Content uploaded by Iain G Cleator
Author content
All content in this area was uploaded by Iain G Cleator on Jun 02, 2014
Content may be subject to copyright.
Anastomotic
Ulceration
1.
G.
M.
CLEATOR,
M.B.,
Ch.B.
Ed.,
F.R.C.S.(C),
F.R.C.S.(E),
F.R.C.S.,
1.
B.
HOLUBITSKY,*
M.D.,
F.R.C.S.(C),
F.A.C.S.,
R.
C.
HARRISON,
M.D.,
M.S.,
F.R.C.S.(C),
F.A.C.S.
A
NASTOMOTIC
ULCER
PROVIDES
THE
CLINICIAN
with
an
even
greater
therapeutic
challenge
than
primary
peptic
ulcer.
Not
only
has
the
original
operation
failed
to
cure
the
patient's
ulcerative
diathesis,
but
subsequent
management
has
been
complicated
by
alterations
in
the
physiology
and
anatomy
of
the
upper
gastrointestinal
tract.
It
is
therefore
hardly
surprising
that
there
is
a
high
mortality
and
morbidity
associated
with
this
complica-
tion.
Although
anastomotic
ulcers
have
been
studied
since
the
first
quoted
case
of
Braun5
in
1891
there
is
still
insufficient
information
to
guide
the
clinician
in
his
man-
agement
of
this
complication.
In
the
period
1957-1968,
331
patients
with
anastomotic.
ulcer
were
admitted
to
the
Vancouver
General
or
Shaughnessy
Veteran's
Hospital.
This
includes
patients
treated
medically
as
well
as
surgically
and
enables
a
comparison
to
be
made
of
the
results
of
treatment
after
conservative
as
well
as
surgical
therapy.
The
primary
operation
for
peptic
ulcer
was
carried
out
in
the
two
hospitals
in
44%.
In
all
but
11%
of
the
total
series,
the
original
case
records
were
available.
This
information
was
supplemented
over
the
years
by
more
detailed
case
studies
at
the
Gastrointestinal
Clinic
of
the
University
of
British
Columbia.
We
do
not
know
the
size
of
the
original
"pool"
of
primary
peptic
ulcer
patients.
Etiological
Factors
We
found
certain
factors
important
in
the
develop-
ment
of
recurrent
ulcers.
Patients
with
a
recurrent
ulcer
often
developed
further
recurrence
following
therapy.
Most
of
the
patients
developed
recurrent
ulcers
following
From
the
Department
of
Surgery,
University
of
British
Columbia,
Vancouver,
B.C.,
Canada
operation
for
duodenal
ulcer,
but
some
developed
recur-
rence
following
operation
for
other
varieties
of
peptic
ulcer.
As
a
group,
the
patients
with
recurrent
ulcer
were
drug
abusers.
Patients
with
anastomotic
ulcer
tended
to
belong
to
blood
group
"O".
There
was
a
relatively
high
incidence
of
women
with
recurrent
ulcer
compared
to
the
incidence
found
in
primary
peptic
ulcer.
In
post-
gastrectomy
anastomotic
ulcer,
only
a
few
recurrences
were
attributable
to
retained
antra.
Post-Billroth
I
recur-
rent
ulcer
patients
had
a
lesser
gastric
resection
than
post-
Billroth
II
anastomotic
ulcer
patients.
Distribution
Fifty-one
of
the
331
patients
developed
recurrent
ulcers
after
surgical
treatment
and
six
of
these
developed
a
third
recurrent
ulcer.
One
patient
went
on
to
have
a
fourth
and
fifth
anastomotic
ulcer
after
surgical
treat-
ment.
There
were
therefore
390
anastomotic
ulcers
in
331
patients
(Table
1).
The
follow-up
varied
from
one
to
14
years,
half
being
followed
six
years
or
more.
Only
20%
of
331
patients
were
regular
clinic
attenders,
and
the
other
patients
were
followed-up
by
questionnaire
or
interview
in
1971.
Of
the
331
patients,
22%
were
untraced
but
many
of
these
had
been
previously
followed-up
for
several
years,
16%
of
the
74
untraced
patients
having
been
followed
for
six
years
or
more.
Site
of
the
Primary
Ulcer
This
reflects
the
usual
finding
that
the
primary
opera-
tion
was
for
duodenal
ulcer
in
the
majority
of
cases,
88%
in
our
study
(Table
2).
339
Submitted
for
publication
April
10,
1973.
Reprint
requests:
Dr.
R.
C.
Harrison,
Department
of
Surgery,
Vancouver
General
Hospital,
Vancouver
9,
B.C.,
Canada.
*
Deceased
Ann.
Surg.
*
March
1974
CLEATOR,
HOLUBITSKY
AND
HARRISON
TABLE
1.
Distribution
of
Anastomotic
Ulcers
by
Primary
Operation
No.
with
one
No.
with
Two
No.
with
Three
No.
with
Five
Primary
Operation
Anastomotic
Ulcer
Anastomotic
Ulcers
Anastomotic
Ulcers
Anastomotic
Ulcers
Billroth
I
48
5
(90.6)
(9.4)
Billroth
II
124
15
1
(88.6)
(10.7)
(0.7)
Vagotomy
and
Pyloroplasty
30
4
1
(85.7)
(11.4)
(2.9)
Vagotomy
and
Gastroenterostomy
6
3
(66.7)
(33.3)
Vagotomy
and
Partial
Gastrectomy
14
1
(93.3)
(6.7)
Gastroenterostomy
58
17
3
1
(73.4)
(21.5)
(3.8)
(1.3)
Total
280
45
5
1
(84.6)
(13.6)
(1.5)
(0.3)
Figures
in
parentheses
are
percentages
Results
of
Primary
Operation
avoided
aspirin
products
or
took
them
rarely.
More
than
The
immediate
post-operative
results
of
surgery
for
eight
fluid
ounces
of
spirits
or
the
equivalent
amount
of
primary
peptic
ulcer
disease
were
satisfactory
in
all
but
alcohol
per
day
was
drunk
by
19%,
14%
were
regular
19%
of
operations
(Table
3).
The
early
results
of
gastric
drinkers
and
the
rest
avoided
alcohol
or
drank
it
rarely
surgery
therefore
gave
no
prediction
of
the
eventual
out-
These
results
are
startlingly
different
from
a
similar
come
in
terms
of
ulcer
recurrence.
The
excellent
im-
survey
by
Small2l
in
1964,
who
found
no
aspirin
ingestion
mediate
results
of
simple
gastroenterostomy
(79%)
are
of
in
his
series
of
83
patients
with
jejunal
ulcer,
but
a
higher
particular
interest. proportion
of
tobacco
smoking
than
in
our
series.
Drug
Habits
Blood
Group
The
incidence
of
tobacco
smoking,
and
ingestion
of
The
blood
group
was
recorded
in
324
of
the
331
pa-
aspirin
products,
alcohol
and
coffee
and
tea
were
studied
tients
with
anastomotic
ulcer:
55%
belonged
to
Blood
in
the
anastomotic
ulcer
patients.
Of
the
331
patients
Group
0.
In
a
study
of
blood groups
of
more
than
28%
smoked more
than
one
package
of
cigarettes
per
day.
15,000
donors
at
randomly
selected
clinics
in
the
greater
More
than
five
cups
of
tea
or
coffee
per
day
were
drunk
Vancouver
area,
the
frequency
of
Blood
Group
0
was
by
11%
of
the
patients.
Aspirin
products
were
ingested
found
to
be
44%.
The
increased
frequency
of
Blood
every
day
for
weeks
at
a
time
by
15%
of
patients,
11%
Group
0
in
anastomotic
ulcer
has
been
noted
in
the
took
aspirin
products
more
than
once
a
week,
and
the
rest
past
but
the
reason
for
this
finding
is
still
unclear.
TABLE
2.
Site
ot
Ulcers
at
Primary
Operations
One
Two
Duodenal
Duodenal
Pyloric
Pyloric
Primary
Operation
Duodenal
Duodenal
&
Gastric
&
Pyloric
&
Gastric
Gastric
Not
Known
Billroth
I
32
2
11 7
1
(60.4)
(3.8)
(20.7) (13.2)
(1.9)
Billroth
II
117
3
7
2
3
7
1
(83.6)
(2.1)
(5.0)
(1.4)
(2.1)
(5.0)
(0.7)
Vagotomy
and
Pyloroplasty
28
1
1
3
2
(80.0)
(2.9)
(2.9)
(8.6)
Vagotomy
and
Gastroenterostomy
8
1
(88.9)
(11
.1)
Vagotomy
and
Partial
Gastrectomy
13
11
(86.7)
(6.7)
(6.7)
Gastroenterostomy
75
3
1
(94.9)
(3.8)
(1.3)
Total
273
4
11
2
21
2
15
3
(82.5)
(1.2)
(3.3)
(0.6)
(6.3)
(0.6)
(4.5)
(0.9)
Figures
in
parentheses
are
percentages
340
ANASTOMOTIC
ULCERATION
341
TABLE
3.
