with
low
basal
LOS
pressure.
Based
on
the
findings
of
this
study,
we
propose
that
excessive
reflux
in
gastrooesophageal
reflux
disease
may
be
caused,
at
least
in
part,
by
a
spectrum
of
deranged
neural
control
of
LOS
pressure.
At
one
end
of
the
spectrum,
basal
LOS
pressure
is
normal
but
an
abnormally
high
reflux
rate
occurs
because
of
an
increased
sensitivity
of
the
neural
control
mech-
anisms
that
trigger
transient
LOSRs.
At
the
other
end
of
the
spectrum,
gross
LOS
hypotonia
reflects,
at
least
in
part,
a
persistent
defect
in
the
neural
mechanisms
which
maintain
basal
LOS
pressure.
Between
these
poles
of
the
spectrum,
are
patients
who
have
a
partial
defect
in
the
neural
mechanisms
that
normally
maintain
basal
LOS
pressure
resulting
in
intermittent
episodes
of
absent
basal
LOS
pressure
and
an
increased
tendency
for
LOS
pressure
to
drift
transiently
to excessively
low
values.
This
work
was
presented
in
part
at
the
Annual
Meeting
of
the
American
Gastroenterological
Association,
Washington,
DC,
1983
and
published
in
abstract
form
in
Gastroenterology
1983;
84:
1135.
This
work
was
supported
by
a
project
research
grant
from
the
Australian
National
Health
and
Medical
Research
Council.
We
gratefully
acknowledge
the
technical
assistance
of
Ian
Lewis
and
the
late
Beverley
Barnes.
References
1
Wankling
WJ,
Warrian
WG,
Lind
JF.
The
gastro-
esophageal
sphincter
in
hiatus
hernia.
Can
J
Surg
1965;
8:61-7.
2
Pope
CE
II.
A
dynamic
test
of
sphincter
strength:
its
application
to
the
lower
esophageal
sphincter.
Gastro-
enterology
1967;
52:
779-86.
3
Cohen
S,
Harris
LD.
Does
hiatus
hernia
affect
-com-
petence
of
the
gastroesophageal
sphincter?
N
Engl
J
Med
1971;
284:
1053-6.
4
Fisher
RS,
Malmud
LS,
Roberts
GS,
Lobis
IF.
The
lower
esophageal
sphincter
as
a
barrier
to
gastro-
esophageal
reflux.
Gastroenterology
1977;
72:
19-22.
5
Behar
J,
Biancani
P,
Sheahan
DG.
Evaluation
of
esophageal
tests
in
the
diagnosis
of
reflux
esophagitis.
Gastroenterology
1976;
71:
9-15.
6
Krejs
GJ,
Seefeld
V,
Brandlii
HH,
et
al.
Gastro-
oesophageal
reflux
disease:
correlation
of
subjective
symptoms
with
7
objective
oesophageal
function
tests.
Acta
Hepato-Gastroenterol
1976;
23:
130-40.
7
Stanciu
C,
Hoare
RC,
Bennett
JR.
Correlation
between
manometric
and
pH
tests
for
gastro-oesophageal
reflux.
Gut
1977;
18:
536-40.
8
Dent
J,
Dodds
WJ,
Friedman
RH,
et
al.
Mechanism
of
gastroesophageal
reflux
in
recumbent
asymptomatic
human
subjects.
J
Clin
Invest
1980;
65:
256-67.
9
Dodds
WJ,
Dent
J.
Hogan
WJ,
et
al.
Mechanisms
of
gastroesophageal
reflux
in
patients
with
reflux
eso-
phagitis.
N
Engl
J
Med
1982;
307:
1547-52.
10
Dent
J.
A
new
technique
for
continuous
sphincter
pressure
measurement.
Gastroenterology
1976;
71:
263-7.
11
Dent
J,
Dodds
WJ,
Sekiguchi
T,
Hogan
WJ,
Arndorfer
RC.
Interdigestive
phasic
contractions
of
the
human
lower
esophageal
sphincter.
Gastroenterology
1983;
84:
453-60.
12
Orr
WC,
Robinson
MG,
Johnson
LF.
Acid
clearance
during
sleep
in
the
pathogenesis
of
reflux
esophagitis.
Dig
Dis
Sci
1981;
26:
423-7.
13
Hyams
JS,
Leichtner
AM,
Zamett
LO,
Walters
JK.
Effect
of
metoclopramide
on
prolonged
intraesophageal
pH
testing
in
infants
with
gastroesophageal
reflux.
J
Pediatr
Gastroenterol
Nutr
1986;
5:
716-20.
14
Mittal
RK,
McCallum
RW.
Characteristics
of
transient
lower
esophageal
sphincter
relaxation
in
humans.
Am
J
Physiol
1987;
252:
G636-41.
15
Corazziari
E.
Bontempo
I,
Anzini
F,
Torsoli
A.
Motor
activity
of
the
distal
oesophagus
and
gastro-oesophageal
reflux.
Gut
1984;
25:
7-13.
16
Baldi
F,
Ferrarini
F,
Balestra
R,
et
al.
Oesophageal
motor
events
at
the
occurrence
of
acid
reflux
during
endogenous
acid
exposure
in
healthy
subjects
and
in
patients
with
oesophagitis.
Gut
1985;
26:
336-41.
17
Dodds
WJ,
Stewart
ET,
Hodges
D,
Zboralske
FF.
Movement
of
the
feline
esophagus
associated
with
respiration
and
peristalsis.
J
Clin
Invest
1973;
52:
1-13.
18
Martin
CJ,
Patrikios
J,
Dent
J.
Abolition
of
gas
reflux

1028
Dent,
Holloway,
Toouli,
and
Dodds
and
transient
lower
esophageal
sphincter
relaxation
by
vagal
blockade.
Gastroenterology
1986;
91:
890-6.
19
Cox
MR,
Martin
CJ,
Westmore
M,
Dent
J.
Effect
of
general
anaesthesia
on
transient
lower
esophageal
sphincter
relaxation
in
the
dog
[Abstract].
Gastro-
enterology
1987;
92:
1357.
20
Holloway
RH,
Dent
J,
Wyman
JB.
Impairment
of
belching
in
achalasia:
evidence
for
neural
mediation
of
transient
lower
esophageal
sphincter
(LES)
relaxations
[Abstract].
Gastroenterology
1986;
91:
1055.
21
Wyman
JB,
Dent
J,
Heddle
R,
Dodds
WJ,
Toouli
J,
Lewis
I.
Belching:
a
clue
to
understanding
of
patho-
logical
gastroesophageal
reflux
[Abstract].
Gastro-
enterology
1984;
86:
1303.
22
Holloway
RH,
Hongo
M,
Berger
K,
McCallum
RW.
Gastric
distension:
a
mechanism
for
post-prandial
gastroesophageal
reflux.
Gastroenterology
1985;
89:
779-84.
23
Paterson
WG,
Rattan
S,
Goyal
RK.
Experimental
induction
of
isolated
esophageal
sphincter
relaxation
in
anaesthetised
opossums.
J
Clin
Invest
1986;
77:
1187-93.
24
Grossman
MI.
Do
refluxers
have
sick
sphincters?
N
Engl
J
Med
1974;
290:
167.
25
Higgs
RH,
Castell
DO,
Eastwood
GL.
Studies
on
the
mechanism
of
esophagitis
induced
lower
esophageal
sphincter
hypotension
in
cats.
Gastroenterology.1976;
71:
51-7.