Visick
Grading
after
Primary
Operation.
(In
this,
and
the
Tables
Following,
Grade
I,
II
and
IIIS
are
Considered
Satisfactory
Results.
Grades
III
U
and
IV
are
Unsatisfactory)
Primary
Operation
Not
Known
I
II
IIIS
IIIU
IV
Billroth
I
1
20
15
3
1
13
(1.9)
(37.7)
(28.3)
(5.7)
(1.9)
(24.5)
Billroth
II
40
63
9
7
21
(28.6) (45.0)
(6.4)
(5.0)
(15.0)
Vagotomy
and
5
3
1
Gastroenterostomiy
(55.6)
(33.3)
(11.1)
Vagotomy
and
Pyloroplasty
1
16
10
2
2
4
(2.9)
(45.7)
(28.6)
(5.7)
(5.7)
(11.4)
Vagotomy
anid
Partial
1
3
2
6
3
Gastrectomy
(6.7)
(20.0)
(13.3)
(40.0)
(20.0)
Gastroenterostomy
1
62
7
2
1
6
(1.3)
(78.5)
(8.9)
(2.5)
(1.3)
(7.6)
Total
3
144
101
19
17
47
(0.9)
(43.5)
(30.5)
(5.7)
(5.1)
(14.2)
Figuires
in
parentheses
are
percenitages
Sex
Incidence
In
this
study
21%
of
the
patients
were
females.
We
have
no
means
of
ascertaining
the
size
of
the
population
from
which
these
patients
were
drawn
but
there
does
seem
to
be
a
higher
percentage
of
females
than
one
could
expect
from
the
relative
frequency
of
surgical
interven-
tion
for
primary
peptic
ulcer
disease
which
is
15%
for
women
in
the
Vancouver
General
Hospital.
A
possible
explanation
for
this
is
that
surgeons
tend
to
be
more
conservative
in
the
surgery
of
women
both
in
terms
of
gastric
resection
and
a
tendency
to
carry
out
a
Billroth
I
anastomosis
rather
than
a
Billroth
II
type.
Retained
Antrum
in
the
Duodenal
Stump
Following
Partial
Gastrectomy
The
pathology
reports
from
the
original
operation
were
available
for
most
of
the
gastrectomy
specimens.
Many
of
the
reports
did
not
mention
the
presence
or
absence
of
duodenal
mucosa
in
the
excised
specimen
but
it
was
pos-
sible
to
exclude
antral
tissue
in
the
duodenum
in
67%
of
Billroth
II
gastrectomies.
Of
the
remaining
46
post
Billroth
II
anastomotic
ulcer
patients,
re-operation
was
carried
out
in
34
and
the
duodenal
stump
was
examined
in
31
of
these,
16
of
whom
had
a
biopsy
or
excision
done:
six
of
these
patients
were
found
to
have
retained
antra.
Twelve
patients
were
treated
medically
and
we
have
no
histologic
evidence
in
11
of
these,
but
one
autopsy
showed
no
retained
antrum
in
the
duodenal
stump.
In
the
remaining
patients
we
have
no
clinical
evidence
of
retained
antra
but
this
has
not
been
checked
by
histology
or
at
operation.
In
summary,
of
the
total
140
Billroth
II
gastrectomies
retained
antral
tissue
was
present
in
the
duodenal
stump
in
six
(4%)
and
there
is
good
evidence
that
there
was
no
retained
antrum
in
the
remaining
96%.
This
is
a
much
lower
incidence
of
retained
antrum
than
that
found
by
Boles4
and
his
colleagues
in
1960
who
reported
retained
antra
in
14
of
36
patients.
This
indi-
cates
improvement
of
surgical
technique
in
this
respect
in
the
last
20
years
due
to
awareness
of
this
problem.
Of
the
six
patients
with
retained
antra,
three
had
retained
antra
detected
at
their
first
anastomotic
opera-
tion
and
three
had
the
retained
antra
detected
only
after
developing
a
second
anastomotic
ulcer.
Table
4
summarizes
the
treatment
of
the
patients
in
this
group.
One
of
the
six
patients
has
had
an
ulcer
recurrence
demonstrated
at
operation
since
removal
of
retained
antrum.
Adequacy
of
Partial
Gastrectomy
The
adequacy
of
gastrectomy
in
terms
of
percentage
of
stomach
resected
must
always
be
a
compromise
be-
tween
prevention
of
anastomotic
ulcer
and
the
increased
incidence
of
malnutrition,
malabsorption
and
tuberculosis
which
follows
a
high
gastrectomy.
On
the
basis
of
the
surgeon's
estimate
of
percentage
of
stomach
resected,
76%
had
a
resection
of
66%
or
more
and
only
3%
had
a
resection
of
less
than
50%
of
the
stomach
(Table
5).
It
may
be
important,
however,
to
note
that
resection
for
a
Billroth
I
was
significantly
less
than
resection
for
a
Billroth
II.
This
was
presumably
due
to
the
fear
of
the
surgeon
that
insufficient
gastric
remnant
would
be
avail-
able
for
anastomosis
to
the
duodenum
without
tension.
This
may
be
one
of
the
reasons
why
Billroth
I
gastrec-
tomy
is
more
liable
to
result
in
recurrent
ulceration.
Clinical
Characteristics
We
must
stress
that
the
clinical
characteristics
are
still
the
best
single
method
of
diagnosing
anastomotic
ulcer.
We
found
that
haemorrhage
was
a
prominent
symptom
of
most
of
the
patients
with
anastomotic
ulcer.
The
site
of
the
ulcer
was
influenced
by
the
type
of
antecedent
operation.
The
age
of
development
of
anast:omotic
ulcer
varied
with
the
type
of
antecedent
Vol.
179
*
No.
3
Ann.
Surg.
*
March
1974
CLEATOR,
HOLUBITSKY
AND
HARRISON
TABI,
4.
Retained
Antrium
in
the
Dutodenal
Stump.
Sutmmary
of
the
Clinical
Course
in
the
Six
Patients
with
Retained
Antra
Following
Billroth
II
Gastrectomy.
Sex
Date
of
Birth
Antecedent
Operations
Findings
Follow-up
M
1897
1964,
60%
BIIA
for
D.U.
of
13
Nov.
1965,
retained
antrum
and
1966,
haemorrhage.
Jejtunal
ulcer
years
dtiration.
60
ml.
blood
in
stomach.
at
operation
and
vagotomy
with
Oct.
1965,
75%
BIIA
for
massive
No
bleeding
point
found.
85%
BIIA.
haemnorrhage,
3
ulcers
present,
Antrum
excised
and
anastomosis
1971,
Visick
I
on
F.U.
(2
stomal
and
1
gastric).
revised
as
Roux-en-Y.
M
1905
1962,
70%
BIIR
aind
vagotomy
1966,
retained
antrum
and
1971,
Visick
I
on
F.U.
for
D.U.
of
1
year
dLuration.
efferent
jejunal
uilcer.
1963,
prandial
pain.
Antrum
excised
and
95%
BIIR.
1963,
efferent.
jej
unal
uilcer,
90%
BII
and
Roux-eni-Y.
1964,
paini
and
vomiting.
M
1926
1957,
66%
BIIR
and
Rouix-en-Y
1961,
stomal
ulcer
and
retained
1971,
Visick
II
oni
F.U.
for
D.U.
of
7
years
duration.
antrum.
Antrum
excised
and
1958,
paini
anid
haemorrhage.
vagotomy.
F
1924
1956,
66%
BIIR
for
D.U.
of
17
1961,
jejunal
ulcer
and
retained
1962,
repeated
episodes
of
bleeding
years
duiration.
antrum.
Antrum
excised
and
eoesophageal
varices.
1958,
prandial
pain.
vagotomy.
On
portogram
pressure
in
portal
vein
was
340
mm
Hg.
Lost
to
follow-uLp.
F
1942
1962,
66%
BIIA
for
D.U.
of
12
1968,
distended
afferent
loop
with
1970,
prandial
pain
and
vomiting.
years
duiration.
jejunal
ulcer
on
afferent
side
and
1971,
Barium
meal
showed
stomal
1965,
pain
aid
vomiting,
retained
antrum.
Antruni
ulcer
crater.
excised
and
75%
BIIA.
M
1932
1968,
V.
and
P.
for
D.U.
of
1
1970,
retained
antrum-no
1971,
Visick
II
on
F.U.
year
dturationi.
ulcer
foutnd.
Antrum
excised
1968.
prandial
pain.
and
66%
BI
with
vagotomy
1969,
50%
BIIA
for
perforated
D.L.
1969,
pain
and
haenmorrhage.
operation
as
did
the
interval
between
antecedent
opera-
2%
gastro-jejuno-colic
fistulae
in
this
study,
indicative
of
tion
and
definitive
treatment
of
recurrent
ulcer.
the
decreasing
frequency
of
this
dangerous
complication
in
recent
studies.
Symptoms
It
is
noteworthy
that
56%
of
the
390
anastomotic
ulcer
The
pattem
of
symptoms
of
anastomotic
ulcer
was
patients
had
one
or
more
episodes
of
haemorrhage.
There
similar
to
that of
the
primary
operation.
Post-prandial
was
a
perforation
of
an
anastomotic
ulcer
in
5%
and
this
is
pain
occurred
in
74%
of
the
patients,
24%
had
no
pain
and
a
standard
observation.
The
incidence
of
stenosis
was
9%.
there
was
insufficient
information
concerning
symptoms
in
2%.
Site
of
Anastomotic
Ulcers
Twelve
patients
had
diarrhoea
and
weight
loss
and
Table
6
shows
the
site
of
the
390
anastomotic
ulcers.
seven
of
these
were
found
to
have
gastro-jejuno-colic
Clearly
the
site
at
which
an
anastomotic
ulcer
develops
fistula.
These
symptoms
of
diarrhoea
and
weight
loss
depends
partly
on
the
primary
operation.
The
ulcers
seen
were
therefore
considered
to
be
highly
suggestive
of
the
at
operation
were
chronic
in
91%,
7%
were
acute,
and
diagnosis
of
gastro-jejuno-colic
fistula.
There
were
only
2%
were
suture
ulcers.
TABiEF
5.
Percentage
oj
Stomach
Resected
(Estimated)
at
the
Primary
Operation
Antecedent
Operation
Not
Known
1/3
P.
G.
1/2
P.
G.
2/3
P.
G.
3/4
P.
G.
S.
T.
Billroth
I
1
1
28
18
5
(1.9)
(1.9)
(52.8)
(34.0)
(9.4)
Billroth
II
5
9
116
10
(3.6)
(6.4)
(82.9)
(7.1)
Vagotomy
and
6
7
1
1
Partial
Gastrectomny
(40.0)
(46.7)
(6.7)
(6.7)
Total
1
6
43
141
16
1
(0.5)
(3.0)
(20.5)
(67.5)
(8.0)
(0.5)
P.
G.-Partial
gastrectomy
S.
T.-Subtotal
gastrectomy
Figures
in
parentheses
are
percentages
342
TABLE
6.
Site
of
A
nastomotic
Ulcers
Jejunal
Jejunal
Jejunal
&
Jejunal
&
Antecedent
Operation
(s)
Gastric
Stomal
Efferent Afferent
Duodenal
Gastric
Duodenal
Not
Known
Billroth
I
11
20
0
0
15
0
0
10
(19.6)
(35.7)
(0) (0)
(26.8)
(0) (0)
(17.9)
Billroth
II
14
36
74
1
0
3
0
43
(8.2)
(21.1)
(43.3)
(0.6)
(0)
(1.7)
(0)
(25.1)
Vagotomy
and
Pyloroplasty
7
0
0
0
18
0
0
11
(19.4)
(0)
(0) (0)
(50.0)
(0j
(0)
(30.6)
Vagotomy
and
1
3
4
00
0 0
3
Gastroenterostomy
(9.1)
(27.3)
(36.4)
(0)
(0)
(0)
(0)
(27.3)
Vagotomy
and
Partial
6
8
10
0
2
0
0
10
gastrectomy
(16.7)
(22.2) (27.8)
(0)
(5.6)
(0)
(0)
(27.8)
Gastroenterstomy
17
10
32
0
6
0
3
12
(31.2)
(12.5)
(40.0)
(0)
(7.5)
(0)
(3.8)
(15.0)
Total
56
77
120
1
41
3
3
89
(14.4)
(19.7)
(31.1)
(0.3)
(10.5)
(0.7)
(0.7)
(22.8)
Figures
in
parentheses
are
percentages
Gastric
ulcer
was
significantly
commoner
after
gastro-
enterostomy
alone
than
after
Billroth
II
gastrectomy
when
numbers
observed
were
compared
to
numbers
expected
by
chance
alone.
A
possible
reason
for
this
is
that
the
area
of
the
stomach
at
risk
is
larger
in
gastro-
enterostomy
alone
than
after
Billroth
II
gastrectomy.
However,
when
the
site
of
recurrence
after
Billroth
I
was
compared
to
that
after
vagotomy
with
pyroroplasty
there
was
no
difference
in
the
rate
of
gastric
ulceration.
Golligher10
has
noted
increased
frequency
of
develop-
ment
of
gastric
ulcer
following
the
Billroth
I
procedure
and
considers
this
operation
may,
predispose
to
the
development
of
gastric
ulceration.
However,
in
our
series,
the
vagotomy
with
pyloroplasty
group
has
been
followed
for
a
much
shorter
time
than
the
Billroth
I
group
and
the
incidence
of
gastric
ulcer
after
vagotomy
and
pyloro-
plasty
procedures
may
rise
as
the
results
become
avail-
able
for
long
term
follow-up.
Another
factor
might
be
that
the
mean
age
following
gastroenterostomy
(61
+
14)
is
higher
than
after
the
Billroth
II
operation
(50
+
11)
and
the
mean
age
of
the
patients
with
a
Billroth
I
procedure
(52
11)
is
not
significantly
higher
than
that
of
the
vagotomy
with
pyloroplasty
group
(48
+
13).
It
is
known
that
gastric
ulcers
are
more
freqeunt
in
the
elderly
and
this
would
explain
our
findings.
Age
at
Operation
for
First
Anastomotic
Ulcer
This
was
calculated
only
for
those
undergoing
surgical
treatment.
The
age
of
the
patients
with
anastomotic
ulceration
following
gastroenterostomy
alone
(61
14)
is
greater
on
average
than
the
patients
with
anastomotic
ulceration
following
other
procedures
(54
11)..This
reflects
the
changing
fashion
in
surgical
operations
over
the
last
20
years
and
the
decline
in
frequency
of
gastro-
enterostomy
as
a
routine
treatment
of
duodenal
ulcer.
Interval
Prior
to
Treatment
of
Anastomotic
Ulcer
The
interval
between
the
primary
operation
of
peptic
ulcer
and
year
of
treatment
of
anastomotic
ulcer
varied
markedly
depending
on
the
type
of
antecedent
opera-
tion
(Table
7).
Although
there
is
general
agreement
that
after
gastroenterostomy
an
anastomotic
ulcer
can
occur
many
years
after
the
initial
treatment,
it
has
been
held
that
after
gastrectomy
an
anastomotic
ulcer
tends
to
recur
much
earlier.
Our
study
shows
a
long
interval
after
gastrectomy
does
not
convey
immunity
from
sub-
sequent
anastomotic
ulcer.
We
would
suggest
that
this
is
due
to
the
late
results
of
gastr
tomy
now
becoming
available
and
predict
that
the
prsent
short
intervals
prior
to
the
development
of
anastomotic
ulcer
following
the
vagotomy
procedures
will
prove
artifactual,
and
as
long
term
studies
become
available
more
and
more
patients
will
be
found
to
develop
-anastomotic
ulcer
at
long
intervals
after
these
procedures.
Diagnostic
Procedures
We
define
anastomotic
ulcer
as
a
benign
peptic
ulcer
which
develops
after
definitive
gastric
surgery.
Two
TABLE
7.
Interval
between
Primary
Operation
and
Operation
for
First
Anastomotic
Ulcer
I
nterval-
Mean
&
S.D.
Range
Antecedent
Operation
No.
(In
Years)
in
years
Billroth
I
53
5.40
5.60
0-33
Billroth
II
140
3.39
3.04
0-24
Vagotomy
and
Pyloroplasty
35
1.48
4
0.77
0-4
Vagotomy
and
Gastroenterostomy
9
3.56
i
2.30
0-7
Vagotomy
and
Partial
gastrectomy
15
2.78
i
2.91
0-10
Gastroenterostomy
79
17.27
+
11.
18
0-40
Vol.
179
*
No.
3
ANASTOMOTIC
ULCERATION
343
Ann.
Surg.
*
March
1974
CLEATOR,
HOLUBITSKY
AND
HARRISON
hundred
and
forty-four
of
the
390
anastomotic
ulcers
in
the
series
were
visualized
at
operation
or
autopsy.
An
ulcer
crater
was
detected
on
barium
meal
examination
in
a
further
64
instances.
An
ulcer
was
seen
on
gastro-
scopy
in
15,
and
in
67
remaining
cases,
the
diagnosis
was
made
on
clinical
grounds
alone.
Acid
secretory
tests,
both
maximal
acid
secretion
and
Hollander
tests,
were
of
value.
Bariuim
Meal
X-Ray
The
diagnostic
value
of
this
test
was
low.
In
331
barium
meal
examinations,
52%
showed
an
ulcer
crater.
We
consider
the
results
indefinite
in
a
further
l9%
since
distortion
or
spasm
were
the
only
abnormalities
demon-
strated.
No
abnormality
was
demonstrated
in
the
remain-
ing
29%.
These
results
agree
with
the
findings
of
Walters
and
Chance27
in
1955.
Gastroscopy
Gastroscopy
proved
disappointing
in
determining
the
presence
of
an
anastomotic
ulcer.
Of
72
gastroscopies,
an
ulcer
crater
was
seen
in
37%.
In
a
further
25%,
spasm
or
irritability
of
the
anastomosis
or
inflammation
around
the
anastomosis
was
noted
which
would
have
given
rise
to
suspicion
of
an
ulcer
beyond
the
range
of
the
gastroscope.
However,
it
must
be
noted
that
the
gastroscopies
per-
formed
in
this
series
were
carried
out
prior
to
the
use
of
fiberoptic
endoscopy
in
our
department.
It
is
probable
therefore
that
with
the
use
of
the
more
modern
instru-
ments
now
available,
the
incidence
of
positive
diagnosis
will
improve.
Maximal
Acid
Secretory
Studies
The
peak
acid
output
in
respoinse
to
Histalog*
or
more
recently
Pentagastrin
was
measured
in
142
of
the
anastomotic
ulcers.
We
found
that
the
peak
acid
output
measurement
in
patients
after
partial
gastrectomy
with
or
without
vagotomy,
was
of
some
value
in
discriminating
between
patients
with
anastomotic
ulcer
and
controls.
More
than
95%
of
the
control
values
(110
controls)
were
below
12
mEq.
acid
output
in
the
peak
post-stimulation
hours
and
more
than
40%
of
the
anastomotic
ulcers
had
acid
secretory
results
above
this
value.
Hollander
Test
This
proved
most
valuable
in
the
post-vagotomy
with
pyloroplasty
group.
Insulin
testing
was
carried
out
in
22
of
these
patients,
and
the
test
was
positive
on
Hol-
lander's
criteria
in
21
of
these,
indicating
an
incomplete
vagotomy.
The
solitary
negative
test
occurred
in
a
patient
with
a
gastric
anastomotic
ulcer,
but
was
not
repeated
prior
to
reoperation.
*
Eli
Lilly
and
Co.
Serum
Gastrin
Serum
gastrin
estimations
should
now
be
available
for
every
patient
with
anastomotic
ulcer,
using
either
radio-immune
assay14
or,
if
not
available,
bioassay.625
This
would
serve
the
two-fold
purpose
of
excluding
the
Zollinger-Ellison
syndrome,
and
excluding
the
possibility
of
retained
antra
in
the
duodenal
stump
of
post
gastrec-
tomy
patients.
Although
we
have
been
carrying
out
radio-
immune
assay
and
bioassay
of
gastrin
for
the
last
year,
we
do
not
yet
have
the
serum
gastrin
levels
for
the
patients
in
this
study.
Therapy
Employed
The
effectiveness
of
treatment
of
anastomotic
ulcer
is
usually
judged
by
the
mortality
rate
from
the
operation
or
treatment
and
the
recurrent
ulcer
rate.
In
this
study
we
have
included
deaths
due
to
ulcer
and
deaths
due
to
operation
in
each
group.
For
medical
treatment
the
mor-
tality
rate
is
deaths
due
to
ulcer,
and
deaths
from
other
causes
are
not
included.
Following
surgical
treatment,
the
deaths
represent
the
operative
mortality,
and
after
the
first
month
deaths
from
incidental
causes
are
not
included.
The
period
of
follow-up
is
critical
to
any
assessment
of
the
results.
When
our
results
were
grouped
into
follow-up
periods
of
five
years
or
less
and
follow-up
periods
of
six
years
or
more
there
was
a
significantly
higher
recurrence
rate
in
the
group
with
longer
follow-up.
Of
the
194
patients
followed
for
five
years
or
less,
18
had
one
recurrent
ulcer
(9%).
Of
the
137
patients
fol-
lowed
six
years
or
more,
27
had
one
recurrent
ulcer
and
six
had
two
or
more
recurrent
ulcers
(24%).
In
the
fur-
ther
discussion
of
the
results
of
treatment
therefore,
our
results
have
been
broken
down
into
follow-up
periods
of
greater
than
one
year
to
permit
comparison
with
the
results
of
other
studies,
and
six
years
or
more
to
give
the
recurrence
rate
on
long
term
follow-up.
Post-Gastrectomy
Anastomotic
Ulcer
These
ulcers
should
be
treated
by
transabdominal
vagotomy.
Seventy-one
patients
in
our
study
were
treated
in
this
way
with
no
operative
deaths,
seven
proven
recur-
rent
ulcers
and
seven
suspected
recurrent
ulcers
(Tables
8-10).
A
further
ten
patients
required
some
modification
of
the
anastomosis
in
addition
to
vagotomy
and
of
these
patients
there
were
no
operative
deaths.
One
had
a
recurrent
ulcer
and
one
had
a
suspected
recurrent
ulcer.
A
more
proximal
or
higher
gastrectomy
was
carried
out
in
78
patients,
40
of
whom
had
a
vagotomy
in
addition
to
the
gastrectomy.
Compared
to
vagotomy
alone,
there
was
no
improvement
in
the
recurrence
rate
which
was
15
with
recurrent
ulcer,
five
with
suspected
recurrent
ulcer,
but
there
was
a
higher
death
rate.
Six
of
the
71
patients
died
as
a
result
of
complications
of
the
gastrectomy.
344
ANASTOMOTIC
ULCERATION
345
A
study
of
the
literature
confirms
our
views
on
the
other
current
treatment
is
below
5%
(Table
11).
The
re-
treatment
of
this
group
of
ulcers.
The
death
rate
follow-
currence
rate
on
short-term
follow-up
is
less
than
16%
ing
vagotomy
alone
as
a
treatment
for
post-partial
gas-
and
on
long-term
follow-up
less
than
19%
(Table
12).
trectomy
anastomotic
ulcers
is
less
than
1%,
while
no
Only
vagotomy
with
partial
gastrectomy
has
a
lower
TABLE
8.
Treatment
of
the
Complete
Series
of
Anastomotic
Ulcers
Followed
from
one
to
14
Years
Antecedent
Treatment
of
Related
Unrelated
I
II
IIIS
IIIU
IV
Recurrent
Operation
Anastomotic
Ulcer
Death
Death
(Visick
Grade)
Ulcer
Lost
Total
BI
Med
Rx
2nd
op-BI
-BII
-Vag
-V
&
PG
-Other
BII
Total
Med
Rx
2nd
op-BI
-BII
-Vag
-V
&
0
-V
&
PG
-Other
Total
V
&
Pyl
Med
Rx
2nd
op-BI
-BII
-Vag
-V
&
0
-V
&
PG
-Other
Total
V
&
G-E
Med
Rx
2nd
op-BI
-BII
-V
&
PG
-Other
Total
V
&
PG
Med
Rx
2nd
op-BII
-Vag
-V
&
0
-V
&
PG
-Other
Total
G-E
Med
Rx
2nd
op-BI
-BII
-Vag
-V
&
0
-V
&
PG
-Other
Total
1
1
4
1
0
4
1
0
4
O
0
0
0
0
1
0
0
0
2
22
2
2
0
1
4
3
O
0
3
4
2
0
0
0
2
0
2
2
3
0
1
0
0
1
O
0
0
0
0
0
0
1
0
3
5
1
1
10
46
25
10
2
6
11
8
1
3
1
0
14
O
0
0
0
0
0
0
1
0
2
1
4
0
2
1
0
7
1
04
15
9
3
6
1
7
15
0
0
4
2
1
1
0
1
1
2
1
14
3
1
1
0
3
6
0
0
0
1
0
0
0
3
1
6
12
48
23
8
12
2
22
38
0
1
5
3
2
0
0
0
2
0
0
0
0
0
0
0
1
0
0
1
4
3
0
1
0
3
1
0
0
2
1
0
0 0
1
1
0
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
14
7
2
1
0
5
5
0
3
0
0
0
0
0
0
0
0
00
0
0
0
0
1
0
0
1
3
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
4
3
0
0
0
0
3
1
0
1
1
2 3
1
0
0
3
0
1
2
1
1
0
0
3
1
0
1
2
1
2
0
0
0
1
0
0
1
0
00
0
0
0
0
0
1
1
0
1
1
0
0
0
0
4
0 0
0
0
00
0
7
1
2
0
0
0
1
I
3
6
0
10
0
0
0
0
16
0
0
6
0
2
1
0
9
5
0
0
1
0
2
1
0
4
6
1
0
1
1
1
0
0
4
2
1
0
0
0
0
0
0
1
1
3
0
0
0
1
0
17
0
2
0
0
0
0
0
1
0
21
5
4
1
8
0
0
1
0
14
Med
Rx-Treated
medically
only
BI-Partial
gastrectomy
of
the
Billroth
I
type
BII-Partial
gastrectomy
of
the
Billroth
II
type
Vag-Vagotomy
only
Vag
&
PG-Vagotomy
combined
with
partial
gastrectomy
O-Other
operations
(Non-definitive
surgical
procedures
such
as
closure
of
perforation,
undersewing
of
bleeding
vessel,
laparotomy,
any
excision
or
take-down
of
anastomosis)
Related
deaths-Those
occuring
in
the
immediate
postoperative
period
or,
when
following
medical
treatment,
caused
by
complications
of
a
persistent
ulcer
Unrelated
deaths-From
causes
unrelated
to
the
ulcer
or
its
surgical
or
medical
treatment
16
1
18
11
9
1
56
46
1
18
60
10
31
5
171
13
1
13
5
0
3
1
36
3
l
5
1
11
11
9
7
1
4
4
36
19
3
45
3
5
3
2
80
Vol.
179
*
No.
3
346
CLEATOR,
HOLUBI1
recurrence
rate
than
this
and
that
is
only
on
short-term
follow-up.
Exceptionally,
additional
surgery
may
be
required.
Only
six
of
our
patients
required
excision
of
the
re-tained
antra
in
the
duodenal
stump
and
every
effort
should
be
made
to
exclude
this
condition
pre-operatively
rather
than
subject
the
patient
to
unnecessary
exploration.
Stenosis
of
the
anastomosis
is
rare,
but
this
may
require
treatment
by
excision
and
reconstitution
of
the
anasto-
mosis.
Post-Gastroenterostomy
Anastomotic
Ulcer
This
group
should
be
treated
by
two-thirds
partial
gastrectomy.
Our
results
for
all
forms
of
treatment
were
unsatisfactory
in
this
group
of
patients.
Forty-eight
pa-
tients
were
treated
with
two-thirds
partial
gastrectomy
and
three
died.
The
cause
of
death
in
these
three
patients
was,
however,
from
embolism
in
two
and
pneumonia
in
TSKY
AND
HARRISON
the
other,
and
we
feel
that
this
is
more
a
reflection
on
the
age
and
debility
of
the
patients
than
on
the
choice
of
operation.
The
recurrence
was
high,
with
18
patients
having
a
proven
recurrent
-ulcer
and
none
suspected
of
recurrence.
Only
eight
patients
were
treated
by
vagot-
omy,
but
five
of
these
required
some
reconstruction
of
the
anastomosis.
There
were
no
deaths
but
two
patients
developed
recurrent
ulcers
(Table
8).
Our
views
agree
with
those
in
the
literature.
The
mortality
rate
for
partial
gastrectomy
is
the
same
as
that
following
vagotomy
for
post-gastroenterostomy
ulcers
(Table
13).
The
rate
of
ulcer
recurrence
following
partial
gastrectomy
is
less
than
half
that
of
vagotomy
alone
in
that
group
(Table
13
and
14).
Curiously,
the
addition
of
vagotomy
to
the
gastrec-
tomy
in
post-gastroenterostomy
ulcers
appears
to
result
in
an
increased
operative
mortality
in
the
literature,
although
we
ourselves
have
had
no
deaths
in
this
group
in
our
study.
TABLF,
9.
Late
Results
of
Treatment
of
Anastomotic
Ulcer,
Including
Only
the
Patients
Followed
for
Six
Years
or
More
Antecedent
Treatment
of
Related
Unrelated
I
II
IllS
IIIU
IV
Recurrent
Operation
Anastomotic
Ulcer
Death
Death
(Visick
Grade)
Ulcer
Lost
Total
BI
Med
Rx
1
0
2
1
0
4
1
0
1
10
2nd
op-BI
0
0
0
0 0
1
0
0
0
1
-BII
0
0
2
1
1
0
1
2
2
9
-Vag
0
0
2
0
0
0
0
0
0
2
-V&PG
0
0
0
1
0
0
0 0
0
1
Total
1
06
3
1
5
22
3
23
BIT
Med
Rx
1
3
8
8
1
2
1
0
4
28
2nd
op-BI
0
00
0
000
1
0
1
-BII
0
0
3
0
1
0
0
4
.0
8
Vag
0
0
5
3
0
2
0
5
3
18
-V&O
0
0
2
0
0
0
0
1
0
3
-V
&
PG
1
1
6
1
1
1
0
3
0
14
-Other
0
0
0
0
00
0
3
0
3
Total
2
4
24
12
3
5
1
17
7
75
V
&
Pyl
Med
Rx
00
1
1
00
0
0
0
2
2nd
op-BII
00
0
0
0
1
0
1
0
2
Total
0
0
1
1
0
1
0
1
0
4.
V&G-E
Med
Rx
0
1
0
0
0
0 0
0
0
1
2nd
op-BI
0
0
0
0
000
1
0
1
-BIT
0
0
2
0
00
00
0
2
Total
0
1
2
0
000
1
0
4
V
&
PG
Med
Rx
0
1
0
1
2
1
0
0 0
5
2nd
op-BII
0
1
1
1
0
0
0
3
0
6
-V&
PG
0
0
0
0
0
1
0
0
0
1
-Vag
0
0
1
1
0
0
0
0
0
2
-Other
0
0
2
0
0
0
0
0
0
2
Total
0
2
4
3
22
0
3
0
16
G-E
Med
Rx
6
3
0
0 0
1
0
0.
3
13
2nd
op-BI
00
0
0
0
0
0
1
0
1
-BIT
1
2
6
0
1
0
0
13
2
25
-Vag
0
0
0
0
1
0
0 0
0
1
-V&O
00
2
0
0
0
0
0
0
2
-V&PG
0
0
0
0
0
0 0
0
1
1
-Other
0
0
0
0
0
0
0
1
0
1
Total
7
5
8
0
2
1
0
15
6
44
Ann.
Surg.
*
March
1974
ANASTOMOTIC
ULCERATION
347
TABLE
10.
Deaths
due
to
Operation
or
Ulcer
A.
Immediate
Post-operative
Deaths
(10
Patients)
Antecedent
Operation
for
Age
(Years)
Operation
(s)
Anastomotic
ulcer
Cause
of
Death
Contributing
Cause
71
BI
BIT
Duodenal
fistula
None
70
BI
BII
Anastomotic
dehiscence
None
45
BI
&
BII
V
&
PG
Peritonitis
None
75
BIT
BIT
Cardiac
arrest
Long
operation
55
BII
V
&
PG
Anastomotic
dehiscence
None
67
G-E
BI
Pneumonia
Emergency
operation
for
haemorrhage
53
G-E
BIT
Pulmonary
embolus
None
70
G-E
Closure
Cardiac
arrest
Perforated
ulcer,
18
hours
69
G-E
&
BIT
BII
Anastomotic
dehiscence
Perforated
uilcer
71
G-E
BII
Pulmonary
embolus
None
B.
Late
Deaths
(10
Patients)
Antecedent
Interval
from
antecedent
Age
(Years)
Operation
(s)
Treatment
Operation
to
death
(years)
Cause
of
Death
49
BI
Medical
1
Haemorrhage
81
BIT
Medical
4
Haemorrhage
75
G-E
Medical
10
Haemorrhage
79
G-E
Medical
37
G-J-C
fistula
74
G-E
Medical
22
Haemorrhage
63
G-E
Medical
27
Haemorrhage
70
G-E
Medical
32
Haemorrhage
78
G-E
&
BII
Medical
3
Haemorrhage
72
G-E
Medical
20
Haemorrhage
76
G-E
Medical
29
Haemorrhage
The
cause
of
death
in
those
treated
medically
was
haemorrhage,
except
for
one
patient
with
a
gastro-jejunocolic
(GJC)
fistuila.
Treatment
of
Post-Vagotomy
and
Drainage
of
Anastcnmotic
Ulcers
We
would
suggest
repeat
transabdominal
vagotomy
alone
for
this
group.
This
is
probably
the
most
difficult
group
to
make
a
decision
on
as
such
small
numbers
are
involved
in
our
study
and
in
other
series.
Five
patients
were
treated
with
vagotomy
in
our
study
and
there
were
no
deaths
and
one
recurrence
on
short-term
follow-up.
Of
20
patients
treated
with
partial
gastrectomy,
there
were
no
deaths
but
six
patients
developed
recurrent
ulcers
on
short-term
follow-up.
Review
of
the
literature
confirms
that
vagotomy
is
the
operation
of
choice.
There
was
a
mortality
rate
of
one
percent
following
vagotomy
and
the
recurrence
rate
was
approximately
26%
on
short-term
follow-up
(Table
15).
Following
partial
gastrectomy,
the
death
rate
was
6%
and
the
recurrence
rate
24%.
Two-thirds
of
the
deaths
following
gastrectomy
were
in
the
post-vagotomy
and
pyloroplasty
group,
although
less
than
half
of
the
patients
had
had
a
vagotomy
and
pyloroplasty.
This
would
tend
to
indicate
that
gastrec-
tomy
is
particularly
dangerous
as
a
treatment
for
post-
vagotomy
and
pyloroplasty
anastomotic
ulcers.
The
mortality
rate
in
the
literature
of
gastrectomy
for
ulcer
following
vagotomy
with
pyloroplasty
is
9%
(six
deaths
in
63
procedures).
This
high
death
rate
may
be
caused
by
the
difficulty
in
closing
the
duodenal
stump
in
patients
who
have
a
pyloroplasty
and
a
recurrent
ulcer.
As
most
of
the
pyloroplasties
in
the
literature
were
of
the
Heineke
Mikulicz
type,
a
Finney
or
Jaboulay
pyloroplasty
may
have
an
even
higher
mortality
when
a
gastrectomy
is
carried
out
as
a
second
operation.
Experience
of
anastomotic
ulcer
following
vagotomy
and
drainage
is
at
the
present
time
limited,
but
we
would
expect
as
found
by
Fawcett
and
his
colleagues,9
that
the
success
rate
of
vagotomy
would
be
much
lower
in
this
group
of
recurrent
ulcers
than
when
vagotomy
was
car-
ried
out
for
primary
peptic
ulcer.
We
feel
it
is
wrong,
however,
to
suggest
gastrectomy
as
a
routine
procedure
for
ulcers
following
vagotomy
and
drainage
because
gastrectomy
is
attended
by
an
increased
mortality
rate
and
also
because
there
is
no
advantage
to
be
shown
for
gastrectomy
in
terms
of
prevention
of
ulcer
recurrence
compared
with
vagotomy.
In
a
patient
in
whom
satis-
factory
closure
of
the
duodenal
stump
could
be
assured,
it
would
be
logical
to
suggest
vagotomy
and
gastrectomy,
but
if
these
conditions
cannot
be
met
our
first
choice
would
remain
vagotomy
alone.
Jejunal,
stomal,
or
gastric
ulceration
after
gastroenterostomy
with
vagotomy
may
meet
these
conditions.
The
mortality
rate
in
the
literature
is
4%
(three
deaths
out
of
77
patients)
for
gastrectomy
for
recurrent
ulcer
following
vagotomy
with
gastro-
enterostomy,
but
it
might
be
possible
to
reduce
this
mortality
by
selection
of
patients
with
easily
closed
duodenal
stumps.
Vol.
179
*
No.
3
CLEATOR,
HOLUBITSKY
AND
HARRISON
TABLE
11.
Summary
of
the
Literature
Regarding
the
Results
of
Treatment
ot
Post
Partial
Gastrectomy
Ulcer
Present
Total
Operation
Result
Study
Literature
Per
cent
More
proximal
Excellent
15
98
65.3
Gastrectomy
(30)
(143)
Recurrent
Ulcer
12
40
30.1
(30)
(143)
Dead
4
10
7
.0
(38)
(161)
Vagotomy
alone
Excellent
40
215
73.9
(54)
(291)
Recurrent
Ulcer
7
47
15.7
(54)
(291)
Dead
0
3
0.8
(71)
(301)
Vagotomy
and
Other
Excellent
7
1
(9)
(3)
Recurrent
Ulcer
1
0
(9)
(3)
Dead
0
0
(10)
(3)
Vagotomy
and
Excellent
26
66
78.6
Partial
(31)
(86)
Gastrectomy
Recurrent
Ulcer
3
11
12.0
(31)
(86)
Dead
2
5
5.2
(40)
(94)
Other
Excellent
1
4
(5)
(14)
Recurrent
Ulcer
4
9
(5)
(14)
Dead
0
3
(6)
(16)
Medical
Treatment
Excellent
32
16
62.3
(41)
(36)
Recurrent
Ulcer
9
20
37.7
(41)
(36)
Dead
3
2
5.2
(62)
(35)
Note:
Lost
patients
are
included
in
the
operative
mortality
statistics,
but
excluded
from
all
other
statistics
in
this
and
subsequent
tables.
Patients
who
died
of
unrelated
causes
are
included
as
excellent
results
if
the
treatment
controlled
their
ulcer
disease
and
if
ulcer
disease
did
not
contribute
to
their
deaths.
Follow-up
period
varied
0-16
years.
1-4,7,8,12,13,15-18,21,22,24,28-30
Figures
in
parentheses
are
the
number
of
patients
in
each
group.
Post-Vagotomy
and
Partial
Gastrectomy
We
recommend
repeat
transabdominal
vagotomy
in
this
group.
It
is
essential
to
screen
patients
with
ulcer
following
vagotomy
with
gastrectomy
for
ulcer
diathesis
due
to
endocrine
tumours
or
retained
antrum
in
the
duodenum.
Although
none
of
our
patients
had
a
non-
beta
cell
tumour
of
the
pancreas,
nor
a
parathyroid
tumour,
two
of
our
36
patients
had
retained
antra
in
the
duodenal
stump.
In
both
patients
excision
of
the
antral
tissue
(with
more
proximal
gastrectomy
in
one)
cured
the
ulcers.
It
has
been
the
experience
of
others
too
that
this
group
of
patients
tends
to
have
an
appreciable
incidence
of
high
gastrin-secreting
or
parathormone-
secreting
conditions.
Of
12
patients
with
ulceration
after
gastric
resection
and
vagotomy,
performed
separately,
Wychulis30
found
two
patients
with
multiple
endocrine
adenomas.
Of
14
patients
with
ulceration
after
gastric
resection
and
vagotomy
(ten
were
one
stage
and
four
were
two
stage
procedures)
Stuart23
found
two
patients
with
the
Zollinger-Ellison
syndrome.
Only
13
of
1600
patients
with
a
one
stage
vagotomy
and
antrectomy
pro-
cedure
developed
recurrent
ulcers
in
Scott's19
study,
but
three
of
the
13
had
a
Zollinger-Ellison
syndrome.
The
Zollinger-Ellison
syndrome
requires
a
total
gastrectomy
to
control
the
ulcer
diathesis.
If
an
ulcer
diathesis
such
as
an
endocrine
adenoma
or
retained
antrum
can
be
excluded,
it
is
our
experience
that
the
anastomotic
ulcer
following
vagotomy
with
gastrectomy
is
a
benign
disease
attended
by
no
deaths
and
a
low
recurrence
rate.
Treatment
by
repeat
vagotomy
gave
excellent
results:
there
was
no
mortality,
and
six
traced
patients
of
the
seven
who
had
a
vagotomy
all
had
excellent
results
on
follow-up.
Study
of
the
literature
shows
few
reports
of
repeat
vagotomy
for
this
type
of
ulcer
(Table
16)
although
Jaffe"1
had
excellent
results
in
the
three
patients
who
had
vagotomy
in
his
study.
Reresection
gave
worse
results
in
our
hands;
three
of
the
eight
patients
followed
had
ulcer
recurrence.
There
are
more
reports
available
for
this
operation
and
these
show
a
recurrence
rate
in
line
with
our
results.
Even
medical
treatment
in
11
patients
in
our
study
gave
reasonable
results
with
only
one
of
the
eight
patients
TABLE
12.
Results
of
Treatment
of
Post
Partial
Gastrectomy
Ulcers
in
the
Literature
on
Follow-up
of
Five
Years
or
Morel
3,13,18
Present
Total
Operation
Result
Study
Literature
Per
cent
More
proximal
Excellent
8
19
61
.4
Gastrectomy
(17)
(27)
Recurrent
Ulcer
7
8
34.1
(17)
(27)
Vagotomy
alone
Excellent
10
58
77.3
(17)
(71)
Recurrent
Ulcer
5
11
18.2
(17)
(71)
Vagotomy
and
Partial
Excellent
10
-
-
Gastrectomy
(14)
Recurrent
Ulcer
3
(14)
Other
Excellent
0 0
(3)
(1)
Recurrent
Ulcer
3
1
(3)
(1)
Medical
Treatment
Excellent
23
2
(31)
(9)
Recurrent
Ulcer
8
7
(31)
(9)
Figures
in
parentheses
are
the
number
of
patients
in
each
group.
Ann.
Surg.
-
March
1974
ANASTOMOTIC
ULCERATION
followed
having
a
suspected
persistent
ulcer.
We
do
not,
however,
recommend
medical
treatment
for
the
definitive
treatment
of
this,
or
any
other
type
of
anastomotic
ulcer
at
this
time.
Repeat
vagotomy
with
more
proximal
gastrectomy
was
carried
out
in
four
patients,
and
excellent
results
were
found
in
two
on
follow-up,
with
unsatisfactory
results
in
the
other
two,
although
there
were
no
proven
recur-
rences.
We
feel
that
more
proximal
gastrectomy
adds
little
to
revagotomy
for
this
type
of
ulcer.
If
there
is
no
ulcer
diathesis,
more
proximal
gastrectomy
is
probably
unnecessary
and
adds
additional
risk
for
no
proven
gain.
If
there
is
an
ulcer
diathesis,
then
the
operation
is
inade-
quate
to
control
the
disease.
Deaths
Sixty-two
patients
are
known
to
have
died.
All
the
ten
patients
who
died
in
the
immediate
post-operative
period
TABLE
13.
Summary
of
the
Literature
Regarding
the
Results
of
Treatment
of
Post
Simple
Gastroenterostomy
Ulcersl-4,7,8,11,13,15-18,21,22,24,28-30
(Follow-up
Period
0-16
Years)
Present
Total
Operation
Result
Study
Literature
Per
cent
Partial
Gastrectomy
Excellent
18
543
80.1
(36)
(664)*
Recurrent
Ulcer
18
98
16.6
(36)
(664)
Dead
3
18
2.7
(48)
(773)
Vagotomy
alone
Excellent
1
51
57.8
(3)
(87)
Recurrent Ulcer
2
34
40.0
(3)
(87)
Dead
0
2
2.3
(3)
(85)
Vagotomy
and
Excellent
5
0
Other
(5)
(2)
Recurrent
Ulcer
0
2
(5)
(2)
Dead
0
0
(5)
(2)
Vagotomy
and
Partial
Excellent
2
25
75.0
Gastrectomy
(2)
(34)
RecuLrrent
Ulcer
0
4
11.1
(2)
(34)
Dead
04
9.8
(3)
(38)
Other
Excellent
0
24
23.3
(1)
(102)
Recurrent
Ulcer
1
76
74.8
(1)
(102)
Dead
1
14
13.6
(2)
(108)
Medical
Treatment
Excellent
7
14
21.9
(15)
(81)
Recurrent
Jlcer
8
67
78.1
(15)
(81)
Dead
7
11
18.6
(19)
(78)
*
Three
patients
also
had
vagotomy.
Figuires
in
parentheses
are
the
number
of
patients
in
each
grotup.
349
TABLE
14.
Summary
ofthe
Literature
Regarding
Results
of
Treatment
of
Post
Gastroenterostomy
Ulcers
on
Follow-utp
of
Five
Years
or
Morel,"13,318
Present
Total
Operation
Result
Study
Literature
Per
cent
Partial
Gastrectomy
Excellent
9
184
73.6
(23)
(239)
Recurrent
Ulcer
14
55
26.4
(23)
(239)
Vagotomy
alone
Excellent
1
10
(1)
(20)
Recurrent
Ulcer
0
10
(1)
(20)
Other
Excellent
0
14
17.1
(1)
(81)
Recurrent
Ulcer
1
67
82
.9
(1)
(81)
Medical
Treatment
Excellent
3
3
13.0
(10)
(36)
Recurrent
Ulcer
7
33
87.0
(10)
(36)
Figures
in
parentheses
are
the
number
of
patients
in
each
group
had
an
autopsy,
and
in
many
cases
autopsy
was
carried
out
in
those
who
died
of
incidental
causes.
Autopsy
infor-
mation
on
the
cause
of
death
and
the
presence
or
absence
of
stomal
ulceration
is
available
in
41
of
the
62
pa-
tients.
Forty
patients
died
of
incidental
causes
and
are
judged
to
have
good
results
following
their
gastric
sur-
gery
since
no
evidence
of
recurrence
was
noted
clinically
or
at
autopsy
in
these
patients,
and
their
gastric
surgery
was
adequate
to
last
them
until
the
end
of
their
natural
life
span.
Post-operative
Deaths
During
the
post-operative
period
seven
patients
died
following
a
partial
gastrectomy;
two
died
following
a
partial
gastrectomy
with
vagotomy
and
one
died
during
closure
of
a
neglected
perforated
jejunal
ulcer.
The
cause
of
death
following
re-gastrectomy
for
a
post-gastrectomy
anastomotic
ulcer
differed
from
the
cause
of
death
fol-
lowing
gastrectomy
for
a
post-gastroenterostomy
anas-
tomotic
ulcer.
Of
the
six
patients
with
post-gastrectomy
ulcers,
five
died
due
to
intra-abdominal
sepsis
and
one
died
of
cardiac
arrest
during
a
prolonged
and
difficult
gastrectomy.
On
the
other
hand,
the
deaths
in
the
three
patients
with
post-gastroenterostomy
ulcers
were
due
to
pulmonary
embolus
in
two
and
pneumonia
in
one,
the
anastomosis
being
sound
in
all
these
patients.
This
would
confirm
the
clinical
impression
that
in
stomal
ulcer
pa-
tients
gastrectomy
is
much
more
difficult
technically
in
a
patient
with
a
previous
gastrectomy.
Late
Deaths
All
these
ten
patients
were
on
medical
treatment.
We
found
that
these
patients
had
died
of
complications
of
anastomotic
ulcer,
six
months
to
seven
years
after
treat-
Vol.
179
*
No.
3
Ann.
Surg.
*
March
1974
CLEATOR,
HOLUBITSKY
AND
HARRISON
TABLE
15.
Summary
of
the
Literature
Regarding
the
Results
of
Treatment
of
Post
Vagotomy
with
Drainage8,9,11,12l'7,20-23,30
(Follow-up
Period
0-16
Years)
Present
Total
Operation
Result
Study
Literature
Per
cent
Partial
Gastrectomy
Excellent
12
56
67.3
(19)
(82)
Recurrent
Ulcer
6
18
23.8
(19)
(82)
Dead
0
9*
6.3
(20)
(123)
Vagotomy
alone
Excellent
3
44
68.1
(4)
(65)
Recurrent
Ulcer
1
17
26.
1
(4)
(65)
Dead
0
1
1.1
(15)
(88)
Vagotomy
and
Excellent
1
Other
(1)
Recurrent
Ulcer
0
(1)
Dead
0
(4)
Vagotomy
and
Excellent
3
2
Partial
(4)
(2)
Gastrectomy
Recurrent
Ulcer
1
0
(4)
(2)
Dead
0
0
(4)
(2)
Other
Excellent
2
(2)
Recurrent
Ulcer
-
0
(2)
Dead
0
0
(1)
(11)
Medical
Treatment
Excellent
14
3
(14)t
(11)
Recurrent
Ulcer
0
8
(14)
(11)
Dead
0
0
(14)
(11)
*
Six
of
these
deaths
followed
vagotomy
with
pyloroplasty
ulcers.
The
mortality
rate
for
gastrectomy
was
9.5
per
cent
if
a
pyloroplasty
was
the
drainage
procedure,
and
3.9
per
cent
if
gastroenterostomy
was
the
drainage
operation.
t
Three
dead
of
related
causes.
Figures
in
parentheses
are
the
number
of
patients
in
each
group.
ment
of
their
anastomotic
ulcer.
In
all
of
these
a
clear
history
of
anastomotic
ulcer
symptoms
was
available,
and
medical
treatment
was
employed
with
varying
degrees
of
vigour.
At
the
time
of
initial
assessment,
the
clinicians
chose
medical
treatment
either
because
the
clinical
diag-
nosis
was
not
supported
by
barium meal
examination,
or
because
of
the
advanced
age
of
the
patients.
In
one
further
patient
an
unhealed
anastomotic
ulcer
was
found
at
autopsy,
but
this
had
not
contributed
to
the
patient's
death
of
pulmonary
edema
at
the
age
of
86
years.
Summary
Three
hundred
and
thirty-one
patients
were
treated
for
390
episo-des
of
anastomotic
ulceration
at
two
Van-
couver
hospitals.
These
patients
were
followed
and
the
recurrence
rate
after
varying
treatment
for
their
anas-
tomotic
ulceration
was
evaluated
in
terms
of
mortality
and
recurrence
rate.
The
overall
death
rate
from
ulcer
or
its
treatment
was
6%
of
the
331
patients,
and
15%
had
a
further
ulcer
recurrence
after
surgical
treatment.
After
surgical
treatment
of
the
second
anastomotic
ulcer,
12%
of
that
group
went
on
to
develop
a
third
ulcer.
One
patient
developed
a
fourth
and
fifth
anastomotic
ulcer.
Barium
meal
was
effective
in
diagnosing
51%
of
the
321
anastomotic
ulcers
in
which
this
examination
was
carried
out.
Peak
acid
output
studies
were
most
useful
in
the
post-gastrectomy
with
or
without
vagotomy
groups:
a
level
of
over
12
mEq.
was
diagnostic
of
hypersecretion
and
was
present
in
40%
of
these
anastomotic
ulcers.
Gastroscopy
gave
a
positive
result
in
only
33%
of
the
72
instances
in
which
it
was
employed;
this
was
in
the
pre-
fiber-optic
era.
TABLE
16.
Summary
of
the
Literature
Regarding
Results
of
Treatment
of
Post
Vagotomy
with
Gastrectomy
Ulcers
2,11,21-23,30
(Follow-up
Period
0-16
Years)
Present
Operation
Result
Study
Literature
Total
Partial
Excellent
5
14
19
Gastrectomy
(8)
(26)
(34)
Recurrent
Ulcer
3
8
11
(8)
(26)
(34)
Dead
0
1
1
(9)
(28)
(37)
Vagotomy
alone
Excellent
6
3
9
(6)
(3)
(9)
Recurrent
Ulcer
0
0
0
(6)
(3)
(9)
Dead
0
1
1
(7)
(4)
(11)
Vagotomy
and
Excellent
I
-
1
Other
(1)
(1)
Recurrent
Ulcer
0
-
0
(1)
(1)
Dead
0
-
0
(1)
(1)
Vagotomy
and
Partial
Excellent
2
1
3
Gastrectomy
(4)
(1)
(5)
Recurrent
Ulcer
00 0
(4)
(1)
(5)
Dead
000
(4)
(2)
(6)
Other
Excellent
4
4
(4)
(4)
Recurrent
Ulcer
0
-
0
(4)
(4)
Dead
0
1
1
(4)
(1)
(5)
Medical
Treatment
Excellent
7
7
(8)
(8)
Recurrent
Ulcer
1
1
(8)
(8)
Dead
0
-
0
(11)
(11)
Figures
in
parentheses
are
the
number
of
patients
in
each
group.
350
Vol.
179
*
No.
3
ANASTOMOTIC
ULCERATION
351
Of
the
patients
with
anastomotic
ulcer,
74%
had
prandial
pain
and
56%
had
one
or
more
episodes
of
haemorrhage.
Perforation
of
an
anastomotic
ulcer
was
present
in
5%,
9%
had
obstructive
symptoms
and
2%
had
gastro-jejuno-colic
fistulae.
We
found
no
patients
with
secreting
tumours
of
the
pancreas
or
parathyroid.
Of
140
Billroth
II
anastomotic
ulcers,
6%
had
retained
an-
trum
in
the
duodenal
stump.
After
comparison
of
the
mortality
and
recurrence
rate
of
the
various
treatment
options
available
in
this
and
other
studies,
we
conclude
that
vagotomy
is
the
treatment
of
choice
for
post-gastrectomy
anastomotic
ulcer,
for
post-vagotomy
and
drainage
anastomotic
ulcer
and
for
post-gastrectomy
with
vagotomy
anastomotic
ulcer.
We
also
conclude
that
partial
gastrectomy
alone
is
the
treat-
ment
of
choice
for
post-gastroenterostomy
anastomotic
ulcer.
References
1.
Andros,
G.,
Donaldson,
G.
A.,
Hedberg,
S.
E.
and
Welch,
C.
E.:
Anastomotic
Ulcers.
Ann.
Surg.,
165:955,
1967.
2.
Balint,
J.
A.,
Cooper,
G.
W.,
Price,
E.
C.
V.,
Pulvertaft,
C.
N.
and
Swynnerton,
B.
F.
A.:
The
Management
of
Anas-
tomotic
Ulcer.
Lancet,
1:55,
1957.
3.
Beal,
J.
M.:
The
Surgical
Treatment
of
Marginal
Ulcer.
Amer.
Surg.,
25:1,
1959.
4.
Boles,
R.
S.,
Marshall,
S.
F.
and
Bersoux,
R.
V.:
Follow-up
Study
of
127
Patients
with
Stomal
Ulcer.
Gastroenterology,
38:763,
1960.
5.
Braun,
H.:
Demonstration
Eines
Praparetes
Einer
1
Monate
Nach
der
Ausfuhrung
der
Gastro-Enterostomie
Entstan-
deden
Perforation
des
Jejunum.
Verh.
Ges.
Chir.,
28:94,
1899.
6.
Cleator,
I.
G.
M.,
Thompson,
C.
G.,
Sircus,
W.
and
Combes,
M.:
Bioassay
Evidence
of
Abnormal
Secretin-like
and
Gastrin-like
Activity
in
Tumour
and
in
Blood
in
Cases
of
Choleraic
Diarrhoea.
Gut,
11:206,
1970.
7.
Edwards,
L
.W.,
Herrington,
J.
L.
Jr.,
Cate,
W.
R.
Jr.,
et
al:
Gastrojejunal
Ulcer:
Problems
in
Surgical
Management.
Ann.
Surg.,
143:235,
1956.
8.
Everson,
T.
C.
and
Allen,
M.
J.:
Gastrojejunal
Ulceration.
A.M.A.
Arch.
Surg.,
69:140,
1954.
9.
Fawcett,
A.
N.,
Johnston,
D.
and
Duthie,
H.
L.:
Vagotomy
for
Recurrent
Ulcer
After
Vagotomy
and
Drainage
for
Duodenal
Ulcer.
Brit.
J.
Surg.,
56:111,
1969.
10.
Goligher,
J.
C.:
The
Comparitive
Results
of
Different
Opera-
tions
in
the
Elective
Treatment
of
Duodenal
Ulcer.
Brit
J.
Surg.,
57:780,
1970.
11.
Jaffe,
B.
M.,
Newton,
W.
T.,
Judd,
D.
R.
and
Ballinger,
W.
F.:
Surgical
Management
of
Recurrent
Peptic
Ulcers.
Amer.
J.
Surg.,
117:214,
1969.
12.
Judd,
D.
R.,
Starkloff,
G.
B.,
Morioka,
W.,
et
al.:
Vagotomy
and
Drainage
Procedures.
Arch.
Surg.,
102:242,
1971.
13.
Knox,
W.
G.
and
West,
J.
P.:
Vagus
Section
in
the
Treatment
of
Gastrojejunal
Ulcer.
Ann.
Surg.,
149:481,
1959.
14.
McGuigan,
J.
E.
and
Trudeau,
W.
C.:
Immunochemical
Measurement
of Elevated
Levels
of
Gastrin
in
the
Serum
of
Patients
with
Pancreatic
Tumours
of
the
Zollinger-
Ellison
Variety.
N.
Engl.
J.
Med.,
278:1308,
1968.
15.
Marshall,
S.
F.
and
Terrell,
G.
K.:
Postoperative
Recurrent
Ulcer.
Surg.
Clin.
N.
Amer.,
37:653,
1957.
16.
Osnes,
S.
V.:
Transthoracic
Resection
of
Vagi
in
the
Treat-
ment
of
Gastrojejunal
Ulcer.
Acta
Chir.
Scand.,
110:373,
1956.
17.
Postlethwait,
R.
W.:
Results
of
Surgery
for
Peptic
Ulcer.
Philadelphia
and
London,
W.
B.
Saunders
Co.,
1963.
18.
Priestley,
J.
T.
and
Gibson,
R.
H.:
Gastrojejunal
Ulcer:
Clinical
Features
and
Late
Results.
Arch.
Surg.,
56:625,
1948.
19.
Scott,
H.
W.,
Sawyers,
J.
F.,
Gobbel,
W.
G.,
Herrington,
J.
L.,
Edwards,
W.
H.
and
Edwards,
L.
W.:
Vagotomy
and
Antrectomy
in
Surgical
Treatment
of
Duodenal
Ulcer
Disease.
J.
Clin.
North
America,
46:349,
1966.
20.
Shiffman,
M.
A.:
Evaluation
of
Recurrence
Following
Vagot-
omy
and
Pyloroplasty.
Int.
Surg.,
48:574,
1967.
21.
Small,
W.
P.:
The
Recurrence
of
Ulceration
After
Surgery
for
Duodenal
Ulcer.
J.
Roy.
Col.
Surg.,
Edinb.,
9:255,
1964.
22.
Small,
W.
P.,
Smith,
A.
N.,
Ruckley,
C.
V.,
Falconer,
C
W.
A.,
Sircus,
W,
McManus,
J.
P.
A.
and
Bruce,
J.:
The
Con-
tinuing
Problem
of
Jejunal
Ulcer.
Amer.
J.
Surg.,
121:541
1971.
23.
Stuart,
M.
and
Hoerr,
S.
O.:
Recurrent
Peptic
Ulcer
Follow-
ing
Primary
Operations
with
Vagotomy
for
Duodenal
Ulcer.
Arch.
Surg.,
103:
129,1971.
24.
Thompson,
J.
E.:
Stomal
Ulceration
after
Gastric
Surgery.
Ann.
Surg.,
143:697,
1956.
25.
Thomson,
C.,
Cleator,
I.
G.
M.
and
Sircus,
W.:
Experiences
with
a
Rat
Bioassay
in
the
Diagnosis
of
the
Zollinger-
Ellison
Syndrome.
Gut,
11:409,
1970.
26.
Visick,
A.
H.:
A
Study
of
the
Failures
After
Gastrectomy.
Ann.
R.
Coll.
Surg.,
3:266,
1948.
27.
Walter,
W.,
Chance,
D.
P.
and
Berkson,
J.:
A
Comparison
of
Vagotomy
and
Gastric
Resection
for
Gastro-Jejunal
Ulceration:
A
Follow-up
Study
of
301
Cases.
Surg.
Gynecol.
Obstet.,
100:1,
1955.
28.
Wells,
B.
W.:
Result
of
Vagotomy
in
Treatment
of
Anasto-
motic
Ulcer.
Lancet,
1:598,
1954.
29.
Wells,
C.
and
Silberman,
R.:
Transthoracic
Vagotomy
for
Stomal
Ulceration.
Lancet,
1:406,
1960.
30.
Wychulis,
A.
R.,
Priestley,
J.
T.
and
Foulk,
W.
T.:
A
Study
of
360
Patients
with
Gastrojejunal
Ulceration.
Surg.
Gyne-
col.
Obstet.,
122:89,
1966.
